Content uploaded by Timothy L Hubbard
Author content
All content in this area was uploaded by Timothy L Hubbard on Feb 11, 2014
Content may be subject to copyright.
The Flash-Lag Effect and Related Mislocalizations: Findings,
Properties, and Theories
Timothy L. Hubbard
Texas Christian University
If an observer sees a flashed (briefly presented) object that is aligned with a moving target, the perceived
position of the flashed object usually lags the perceived position of the moving target. This has been
referred to as the flash-lag effect, and the flash-lag effect has been suggested to reflect how an observer
compensates for delays in perception that are due to neural processing times and is thus able to interact
with dynamic stimuli in real time. Characteristics of the stimulus and of the observer that influence the
flash-lag effect are reviewed, and the sensitivity or robustness of the flash-lag effect to numerous
variables is discussed. Properties of the flash-lag effect and how the flash-lag effect might be related to
several other perceptual and cognitive processes and phenomena are considered. Unresolved empirical
issues are noted. Theories of the flash-lag effect are reviewed, and evidence inconsistent with each theory
is noted. The flash-lag effect appears to involve low-level perceptual processes and high-level cognitive
processes, reflects the operation of multiple mechanisms, occurs in numerous stimulus dimensions, and
occurs within and across multiple modalities. It is suggested that the flash-lag effect derives from more
basic mislocalizations of the moving target or flashed object and that understanding and analysis of the
flash-lag effect should focus on these more basic mislocalizations rather than on the relationship between
the moving target and the flashed object.
Keywords: flash-lag effect, representational momentum, Fröhlich effect, spatial representation, motion
perception
When we perceive a visual stimulus, neural processes pass that
information from the retina to and throughout the cortex. These
processes are fast but not instantaneous, and there is a delay of at
least 100 ms between when the retina is stimulated and when
conscious perception of the stimulus occurs (De Valois & De
Valois, 1991; Nijhawan, 2008). Real-time interaction with stimuli
must involve some form of compensation for these neural delays,
or else our responses to dynamic and changing (e.g., moving)
stimuli would be too late (e.g., as noted in Nijhawan, 1994, neural
delays would result in an object moving at 30 miles per hour
appearing 4.4 feet behind its actual location). Several types of
mislocalization potentially related to compensation for neural de-
lays have been studied, and the type of mislocalization that has
received the most attention is the flash-lag effect. In the flash-lag
effect, a flashed (i.e., briefly presented) object that is aligned with
a moving target is perceived to lag behind the position of that
target. Previous reviews of the flash-lag effect focused on histor-
ical antecedents (Maus, Khurana, & Nijhawan, 2010) or specific
potential theories (Krekelberg & Lappe, 2001; Nijhawan, 2002;
Schlag & Schlag-Rey, 2002; Whitney, 2002). In addition to con-
sidering potential theories, this review presents the first catalog of
variables that influence the flash-lag effect and examines potential
connections between the flash-lag effect and other phenomena.
The most common types of stimuli used to investigate the
flash-lag effect are illustrated in Figure 1.InFigure 1A, a rotating
bar is shown, and two sets of dashed lines are briefly presented in
alignment with the bar (e.g., Nijhawan, 1994). In Figure 1B,a
revolving annulus is shown, and a disk is briefly presented inside
the annulus (e.g., Khurana, Watanabe, & Nijhawan, 2000). In
Figures 1C and 1D, a translating bar is shown, and a stationary bar
is briefly presented away from the translating bar (e.g., Kanai,
Seth, & Shimojo, 2004) or within a gap in the translating bar (e.g.,
Murakami, 2001b). There are two common ways to assess whether
a flash-lag effect occurred. One way is to have participants judge
whether the flashed object was presented before or after the
moving target passed the position of the flashed object (e.g., was
the flashed object to the left or right of a horizontally moving
target? e.g., Whitney, Murakami, & Cavanagh, 2000) or whether
the moving target was approaching or had passed the flashed
object when the flashed object was presented (e.g., was the target
ahead of or behind the flashed object? e.g., Moore & Enns, 2004).
The actual position of the flashed object relative to the moving
target varies across trials, and a point of subjective alignment or
simultaneity can be estimated (e.g., López-Moliner & Linares,
2006). A second way is to use a nulling procedure (method of
adjustment) in which participants adjust presentation of the flashed
object relative to the moving target so that the two stimuli appear
aligned or simultaneous (e.g., Lappe & Krekelberg, 1998).
The first account of a flash-lag type of effect was in Mach
(1885/1897), who reported a flash (spark) presented during a
saccadic eye movement appeared displaced. Metzger (1932)
passed a vertical line behind an occluder containing a horizontal
slit and a small hole above the midpoint of the slit. When the line
This article was published Online First June 24, 2013.
Correspondence concerning this article should be addressed to Timothy L.
Hubbard, Department of Psychology, Texas Christian University, 2800 South
University, Fort Worth, TX 761232. E-mail: timothyleehubbard@gmail
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Psychological Bulletin © 2013 American Psychological Association
2014, Vol. 140, No. 1, 308–338 0033-2909/14/$12.00 DOI: 10.1037/a0032899
308
passed behind the hole, it provided a flashed object aligned with
the part of the line visible in the slit; however, the part of the line
visible in the hole appeared to lag the part of the line visible in the
slit. MacKay (1958) presented a self-luminous object and a non-
luminous object in a stroboscopically lit visual field and passively
moved the eye, and he reported a stroboscopically lit object sub-
jectively lagged a self-luminous object. Mitrani and Dimitrov
(1982) and Mateeff and Hohnsbein (1988) reported a similar effect
in which the position of a flashed stimulus was displaced in the
presence of a moving target. However, it was with Nijhawan’s
(1994) rediscovery of this effect that an explosion of interest in the
flash-lag effect, and in the question of how sensory or motor
systems compensate for delays in perception due to neural pro-
cessing times, occurred. Part 1 reviews variables that influence the
flash-lag effect. Part 2 considers general properties of the flash-lag
effect and how the flash-lag effect is related to several other
perceptual and cognitive phenomena. Part 3 reviews theories of the
flash-lag effect, and Part 4 provides a summary and conclusions.
Part 1: Variables that Influence the Flash-Lag Effect
Variables that influence the flash-lag effect are classified as
characteristics of the stimulus or characteristics of the observer.
This distinction is helpful in organization but does not suggest
these categories are exhaustive (i.e., other categories or variables
might be documented by future research) or mutually exclusive
(e.g., presentation of a cue, which is a characteristic of the stim-
ulus, might influence allocation of attention, which is a character-
istic of the observer).
Characteristics of the Stimulus
Characteristics of the stimulus considered here include (a) tim-
ing of the presentation of the flashed object, (b) continuity of target
motion, (c) distance traveled by the target, (d) distance between the
moving target and the flashed object, (e) eccentricity, (f) target
velocity, (g) target direction, (h) binocular disparity, (i) color, (j)
luminance, (k) contrast, (l) spatial frequency, (m) target identity,
(n) pattern entropy, (o) duration of the flashed object, (p) motion
of the flashed object, (q) presence of cues, (r) predictability of the
flashed object, (s) uncertainty of target location, (t) number of
targets, and (u) presence of unrelated stimuli. The effects of these
variables are summarized in Table 1.
Timing of the presentation of the flashed object. Perhaps
the most important stimulus variable in the flash-lag effect is the
timing of the presentation of the flashed object relative to target
motion. If the flashed object is presented at target motion onset
(i.e., moving target and flashed object simultaneously appear), this
is usually referred to as a flash-initiated display. If the flashed
Flashed Object
Flashed Object
Rotating
Target
Flashed Disk
Revolving
Annulus
Flashed Object
Moving Target
A
B
C
Physical Perceived
Flashed Object
Moving Target
D
Moving Target
Figure 1. Illustrations of common stimulus displays in studies of the flash-lag effect. The actual physical stimulus is depicted on the left, and the typical
perceived stimulus is depicted on the right; in each case, the position of the flashed object appears to lag behind the position of the moving target. (A) The
target consists of a bar rotating clockwise, and the flashed object consists of two dashed line segments that are briefly flashed when they are in alignment
with the bar (adapted from Nijhawan, 1994, p. 256). (B) The target consists of a black annulus moving clockwise on a circular path (indicated by the
dashed line), and the flashed object consists of a white disk flashed within the annulus (adapted from Khurana et al., 2000, p. 679). (C) The target
consists of a vertical bar moving from left to right in the upper part of the display, and the flashed object is a vertical bar in the lower part of the
display that is briefly flashed when it is aligned with the moving target (adapted from Kanai et al., 2004, p. 2607). (D) The moving target consists
of two vertically aligned vertical bars moving from left to right, and the flashed object consists of a vertical bar flashed in the gap within the target
(adapted from Murakami, 2001a, p. 126).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
309
FLASH-LAG EFFECT
object is presented at target motion offset (i.e., moving target and
flashed object simultaneously vanish), this is usually referred to as
a flash-terminated display. If the flashed object is presented after
target motion onset but before target motion offset, this has been
referred to as a complete cycle (e.g., Nijhawan, Watanabe,
Khurana, & Shimojo, 2004), full-view (e.g., Watanabe, 2004), or
continuous motion (e.g., Eagleman & Sejnowski, 2000b) display.
However, the latter term is ambiguous (continuous describes mo-
tion of the target as well as timing of the presentation of the flashed
object), and the two former terms are not sufficiently descriptive.
Table 1
Effects of Stimulus Characteristics on the Flash-Lag Effect
Characteristic Flash-lag effect Primary sources
Timing of presentation of flashed
object
Occurs with flash-initiated and flash-midpoint
displays, but not with flash-terminated
displays
Eagleman & Sejnowski (2000b); Khurana & Nijhawan
(1995); Nijhawan et al. (2004); Watanabe (2004)
Continuity of target motion Occurs with sampled, continuous, and
random motion
Arrighi et al. (2005); Fukiage & Murakami (2010); Murakami
(2001b); Rizk et al. (2009)
Distance traveled by moving
target
Decreases if target travels further before the
flashed object appears (and flashed object
is more predictable)
Vreven & Verghese (2005)
Distance between moving target
and flashed object
Increases if moving target and flashed object
are farther apart
Baldo, Kihara, et al. (2002); Baldo & Klein (1995); Kanai et
al. (2004)
Eccentricity of moving target or
flashed object
Increases if flashed object is more eccentric
than moving target
Baldo, Kihara, et al. (2002); Baldo & Klein (1995)
Decreases if moving target is more eccentric
than flashed object
Linares et al. (2007)
Target velocity Increases with increases in target velocity Krekelberg & Lappe (1999); Lee et al. (2008); Nijhawan
(1994); Wojtach et al. (2008)
Target direction Occurs if flashed object is presented
immediately after target changes direction
Eagleman & Sejnowski (2000b); Whitney & Murakami
(1998); Whitney, Murakami, & Cavanagh (2000)
Is larger if target moves toward fixation than
away from fixation
Brenner et al. (2006); Kanai et al. (2004); Mateeff et al.
(1991); Shi & Nijhawan (2008)
Is larger for approaching motion than
receding motion, and with stereomotion
than with looming
Harris et al. (2006); Ishii et al. (2004); Lee et al. (2008)
Binocular disparity Is larger if moving target and flashed object
are defined by disparity as well as by
monocular cues
Harris et al. (2006); Lee et al. (2008); Nieman et al. (2006)
Color Prevents additive color mixing if flashed
object is superimposed on moving target
Nijhawan (1997); Nijhawan et al. (1998)
Occurs for changes in saturation Kreegipuu & Allik (2004); Sheth et al. (2000)
Luminance Decreases if luminance of flashed object is
increased
Ög
˘
men et al. (2004); Purushothaman et al. (1998)
Occurs for changes in luminance Chappell & Mullen (2010); Ichikawa & Masakura (2006);
Sheth et al. (2000)
Contrast Increases with decreases in contrast of
moving target and flashed object with
background
Kanai et al. (2004)
Increases with increases in contrast between
the moving target and flashed object
Arnold et al. (2009)
Spatial frequency Increases with decreases in spatial frequency Cantor & Schor (2007); Fu et al. (2001)
Occurs with changes in spatial frequency Sheth et al. (2000)
Identity of moving target Is disrupted by changes in moving target size Moore & Enns (2004)
Disrupts composite face effect Khurana et al. (2006)
Pattern entropy Occurs with changes in pattern entropy Sheth et al. (2000)
Duration of flashed object Occurs with stationary flashed object
durations under 80 ms
Brenner & Smeets (2000); Eagleman & Sejnowski (2000b)
Occurs with flashed object durations up to
500 ms
Krekelberg & Lappe (1999); Lappe & Krekelberg (1998)
Motion of flashed object Decreases with increased motion of flashed
object
Bachmann & Kalev (1997); Gauch & Kerzel (2008a);
Krekelberg & Lappe (1999)
Presence of cues No effect of cuing Khurana et al. (2000)
Decreases with (valid) cues Brenner & Smeets (2000); Namba & Baldo (2004); Rotman
et al. (2002); Shioiri et al. (2010)
Predictability of the flashed
object
Increases with decreases in predictability of
when the flashed object would appear
Baldo & Namba (2002); Vreven & Verghese (2005)
Uncertainty of target location Increases with increases in uncertainty of
moving target location
Fu et al. (2001); Kanai et al. (2004); Maus & Nijhawan
(2006, 2009)
Number of moving targets Increases with increases in the number of
moving targets
Shioiri et al. (2010)
Presence of unrelated stimuli Increases if another stimulus close to the
moving target moves toward the target
Maiche et al. (2007)
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
310
HUBBARD
The term flash-midpoint is more consistent with flash-initiated and
flash-terminated, and so flash-midpoint is adopted here. Unless
noted otherwise, experiments discussed in this review presented a
flash-midpoint display. Also, the moving (changing) stimulus is
referred to as the target, and the flashed (unchanging or briefly
presented) stimulus is referred to as the object.
Khurana and Nijhawan (1995) and Watanabe (2004) presented
a flash-initiated display and reported a flash-lag effect occurred,
and Nijhawan et al. (2004) replicated this finding and reported a
flash-lag effect in flash-initiated displays increased with faster
target velocities. Watanabe presented a flashed object before, at
the moment of, or after motion onset. A stationary version of the
moving target then appeared, and participants adjusted the location
of a small disk to indicate the location of the flashed object relative
to the moving target. A larger flash-lag effect occurred if the
flashed object was presented near the leading edge than near the
trailing edge of the moving target. Rizk, Chappell, and Hine (2009)
reported a flash-initiated flash-lag effect was larger than a flash-
midpoint flash-lag effect, and Ög
˘
men, Patel, Bedell, and Camuz
(2004) reported the flash-lag effect for a stationary target that
changed in luminance was larger in flash-initiated displays than in
flash-midpoint displays. Existence of a flash-lag effect in flash-
initiated displays seems inconsistent with theories of the flash-lag
effect involving differential latencies to perceive the moving target
and flashed object, perceptual acceleration, or motion extrapola-
tion (see Part 3), as onset times of the moving target and the
flashed object would be the same.
Chappell and Hine (2004) presented a target that was stationary
before the flashed object appeared, and when the flashed object
appeared, target motion began. The flash-lag effect decreased if
the target was stationary for 50 or 250, but not 750, ms prior to
flash-initiated motion. Kreegipuu and Allik (2003) presented a
target that was stationary before the flashed object appeared and a
probe bar that was visible until the target vanished. Participants
judged whether the probe bar appeared before or after the start of
target motion (time judgment) or whether target motion began to
the left or right of the probe bar (position judgment). If the target
was stationary when the probe bar was presented, time judgments
and localization judgments were accurate. If the target was moving
when the probe bar was presented, then in order to achieve
perceptual simultaneity, the probe bar had to be presented 40 –70
ms before motion onset (i.e., a flash-lag effect occurred). Findings
of Chappell and Hine and of Kreegipuu and Allik suggested claims
of Brenner and Smeets (2000) and Eagleman and Sejnowski
(2000b) that information presented prior to the flashed object does
not influence the flash-lag effect were not entirely correct. Chap-
pell and Hine suggested there is a moving temporal window of
integration extending prior and subsequent to the flashed object
(cf. Krekelberg & Lappe, 2000a, 2000b).
Eagleman and Sejnowski (2000b) reported a flash-lag effect
occurred if target motion continued or reversed direction after the
flashed object was presented but did not occur in a flash-
terminated display. Moore and Enns (2004) presented a target
moving along a circular path in a flash-terminated display or a
flash-midpoint display, and the flashed object could be behind,
aligned with, or ahead of the target. A flash-lag effect occurred
with flash-midpoint displays but not flash-terminated displays.
Watanabe (2004) reported a flash-lag effect did not occur with
flash-terminated displays, and Kessler, Gordon, Cessford, and
Lages (2010); Munger and Owens (2004); and Rizk et al. (2009)
reported a flash-lag effect with flash-midpoint displays but not
with flash-terminated displays. Only two studies reported a flash-
lag effect with flash-terminated displays: Kanai et al. (2004) re-
ported a flash-lag effect if the moving target and flashed object
were far apart and in the periphery. Gauch and Kerzel (2008a)
reported a flash-lag effect if the flashed object moved but not if the
flashed object was stationary. In general, a flash-lag effect usually
occurs with flash-initiated and flash-midpoint displays (and is
sometimes larger with flash-initiated displays) but not with flash-
terminated displays.
Continuity of target motion. Rizk et al. (2009) presented
flash-lag stimuli in which target motion was intermittently sam-
pled (referred to as moving “from station to station”) rather than
continuous; in other words, a static target was displayed at one
position, and after a clear delay, another static depiction of the
target was displayed at a different position. A flash-lag effect
occurred for continuous motion and for sampled motion with
flash-initiated displays and flash-midpoint displays, although the
flash-lag effect decreased as continuity of motion decreased.
Lappe and Krekelberg (1998) presented a flash-lag stimulus in
which the moving target consisted of three circles along a single
rotating line and the flashed object consisted of four circles (two
each on opposite ends of the rotating line). After the stimulus was
presented, participants indicated the perceived relative locations of
the moving target and flashed object by pressing the left or right
mouse button to adjust the offset angle between the moving target
and the flashed object. Consistent with MacKay (1958), objects in
stroboscopic motion (i.e., flashed) appeared to lag objects in con-
tinuous motion (but see Vreven & Verghese, 2005, for disruption
of the flash-lag effect with extreme stroboscopic motion).
Murakami (2001b) presented a display in which the moving
target jumped between random locations along the horizontal
meridian of the display. A flashed object was presented within the
gap in the target. Participants judged whether the flashed object
was perceived to the left or right of the target, and a flash-lag effect
occurred. Murakami argued that (a) because motion of the target
was random, motion extrapolation theory could not account for the
flash-lag effect, and (b) a differential latency theory in which
latency fluctuated could account for the flash-lag effect. However,
some direction of motion must be attributed to the target (if only
from the last jump) to distinguish whether the flashed object
lagged or led the moving target, and such an attribution might be
sufficient for extrapolation (cf. Nijhawan’s, 2008, suggestion that
the 100 ms between presentation and perception of a moving target
is enough for extrapolation). Vreven and Verghese (2005) pre-
sented a continuously moving target that unpredictably changed
direction, and the flash-lag effect decreased from that found if the
moving target maintained the same direction. Also, Bachmann and
Põder (2001) reported a flash-lag effect occurred if the moving
target was a sequence of discrete letters and the flashed object was
a single letter.
Arrighi, Alais, and Burr (2005) presented stimuli in which the
moving target was a visual Gaussian blob or auditory white noise,
and the moving target jumped randomly between positions along
the horizontal axis. The flashed object was a white disk or a
400-Hz pure tone. A weak flash-lag effect occurred with visual
moving targets and visual flashed objects, and a stronger flash-lag
effect occurred with visual moving targets and auditory flashed
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
311
FLASH-LAG EFFECT
objects. Latencies for visual stimuli and for auditory stimuli were
not consistent with differential latency theory. Fukiage and Mu-
rakami (2010) presented a moving target composed of a tilted
grating that randomly jumped between orientations and a flashed
object composed of a vertical grating. Presentation of the flashed
object varied relative to the jump of the moving target, and the
effect of timing suggested a flash-lag effect occurred. A tilt after-
effect occurred, and Fukiage and Murakami concluded the flash-
lag effect had a negligible influence on, and was generated at a
different processing stage than, the tilt aftereffect. In general,
findings regarding a lack of continuity of target motion (i.e.,
unpredictable changes in direction of target motion, target motion
involving spatially and temporally discrete positions) suggest con-
tinuity of target motion is not required for a flash-lag effect.
Distance traveled by the target. Vreven and Verghese
(2005) reported increases in the distance traveled by the target
prior to presentation of the flashed object decreased the flash-lag
effect, and they referred to this as an interval effect (as they divided
the target trajectory into intervals). Vreven and Verghese sug-
gested the position of the flashed object becomes more predictable
over time (i.e., if the flashed object did not appear in the first or
second of three possible intervals, then it would appear in the third
interval) and that this is not consistent with differential latency
theory. However, it seems plausible that predictability could prime
an expected appearance and thus influence processing latency.
Linares, López-Moliner, and Johnston (2007) presented a flashed
object at latencies between 0 and 3,000 ms after the moving target
appeared. If the flashed object was close to the target, duration of
preflash trajectory did not influence the flash-lag effect (see also
Brenner & Smeets, 2000; Eagleman & Sejnowski, 2000b; Whit-
ney, Murakami, & Cavanagh, 2000). If the flashed object was
farther from the target, then the flash-lag effect increased with
increases in duration of preflash trajectory. Linares et al. attributed
this to duration of target motion, but an alternative hypothesis
regarding distance traveled by the target cannot be ruled out. Maus
and Nijhawan (2006) reported the threshold for a moving target
was lower if the target traveled farther, and they suggested this
involved larger forward displacement of the target.
Distance between moving target and flashed object.
Increases in distance between the moving target and the flashed
object increased the flash-lag effect in Baldo and Klein (1995);
Baldo, Kihara, Namba, and Klein (2002); and Kanai et al. (2004).
In these experiments, the moving target was located at different
spatial coordinates than the flashed object. However, in other
experiments, a single flashed object was centered at the same
spatial coordinates as the moving target and enclosed within the
moving target (e.g., a flashed disk within an annulus, Becker,
Ansorge, & Turatto, 2009; Eagleman & Sejnowski, 2000b; Nijha-
wan, 2001), or the flashed object was interleaved with the moving
target (e.g., Khurana & Nijhawan, 1995). These different types of
displays appear to have been presumed to provide equally valid
measures of the flash-lag effect, but there are differences between
these types of displays that could potentially influence the flash-
lag effect (e.g., it might be easier to attend to a flashed object
closer to or contained within a moving target, and masking might
be more likely to influence the flash-lag effect if the flashed object
is contained or interleaved within the moving target). Also, many
studies that manipulated distance between the moving target and
the flashed object did not control for eccentricity of the moving
target or flashed object.
Eccentricity. Baldo and Klein (1995) and Baldo, Kihara, et al.
(2002) varied the distance of the flashed object from fixation
(which was centered on a rotating target), and they reported the
flash-lag effect increased with increases in distance of the flashed
object from fixation (i.e., from the moving target). Kanai et al.
(2004) reported a flash-lag effect was more likely to occur with
flash-terminated displays as eccentricity of moving targets in-
creased and as eccentricity of flashed objects decreased. Linares et
al. (2007) reported the flash-lag effect decreased as eccentricity of
the flashed object increased. Linares et al. presented flashed ob-
jects at a smaller eccentricity than moving targets, whereas Baldo
and Klein presented flashed objects at a larger eccentricity than
moving targets; in Linares et al. and in Baldo and Klein, though,
increasing eccentricity led to a larger flash-lag effect. Linares et al.
suggested their data, in conjunction with data of Baldo and Klein,
demonstrated that distance between the moving target and the
flashed object, rather than absolute eccentricity of the moving
target or flashed object, influenced the flash-lag effect. Also,
Lappe and Krekelberg (1998) suggested the effect of eccentricity
in their stimuli was due to differences in tangential velocity rather
than eccentricity per se.
Target velocity. Many researchers referred to the “speed” of
target motion, although “velocity” is more correct (as target mo-
tion exhibited magnitude and direction), and so the latter term is
used here. Several researchers reported the flash-lag effect in-
creased linearly with increases in target velocity in the picture
plane (e.g., Bachmann & Kalev, 1997; Brenner & Smeets, 2000;
Krekelberg & Lappe, 1999, 2000a; López-Moliner & Linares,
2006; Murakami, 2001b; Nijhawan, 1994). Wojtach, Sung,
Truong, and Purves (2008) presented a wider range of velocities
than did previous studies (3°/s–50°/s); they reported the flash-lag
effect was related to target velocity by a logarithmic function, and
they suggested the effect of velocity appeared linear in earlier
studies because those studies used smaller ranges of velocity. Lee,
Khuu, Li, and Hayes (2008) reported increases in target velocity in
depth increased the flash-lag effect in depth. Shioiri, Yamamoto,
Oshida, Matsubara, and Yaguchi (2010) reported increases in the
flash-lag effect with increases in target velocity were smaller in a
multitarget display if the target by which the flashed object would
appear was cued, and they suggested this reflected differences in
attention allocated to a given target if tracking only that target relative
to attention allocated to a given target if tracking multiple targets.
Whitney, Murakami, and Cavanagh (2000) presented displays in
which the moving target unpredictably accelerated or decelerated,
and they suggested the resultant increases and decreases in the
flash-lag effect, respectively, were consistent with a temporal
advantage in processing the moving target relative to the flashed
object and were not consistent with motion extrapolation. How-
ever, whether the flashed object was presented prior to, concurrent
with, or subsequent to a change in target velocity would presum-
ably influence a flash-lag effect and potential extrapolation, but
such analyses were not reported. Brenner and Smeets (2000)
presented a flash-lag display in which target velocity doubled or
halved when the flashed object was presented. The flash-lag effect
was influenced by target velocity after the flashed object was
presented, and Brenner and Smeets suggested this was evidence
against motion extrapolation theory. Kessler et al. (2010) pre-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
312
HUBBARD
sented biological motion stimuli (and control stimuli consisting of
geometric shapes with the same velocity profile as biological
motion), and they reported a robust flash-lag effect. Blohm, Mis-
sal, and Lefèvre (2003) reported the effect of velocity on the
flash-lag effect was larger if target motion was toward fixation
than away from fixation.
There are a few exceptions to the typical finding that increases
in target velocity increase the flash-lag effect. Cantor and Schor
(2007) reported the flash-lag effect decreased with increases in
velocities above 1°/s (0.5 Hz); they suggested this occurred be-
cause their stimuli were Gabor patches without sharp edges,
whereas previous studies generally presented stimuli with sharp
edges that produced relatively high spatial frequency energy (cf.
Fu, Shen, & Dan, 2001). Kanai et al. (2004) found no effect of
velocity on the flash-lag effect in flash-terminated displays if lag
was considered in units of space, whereas the flash-lag effect
decreased with increases in velocity if lag was considered in units
of time. Kreegipuu and Allik (2004) reported velocity did not
influence the flash-lag effect, but in their experiments, participants
judged the spatial or temporal relationship between flashed objects
and color changes in moving targets. Vreven and Verghese (2005)
reported random changes in velocity decreased the flash-lag effect
relative to the flash-lag effect that resulted from a consistent velocity.
In general, faster target velocities are linked with a larger flash-lag
effect, but exceptions occur, especially if uncertainty regarding target
position is high and the flash-lag effect is considered in terms of
spatial differences rather than temporal differences.
Target direction. There have been few studies that examined
effects of absolute target direction on the flash-lag effect. Ichikawa
and Masakura (2006, 2010) reported the flash-lag effect was not
influenced by whether targets ascended or descended. More com-
mon are studies of the flash-lag effect in which target direction
changes during a trial. Eagleman and Sejnowski (2000b) presented
a moving target that reversed direction after presentation of the
flashed object. As latency between the flashed object and reversal
of target motion decreased from 80 ms to 26 ms, the flash-lag
effect decreased, and mislocalization of the flashed object reversed
direction with latencies less than 26 ms. Shen, Zhou, Gao, Liang,
and Shui (2007) presented two targets that moved toward each
other along different curved trajectories and reversed direction
upon contact. A flashed object was presented at a random position
along one of the trajectories. If judgments of alignment with the
flash were used to determine perceived reversal position, targets
appeared to reverse before contact (cf. Whitney, Murakami, &
Cavanagh, 2000). If participants judged whether targets contacted,
the difference between actual reversal position and perceived re-
versal position decreased. Shen et al. argued this was not consis-
tent with a theory of the flash-lag effect based on position inte-
gration.
Whitney and Murakami (1998) presented a display involving a
translating bar, and direction of target motion reversed at a random
position. A flashed object was presented before or after reversal of
target motion. Whitney and Murakami suggested if the flashed
object was presented at the moment of reversal, then motion
extrapolation theory predicted the moving target should be per-
ceived to be beyond the reversal position, but this did not occur.
Whitney, Cavanagh, and Murakami (2000) presented a target that
moved upward and then turned to the left or right. A flash-lag
effect occurred even if the flashed object was presented immedi-
ately after a change in direction. Because the change in target
direction was unpredictable, and motion extrapolation theory is
based on predictability of target motion, Whitney et al. argued a
flash-lag effect immediately after a change in target direction
could not result from motion extrapolation. Similarly, Whitney,
Murakami, and Cavanagh (2000) had the moving target unpredict-
ably reverse direction; judgments of alignment suggested the target
was never perceived to be beyond the reversal point, and they
argued this was evidence against motion extrapolation. Whitney
and colleagues suggested their data were most consistent with
differential latency theory.
Kanai et al. (2004) and Shi and Nijhawan (2008) reported the
flash-lag effect was larger if targets moved toward rather than
away from a fixated region (see also Mateeff & Hohnsbein, 1988;
Mateeff et al., 1991; van Beers, Wolpert, & Haggard, 2001).
Brenner, van Beers, Rotman, and Smeets (2006) explained the
larger flash-lag effect with motion toward fixation as resulting
from a combination of a bias toward the fovea and a delay in
sampling position of the moving target. If a target moved toward
the fixation point, foveal bias and delay in sampling operated in the
same direction, and so they summed, and forward displacement of
that target was relatively larger. If a target moved away from the
fixation point, foveal bias and delay in sampling operated in
opposite directions, and so they partially canceled, and forward
displacement of that target was relatively smaller.
1
Kanai et al.
reported the flash-lag effect was larger if motion was in the left
than in the right visual field, and Shi and Nijhawan reported the
flash-lag effect was more influenced by target direction if target
motion was in the right than in the left visual field. However,
Ichikawa and Masakura (2006, 2010) reported the flash-lag effect
for vertically moving targets was not influenced by whether stim-
uli were in the left or right visual field.
Ishii, Seekkuarachchi, Tamura, and Tang (2004) presented a
moving target composed of two vertical bars that appeared to
move in depth. A stationary vertical bar was flashed. Participants
verbally indicated which stimulus appeared closer or if the stimuli
appeared at equal depths, and participants indicated perceived
depth with a vernier caliper. A flash-lag effect in depth occurred,
and Ishii et al. suggested the flash-lag effect occurred after images
from the two eyes were fused. Harris, Duke, and Kopinska (2006)
presented targets that appeared to move in depth. Some targets
exhibited looming, disparity, and perspective cues, and other tar-
gets involved random dot stereograms in which depth was indi-
cated only by disparity. The flash-lag effect was stronger with
random dot stimuli and for approaching motion than for receding
motion. Lee et al. (2008) presented stereo images of a square that
appeared to approach or recede from the participant. A Gaussian
blob appeared for 33 ms near the midpoint of the target trajectory.
1
Brenner et al.’s (2006) findings of a larger forward displacement if a
target moved toward fixation paralleled findings of a larger representa-
tional momentum (see Part 2) if a target moved toward a landmark in
Hubbard and Ruppel (1999). Furthermore, Brenner et al.’s account of their
data paralleled Hubbard and Ruppel’s account for representational momen-
tum: If a target moved toward a landmark, representational momentum and the
landmark attraction effect operated in the same direction, and so they summed,
and forward displacement of that target was relatively larger. If a target moved
away from a landmark, representational momentum and the landmark attrac-
tion effect operated in opposite directions, and so they partially canceled, and
forward displacement of that target was relatively smaller.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
313
FLASH-LAG EFFECT
After target motion, participants indicated perceived depth of the
blob by pressing different keys to increase or decrease binocular
disparity of a probe blob or by adjusting a probe blob to match the
size of the blob flashed during target motion. Participants indicated
a larger size for approaching motion and a smaller size for reced-
ing motion, and this effect was larger with stereomotion than with
looming.
Binocular disparity. Nieman, Nijhawan, Khurana, and Shi-
mojo (2006) presented moving targets and flashed objects within
random dot stimuli, and stimuli were isoluminant and distin-
guished from the background and from each other by binocular
disparity. A moving bar appeared in the lower part of the display,
and a flashed bar of the same size appeared in the upper part of the
display. A flash-lag effect occurred if participants viewed the
display binocularly. If the moving target was monocularly detect-
able and the flashed object was disparity defined, a flash-lag effect
occurred. If the moving target was disparity defined and the
flashed object was monocularly detectable, a flash-lag effect oc-
curred, but it was weaker than if the flashed object or both the
moving target and the flashed object were disparity defined. Given
that a flash-lag effect occurred if stimuli lacked luminance bound-
aries but were disparity defined, Nieman et al. suggested the
flash-lag effect does not result from retinal mechanisms but instead
results from cortical processes subsequent to spatial pooling of
neurons in V1 (cf. Maus, Ward, Nijhawan, & Whitney, 2013). As
noted earlier, Lee et al. (2008) obtained a flash-lag effect with
stimuli that differed in binocular disparity, and Harris et al. (2006)
obtained a flash-lag effect with random dot stimuli (in which the
only depth cue was disparity).
Color. Nijhawan (1997) presented participants with a moving
target consisting of a green bar, and the flashed object consisted of
a red bar briefly superimposed on the moving target. If participants
integrated the spatially overlapping red and green, then a single
yellow bar should have been observed for the duration of the flash.
However, participants reported perceiving a stationary red bar that
lagged a moving green bar. Nijhawan, Khurana, Kamitani, Wa-
tanabe, and Shimojo (1998) had participants execute smooth pur-
suit eye movements past a continuously visible and stationary
green bar (that appeared to move across the retina in the direction
opposite to eye movement). A red bar was briefly superimposed on
the green bar, and participants reported displacement of the red bar
in the direction of pursuit. Sheth, Nijhawan, and Shimojo (2000)
presented a stationary target that continuously changed in color,
becoming more green or more red. Halfway though the change,
another stationary object was briefly flashed nearby, and partici-
pants judged which of the two stimuli was greener. The flashed
object was the same color as the concurrent color of the changing
target, but participants judged the target to be more extreme (i.e.,
to have changed more). In other words, a flash-lag effect for
perceived color change occurred.
Arnold, Ong, and Roseboom (2009) presented participants with
stimuli consisting of an annulus containing a grating that changed
in orientation and an inner disk containing a stationary grating. The
stimuli could change in color, and participants judged whether
color changes in the moving parts and static parts of the stimulus
were synchronous or whether the moving parts and static parts of
the stimulus were aligned at the time of color change. For judg-
ments of synchronicity, color change of the annulus occurred
slightly before, at the same time as, or slightly after color change
of the inner disk. For judgments of alignment, color change of the
moving annulus occurred slightly before, at the same time as, or
slightly after the grating in the moving annulus was aligned with
the stationary grating in the inner disk. A temporal advantage for
alignment judgments (i.e., a flash-lag effect) occurred, but no
temporal advantage for color judgments occurred. Kreegipuu and
Allik (2004) reported a flash-lag effect if the moving target
changed in color, and Cai and Schlag (2001) reported perceived
color change lagged perceived position change. Gauch and Kerzel
(2008b) reported perceived color change lagged perceived position
change in flash-midpoint displays but not in flash-initiated dis-
plays.
Luminance. Purushothaman, Patel, Bedell, and Ög
˘
men
(1998) and Ög
˘
men et al. (2004) reported the flash-lag effect
decreased if luminance of the flashed object increased. With high
luminance of the flashed object and low luminance of the moving
target, the flash-lag effect reversed and the flashed object led the
moving target (i.e., a flash-lead effect). Sheth et al. (2000) and
Ichikawa and Masakura (2006) reported a flash-lag effect for
changes in luminance, and in Ichikawa and Masakura, the flash-lag
effect decreased if participants controlled changes in target lumi-
nance. Nieman et al. (2006) suggested a motion detector activated
only by luminance differences would not detect motion for stimuli
whose average luminance matched the luminance of the back-
ground (e.g., random dot stimuli). Thus, if retinal mechanisms
were responsible for the flash-lag effect, no effect should occur
with random dot stimuli. However, and as noted earlier, Nieman et
al. obtained flash-lag effects with random dot stimuli. Kerzel
(2003) presented a moving target that changed in luminance with-
out warning, and participants indicated where the luminance
change occurred. The perceived location of luminance change was
displaced in the direction of motion, and Kerzel claimed time-to-
consciousness was longer for luminance than for motion and that
this was consistent with the flash-lag effect.
Chappell and Mullen (2010) presented stimuli in which the
flashed object or the moving target was luminance modulated,
equiluminant, or equiluminant in luminance noise. The flashed
object was an inverted isosceles triangle (pointing downward) in
the upper half of the display, and the moving target was an upright
isosceles triangle (pointing upward) in the bottom half of the
display. Participants judged whether the vertex of the flashed
object was to the left or right of the vertex of the moving target. A
flash-lag effect occurred in all conditions except those containing
a luminance-modulated flashed object and an equiluminant or
equiluminant-in-luminance-noise moving target. Adding lumi-
nance noise to an equiluminant moving target, but not to an
equiluminant flashed object, decreased the flash-lag effect. Chap-
pell and Mullen suggested the magnocellular pathway contributes
to the flash-lag effect. This might confer a temporal advantage for
the moving target (e.g., Purushothaman et al., 1998), but even so,
magnocellular pathway processing was not limited to the moving
target, as the flash-lag effect was influenced by luminance of the
flashed object or the moving target. Overall, luminance influences
a flash-lag effect involving relative spatial position of a moving
target and a flashed object, and a flash-lag effect occurs for
changes in luminance.
Contrast. Kanai et al. (2004) varied contrast of a moving
target and a flashed object to the background. The lowest contrast
resulted in the largest flash-lag effect, and Kanai et al. suggested
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
314
HUBBARD
mislocalization arose from positional uncertainty (cf. blurriness in
Fu et al., 2001). Arnold et al. (2009) varied contrast between two
rotating gratings, one of which surrounded the other. The inner
grating was vertically oriented and static, and the outer grating was
rotating; contrast of the inner grating was constant, and contrast of
the outer grating was higher or lower than contrast of the inner
grating. In the higher contrast condition, a flash-lag effect oc-
curred, whereas in the lower contrast condition, a flash-lead effect
occurred (i.e., the flashed object was perceived in front of the
moving target). Arnold et al. suggested their data were inconsistent
with differential latencies, causing participants to perceive moving
stimuli before they perceived stationary stimuli, and that latency
differences modulate, but do not cause, the flash-lag effect. Maus
and Nijhawan (2009) reported larger forward displacement of the
moving target if contrast between the target and background grad-
ually decreased than if the target abruptly vanished, and this is
consistent with the larger flash-lag effect with decreases in contrast
in Kanai et al.
Spatial frequency. Cantor and Schor (2007) presented partic-
ipants with pairs of Gabor patches that were vertically separated;
one patch was presented for a brief duration, and another patch was
presented for a longer duration. The patches were stationary, but
within the longer duration patch the grating appeared to drift.
Participants judged whether the top grating was offset to the left or
right of the bottom grating. The flash-lag effect decreased as
spatial frequency increased from .25 to 1 cycles per degree, and the
decrease with higher spatial frequencies is consistent with involve-
ment of the magnocellular pathway (cf. Chappell & Mullen, 2010).
Cantor and Schor suggested variations in the flash-lag effect
reflect interactions of velocity, temporal frequency, and spatial
frequency. Sheth et al. (2000) presented a stationary square wave
grating that increased or decreased in spatial frequency, and a
stationary square wave grating of a single spatial frequency was
flashed nearby. Participants judged which grating had higher spa-
tial frequency, and judgments were consistent with a flash-lag
effect. Fu et al. (2001) suggested forward displacement of a
moving target is more likely if the target is blurry (i.e., if high
spatial frequencies have been removed). Spatial frequencies influ-
ence the flash-lag effect for spatial position, and a flash-lag effect
occurs for changes in spatial frequency.
Target identity. Moore and Enns (2004) varied size of the
moving target during presentation of the flashed object. A change
in target size nearly eliminated the flash-lag effect, and Moore and
Enns suggested large and transient changes to the moving target
resulted in perception of two separate targets rather than percep-
tion of change in a single target. A decrease in the flash-lag effect
with perception of an apparent additional target is consistent with
previous suggestions that perceptual organization influences the
flash-lag effect (cf. Watanabe, Nijhawan, Khurana, & Shimojo,
2001). Furthermore, Moore and Enns suggested a flash-lag effect
occurs in flash-midpoint displays but not in flash-terminated dis-
plays because object updating occurs with flash-midpoint displays
but not with flash-terminated displays. Khurana, Carter, Watanabe,
and Nijhawan (2006) presented stimuli consisting of the top half of
a face that was flashed and the bottom half of a face that was
moving or stationary (cf. composite face effect; recognition of
either half of a face composed of two halves from different
individuals is impaired when the halves are aligned; Young, Hel-
lawell, & Hay, 1987). Participants identified the individual in the
top half. Faces in which the bottom half was initially in motion
were recognized better than stationary faces, and this was attrib-
uted to a lack of perceptual alignment due to a flash-lag effect.
Bachmann and Põder (2001; Bachmann, Luiga, Põder, & Kalev,
2003) considered the moving target and flashed object as occupy-
ing different perceptual streams. In the motion stream, a sequence
of repetitions of the letter I was presented, and within this sequence
a single letter Z was presented. In the flashed stream, a single letter
Z was presented. Bachmann and Põder reported the Z in the motion
stream appeared to occur before a simultaneous Z in the flashed
stream, and they suggested this reflected a flash-lag effect. Bach-
mann and Põder argued the flash-lag effect did not depend upon
motion (see also Sheth et al., 2000) or changes in features. How-
ever, the letter Z involves different features than the letter I, and so
their argument about features seems overstated unless features are
limited to those aspects of a stimulus that undergo continuous
change. Also, given that the letter Z is a different stimulus than the
letter I, findings of Bachmann and colleagues suggest maintenance
of a single continuous target identity is not essential for the
flash-lag effect if the task of the participants involves comparing
identities (shapes) of stimuli in the moving and flashed streams
rather than comparing spatial positions (cf. Linares & López-
Moliner, 2007).
Many researchers presented a flashed disk within a moving
annulus (e.g., Becker et al., 2009; Eagleman & Sejnowski, 2000b;
Khurana et al., 2000; Nijhawan, 2001), but Kanai and Verstraten
(2006) reversed this and presented a moving disk and a flashed
surrounding annulus (see also Rotman, Brenner, & Smeets, 2002).
Most theories of the flash-lag effect suggest a flash-lag effect
should occur under these conditions, but Kanai and Verstraten
reported that in addition to a portion of the disk being perceived as
ahead of the annulus (as in a typical flash-lag effect), the entirety
of the interior of the annulus was filled by the color of the moving
disk. If the moving target was a vertically oriented bar that moved
to the right, the part of the bar inside a flashed annulus was
correctly perceived, but the parts of the bar extending above and
below the annulus were perceived to be ahead of the bar. Kanai
and Verstraten suggested the veridical position of the flashed
object can be made visible if the moving target is enclosed by a
transient (flashed) stimulus, and that this involves a filling-in
process that propagates from the interior edges of the transient
stimulus. Thus, whether the target encloses, is enclosed by, or is
spatially separated from the flashed object, and whether a single
target identity is maintained, can influence the flash-lag effect.
Pattern entropy. Sheth et al. (2000) presented a stationary
target square that changed in pattern entropy. The target contained
a fixed number of dots, and the dots were initially aligned to a grid
pattern (0% entropy) or randomly distributed (100% entropy).
Successive presentations of the target varied the number of dots
that deviated from their position on the grid and exhibited increas-
ing (from 0% to 100%) or decreasing (from 100% to 0%) pattern
entropy. Midway through the target presentation, another station-
ary square briefly flashed, and participants judged which of the
two squares appeared more disordered. The flashed square exhib-
ited the same entropy as the target square at the moment the
flashed square was presented, but participants judged the flashed
square to be more disordered if entropy of the target was decreas-
ing and less disordered if entropy of the target was increasing (i.e.,
a flash-lag effect for pattern entropy occurred). Sheth et al. noted
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
315
FLASH-LAG EFFECT
there are no known neural structures in visual cortex dedicated to
processing information regarding pattern entropy, and they sug-
gested a flash-lag effect for a stimulus changing in pattern entropy
demonstrates the flash-lag effect reflects processes widespread
throughout the brain.
Duration of the flashed object. Some researchers did not
report the duration of the flashed object in their stimuli (e.g.,
Chappell & Hine, 2004; Ichikawa & Masakura, 2006; Namba &
Baldo, 2004). Among researchers who did report the duration of
the flashed object, durations of 5 (e.g., Shi & Nijhawan, 2008),
9 (e.g., Rotman et al., 2002), 10 (e.g., Becker et al., 2009; Linares
et al., 2007), 13 (e.g., Durant & Johnston, 2004; Kanai et al., 2004;
Vreven & Verghese, 2005), 15 (e.g., Whitney, Murakami, &
Cavanagh, 2000), 30 (e.g., Harris et al., 2006), 33 (e.g., Bachmann
et al., 2003), 60 (e.g., Whitney & Cavanagh, 2000), and 70 (e.g.,
Moore & Enns, 2004) ms have been reported. Some researchers
used a range of flashed object durations (e.g., 1, 43, 93, and 193 ms
in Rotman, Brenner, & Smeets, 2005; 24, 40, 80, 120, 160, and
200 ms in Cantor & Schor, 2007). Brenner and Smeets (2000) and
Eagleman and Sejnowski (2000b) suggested a flashed object can
be presented for up to 80 ms and produce a flash-lag effect, and
Cantor and Schor (2007) suggested the flash-lag effect plateaued
with flash durations longer than 80 ms. Eagleman and Sejnowski
(2000c) reported a flash-lag effect did not occur with flash dura-
tions of 200 ms, but Lappe and Krekelberg (1998) reported a
flash-lag effect with durations of up to 500 ms (see also Krekel-
berg & Lappe, 1999).
Baldo, Kihara, et al. (2002) reported a flash-lag effect if a
flashed object appeared at the beginning of a trial and vanished in
midtrajectory or appeared at midtrajectory and vanished at the end
of a trial. In this case, the “flash” involved an abrupt change in a
continuously visible stimulus and not a briefly presented stimulus.
Kreegipuu and Allik (2004) reported a flash-lag effect if the
reference object was a stationary (and continuously) visible target
that abruptly changed color. Kreegipuu and Allik suggested a
flashed object was not necessary in order for a flash-lag effect to
occur; even so, it could be argued that an abrupt color change is
functionally equivalent to a flash. Given the results of Kreegipuu
and Allik (and comparisons of position change with color change
in Gauch & Kerzel, 2008b, and with luminance change in Kerzel,
2003), a single onset or offset during target motion might be all
that is necessary in order to induce a flash-lag effect (cf. “offset-
lag” in Maus & Nijhawan, 2009). Duration of the flashed object is
less critical. Rather than reflecting characteristics of the flashed
object or the relationship of the flashed object and moving target,
onset or offset of the flashed object might simply be a marker
indicating when the position of the target is to be sampled (cf.
Brenner & Smeets, 2000), similar to using actual target onset or
offset location as a marker in the Fröhlich effect or in representa-
tional momentum, respectively (see Part 2).
Motion of the flashed object. In most studies of the flash-lag
effect, the flashed object was stationary. Indeed, Eagleman and
Sejnowski (2007, p. 4) claim “flash-lag is turned into flash-drag
when motion is attributed to the flash.” However, such a claim is
too broad, as several studies reported a flash-lag effect occurred if the
flashed object was in motion (i.e., if motion was attributed to the
flashed object). In MacKay’s (1958) report, motion information
was available continuously for self-luminous objects but available
only intermittently for stroboscopically illuminated (flashed) ob-
jects, and stroboscopically illuminated objects appeared to lag
self-luminous objects. Krekelberg and Lappe (1999) presented a
rotating target and a flashed object that rotated in alignment with
the target for varying durations. A flash-lag effect occurred, and
the flash-lag effect decreased with increasing rotation (dura-
tion) of the flashed object and was not eliminated until duration
of the moving flashed object reached 500 ms (cf. flashed object
duration in Eagleman & Sejnowski, 2000b). Curiously, Lappe
and Krekelberg (1998; also Cantor & Schor, 2007) reported the
flash-lag effect increased if duration of an intermittently visible
target increased, and so increasing duration of the flashed object
appears to produce an effect opposite to increasing duration of the
moving target.
Bachmann and Kalev (1997) presented a pair of vertically
separated vertical lines that moved horizontally in the same direc-
tion and at the same velocity. The bottom line was visible for a
relatively long spatial extent, and the top line was only visible
within a narrow window centered on the midtrajectory of the
bottom line. The bottom line started its motion first, and the top
line appeared when the bottom line reached alignment with the
initial edge of the window. A flash-lag effect occurred and de-
creased with increases in the length of the flashed object’s trajec-
tory. Bachmann et al. (2003) suggested the decrease in the flash-
lag effect reflects acceleration of perceptual processing over time;
however, it could be argued this result reflects misperception of
velocity (see Palmer & Kellman, 2002, 2003) or extrapolation/
facilitation due to preceding motion of the flashed object. The
larger flash-lag effect at the beginning of the flashed object’s
motion parallels Watanabe et al.’s (2001) finding that the flash-lag
effect was larger if the flashed object was presented at the leading
edge of the moving target. In Cantor and Schor (2007), a briefly
presented moving target in one condition was the same as a
moving flashed object in another condition, and a flash-lag effect
occurred in both conditions.
Gauch and Kerzel (2008a) presented a moving target and a
flashed object in the upper and lower portions, respectively, of the
visual field, and flash-initiated, flash-midpoint, and flash-
terminated displays were shown. The flashed object was stationary
or moving. For flash-initiated displays, a flash-lag effect occurred
with stationary flashed objects and moving flashed objects, and the
flash-lag effect was larger with moving flashed objects. For flash-
midpoint displays, a flash-lag effect occurred with stationary
flashed objects and moving flashed objects, and the magnitudes
did not differ. For flash-terminated displays, a flash-lag effect did
not occur with stationary flashed objects, but a flash-lag effect did
occur with moving flashed objects. Gauch and Kerzel suggested
motion extrapolation, postdiction, and differential latency theories
cannot account for differences in the flash-lag effect across con-
ditions. In general, a flash-lag effect can occur regardless of
whether the flashed object is stationary or moving. However,
moving flashed objects are typically visible for longer durations
than stationary flashed objects. Given that duration of the flashed
object might influence the flash-lag effect (e.g., Eagleman &
Sejnowski, 2000c; Lappe & Krekelberg, 1998), future studies
should explicitly separate duration and motion of the flashed
object.
Cues. Khurana et al. (2000) presented cues instructing partic-
ipants to expect the flashed object to appear on the left or right side
of the display. Cues were presented 100 ms prior to presentation of
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
316
HUBBARD
the flashed object, and cues were valid on 70% of the trials and
invalid on 30% of the trials. Cue validity did not influence the
flash-lag effect. Khurana et al. then presented cues 500 ms prior to
presentation of the flashed object, and cues were valid on 80% of
the trials and invalid on 20% of the trials. Valid cues led to faster
responses regarding whether the flashed object was above or
below the fixation point, but cue validity did not influence the
flash-lag effect. Khurana et al. also reported that cuing which
target in a multitarget display would be closest to the flashed
object did not influence the flash-lag effect. Namba and Baldo
(2004) presented cues instructing participants to expect the flashed
object on the left or right of the target. Cues were presented
between 3,000 and 5,000 ms prior to presentation of the flashed
object, and cues were valid on 80% of the trials and invalid on 20%
of the trials. A flash-lag effect occurred regardless of cue validity,
and the flash-lag effect was smaller with valid cues than with
invalid cues. Namba and Baldo suggested their findings differed
from those of Khurana et al. because of differences in latency
between presentation of cues and presentation of flashed objects.
Brenner and Smeets (2000) presented a rotating stimulus, and a
stationary bar was flashed. If a cue consisting of a faint outline of
the bar was visible from the beginning of a trial until the bar
appeared, then the flash-lag effect was decreased. Rotman et al.
(2002) reported that the perceived location of a flashed object was
mislocalized in the direction in which participants’ eyes were
moving as they tracked a moving target (see also Rotman, Brenner,
& Smeets, 2004; Rotman et al., 2005).
2
Cues in the form of
warning beeps or warning flashes that predicted when a flashed
object would appear did not influence mislocalization, but if the
stimulus was shown twice and participants responded after the
second showing, mislocalization decreased. Rotman et al. reported
that informing participants the flash was about to appear did not
influence mislocalization, but informing participants where the
flashed object would appear decreased mislocalization, and they
suggested the type of cue rather than the presence of a cue
determined whether a cue influenced mislocalization. Rotman et
al. suggested it was not predictability per se that influenced mis-
localization, but rather whether the target appeared at a spatially
cued position. Consistent with this notion, Shioiri et al. (2010)
reported that cuing which target in a multitarget display would be
nearest the flashed object decreased the flash-lag effect.
Hommuk, Bachmann, and Oja (2008) presented a moving target
in the form of a sequence of discrete presentations of the letter I,
and at one position within this motion stream, the letter Z was
superimposed and was the stimulus to be detected. In a separate
perceptual stream on the other side of fixation, a single flashed
letter Z was presented. Presentation of the Z in the flashed stream
appeared to lag a simultaneously presented Z in the motion stream,
and this is consistent with a flash-lag effect (see also Bachmann &
Põder, 2001). However, if the Z in the flashed stream was precued,
a flash-lead effect occurred if that Z appeared within the first 250
ms, and a flash-lag effect did not occur unless that Z did not appear
for at least 400 ms. Such a finding seems inconsistent with a
flash-lag effect in flash-initiated displays, but the precue might
have provided additional facilitation (cf. Maiche, Budelli, &
Gómez-Sena, 2007) not evoked by flash-initiated displays and that
potentially decreased processing latency. In Hommuk et al., the
precue potentially primed recognition of the Z in both streams, and
intervening objects interfered with this priming in the motion
stream but not in the flashed stream.
Predictability of the flashed object. Vreven and Verghese
(2005) noted the position of the flashed object in studies of the
flash-lag effect was usually quite predictable. They presented a
moving target consisting of a line of three dots that pivoted around
one of the end dots. The flashed object was a dot outside the circle
defined by rotation of the line and presented at one of three
locations; in the absence of other information, participants had no
way to predict in which location the flashed object would be
presented. In a spatial cue condition, a dot indicated where the
flashed object would be presented. In a temporal cue condition,
there was a series of beeps, and one of the beeps occurred at the
same time that the flashed object was presented. In a both-cues
condition, both a spatial cue and a temporal cue were presented.
There was also a no-cue condition in which neither a spatial cue
nor a temporal cue was presented. The flash-lag effect was largest
in the no-cue condition, followed by temporal, spatial, and both-
cues conditions; thus, the flash-lag effect was largest when unpre-
dictability of the flashed object was highest (cf. Rotman et al.,
2002). Consistent with this, the spatial cue unequivocally indicated
where the flashed object would be presented, whereas the temporal
cue was not unequivocal, as only one of the beeps coincided with
presentation of the flashed object.
Baldo and Namba (2002; Namba & Baldo, 2004) presented a
flashed object in one of two positions relative to a rotating target.
The position of the flashed object was fixed at a single position,
alternated between positions, or varied randomly. The flash-lag
effect was larger if position of the flashed object varied randomly
(i.e., was not predictable) than if position was fixed. Baldo, Kihara,
et al. (2002) presented the flashed object at a blocked and highly
predictable eccentricity or allowed eccentricity of the flashed ob-
ject to vary randomly. The flash-lag effect increased with increases
in eccentricity of the flashed object, and if predictability was low,
the increase was larger. López-Moliner and Linares (2006) pre-
sented an auditory tone 300 ms prior to the flashed object, and this
did not decrease the flash-lag effect as much as if participants
triggered occurrence of the flashed object by a keypress. However,
there was no effect of a slight delay between the keypress and
presentation of the flashed object, and López-Moliner and Linares
concluded the keypress was not a temporal marker for the flashed
object. In general, increased predictability of the flashed object
(based on cues or other expectations) might decrease processing
latency of that stimulus, and this would diminish the time required
for perception of the flashed object (and diminish the difference in
processing latencies between the moving target and the flashed
object), and thus diminish the flash-lag effect.
Uncertainty of target location. Fu et al. (2001) presented two
horizontally moving targets that were vertically offset and moved
toward each other; if targets were vertically aligned when motion
2
It should be noted that participants in Rotman et al. (2002) visually
tracked the moving target and that visual tracking of a moving target has
been linked to decreases in the flash-lag effect (e.g., Nijhawan, 2001).
Indeed, displacement of the flashed object in the direction of target motion
that Rotman et al. obtained was more similar to a flash-drag or flash-lead
effect. Also, in Rotman et al. the target would have been stationary on the
retina, and so mislocalization was not due to target motion across the retina
(cf. Blohm et al., 2003; Nijhawan, 2001; van Beers et al., 2001).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
317
FLASH-LAG EFFECT
stopped, then there was perceived misalignment (forward displace-
ment of targets) if target edges were blurry (i.e., more positional
uncertainty), but there was no perceived misalignment if target
edges were sharply defined. Kanai et al. (2004) reported the
flash-lag effect increased with decreases in contrast of the moving
target with the background, and they suggested this reflected an
increase in positional uncertainty. Maus and Nijhawan (2006,
2009) reported forward displacement of a moving target increased
if contrast of the target and the background decreased. The find-
ings of Fu et al., Kanai et al., and Maus and Nijhawan suggest
increases in uncertainty regarding target location increase the
flash-lag effect. Brenner et al. (2006) found judgments of synchro-
nization of stimuli in flash-lag effect displays were more variable
than judgments of localization, and they suggested a limiting factor
in determining a moving target’s location was temporal resolution
of underlying signals. Such a suggestion is consistent with low
temporal precision in judgments of location found by Linares,
Holcombe, and White (2009).
Number of moving targets. In many studies of the flash-lag
effect, a moving target was composed of separate stimuli grouped
as a single target (e.g., three circles along an imaginary line, Baldo
& Klein, 1995; two vertically aligned rectangles, Murakami,
2001b; pairs of lines grouped by color, Watanabe et al., 2001). In
other studies of the flash-lag effect, separate stimuli were not
grouped as a single moving target, but were considered as multiple
targets. Shioiri et al. (2010) presented one, two, or six targets that
revolved along the same circular path. The flashed object consisted
of two black dots adjacent to (and on either side of) a single target.
The flash-lag effect increased with increases in the number of
targets, and the increase was smaller if the moving target that
would be closest to the flashed object was cued in advance (i.e., if
participants could track a single target rather than divide attention
over multiple targets). Khurana et al. (2000; but see Baldo, Kihara,
et al., 2002) presented five equally spaced targets moving along a
circular path but reported no difference in the flash-lag effect as a
function of whether the target closest to where the flashed object
would be presented was cued in advance. Relatedly, Krekelberg
and Lappe (1999) reported the flash-lag effect decreased as the
number of flashed objects increased (to a temporal horizon of
approximately 500 ms).
Presence of unrelated stimuli. Maiche et al. (2007) exam-
ined whether the presence of a stimulus unrelated to the moving
target or flashed object could influence the flash-lag effect. They
presented an annulus moving from left to right in the upper visual
field, and a thin vertical bar flashed over the target (cf. Kanai &
Verstraten, 2006, who presented a moving vertical bar and a
flashed annulus). Additionally, a disk-shaped stimulus moved ver-
tically in the lower visual field (to the right of where the flashed
object was presented). If the disk moved toward the position where
the flashed object appeared, then the flash-lag effect was larger
than if the moving disk was not presented, and the increase was
larger if contrast between the moving disk and the background was
increased. However, if the disk moved away from the position of
the flashed object or if the disk’s path of motion was farther from
the flashed object, then the moving disk did not influence the
flash-lag effect. Maiche et al. suggested there was a general
facilitation of processing along the path of motion by a nearby
stimulus. Such facilitation is consistent with differential latency
theory and appears to exhibit a gradient based on proximity to the
location of the flashed object.
Characteristics of the Observer
There has been relatively little research regarding how charac-
teristics of the observer influence the flash-lag effect. Character-
istics of the observer considered here include (a) allocation of
attention, (b) eye movements and fixation, (c) body movement, (d)
control of the moving target or flashed object, (e) perceptual set,
(f) perceptual organization, and (g) conceptual knowledge. The
effects of these variables are summarized in Table 2.
Allocation of attention. Baldo and Klein (1995) reported the
flash-lag effect increased as distance of the flashed object from the
moving target increased, and they attributed this to the time re-
quired to shift attention from the moving target to the flashed
object. Khurana and Nijhawan (1995) responded to Baldo and
Klein by presenting participants with a display in which a rotating
target consisted of a line of equally spaced rectangles, and the
flashed object consisted of circles within the gaps between the
rectangles. A flash-lag effect occurred, and Khurana and Nijhawan
suggested the flash-lag effect was not due to a shifting of attention,
because interleaving the flashed object and moving target removed
the need for a (spatial) shift of attention. As noted earlier, Khurana
et al. (2000) presented multiple moving targets (annuli) and a
flashed object that filled the inner area of one of the targets. A
flash-lag effect occurred and was not influenced by whether the
target or flash position had been cued or by cue validity, and
Khurana et al. concluded the flash-lag effect is not influenced by
the distribution or allocation of attention (but see Shioiri et al.,
2010). However, confidence in this conclusion is weakened by
potential methodological issues (e.g., inappropriate analyses, in-
sufficient stimulus onset asynchronies; see Baldo, Kihara, et al.,
2002).
Baldo, Kihara, et al. (2002) presented a rotating target and a
flashed object consisting of an extension of the diameter of the
target. The flashed object was visible (a) during a single frame near
the midpoint of target motion (onset– offset condition), (b) from
the midpoint of target motion until the target vanished (onset
condition), (c) from the appearance of the target until the midpoint
of target motion (offset condition), or (d) from the appearance of
the target until the target vanished (moving offset condition). A
flash-lag effect occurred in the onset– offset, onset, and offset
conditions, and a flash-lead effect occurred in the moving offset
condition. The data suggest the flash-lag effect depends upon
abrupt onset or offset of the flashed object during target motion,
and this would involve a shift of attention during target motion
(rather than at target onset or offset). The large flash-lag effect
with random dot stereograms in Nieman et al. (2006) is also
consistent with a role of attention: Tracking the moving target
demanded a high level of attention, leaving less attention available
for detecting the flashed object. Thus, the flashed object would
take even longer to enter perceptual awareness, resulting in a larger
flash-lag effect. Such a notion is consistent with findings of a
larger flash-lag effect if participants engaged in a concurrent task
(e.g., Sarich, Chappell, & Burgess, 2007; Scocchia, Actis-Grosso,
de’Sperati, Stucchi, & Baud-Bovy, 2009).
Chappell, Hine, Acworth, and Hardwick (2006) suggested a
flashed object automatically captures attention (see also Kirschfeld
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
318
HUBBARD
& Kammer, 1999). They presented a target that moved in a single
direction or reversed direction, and a flashed object was presented
at motion onset, offset, or reversal. In a landmark condition,
participants compared the perceived position of the moving target
with the perceived position of a stationary landmark (fixation
crosshairs), and in a flash-irrelevant condition, participants made
the same judgments in the presence of an irrelevant flash. A
significant Fröhlich effect occurred, and an irrelevant flash in-
creased the Fröhlich effect in onset or reversal conditions. Partic-
ipants also judged relative positions of the moving target and the
flashed object. A flash-lag effect occurred; there was no difference
between reversal and midtrajectory conditions, and both were
smaller than the flash-lag effect in the motion-onset condition.
Interestingly, the flash-lag effect was larger than the Fröhlich
effect. Müsseler, Stork, and Kerzel (2002) reported displacement
of the moving target decreased if an irrelevant flash was presented
at motion onset or offset, but they did not assess whether a
flash-lag effect occurred. It is possible that different tasks (i.e.,
judging relative location in Chappell et al., 2006; judging absolute
location in Müsseler et al., 2002) resulted in different effects of the
irrelevant flash.
Shioiri et al. (2010) manipulated allocation of attention to the
flashed object by varying the number of targets and by cuing or not
cuing the target by which the flashed object would be presented.
Increasing the number of targets and not cuing the flashed object
both increased the flash-lag effect, and they suggested decreases in
attention increase the flash-lag effect. Shioiri et al. presented
flashed objects near the cued target and also near other targets, and
the flash-lag effect increased as distance of the flashed objects
from the cued target increased. Shioiri et al. admitted their results
seem inconsistent with Khurana et al.’s (2000) lack of an effect of
cuing, but they suggested a small effect of cuing is apparent in
Khurana et al.’s data. Sarich et al. (2007) had participants identify
a briefly presented numeral while simultaneously judging position
of the flashed object, and they reported the flash-lag effect in-
creased if attention was divided. Scocchia et al. (2009) manipu-
lated allocation of attention to the moving target by having par-
ticipants in some conditions also indicate when the moving target
changed shape. The flash-lag effect in the shape-change condition
did not differ from that in a control condition. Differences in
findings of Sarich et al. and Scocchia et al. might result from
whether the secondary task involved the attended flash-lag stim-
ulus (Scocchia et al., 2009) or a different stimulus (Sarich et al.,
2007).
Eye movements and fixation. Nijhawan (2001) presented a
flashed disk within an annulus revolving along a circular path. If
participants fixated the center of the circular path or fixated the
location where the flashed object would be presented, a flash-lag
effect occurred. However, if participants tracked the annulus, the
flash-lag effect was eliminated (see also Nijhawan, 1997).
3
If
participants tracked a smoothly moving object past a stationary
annulus and a disk flashed inside the annulus, a flash-lag effect
involving the stationary annulus and the flashed object occurred.
Nijhawan argued this latter finding demonstrated an eye move-
3
An anonymous reviewer pointed out that differences between fixation
and pursuit conditions cannot be used in evaluating theories of the flash-lag
effect such as motion extrapolation and differential latency, because local-
ization errors during smooth pursuit movements result from errors in
Table 2
Effects of Observer Characteristics on the Flash-Lag Effect
Characteristic Flash-lag effect Primary sources
Allocation of attention Increases with shifts of attention across a larger distance Baldo & Klein (1995); Shioiri et al. (2010)
Increases if participants attend to multiple targets or multiple
tasks
Sarich et al. (2007); Shioiri et al. (2010)
Eye movements and fixation Is eliminated if participants track a smoothly moving target Nijhawan (2001)
Occurs if flashed object is presented during an eye
movement
Blohm et al. (2003); Nijhawan (2001); van Beers
et al. (2001)
Might reflect allocentric encoding of the target Becker et al. (2009); Blohm et al. (2003)
Body movement Reflects active or passive body movement Cai et al. (2000); Nijhawan & Kirschfeld (2003);
Schlag et al. (2000)
Occurs for biological motion stimuli and nonbiological
motion stimuli
Kessler et al. (2010)
Control Decreases if participants control presentation of the flashed
object
López-Moliner & Linares (2006)
Decreases if participants control movement of target using a
mouse but not keypad or trackball
Ichikawa & Masakura (2006, 2010)
Increases if participants control movement of target with a
robotic arm
Scocchia et al. (2009)
Perceptual set Decreases if participants have perceptual set to attend the
flashed object
Gauch & Kerzel (2009)
Perceptual organization Increases at leading edge and decreases at trailing edge of a
target (or set of stimuli perceptually grouped as a single
target)
Watanabe (2004); Watanabe et al. (2001)
Decreases if participants judge shape rather than judge
location
Linares & López-Moliner (2007)
Conceptual knowledge Decreases if stimuli are semantically meaningful Noguchi & Kakigi (2008)
Is influenced by whether target moves in the “typical”
direction
Nagai et al. (2010)
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
319
FLASH-LAG EFFECT
ment based flash-lag effect. Similarly, Blohm et al. (2003) reported
a flash-lag effect occurred if a flashed object was presented during
a smooth anticipatory eye movement. Van Beers et al. (2001)
presented a display in which participants fixated a stationary point
or pursued a moving target, and then judged whether a flashed
object was aligned with two stationary reference points. Judgments
were consistent with a flash-lag effect. Although Blohm et al. and
van Beers et al. discussed their results as reflecting mislocalization
of the flashed object during eye movements, in their studies the
flashed object was not compared to a tracked target, and so their
findings do not conflict with findings of Nijhawan (2001) or
Rotman et al. (2002).
Findings of Blohm et al. (2003) and van Beers et al. (2001)
suggested participants’ frame of reference influenced the flash-lag
effect. Blohm et al. reported early localization of the flashed object
did not suggest a flash-lag effect, but after the position of the
flashed object was translated from egocentric coordinates into
allocentric coordinates, gaze direction suggested a flash-lag effect.
Blohm et al. speculated the influence on the flash-lag effect of
target motion immediately after the flashed object was presented
(e.g., Eagleman & Sejnowski, 2000b; Krekelberg & Lappe, 2000a)
might reflect the time required to translate from initial egocentric
coordinates to allocentric coordinates (cf. Becker et al., 2009).
Interestingly, a flash-lag effect occurred even though participants
were not aware of their eye movements and reported no sense of
motion. Van Beers et al. reported a flash-lag effect occurred in a
two-dimensional pursuit condition (in which a single flashed ob-
ject was presented above or below the horizontal trajectory of the
pursuit target). In addition to mislocalization along the (horizontal)
axis of motion, there was a smaller mislocalization away from the
trajectory of the pursued target (along the vertical axis), and this
mislocalization was relatively larger if the flashed object was
below the trajectory than if the flashed object was above the
trajectory.
4
The potential importance of the frame of reference for the
flash-lag effect was examined by Becker et al. (2009), who pre-
sented a flashed object that filled or was slightly in front of or
behind an annulus revolving along a circular path. Participants
judged whether the flashed object lagged, was at the same position
as, or led the moving target. After the stimuli vanished, partici-
pants saccaded from the fixation point to the position where the
flashed object appeared or to the position of the moving target at
the time of the flash. Saccades to the position of the flashed object
were not displaced from the actual flash position, but saccades to
the position of the moving target at the time of the flash were
displaced in the direction of target motion. In a subsequent exper-
iment, the flash involved an object at a specific spatial position or
a change in the entire background. After the stimuli vanished,
participants saccaded to the position of the moving target at the
time of the flash. If the flash involved a specific spatial position,
saccades were displaced in the direction of target motion, but if the
flash involved the entire background, saccades were not displaced
from the actual target position. Becker et al. suggested the flash-
lag effect did not show dissociation between perceptual and motor
systems typical of other visual illusions (e.g., Aglioti, DeSouza, &
Goodale, 1995) and that a frame-of-reference theory could account
for the flash-lag effect.
Body movement. Schlag, Cai, Dorfman, Mohempour, and
Schlag-Rey (2000) noted previous studies of the flash-lag effect
relied on retinal motion, and they demonstrated that a flash-lag
effect could result from stimulus motion that was inferred on the
basis of extraretinal signals resulting from active body motion. In
a darkened environment, participants focused on a stationary bar
and horizontally rotated their heads back and forth. Approximately
halfway through the head movements, a flashed object consisting
of a bar aligned with the stationary bar was presented. Participants
reported the flashed object appeared to lag the stationary bar. Cai,
Jacobson, Baloh, Schlag-Rey, and Schlag (2000) placed partici-
pants in a rotating chair in a darkened environment. A vertical line
was continuously lit and rotated with the chair, and below this a set
of five smaller vertical bars was flashed. The continuously lit bar
was aligned with the middle of the flashed object, but at the
beginning of rotation, participants reported the continuously lit bar
appeared to be ahead of the middle of the flashed object. Also,
Nijhawan and Kirschfeld (2003) reported a cross-modal flash-lag
effect occurred if participants in a darkened environment moved
their (nonvisible) hand and a visual flash was presented. Thus, a
flash-lag effect can result from active or passive body movement.
In Schlag et al. (2000), Cai et al. (2000), and Nijhawan and
Kirschfeld (2003), motion involved the participant’s own body.
Kessler et al. (2010) presented flash-lag stimuli in which the
moving target was another person’s body. In a biological motion
condition, participants were shown an overhead view of a table,
and a person was seated at one side of the table and a mug was
placed near the opposite side. The person reached toward the mug,
and a white rectangle (approximately the size of the person’s hand)
was briefly flashed near the center of the table, and the flashed
object was presented slightly before, at the same time as, or
slightly after it would have been aligned with the reaching hand. A
control condition in which the reaching hand and the mug were
replaced by a moving rectangle (the size of the hand and lower
arm) and a circle, respectively, was also presented (and the moving
rectangle exhibited the same velocity profile as the biological
motion stimulus). A robust flash-lag effect occurred in both bio-
logical motion and control conditions. The flash-lag effect in the
sensorimotor integration and thus involve different mechanisms than
would localization errors during fixation. Additionally, it is not clear how
oculomotor behavior would (a) be involved in a flash-lag effect with
changes in visual stimuli that did not involve changes in location (e.g.,
color; Sheth et al., 2000), (b) be involved in a flash-lag effect with nonvisual
stimuli (e.g., auditory stimuli; Alais & Burr, 2003), or (c) account for effects of
higher level processes on the flash-lag effect (e.g., conceptual knowledge;
Nagai et al., 2010; Noguchi & Kakigi, 2008).
4
The larger displacement away from the horizontal trajectory of the
target for flashed objects below the trajectory in van Beers et al. (2001) is
consistent with displacement in the direction of implied gravitational
attraction that has been referred to as representational gravity in studies of
visual (Hubbard, 1997) and auditory (Hubbard & Ruppel, 2013) localiza-
tion. If the flashed object was below the horizontal trajectory, then repre-
sentational gravity and bias away from the trajectory operated in the same
direction, and so they summed, and displacement away from the trajectory
was relatively larger. If the flashed object was above the horizontal
trajectory, then representational gravity and bias away from the trajectory
operated in opposite directions, and so they partially canceled, and dis-
placement away from the trajectory was relatively smaller. Thus, the frame
of reference used by participants in van Beers et al. appeared to include an
external axis aligned with the direction of implied gravitational attraction,
and although not noted as such by van Beers et al., their findings provide
the first evidence that representational gravity can influence the flash-lag
effect.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
320
HUBBARD
biological motion condition, but not in the control condition, was
influenced by whether a first-person (aligned with participants’
perspective) or third-person (opposite to participants’ perspective)
perspective was shown, and this suggests a possible role of per-
ceived agency.
Control. Ichikawa and Masakura (2006) had participants con-
trol with a computer mouse a moving target that ascended or
descended. Hand motion believed by participants to control target
motion decreased the flash-lag effect, and Ichikawa and Masakura
suggested active control of a target in a specific visual field
facilitated processing for that visual field. Ichikawa and Masakura
(2010) found that if participants controlled the target by moving a
computer mouse, the flash-lag effect decreased if the mapping
between movements of the mouse and movements of the target
was typical of computer operating systems with which participants
were familiar (i.e., motion of the mouse away from or toward the
participant produced upward or downward motion, respectively, of
the target) but not if the mapping was unfamiliar. The decrease
with a familiar mapping occurred regardless of whether partici-
pants could view their hand, suggesting proprioceptive information
rather than visual information regarding their hand influenced the
flash-lag effect. Training with an unfamiliar mapping had minimal
effect. If participants controlled the target with a sustained key-
press or manipulation of a trackball, a decrease in the flash-lag
effect did not occur. The latter types of hand movements might not
have mapped as clearly onto target motion.
Participants in Ichikawa and Masakura (2006, 2010) controlled
motion of the target, but participants in López-Moliner and Linares
(2006) controlled presentation of the flashed object. Participants in
López-Moliner and Linares viewed a bar that pivoted around an
end point, and the bar contained a small gap. A small rectangle was
briefly presented in or near the gap. In one condition, participants
triggered presentation of the flashed object by pressing the space
bar, and in other conditions, participants had no control over when
the flashed object was presented. The flash-lag effect decreased if
participants triggered presentation of the flashed object. López-
Moliner and Linares attributed the decrease to control rather than
to predictability, as a similar decrease was not observed if the
flashed object was cued by a sound 300 ms prior to presentation of
the flashed object. The greater importance of control rather than
predictability is consistent with Ichikawa and Masakura’s (2006,
2010) finding that participants’ beliefs regarding control of target
motion, rather than actual control of target motion, was related to
the flash-lag effect. These findings are consistent with the notion
the flash-lag effect is not a low-level phenomenon but depends on
high-level attributions regarding the source of target motion.
Scocchia et al. (2009) presented a moving circular target, and a
flashed object was displayed at unpredictable times. The flash-lag
effect was larger if participants controlled the target (via a robotic
arm) than if participants did not control the target. Scocchia et al.
suggested motor information provided by control of the target
reduced processing latency of the target. Findings of Scocchia et
al. seem inconsistent with findings of Ichikawa and Masakura
(2006, 2010), but as noted by Scocchia et al., there were numerous
methodological differences between their experiments and those of
Ichikawa and Masakura. Scocchia et al. also noted presentation of
the flashed object in Ichikawa and Masakura was triggered when
the moving target passed a predetermined location, thus partici-
pants (indirectly) controlled timing of the flashed object (but
whether participants were aware of this is not clear), which López-
Moliner and Linares (2006) demonstrated decreased the flash-lag
effect. Active control of the target would require more attention,
and so effects of control are consistent with proposals that atten-
tion modulates the flash-lag effect (e.g., Baldo, Kihara, et al.,
2002; Sarich et al., 2007; Shioiri et al., 2010). Overall, participant
control of the moving target or flashed object can influence the
flash-lag effect, but the type of control determines whether the
flash-lag effect is increased or decreased.
Perceptual set. Gauch and Kerzel (2009) suggested partici-
pants respond differently if they are certain a flashed object will be
presented than if they are uncertain whether a flashed object will
be presented. Gauch and Kerzel presented a moving target accom-
panied by a flashed object or by a stationary object that was visible
until the end of target motion. In a pure condition, only flashed
object trials were presented. In a mixed condition, flashed object
trials and stationary object trials were presented in a random order.
A flash-lag effect occurred in the pure condition but not in the
mixed condition with both flash-initiated displays and flash-
midpoint displays. Gauch and Kerzel suggested mixed trials in-
duced a perceptual set to attend the onset position because partic-
ipants did not know how long the stimulus would remain visible.
However, given the brief presentation in flashed object trials, it
might be argued that pure trials should also induce a perceptual set
to attend onset position. A role of perceptual set in the flash-lag
effect distinguishes the flash-lag effect from illusions that are not
influenced by knowledge and expectation (e.g., subjective con-
tours) but not from mislocalizations that can be influenced by
knowledge and expectation (e.g., representational momentum).
Perceptual organization. Watanabe et al. (2001) investigated
how perceptual organization of the moving target influenced the
flash-lag effect. Moving targets consisted in part of (a) a square,
(b) two parallel vertical bars, (c) four parallel vertical bars, (d) two
squares created by connecting the first and second vertical bars and
by connecting the third and fourth vertical bars, or (e) four parallel
vertical bars in which the first and second bars were one color and
the third and fourth bars were a different color. Manipulation of
color and of whether bars or squares were presented was designed
to influence whether or which stimuli were perceptually grouped
with other stimuli. A flashed object near the leading edge of a
moving target (e.g., first of four bars, leading edge of a square)
resulted in a larger flash-lag effect than a flashed object near the
trailing edge of a moving target (e.g., fourth of four bars, trailing
edge of a square). Watanabe et al. suggested differences in the
flash-lag effect were related to perceptual grouping. Watanabe
(2004) found differences in the flash-lag effect as a function of
perceptual grouping in Watanabe et al. (2001) occurred with
flash-initiated and flash-midpoint displays but not with flash-
terminated displays. Watanabe suggested relative positions be-
tween moving targets and flashed objects are computed after
motion grouping.
Linares and López-Moliner (2007) presented Glass patterns
(pairs of dots in which members of each pair are separated by
rotating the radius connecting one dot to the center of the pattern
by a fixed amount). One member of each pair was the moving
target, and the other member of that pair was the flashed object. If
a flash-lag effect occurred, then the best recognition of the global
shape of the Glass pattern should have occurred if the flashed
object was presented prior to when the moving target arrived at the
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
321
FLASH-LAG EFFECT
position that would have resulted in a Glass pattern. However, the
best recognition of the Glass pattern occurred if the flashed object
was presented at the time the moving target occupied the position
that resulted in a Glass pattern. Linares and López-Moliner sug-
gested a flash-lag effect did not occur; more specifically, they
suggested that local spatial relationships between moving targets
and flashed objects were preserved if those differences are used to
detect global shape.
5
The lack of a flash-lag effect if participants
did not engage in judgment of position is consistent with the view
that high-level processes contribute to the flash-lag effect; if the
flash-lag effect resulted solely from low-level processes, it would
have occurred regardless of participant intent (cf. Gauch & Kerzel,
2009).
Conceptual knowledge. Noguchi and Kakigi (2008) pre-
sented native Japanese speakers and non-Japanese English speak-
ers with moving targets and flashed objects that were segments of
Kanji (ideographic) letters, and performance with Kanji segments
was compared to performance if moving targets and flashed ob-
jects were composed of bars, grids, gratings, or pseudo-Kanji
shapes. The flash-lag effect decreased with Kanji segments for
Japanese speakers knowledgeable of Kanji but not for non-
Japanese English speakers not knowledgeable of Kanji. The flash-
lag effect was influenced by conceptual knowledge of Kanji let-
ters. Also, differences in visual evoked fields in participants
knowledgeable of Kanji occurred as early as 160 ms after presen-
tation of the flashed object, and this suggested a substantial effect
of top-down knowledge on the flash-lag effect. Nagai et al. (2010)
presented a flash-lag display in which the moving target was a
picture of an automobile that moved forward or backward, and the
flashed object was a white dot above the roof of the automobile.
The flash-lag effect was larger for backward motion than for
forward motion,
6
and the flash-lag effect was larger for backward
or forward motion than if the automobile was stationary. The
flash-lag effect was influenced by conceptual knowledge of the
typical direction of motion of an automobile.
Part 2: Properties and Related Phenomena
In addition to questions regarding effects of specific variables
on the flash-lag effect that were addressed in Part 1, there are
questions regarding more general properties of the flash-lag effect
and the relationship of the flash-lag effect to other perceptual and
cognitive phenomena that can be addressed. These include whether
the flash-lag effect (a) is temporal or spatial, (b) results from
mislocalization of the moving target or the flashed object, (c) is
limited to visual stimuli, (d) reflects low-level or high-level pro-
cesses, (e) is related to flash-drag and flash-lead effects, (f) is
related to temporal order judgments, (g) is related to the Fröhlich
effect, (h) is related to representational momentum, (i) is related to
backward referral, (j) is related to anisotropic distortion, and (k) is
related to flag errors in football.
Temporal or Spatial?
There has been debate regarding whether the flash-lag effect is
a spatial or temporal phenomenon (e.g., Eagleman & Sejnowski,
2002; Krekelberg & Lappe, 2002). Theories based on motion
extrapolation (e.g., Nijhawan, 1994, 2008) suggest the flash-lag
effect is a spatial phenomenon. Theories based on latency differ-
ences (e.g., Whitney & Cavanagh, 2000; Whitney & Murakami,
1998) suggest the flash-lag effect is a temporal phenomenon.
Theories based on postdiction suggest the flash-lag effect is a
spatial phenomenon (e.g., Eagleman & Sejnowski, 2002)ora
temporal phenomenon (e.g., Whitney, 2002). Ichikawa and Ma-
sakura (2006) defined the flash-lag effect as a temporal lag be-
tween the moving target and the flashed object, and Shen et al.
(2007) suggested the flash-lag illusion is a temporal phenomenon.
Murakami (2001a) suggested the flash-lag effect is a “spatiotem-
poral correlation structure” in which the spatial position of a
flashed object in the past is compared to the spatial position of a
moving target in the present. As space cannot be crossed without
passing through time (and for a moving target, elapsed time relates
to crossed space), Murakami’s suggestion that the flash-lag effect
involves spatial and temporal components seems appropriate. In-
deed, there is precedent for interaction of spatial information and
temporal information in perception (e.g., tau and kappa effects;
Collyer, 1977; Jones & Huang, 1982).
Kreegipuu and Allik (2004) noted the flash-lag effect is usually
measured in terms of spatial offset (between the moving target and
the flashed object), and they attempted to separate spatial offset
and temporal offset. The moving target was a bar that changed
color. Participants judged whether the color change occurred after
(a) the reference object flashed (time judgment) or (b) the target
passed a reference object that flashed (position judgment). In time
judgment, there was no systematic bias, but in position judgment,
responses were displaced further along the path of motion. If the
flashed reference object was replaced with a stationary bar that
changed color, the results were similar: no bias for time judgment,
and displacement further along the path of motion for position
judgment. A similar dissociation between perceived color and
perceived position was reported by Gauch and Kerzel (2008b), and
Kerzel (2003) reported a similar dissociation between perceived
luminance and perceived position. Kreegipuu and Allik (2003)
5
Linares and López-Moliner (2007) also suggested judgment of position
is necessary in order for a flash-lag effect to occur. However, this sugges-
tion seems overly broad given that a flash-lag effect can be found with
changes in color or in luminance (e.g., Sheth et al., 2000), as those
dimensions do not involve changes in (spatial) position per se. It might be
that a flash-lag effect occurs if the changing quality of the target can be
described in terms of a position along a continuous dimension (e.g., smaller
to larger, dimmer to brighter, left to right) in some (potentially abstract)
feature space rather than as an entry in a discontinuous list of discrete
categories (e.g., shapes). However, occurrence of a flash-lag effect if
stimuli are composed of discrete letters (Bachmann et al., 2003; Bachmann
& Põder, 2001; Hommuk et al., 2008) suggests continuity of change in
some feature space is not necessary for a flash-lag effect.
6
Nagai et al. (2010) referred to this result as “surprising” (p. 370)
because it differed from previous findings of a larger representational
momentum for forward motion than for backward motion. One possible
explanation for the pattern in Nagai et al. is that forward displacement of
the moving target in the flash-lag effect display decreased with backward
motion (consistent with previous studies of representational momentum),
but that forward displacement of the flashed object decreased even more
with backward motion (e.g., perhaps the relative lack of experience with
backward motion resulted in a relatively smaller [percentage of] spreading
activation from the representation of the position of the target to the
representation of the position of the flashed object); therefore, relative
differences between the target and the dot looked larger for backward
motion than for forward motion, and so the flash-lag effect appeared larger
for backward motion.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
322
HUBBARD
noted performance in timing judgments and performance in local-
ization judgments were not equivalent. Consistent with this, Kanai
et al. (2004) noted a dissociation between the flash-lag effect
measured in units of space and measured in units of time. The
flash-lag effect involves temporal information and spatial infor-
mation, but the relationship between these types of information is
not clear.
Moving Target or Flashed Object?
The flash-lag effect reflects bias in the relative positions of the
moving target and the flashed object, but as traditionally described
and measured, it is not clear whether this bias in relative position
results from bias in perception of the absolute position of the
flashed object or from bias in perception of the absolute position of
the moving target. Hazelhoff and Wiersma (1924) reported ob-
servers misperceived the position of a flashed object as further in
the direction of an eye movement, and van Beers et al. (2001)
reported a flashed object was displaced forward. Whitney and
Cavanagh (2000) reported displacement in the direction of motion
for horizontal lines flashed on either side of a rotating grating or
pairs of vertically moving gratings. Durant and Johnston (2004)
reported displacement in the direction of motion for stationary bars
flashed on either side of a rotating bar. Rotman et al. (2002, 2004,
2005) reported a flashed object was displaced in the direction of
motion or in the direction of gaze. Hubbard (2008) reported a
briefly presented stationary object near the end of a moving
target’s trajectory was displaced in the direction of motion.
Brenner et al. (2006) and Maus and Nijhawan (2006, 2008, 2009)
reported a moving target could be displaced in the direction of
motion. Also, large literatures on the Fröhlich effect and on rep-
resentational momentum provide ample evidence that the repre-
sented position of a moving target is displaced in the direction of
motion.
Many researchers reported displacement for the flashed ob-
ject or for the moving target, but these two types of displace-
ment have rarely been directly compared. A notable exception
is Shi and de’Sperati (2008), who presented a moving target in
the form of an arc that traveled a circular path and a flashed
object in the form of a dot along the circular path in front of or
behind the leading edge of the moving target. Participants
indicated (a) the position of the moving target at the time the
flashed object was presented or the position of the flashed
object or (b) whether the flashed object was ahead of or behind
the leading edge of the moving target. A flash-lag effect oc-
curred. Judgment of position was displaced in the direction of
motion for moving targets and for flashed objects, and displace-
ment was larger for moving targets than for flashed objects (cf.
displacements in Hubbard, 2008; saccades in Becker et al.,
2009). One possible explanation is that the flash-lag effect
results from mislocalization of the absolute positions of the
moving target and of the flashed object, with larger mislocal-
ization for the moving target resulting in an apparent lagging of
the flashed object. Given this, the flash-lag effect would be a
derived or second-order illusion that emerges from more basic
illusions involving perceived absolute positions of the moving
target and/or the flashed object.
Limited to Vision?
Research on the flash-lag effect generally used visual stimuli,
and given the effects of oculomotor behavior on the flash-lag
effect discussed earlier, an appropriate question is whether the
flash-lag effect requires or is based on visual or visuospatial
processes or representations. If the flash-lag effect is limited to
vision, that might suggest the flash-lag effect results from isolated
low-level perceptual processes or representations rather than from
high-level cognitive processes or representations. Nijhawan and
Kirschfeld (2003) had participants move their (nonvisible) hand
within a darkened environment and judge the felt position of their
hand relative to the position of a visual flash. If the flash was
aligned with their hand, participants perceived the flash as lagging
their hand (cf. body motion and visual flash-lag in Cai et al., 2000;
Schlag et al., 2000). Nijhawan and Kirschfeld referred to this as a
motor flash-lag, but it is more appropriately described as a cross-
modal flash-lag, as the moving target and the flashed object were
in different modalities (cf. cross-modal flash-lag in Alais & Burr,
2003; Arrighi et al., 2005); a true motor flash-lag would require
that both the moving target and the flashed object involved motor
activity. Nijhawan and Kirschfeld suggested visual flash-lag and
motor flash-lag resulted from a common mechanism involving
forward models in the motor system that compensated for neural
processing delays.
Alais and Burr (2003) presented a moving target consisting of
an auditory frequency sweep and a flashed object consisting of a
brief tone, and a flash-lag effect occurred. This finding is consis-
tent with previous findings of forward displacement of the initial
(Hubbard & Ruppel, 2013) or final (Johnston & Jones, 2006) pitch
of an auditory target changing in frequency. Alais and Burr also
presented a constant frequency translating across space and a brief
static sound, and a flash-lag effect occurred. This finding is con-
sistent with previous findings of forward displacement of a moving
sound source in spatial hearing (Getzmann, 2005). Alais and Burr
then combined an auditory translating stimulus with a flashed
visual stimulus or a visual translating stimulus with a flashed
auditory stimulus. Participants judged whether the spatial position
of the flashed stimulus was ahead of or behind the spatially
translating stimulus. Flash-lag effects resulting from cross-modal
stimuli were smaller than if the moving target and the flashed
object were both auditory, but larger than if the moving target and
the flashed object were both visual. Alais and Burr argued cross-
modal flash-lag effects were inconsistent with differential latency
theory; latencies for auditory stimuli are shorter than latencies for
visual stimuli, and so differential latency theory predicts a flash-
lead effect in the auditory-flash/visual-motion condition (see also
Arrighi et al., 2005; Krekelberg, 2003).
Vroomen and de Gelder (2004) presented a visual moving target
and a visual flashed object. Presentation of the flashed object on
some trials was accompanied by an auditory tone synchronized
with the flashed object. A flash-lag effect occurred regardless of
whether a tone was presented, but if a tone was presented, mag-
nitude and variability of the flash-lag effect decreased. Vroomen
and de Gelder suggested presentation of the tone facilitated pro-
cessing of the flashed object, and this reduced the difference
between processing latencies for the moving target and the flashed
object. Vroomen and de Gelder then presented the tone slightly
before, at the same time as, or slightly after the flashed object was
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
323
FLASH-LAG EFFECT
presented. If the tone was presented before or after the flashed
object, the flash-lag effect decreased or increased, respectively.
Stekelenburg and Vroomen (2005) presented similar stimuli and
recorded event-related potentials. A sound presented before or
after the flashed object decreased or increased, respectively, the
flash-lag effect and amplitude of the N1 component of the event-
related potential. Timing of the sound relative to the flashed object
did not influence latency of the N1. Overall, the flash-lag effect is
not limited to visual or visuospatial processes or representations,
but occurs in multiple modalities, cross-modally, and is modulated
by cross-modal information.
Low Level or High Level?
The flash-lag effect does not occur before perceptual grouping
(Watanabe, 2004; Watanabe et al., 2001) or pattern recognition
(Linares & López-Moliner, 2007), and this suggests the flash-lag
effect results from relatively high-level processes. Influences of
conceptual knowledge (Nagai et al., 2010; Noguchi & Kakigi,
2008), beliefs regarding control of the target (Ichikawa & Ma-
sakura, 2006, 2010), predictability of the flashed object (Baldo &
Namba, 2002; Vreven & Verghese, 2005), and perceptual set
(Gauch & Kerzel, 2009) also suggest high-level processes contrib-
ute to the flash-lag effect. Nieman et al. (2006) reported a flash-lag
effect in the absence of luminance boundaries (with random dot
stereograms), and this would require cortical processes sensitive to
binocular disparities and be consistent with high-level processes.
Neurons in cat V1 (Jancke, Erlhagen, Schöner, & Dinse, 2004) and
in monkey V4 (Sundberg, Fallah, & Reynolds, 2006) respond
more quickly to moving targets than to flashed objects, and trans-
cranial magnetic stimulation of human MT⫹, but not V1/V2,
reduces the flash-lag effect (Maus et al., 2013). A temporal ad-
vantage for moving stimuli relative to stationary stimuli occurs in
cat LGN (Orban, Hoffman, & Duysens, 1985), but this advantage
is only 15 ms, and so is smaller than the 45- to 80-ms difference
in a typical visual flash-lag effect.
Arnold, Durant, and Johnston (2003) examined the flash-lag
effect relative to a perceptual phenomenon known to be cortical in
origin: the tilt illusion (a vertical grating is perceived to tilt in the
direction opposite to the tilt of a surrounding grating; e.g.,
Schwartz, Sejnowski, & Dayan, 2009). Participants viewed a cir-
cular stimulus divided into an annulus and an inner disk. The
annulus consisted of a grating that rotated clockwise or counter-
clockwise; during most of the trial, the inner disk consisted of a
solid gray stimulus, but a test stimulus consisting of a stationary
vertical grating was briefly flashed during each trial. The test
grating could be flashed before the rotating grating reached align-
ment, at the moment of alignment, or after the rotating grating
passed alignment. If a flash-lag effect occurred, then a tilt illusion
should not have occurred if the test grating was presented before
the rotating annulus reached alignment; however, a robust tilt
illusion occurred. Indeed, the strength of the tilt illusion at differ-
ent latencies of test presentation did not differ from previous
reports of the tilt illusion with static targets. Arnold et al. suggested
it is unlikely the flash-lag effect arises before the tilt illusion, and
this suggestion is consistent with the notion that the flash-lag effect
does not result from low-level processes (cf. Fukiage & Murakami,
2010).
A visual flash-lag effect can be modified by presentation of an
auditory tone (Vroomen & de Gelder, 2004), a flash-lag effect
occurs if the moving target is visual and the flashed object is
auditory (Alais & Burr, 2003; Arrighi et al., 2005), and a flash-lag
effect occurs if the moving target is a participant’s (nonvisible)
hand and the flashed object is visual (Nijhawan & Kirschfeld,
2003). Findings of such cross-modal contributions to the flash-lag
effect are not consistent with the hypothesis that the flash-lag
effect results from low-level mechanisms, as the flash-lag effect in
such cases involves cross-modal or multisensory processes that
presumably occur at a higher level. Whitney and Cavanagh (2000)
reported a flash-lag effect occurred with dichoptic displays in
which moving targets were presented to one eye and flashed
objects were presented to the other eye, and coupled with findings
of a flash-lag effect with random dot stereograms (Harris et al.,
2006; Nieman et al., 2006), this suggests the flash-lag effect
involves processing at or beyond the level of binocularly sensitive
neurons. Krekelberg and Lappe (2001) suggested the flash-lag
effect is unlikely to be understood in terms of low-level processing
and that the origin of the flash-lag effect is likely to be beyond the
point of integration of retinal and extraretinal signals (cf. Cai et al.,
2000; Schlag et al., 2000).
Although findings from most studies suggest the flash-lag effect
involves high-level processes, a few exceptions have been re-
ported. Activation patterns of retinal ganglion cells can appear to
extrapolate a moving target’s trajectory (for review, see Gollisch &
Meister, 2010), although whether this extrapolation matches the
magnitude of the flash-lag effect is not clear. Neurons in monkey
V5/MT respond more rapidly to transient (flashed) stimuli than to
moving stimuli (Raiguel, Lagae, Gulya
`
s, & Orban, 1989). Anstis
(2007, 2010) presented participants with a “chopsticks illusion” in
which a horizontal bar and an intersecting vertical bar moved in
different circular motions. Participants adjusted the perceived po-
sition of a flashed dot to coincide with the moving intersection of
the two bars. A flash-lag effect occurred. Anstis (2010) presented
a reversed phi stimulus consisting of radii that rotated around a
common center and alternated between black and white. A flashed
object consisting of arrows at the 12 and 6 o’clock positions was
presented, and participants judged whether the arrows aligned with
a radius. A flash-lag effect occurred. Anstis argued responses to a
flashed object with chopsticks illusion stimuli or reversed phi
stimuli were driven by the physical, rather than the perceived,
direction of motion. Overall, it appears that high-level processes
and low-level processes both contribute to the flash-lag effect (cf.
Erlhagen, 2003; Jancke & Erlhagen, 2010).
Flash-Drag and Flash-Lead?
Investigation of the flash-lag effect revealed other types of flash
mislocalization, but it is not clear why these rather than a flash-lag
effect sometimes occur. One such type of flash mislocalization is
the flash-drag effect, in which perceived position of a briefly
flashed stimulus is shifted in the direction of nearby motion (cf.
Hubbard, 2008; Shi & de’Sperati, 2008). Hine, White, and Chap-
pell (2003) had participants judge the location of a moving visual
stimulus relative to a stationary visual stimulus when a click
(“auditory flash”) occurred, and they reported a flash-drag effect.
Fukiage, Whitney, and Murakami (2011) presented a phase-shifted
grating at a different random location every 125 ms, and they
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
324
HUBBARD
reported a larger flash-drag effect if the flashed object was pre-
sented 50 to 150 ms before the grating changed location; however,
Murakami (2001b) previously reported a robust flash-lag effect
with a randomly moving stimulus and a flashed object, and the
reason for the discrepancy between Fukiage et al. and Murakami’s
earlier article is not clear. Murai and Murakami (2012) reported the
flash-drag effect started to increase about 100 ms prior to onset of
target motion and decreased 100 ms prior to disappearance of a
moving target. Eagleman and Sejnowski (2007) suggested the
flash-lag effect and the flash-drag effect worked in opposite direc-
tions and that an increased flash-drag effect diluted the flash-lag
effect.
A second such type of flash mislocalization is the flash-lead
effect, in which a flashed object leads rather than lags a moving
target. Purushothaman et al. (1998) and Ög
˘
men et al. (2004)
reported that increasing luminance of a flashed object resulted
in a flash-lead effect. Hommuk et al. (2008) reported that
precuing a flashed object resulted in a flash-lead effect if the
flashed object was presented within the first 250 ms of the
moving target, and Baldo, Kihara, et al. (2002) reported a
flash-lead effect if the flashed object was visible during target
motion and vanished at target offset. Arnold et al. (2009)
reported a flash-lead effect if contrast between stimuli de-
creased. Whitney and Cavanagh (2000) presented flashed ob-
jects on either side of a rotating target, and participants judged
whether the flashes were aligned. Judgments were displaced in
the direction of target motion but were not influenced by
eccentricity. Durant and Johnston (2004) presented two grat-
ings, one of which drifted upward and one of which drifted
downward. Two flashes, one on either side of the gratings, were
presented. Participants judged alignment of the flashes. Judg-
ments were displaced in the direction of motion of the nearest
grating, and perceived misalignment decreased with increases
in eccentricity. However, in both the flash-drag and flash-lead
effects, the flashed object is shifted in the direction of target
motion, and so whether these are actually different effects is not
clear.
Temporal Order Judgment?
The relationship between the flash-lag effect and temporal
order judgment of the moving target and flashed object is
critical for theories of the flash-lag effect involving differential
latencies for moving targets and flashed objects. Nijhawan et al.
(2004) presented a flash-initiated display and varied whether
the flashed object appeared slightly before, at the same time as,
or slightly after the moving target appeared. Participants judged
which stimulus appeared first. Nijhawan et al. suggested latency
for perceiving a flashed object was shorter than latency for
perceiving a moving target (cf. Raiguel et al., 1989), and
although data figures were presented, statistical tests supporting
this claim were not reported. Chappell et al. (2006) had partic-
ipants judge whether a flashed object was presented before or
after target motion onset, reversal of target direction, or target
motion offset. A point of subjective simultaneity for each
condition was calculated, but none of these differed from zero.
Cravo and Baldo (2008) presented moving targets and flashed
objects in opposite hemifields, and temporal onset asynchrony
of moving targets and flashed objects varied. The mean point of
subjective simultaneity did not differ from zero. All these
findings challenge the claim of differential latency theory that
the flash-lag effect results from shorter latencies in processing
moving targets than in processing flashed objects.
Fröhlich Effect?
The perceived initial (onset) position of a moving target is
displaced forward from the actual initial position of that target,
and this is referred to as the Fröhlich effect (Fröhlich, 1923; for
review, see Kerzel, 2010). Forward displacement of the moving
target in the Fröhlich effect appears similar to forward displace-
ment of the moving target in a flash-initiated display; indeed, if
the flashed object is not considered, then displacement of initial
target position in a flash-initiated display would seem to be
nothing more than a Fröhlich effect. Given this, a flash-lag
effect in a flash-initiated display might be a Fröhlich effect in
which perceived initial target position is measured relative to an
external stimulus (the flashed object) rather than relative to the
actual initial target position. Several variables have similar
influences on the flash-lag effect and Fröhlich effect (see also
Kreegipuu & Allik, 2003). The flash-lag effect (Wojtach et al.,
2008) and Fröhlich effect (Müsseler & Aschersleben, 1998)
increase with increases in target velocity. The flash-lag effect
(Ög
˘
men et al., 2004) and Fröhlich effect (Carbone & Ansorge,
2008) are influenced by luminosity. The flash-lag effect
(Namba & Baldo, 2004) and Fröhlich effect (Müsseler & As-
chersleben, 1998) are decreased by a valid cue prior to stimulus
presentation. However, the flash-lag effect (Baldo, Kihara, et
al., 2002; Lappe & Krekelberg, 1998) but not Fröhlich effect
(Müsseler & Aschersleben, 1998) increases with increases in
eccentricity.
Some mechanisms proposed for the flash-lag effect are sim-
ilar to mechanisms proposed for the Fröhlich effect (e.g., the
time required to shift attention to the onset position of the
flashed object, Baldo & Klein, 1995, or moving target, Müs-
seler & Aschersleben, 1998). Kirschfeld and Kammer (1999)
and Eagleman and Sejnowski (2007) suggested the same mech-
anisms underlie the flash-lag effect and the Fröhlich effect.
However, Kreegipuu and Allik (2003) reported dissociation
between the flash-lag effect and Fröhlich effect if a flashed
object was presented near the time a previously stationary target
began moving and participants estimated whether the flashed
object appeared (a) before or after motion began or (b) to the
left or right of where motion began. It is not clear, though,
whether this reflected dissociation of (a) spatial and temporal
components of the flash-lag effect or (b) the flash-lag effect and
Fröhlich effect. Whitney and Cavanagh (2000) reported presen-
tation of a cue decreased the Fröhlich effect but not the flash-
lag effect; however, Brenner and Smeets (2000), Namba and
Baldo (2004), and Vreven and Verghese (2005) reported pre-
sentation of a cue decreased the flash-lag effect. Chappell et al.
(2006) reported the flash-lag effect was larger than the Fröhlich
effect; they suggested temporal integration might underlie the
flash-lag effect and the Fröhlich effect but that the window of
integration began later in the flash-lag effect than in the
Fröhlich effect.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
325
FLASH-LAG EFFECT
Representational Momentum?
The perceived final (offset) position of a moving target is
displaced forward from the actual final position of that target, and
this is referred to as representational momentum (Freyd & Finke,
1984; for review, see Hubbard, 2005). Many studies in the flash-
lag effect literature suggest perceived position of the moving target
is displaced forward, and this seems equivalent to representational
momentum. However, literature on the flash-lag effect usually
does not consider representational momentum or else dismisses
representational momentum based on incomplete or incorrect in-
formation.
7
In a notable exception, Munger and Owens (2004)
presented participants with a rotating bar and a flashed square, and
they found (a) a flash-lag effect occurred with flash-midpoint but
not flash-terminated displays and (b) robust representational mo-
mentum in a flash-terminated display (cf. Müsseler et al., 2002)or
if a flashed object was not presented. Thus, for the same moving
target, typical representational momentum findings and typical
flash-lag effect findings were obtained (see also Shi & de’Sperati,
2008). Munger and Owens suggested robust representational mo-
mentum in a flash-terminated display might reflect a contribution
of a flash-lag effect: If a flashed object aligned with the target is
perceived to be behind the target, participants might be even more
likely to accept a probe in front of the target’s actual location as
reflecting the target’s actual location.
Table 3 lists similarities of representational momentum and the
flash-lag effect. The number of similarities suggests there might be
a common mechanism underlying representational momentum and
the flash-lag effect, or more radically, that the flash-lag effect is an
example of representational momentum in which represented tar-
get location is measured relative to an external stimulus (the
flashed object) rather than relative to actual target location. How-
ever, lack of a flash-lag effect in flash-terminated displays initially
appears inconsistent with representational momentum. Hubbard
(2008) reported displacement in the direction of target motion for
a stationary object briefly presented near the end of target motion,
and this displacement was not significantly different from dis-
placement of the target; thus, it could be speculated that relative
locations of a briefly presented object and a moving target were
veridical (i.e., a flash-lag effect did not occur) even though abso-
lute locations of that briefly presented object and moving target
were displaced forward (i.e., representational momentum oc-
curred). Also, effects of oculomotor behavior on representational
momentum and on the flash-lag effect for continuous motion differ
(e.g., pursuit eye movements increase representational momentum,
Kerzel, 2000, but decrease the flash-lag effect, Nijhawan, 2001),
but such differences do not rule out a common higher level
mechanism (cf. differences in oculomotor behavior on representa-
tional momentum with continuous or implied motion; Hubbard,
2005, 2006b, 2010).
The flash-lag effect and representational momentum both in-
volve forward displacement of a moving target, and the role, if
any, the flashed object plays in displacement of the moving target
in the flash-lag effect is not clear. As forward displacement in
representational momentum occurs without reference to an exter-
nal stimulus, representational momentum offers a more parsimo-
nious explanation for many findings in flash-lag effect literature.
In arguing for motion extrapolation and against differential laten-
cies, Nijhawan et al. (2004, p. 278) stated, “A newer interpretation
of a given phenomenon can be accepted over and above an existing
one only if the newer interpretation is conceptually simpler (re-
quires fewer assumptions) and/or is capable of explaining a wider
class of empirical findings.” Representational momentum is con-
ceptually simpler (involving one stimulus rather than the relation-
ship between two stimuli) and accounts for a wider class of
empirical findings (displacement of moving targets when a flashed
object is presented and when a flashed object is not presented).
Many cases of what has been identified as a flash-lag effect might
result from a combination of representational momentum of the
moving target and less (no) displacement of the flashed object (cf.
Shi & de’Sperati, 2008); assessment of displacement of the mov-
ing target by reference to an external (flashed) object rather than to
the actual target location might just be a different way to measure
representational momentum.
Backward Referral?
Libet (1985; Libet, Gleason, Wright, & Pearl, 1983; Libet,
Wright, & Gleason, 1982) asked participants to move a finger at a
time of their choosing, and participants viewed a clock face in
which a rotating hand was displayed. When participants made a
7
For example, Kanai and Verstraten (2006, p. 453) stated, “When a
flash physically coincides with a continually moving object, the position of
that moving object is perceived to be ahead of the flash. This visual
phenomenon is called the flash-lag effect,” and representational momen-
tum was not identified by name, nor were any articles on representational
momentum cited. Maus and Nijhawan (2009, p. 611) stated, “One hypoth-
esis, termed motion extrapolation, states that moving objects are spatially
shifted forward to counteract the influence of neural delays in the visual
pathways on the perceived position of moving objects.” Nijhawan et al.
(2004, p. 296) stated, “The flash-lag phenomenon is: the position in which
a moving item (visual object or limb) is sensed is not where the item was
in the recent past, but closer to where the item probably is.” These
statements are quite similar to statements characterizing representational
momentum and the purpose of representational momentum. For example,
Finke et al. (1986, p. 176; see also Freyd, 1987) stated that representational
momentum is useful for “anticipating the future positions of objects . . .
[and] contributes to regulation and control of bodily movements,” and
Hubbard (2005, p. 847) suggested representational momentum and related
types of displacement “adjusts the representation of a target to reflect
where that target would (most likely) be at the moment an immediate motor
response from the observer would reach the target” (see also Hubbard,
1995, 2006a, 2006b, and the discussion “Toward a Computational Theory
of Displacement” in Hubbard, 2005). Although Maus and Nijhawan (2006,
p. 4375) admitted “in representational momentum observers perceive the
final position of a moving object as shifted in the direction of motion,” their
investigation of forward displacement of a moving target made no other
reference to or contact with the representational momentum literature, even
though they investigated effects (e.g., velocity) well investigated within the
representational momentum literature. Finally, Maus and Nijhawan (2009,
p. 612) stated, “Although representational momentum . . . states that mov-
ing objects are remembered to disappear beyond their final position, these
findings can be explained by cognitive processes or eye movements and
visual persistence,” but this is not entirely correct. Although Maus and
Nijhawan’s statement might have been aimed at representational momen-
tum that arises from smooth target motion, representational momentum
also arises from implied motion and with single frozen-action photographs,
and so cannot be explained by eye movements or visual persistence (for
discussion, see Hubbard, 2005, 2006b, 2010). Furthermore, it is unclear
why contributions of cognitive processes to representational momentum
are problematic, as ample evidence suggests high-level (cognitive) pro-
cesses contribute to the flash-lag effect (e.g., voluntary attention, Namba &
Baldo, 2004; Shioiri et al., 2010; perceptual organization, Watanabe et al.,
2001; conceptual knowledge, Nagai et al., 2010; Noguchi & Kakigi, 2008).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
326
HUBBARD
decision to move a finger, they noted the time on the clock face.
Libet also collected simultaneous electroencephalograms and ex-
amined the readiness potential. The average reported time of the
decision to move was 200 ms prior to movement, but the readiness
potential appeared up to 500 ms prior to movement. In one
interpretation, a decision to move was made up to 300 ms before
participants had a conscious experience of making a decision to
move, and the time of the subsequent conscious intent was then
“adjusted” backward in time to reflect the time of the unconscious
decision (cf. postdiction). Libet referred to this as backward re-
ferral. Van de Grind (2002) and Klein (2002) suggested apparent
backward referral reflected a flash-lag effect: Participants reported
the position of a rapidly rotating hand on a clock face (the moving
target) at the moment of a brief and transient event (the decision to
move). Similarly, Joordens, Spalek, Razmy, and van Duijn (2004)
suggested apparent backward referral reflected representational
momentum for the clock hand. In explanations of backward refer-
ral based on the flash-lag effect or representational momentum, the
perceived position of the hand on the clock face is displaced
forward, and so the decision to move appeared to occur at a later
time than it actually did.
Anisotropic Distortion?
Watanabe and Yokoi (2006, 2007) suggested the flash-lag effect
reflected anisotropic distortion of two-dimensional spatial repre-
sentation; specifically, the perceived position of the flashed object
was displaced toward a point of convergence that followed the
moving target at a fixed distance. Anisotropic representation of
space is consistent with differences in the flash-lag effect as a
function of perceptual grouping in Watanabe et al. (2001; Wa-
tanabe, 2004), a larger flash-lag effect if targets moved toward
than away from fixation (Shi & Nijhawan, 2008), and a larger
flash-lag effect at the onset than at the offset of a moving flashed
object (Bachmann & Kalev, 1997). Kanai et al. (2004) reported a
larger flash-lag effect for targets in the left than in the right visual
field and for targets in the upper than in the lower visual field, and
Shi and Nijhawan (2008) reported a larger effect of direction
(toward or away from fixation) if motion was in the right than in
the left visual field. It is not clear if anisotropy of visual space is
consistent with a flash-lag effect involving nonspatial visual (e.g.,
Sheth et al., 2000), auditory (Alais & Burr, 2003), or motor
(Nijhawan & Kirschfeld, 2003) stimuli, unless similar anisotropies
occur in representation of nonspatial or nonvisual stimuli. Whether
the flash-lag effect causes or results from these anisotropic distor-
tions is not clear.
Flag Errors in Football?
Baldo, Ranvaud, and Morya (2002) suggested the flash-lag
effect might contribute to a bias toward an offsides flag (i.e., an
offsides penalty) in football (referred to as soccer in North Amer-
ica). In an offside situation, the attacker receiving the ball is
running toward the opponent’s goal and is equivalent to the mov-
ing target in a flash-lag effect. The passing of the ball is a transient
event equivalent to the flashed object in a flash-lag effect. The
attacker is therefore perceived to be ahead of his actual position
when the ball is passed, and so an offsides flag is more likely.
Gilis, Helsen, Catteeuw, and Wagemans (2008) found Fédération
Internationale de Football Association and Belgium Association
assistant referees were more likely to raise an offsides flag in
ambiguous situations, and they suggested this bias was due to a
flash-lag effect. Gilis et al. also pointed out that forward displace-
ment of the attacker is consistent with representational momentum
and that it is not clear whether bias toward an offsides flag is due
to a flash-lag effect or to representational momentum. However,
Table 3
Similarities Between Representational Momentum and the Flash-Lag Effect
Characteristic Representational momentum Flash-lag effect
Increases with increases in target velocity Hubbard & Bharucha (1988); Hubbard (1990) Krekelberg & Lappe (1999); Nijhawan
(1994); Wojtach et al. (2008)
Increases when attention is divided Hayes & Freyd (2002) Sarich et al. (2007); Shioiri et al. (2010)
Is disrupted if continuity of identity of moving
target is disrupted
Kelly & Freyd (1987) Moore & Enns (2004)
Larger effects in the left visual field Halpern & Kelly (1993); White et al. (1993) Kanai et al. (2004)
Participation of cortical area MT/MST Kourtzi & Kanwisher (2000); Senior et al. (2000,
2002)
Maus et al. (2013)
Decreases if preceded by a valid cue
a
Hubbard et al. (2009) Namba & Baldo (2004); Shioiri et al. (2010);
Vreven & Verghese (2005)
Is influenced by conceptual knowledge of the
target
Reed & Vinson (1996); Vinson & Reed (2002) Nagai et al. (2010); Noguchi & Kakigi
(2008)
Decreases if observers have control over the
stimulus
b
Jordan & Knoblich (2004) López-Moliner & Linares (2006)
Is influenced by attributions regarding source
of target motion
Hubbard (2013); Hubbard et al. (2001) Ichikawa & Masakura (2006, 2010)
Increases with motion toward
landmark/fixation
Hubbard & Ruppel (1999) Brenner et al. (2006); Kanai et al. (2004);
Shi & Nijhawan (2008)
a
The data on effects of cuing are mixed. Brenner and Smeets (2000), Hommuk et al. (2008), Namba and Baldo (2004), Shioiri et al. (2010), and Vreven
and Verghese (2005) all found an effect of cuing on the flash-lag effect, but Khurana et al. (2000) did not find an effect of cuing on the flash-lag effect.
On balance, the evidence suggests there is an effect of cuing.
b
If the position of the flash is used as the marker in judging location of the target, then
the decline in the flash-lag effect when participants control the flashed object is similar to the decline in representational momentum if participants control
the moving target. However, control of the moving target has been shown to result in a larger (Scocchia et al., 2009) and smaller (Ichikawa & Masakura,
2006) flash-lag effect, and the relationship between control of the target and the size of the flash-lag effect is not clear.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
327
FLASH-LAG EFFECT
given that position of the attacker is judged relative to an external
transient stimulus (the passed ball), an explanation based on the
flash-lag effect might be more appropriate, although it is possible
there is (also) representational momentum for the position of the
attacker.
Catteeuw, Helsen, Gilis, van Roie, and Wagemans (2009) sug-
gested experienced assistant referees partially compensated for the
flash-lag effect. Catteeuw, Gilis, Wagemans, and Helsen (2010)
reported a bias toward an offsides flag consistent with the flash-lag
effect in less successful assistant referees, and they reported train-
ing and feedback improved accuracy in play calling but not accu-
racy in memory for player positions. The improved accuracy in
play calling suggests the flash-lag effect might be influenced by
experience or expertise. Catteeuw, Gilis, Jaspers, Wagemans, and
Helsen (2010) reported simulations and computer animations were
equally effective in training to compensate for the flash-lag effect.
In such training studies, it would have been useful to examine
whether the flash-lag effect (and perhaps representational momen-
tum) also decreased on more traditional measures or whether
learning was more stimulus specific, but such comparisons await
future studies. Experimental design and analysis in Catteeuw,
Gilis, Jaspers, et al. were actually more consistent with measures
of representational momentum (cf. Thornton & Hayes, 2004) than
with measures of the flash-lag effect, and it is possible displace-
ment in Catteeuw, Gilis, Jaspers, et al. was due to representational
momentum rather than to a flash-lag effect (cf. Gilis et al., 2008).
Part 3: Theories of the Flash-Lag Effect
Despite many years of study and debate, there is little consensus
regarding potential mechanisms of the flash-lag effect. Theories
considered here include (a) motion extrapolation, (b) attention
shift, (c) visible persistence, (d) attention and metacontrast, (e)
differential latency, (f) temporal integration, (g) postdiction, (h)
temporal sampling, (i) perceptual acceleration, (j) frame of refer-
ence, and (k) misbinding. The major theories and evidence incon-
sistent with each of the major theories are summarized in Table 4.
Motion Extrapolation
Nijhawan (1994) suggested the flash-lag effect occurs because
the visual system extrapolates the trajectory of a moving object to
compensate for delays in perception due to neural processing times
(cf. representational momentum occurs because of “a natural ten-
dency to mentally extrapolate implied motion into the future”;
Finke, Freyd, & Shyi, 1986, p. 176). Without such compensation,
perceived position of a target would lag actual position. This
extrapolation is based on previous motion of the target and biases
perceived position forward to correspond to the target’s position in
real time. The flashed object has no such history, and so it is not
extrapolated. As a result, the neural delay in processing the flashed
object results in the flashed object appearing to lag the forward-
extrapolated moving target. Effects of target velocity (Nijhawan,
1994) and spatial facilitation by a nearby stimulus (Maiche et al.,
Table 4
Major Theories of the Flash-Lag Effect
Theory Explanation of flash-lag effect Inconsistent evidence
Motion extrapolation Trajectory of a moving target is extrapolated
forward.
Flash-lag effect occurs with random motion or unpredictable
changes in direction or velocity, presence of flash-lag effect in
flash-initiated displays, lack of flash-lag effect in flash-
terminated displays
Attention shift Shift of attention to the flashed object takes
time, during which target continues to move.
Occurrence of flash-lag effect if moving target and flashed
objects are interleaved, occurrence of flash-lag in effect in
flash-initiated displays
Visual persistence Visual persistence of a moving target is
decreased because of processes that deblur
that image.
Masking of flashed object does not influence flash-lag effect,
effects of conceptual knowledge or perceptual grouping,
auditory and cross-modal flash-lag effects, effects of control
Attention and metacontrast Increased activation near leading edge of target
and metacontrast masking of previous
locations shifts the representation of the target
forward.
Existence of flash-lag effect in flash-initiated displays, effects of
conceptual knowledge or perceptual grouping, effects of control
Differential latency Processing time for moving targets is less than
for flashed objects, and by the time a flashed
object enters awareness, the target is
perceived as further along its trajectory.
Response times to moving targets or flashed objects do not differ,
temporal order judgments suggest flashed objects might be
processed faster than moving targets, effects of knowledge or
perceptual grouping, auditory and cross-modal flash-lag effects,
occurrence of flash-lag effect in flash-initiated displays, flash-
lag effect with moving flashed objects
Postdiction Appearance of the flashed object resets temporal
integration, and representation of moving
target position reflects an integration of target
locations after the flashed object was
presented.
Effect of preflash target behavior, effect of preflash cues, flash-
lag effect with moving flashed objects
Perceptual acceleration The initial stimuli in a perceptual stream are
processed more slowly than subsequent
stimuli. Moving target is presented later in its
respective stream, and so it is processed more
quickly.
Flash-lag effect in flash-initiated displays, lack of a flash-lag
effect in flash-terminated displays
Note. Recent theories that have not been addressed or evaluated by researchers or laboratories other than those that initially proposed those theories (frame
of reference, misbinding) are not included, as there is insufficient evidence to evaluate those theories.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
328
HUBBARD
2007) seem consistent with motion extrapolation theory. Nijhawan
and colleagues have held to motion extrapolation theory (e.g.,
Maus & Nijhawan, 2008; Nijhawan, 2008; Nijhawan et al., 2004)
despite numerous findings that appear inconsistent with this theory
(e.g., Brenner & Smeets, 2000; Eagleman & Sejnowski, 2000b;
Murakami, 2001b; Whitney, Murakami, & Cavanagh, 2000). Ni-
jhawan (2008) now refers to motion extrapolation as visual pre-
diction, which is curious in light of previous findings of a flash-lag
effect with auditory stimuli and with motor stimuli.
Occurrence of a flash-lag effect with random motion (Mu-
rakami, 2001b), or with unpredictable changes in direction (Eagle-
man & Sejnowski, 2000b; Whitney, Cavanagh, & Murakami,
2000; Whitney & Murakami, 1998; Whitney, Murakami, & Ca-
vanagh, 2000) or velocity (Brenner & Smeets, 2000), is inconsis-
tent with motion extrapolation; motion of a target that moved
randomly or unpredictability would not (by definition) be predict-
able and thus could not be extrapolated. Occurrence of a flash-lag
effect in flash-initiated displays, and lack of a flash-lag effect in
flash-terminated displays, are also problematic (e.g., Eagleman &
Sejnowski, 2000b). Nijhawan et al. (2004; Nijhawan, 2008) sug-
gested motion extrapolation in flash-initiated displays is based on
motion of the target during the interval between presentation and
perception of the flash: Because visual awareness of a flashed
object does not occur for at least 100 ms after the flashed object is
presented, the observer has at least 100 ms worth of trajectory to
use in extrapolation. Nijhawan (2008; Maus & Nijhawan, 2006)
suggested the lack of a flash-lag effect in flash-terminated displays
reflects a correction-for-extrapolation based on offset transients.
Lankheet and van de Grind (2010) suggested lack of a flash-lag
effect in flash-terminated displays is not evidence against motion
extrapolation in other types of displays; rather, motion just is not
extrapolated if the target stops. However, this suggestion does not
appear consistent with the large literature on representational mo-
mentum.
Attention Shift
Attention shift theory suggests the flash-lag effect occurs be-
cause attention is initially focused on the moving target, and when
the flashed object is presented, attention is shifted to the flashed
object. This shift takes time, and during this time, the target
continues to move. By the time the flashed object is perceived, the
target has already moved some distance, and so the flashed object
is perceived as lagging the moving target. Effects of cuing (Namba
& Baldo, 2004), target velocity (Nijhawan, 1994), and predictabil-
ity of the flashed object (Baldo & Namba, 2002; Vreven &
Verghese, 2005) are consistent with attention shift theory. Baldo
and Klein (1995) suggested differences in allocation of attention
caused the flash-lag effect, but Baldo and colleagues (Baldo,
Kihara, et al., 2002; Baldo & Klein, 2010; Namba & Baldo, 2004)
have subsequently taken a more conservative view that distribution
and allocation of attention modulate, but probably do not cause,
the flash-lag effect. The existence of a flash-lag effect if flashed
objects and moving targets are interleaved (Khurana & Nijhawan,
1995), and the lack of an effect of spatial cuing and cue validity in
Khurana et al. (2000; but see Baldo, Kihara, et al., 2002), are not
consistent with attention shift theory. The existence of a flash-lag
effect in flash-initiated displays (Nijhawan et al., 2004; Rizk et al.,
2009; Sheth et al., 2000) is not consistent with attention shift
theory, as there should be no difference between onset of the
moving target and onset of the flashed object.
Visual Persistence
Visual persistence of a moving target might be decreased be-
cause of processes that deblur a changing image (Burr, 1980),
whereas visual persistence of a flashed object would not be sim-
ilarly decreased. Thus, visual persistence of a flashed object would
last longer than visual persistence of a moving target (Efron, 1970;
Hogben & Di Lollo, 1974), and this could result in a flashed object
appearing to lag a moving target (Walker & Irion, 1982). Gauch
and Kerzel (2009) suggested visual persistence, in conjunction
with perceptual set, could account for most instances of the flash-
lag effect. Whitney, Murakami, and Cavanagh (2000) presented a
mask at the position of the flashed object immediately after the
flashed object was presented. If visual persistence contributed to
the flash-lag effect, then the mask should have influenced the
flash-lag effect; however, the mask did not influence the flash-lag
effect. Watanabe et al. (2001; Watanabe, 2004) suggested effects
of perceptual grouping are not consistent with visual persistence.
Visible persistence does not appear consistent with effects of
conceptual knowledge on the flash-lag effect (Nagai et al., 2010;
Noguchi & Kakigi, 2008) or with flash-lag effects involving au-
ditory (Alais & Burr, 2003) or motor (Nijhawan & Kirschfeld,
2003) stimuli.
Attention and Metacontrast
Kirschfeld and Kammer (1999) suggested a moving target acts
as a cue to its next position and metacontrast masking suppresses
activation for previous positions of the moving target (cf. Erlha-
gen, 2003). Increased activation toward the leading edge of the
target, coupled with suppression of activation for previous posi-
tions, could result in forward displacement of the moving target
(see also Baldo & Caticha, 2005; Erlhagen & Jancke, 2004;
Hubbard, 1995, 2005, 2006a, 2008; Jancke & Erlhagen, 2010;
Müsseler et al., 2002). Although attention and metacontrast theory
focused on the Fröhlich effect, Kirschfeld and Kammer discussed
how such a framework could account for findings of Nijhawan
(1994), and Kirschfeld (2006) discussed how such a framework
could account for findings of Kanai et al. (2004). It is not initially
clear how such a framework could account for the lack of a
flash-lag effect in flash-terminated displays (as increased activa-
tion of the leading edge and suppression of previous positions
should result in forward displacement of the target), but Kirschfeld
suggested signals involving attention and metacontrast do not
induce perception per se and so are not detected unless another
stimulus is presented. It is not clear how such a framework could
account for the flash-lag effect in flash-initiated displays (as there
would not yet be increased activation of the leading edge or
suppression of previous positions of the target).
Differential Latency
Metzger (1932) was among the first to suggest a flashed object
aligned with a moving target appears to lag the moving target
because the flashed object takes longer to be perceived. Differen-
tial latency theory suggests a moving target is processed more
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
329
FLASH-LAG EFFECT
quickly than a stationary flashed object, and so the moving target
reaches perceptual awareness more quickly. Thus, if a moving
target and a flashed object were aligned, information regarding the
moving target would reach perceptual awareness more quickly
than information regarding the flashed object; as a result, when the
flashed object entered awareness, the moving target would already
be perceived as further along its trajectory (Ög
˘
men et al., 2004;
Purushothaman et al., 1998; Whitney & Murakami, 1998; Whit-
ney, Murakami, & Cavanagh, 2000). A more developed form of
differential latency theory is due to Ög
˘
men et al. (2004; see also
Kafaligönül, Patel, Ög
˘
men, Bedell, & Purushothaman, 2010) and
referred to as a multichannel differential latency model. This
model suggests the (a) flashed object and moving target are pro-
cessed by different but interacting neural systems (i.e., channels),
(b) latency of each neural system depends upon intrinsic dynamic
properties of that system and on attributes of the stimulus, and (c)
computation of stimulus position and visibility are different pro-
cesses with different dynamics (cf. Maus & Nijhawan, 2006,
2009).
Findings that response time to presentation of a flashed object
does not differ from response time to presentation of a moving
target (Chappell et al., 2006; Cravo & Baldo, 2008; Fouriezos et
al., 2007), that a flashed object is processed more quickly than a
moving target (Nijhawan et al., 2004; Raiguel et al., 1989), and
that perceptual grouping (Watanabe, 2004; Watanabe et al., 2001)
and semantic meaningfulness (Nagai et al., 2010; Noguchi &
Kakigi, 2008) influence the flash-lag effect are not consistent with
differential latency theory. Finding a flash-lag effect with se-
quences of letters (Bachmann et al., 2003) appears inconsistent
with differential latency theory, as no perceived motion per se
occurred. A flash-lag effect at the end of motion of a moving
flashed object (e.g., Bachmann & Kalev, 1997) appears inconsis-
tent with differential latency theory, because processing of the
moving target and flashed object should both be facilitated. Becker
et al. (2009) suggested use of a black annulus as a moving target and a
white disk as a flashed object by many researchers is not consistent
with differential latency theory, as the flashed object would exhibit
greater luminance than the moving target and result in a flash-lead
effect if mislocalization was due to differences in processing
latencies. Differences in latencies cannot account for a flash-lag
effect with visual–auditory cross-modal flash-lag stimuli (Alais &
Burr, 2003; Arrighi et al., 2005).
Temporal Integration
Temporal integration theories, also referred to as position inte-
gration theories, of the flash-lag effect suggest that perceived
position of a stimulus is based on an integration of the positions
that stimulus occupies over some period of time. If positions of a
moving target and a flashed object are compared, their relative
positions are estimated by temporal averaging across persisting
position signals for each stimulus (Krekelberg, 2001; Krekelberg
& Lappe, 1999, 2000a, 2000b, 2001). For the flashed object, the
averaged position is the same as its actual position. However, for
the moving target, the averaged position is ahead of where the
target was when the flashed object appeared, because the average
includes not only the position of the target when the flashed object
appeared but also subsequent positions of the target. Chappell et al.
(2006) suggested temporal integration is consistent with the flash-
lag effect and the Fröhlich effect. Rizk et al. (2009) suggested a
flash-lag effect with sampled motion involves temporal integration
(across “stations”). Shen et al. (2007) pointed out position inte-
gration theory predicts participants should not accurately perceive
the reversal location of a moving target, and they found perceived
reversal location occurred before actual reversal location (cf. Whit-
ney, Murakami, & Cavanagh, 2000). Shen et al. suggested infor-
mation regarding actual reversal location was potentially available
but not integrated, and they rejected position integration theory.
8
Postdiction
Postdiction theory of Eagleman and Sejnowski (2000b; Rao,
Eagleman, & Sejnowski, 2001) is a special case of temporal
integration. Eagleman and Sejnowski presented a moving target
that after presentation of the flashed object continued in the same
direction of motion, stopped, or reversed. A flash-lag effect oc-
curred if the target continued in the same direction, but not if the
target stopped, and if the target reversed, mislocalization was in
the opposite direction (i.e., a flash-lead effect occurred). Preflash
target behavior was the same across conditions, and Eagleman and
Sejnowski argued the flash-lag effect depended on events follow-
ing presentation of the flashed object and that the flash “reset”
integration regarding target position. Use of information (i.e.,
target positions) presented after the flashed object to “adjust”
perception of the target at the time the flashed object was presented
was referred to as postdiction (cf. backward referral). Eagleman
and Sejnowski suggested abrupt onset of a moving target acted like
a flashed object, and so postdiction could also account for the
Fröhlich effect. Eagleman and Sejnowski (2007) updated this
approach to include other types of flash mislocalization, and this
updated approach was referred to as a motion-biasing model.
Brenner and Smeets (2000) and Rotman et al. (2005) similarly
suggested target localization depended upon information available
after presentation of the flashed object.
Krekelberg and Lappe (2000b) suggested temporal integration
of motion signals from the target over a range of 500 ms can
account for the observed data without a need to reset integration of
motion signals and that without a reset, there is no need for
postdiction. Whitney and Cavanagh (2000) suggested if a flashed
object resets motion signals, then a moving target accompanied by
a flashed object presented every 80 ms should be invisible along its
entire trajectory, but such a result has not been reported. Postdic-
8
Roulston, Self, and Zeki (2006) also discussed a position integration
theory of localization. In their experiments, they reported a moving target
was displaced behind a stationary flashed object (a flash-lead effect) and
that two moving targets that approached each other were displaced back-
ward (which they referred to as reverse-repmo). These displacements are in
the direction opposite to displacements typically obtained in studies on the
flash-lag effect and in studies on representational momentum, and the
reasons for these differences are not clear. One possible explanation
involves the time course of displacement. Freyd and Johnson (1987)
reported forward displacement peaked after a few hundred milliseconds
and then decreased, and they attributed this pattern to two distinct pro-
cesses: an initial forward extrapolation process that displaced represented
location in the direction of target motion (representational momentum) and
a subsequent memory averaging process that displaced represented loca-
tion toward an average of the stimulus locations. Depending upon the
latency of judgment in Roulston et al. (which was not reported), the
apparent reverse-repmo might reflect this subsequent memory averaging.
This remains an issue for further research.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
330
HUBBARD
tion theory suggests cues provided prior to presentation of the
flashed object should not influence mislocalization, but effects of
preflash cuing (e.g., Baldo, Kihara, et al., 2002; Namba & Baldo,
2004; Vreven & Verghese, 2005; but see Khurana et al., 2000;
Whitney & Cavanagh, 2000) and other preflash information or
experience (e.g., Chappell & Hine, 2004) have been reported.
Patel, Ög
˘
men, Bedell, and Sampath (2000) pointed out postdiction
cannot account for a flash-lead effect. Eagleman and Sejnowski
(2000a, 2000c) responded to these types of criticisms by suggest-
ing reset did not necessarily disregard all preflash information;
rather, the extent to which reset occurred was related to the amount
of “surprise” in the stimulus, with greater surprise resulting in
greater reset and less use of preflash information.
Temporal Sampling
Brenner and Smeets (2000) suggested the visual system has to
select a moment at which to “sample” the moving target’s position,
and Brenner et al. (2006) suggested a moving target’s position
would be judged only if a specific moment of interest is specified.
In the case of the flash-lag effect, this moment is specified by
presentation of the flashed object, and the time required to initiate
this sampling results in the actual sampled position corresponding
to a later moment in time than when the flashed object was
presented. Effects of cuing (e.g., Brenner & Smeets, 2000; Namba
& Baldo, 2004; but see Khurana et al., 2000) are consistent with
temporal sampling, as cuing would decrease the time required to
initiate sampling and so decrease the flash-lag effect. The temporal
sampling theory is similar to theories of Krekelberg and Lappe
(2000a, 2000b) and Eagleman and Sejnowski (2000b), but sam-
pling in Brenner and Smeets involves a single time or position,
whereas sampling in Krekelberg and Lappe or in Eagleman and
Sejnowski involves integrating multiple times or positions. Per-
haps not surprisingly, many objections that apply to Krekelberg
and Lappe’s theory and to Eagleman and Sejnowski’s theory also
apply to temporal sampling theory (e.g., it is not clear how infor-
mation presented prior to the flashed object could influence a
sample of target motion from after the flashed object; e.g., Chap-
pell & Hine, 2004).
Perceptual Acceleration
Bachmann et al. (2003) suggested a moving target and a flashed
object in a flash-lag effect are in different perceptual streams.
When a perceptual stream first appears, latency to explicit (con-
scious) perception of the initial stimuli in that stream is relatively
long, but this latency decreases for subsequent stimuli. In other
words, when the moving target stream first appears, the first few
stimuli are processed relatively slowly, and subsequent stimuli are
processed more quickly. This is referred to as perceptual acceler-
ation. If a flashed object is presented in a second stream that begins
after the onset of the moving target stream, that flashed object
would be processed more slowly than a simultaneous stimulus in
the moving target stream, and so a flash-lag effect occurs. This
appears similar to differential latency theory, but differential la-
tency theory specifies processing time as a function of stimulus
type and not as a function of the amount of preceding stimuli. It is
not clear whether perceptual acceleration might account for a
flash-lag effect if the moving target and flashed object spatially
overlapped (e.g., a flashed disk in a moving annulus). Perceptual
acceleration theory does not seem consistent with a flash-initiated
flash-lag effect (as processing speed is presumably the same for
onset of the motion stream and for onset of the flashed stream; but
see Bachmann, 2007, 2010, on perceptual “retouch”) or consistent
with the lack of a flash-lag effect in flash-terminated displays.
Frame of Reference
Becker et al. (2009) suggested the flash-lag effect arises because
a saccade to the flashed object is programmed with egocentric
encoding, whereas a saccade to the moving target is programmed
relative to the location of the flashed object and with allocentric
encoding. That is, the visual system first computes the location of
the flashed object and, using that information as a reference or
anchor, then computes the location of the moving target (cf. van de
Grind, 2008; a moving target is a reference for a stationary object).
By the time the visual system computes the location of the moving
target, that target has moved beyond its location at the time the
flashed object was presented. If the flashed object is not spatially
localized (i.e., if the entire background flashed), then it is not used
as a reference or anchor, and no flash-lag effect occurs. Thus,
differences in encoding related to the frame of reference might
account for mislocalization of the moving target. However, it
seems possible that the sequence, rather than the encoding format,
is more critical (i.e., in Becker et al., 2009, the flashed object is
always encoded before the moving target is encoded).
9
Also, it is
not clear whether differences between allocentric encoding and
egocentric encoding could account for a flash-lag effect in non-
spatial visual stimuli (e.g., Sheth et al., 2000) or nonvisual stimuli
(e.g., Alais & Burr, 2003).
Misbinding
Gauch and Kerzel (2008a) reported a larger flash-lag effect
occurred with a moving flashed object than with a stationary
flashed object in flash-initiated or flash-terminated displays, but no
differences in the flash-lag effect occurred as a function of whether
the flashed object was moving or stationary in flash-midpoint
displays. Thus, moving flashed objects produced a larger flash-lag
effect if there was an abrupt change (e.g., onset, offset) in the
target than if there was a continuous change in the target. This
difference might involve differences between transient and sus-
tained channels (cf. Maus & Nijhawan, 2006, 2009) that underlie
processing of abrupt and continuous changes, respectively. Gauch
and Kerzel suggested an abrupt change is (mis)bound to a contin-
uous change that follows abrupt change, that onset of a moving
target presents an abrupt change akin to a flash, and that asyn-
chronous binding of abrupt and continuous changes produces a
flash-lag effect. Alternatively, Sheth et al. (2000) suggested abrupt
and continuous changes differ because of masking or priming from
previous target presentations and because of attention shifts to the
9
An interesting test of this hypothesis would be to vary instructions
regarding whether the position of the moving target should be compared to
the position of the flashed object or the position of the flashed object
should be compared to the position of the moving target, and to examine
whether there is an effect of instruction on the flash-lag effect or eye
movement patterns.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
331
FLASH-LAG EFFECT
flashed object. A potential temporal misbinding was also ad-
dressed by van de Grind (2002), who suggested differences in
latencies are experienced as differences in space rather than as
differences in time.
Part 4: Summary and Conclusions
The flash-lag effect is robust and occurs with many variations
in variables and stimuli. A flash-lag effect occurs with linear or
curvilinear motion and if the flashed object spatially overlaps or
is spatially separated from the moving target. A flash-lag effect
occurs with flash-initiated or flash-midpoint displays, but does
not usually occur with flash-terminated displays. A flash-lag
effect occurs with continuous, sampled, or random motion, and
the flashed object can be stationary or moving. The flash-lag
effect occurs with changes in (a) spatial location, (b) luminance,
(c) color, (d) spatial frequency, (e) pattern entropy, (f) binocular
disparity, (g) auditory frequency, and (h) body position. The
flash-lag effect occurs (a) with visual movement in the picture
plane or in depth, (b) with auditory movement in frequency or
in space, and (c) cross-modally with visual, auditory, and motor
stimuli. The flash-lag effect is larger if (a) distance between the
moving target and the flashed object is larger, (b) eccentricity
of the moving target or flashed object is larger, (c) target
motion is toward fixation, (d) target velocity is faster, (e)
presentation of the flashed object is less predictable, (f) target
position is less certain, (g) participants divide attention over
multiple targets or tasks, (h) a nearby stimulus facilitates pro-
cessing along the path of motion, (i) participants do not control
presentation of the flashed object, (j) the flashed object is
aligned with the leading edge of the moving target, (k) mea-
sured at the onset of flashed object motion, and (l) target size
remains constant.
Several unresolved empirical issues remain. There are incon-
sistent findings regarding effects on the flash-lag effect of (a)
preflash target behavior, (b) whether target motion is in the left
or right visual field, (c) duration or distance of target motion,
and (d) participant control of target motion. Dissociation of
spatial aspects and temporal aspects of the flash-lag effect, and
potential differences between types of stimulus displays (e.g.,
ease in distributing attention, susceptibility to masking), are not
clear. Although most studies reported data consistent with or
requiring high-level processes in the flash-lag effect, data from
a few studies are consistent with operation of only low-level
processes in the flash-lag effect. It is possible a flash-lag effect
might be created in low-level structures by top-down processes
based on previous experience or might arise in low-level struc-
tures and be modified by top-down connections if relevant
information is available. The relationship of the flash-lag effect
to representational momentum and to the Fröhlich effect is not
clear, and it is possible that many examples of what has been
considered a flash-lag effect are examples of representational
momentum or a Fröhlich effect in which the represented posi-
tion of a moving target is measured relative to the perceived
position of an external stimulus (the flashed object) rather than
relative to the actual final or initial position of that moving
target, respectively. Also, principles specifying whether mislo-
calization related to a flashed object involves a perceived lag,
lead, or drag are not clear.
Several theories of the flash-lag effect have been proposed,
and many appear to suggest that the flash-lag effect results from
a single mechanism. This has led to polarization among some
researchers and to overly strong claims; indeed, titles of many
articles bluntly state some single mechanism is or is not re-
sponsible for the flash-lag effect (e.g., Arrighi et al., 2005;
Becker et al., 2009; Brenner & Smeets, 2000; Shen et al., 2007;
Whitney, Murakami, & Cavanagh, 2000). There is significant
disconfirming evidence for each of the major and well-known
theories, and more recent theories have not yet been thoroughly
tested. Some theories of the flash-lag effect seem to assume the
correct level of explanation involves the relationship between
the moving target and the flashed object, but as noted earlier,
the flash-lag effect appears to arise from or be dependent upon
more basic mislocalizations of the moving target and/or flashed
object. An explanation that focuses on these more basic mislo-
calizations might be more appropriate. Also, it would not be
surprising if multiple mechanisms contributed to the flash-lag
effect (e.g., neither attention shifts nor differential latencies
solely account for the flash-lag effect, but each appears capable
of modulating the flash-lag effect). There is no reason why the
proposed mechanisms must be mutually exclusive; rather, dif-
ferent mechanisms might contribute in different ways to differ-
ent examples of the flash-lag effect.
In the flash-lag effect, a briefly presented object spatially
aligned with a moving target is perceived to lag that target. This
observation has attracted the attention and efforts of numerous
researchers and laboratories, and research on the flash-lag ef-
fect has highlighted a fundamental question regarding percep-
tual, cognitive, and motor functioning: How do we compensate
for neural processing times so that we can successfully interact
with dynamic objects in real time? The answer to this question
has significant theoretical and empirical implications, not just
for understanding the flash-lag effect, but for understanding
related phenomena such as the Fröhlich effect, representational
momentum, backward referral, and anisotropic distortions in
spatial representation. Moving stimuli are often more salient
than stationary stimuli, and a representational system might be
more useful if it was preferentially tuned to anticipate (and thus
more accurately represent and respond to) behaviors of moving
stimuli (cf. van de Grind, 2008; discussion of computational
theory in Hubbard, 2005). The flash-lag effect and related
mislocalizations might be one consequence of this tuning, and
given that a moving stimulus is probably more likely to impact
an observer than is a stationary stimulus, such a tuning would
presumably be adaptive. Despite its apparent simplicity, the
flash-lag effect reveals important properties of our representa-
tional system and provides insight into how our representational
system is adapted to allow accurate perception of and interac-
tion with stimuli in the environment.
References
Aglioti, S., DeSouza, J. F. X., & Goodale, M. A. (1995). Size-contrast
illusions deceive the eye but not the hand. Current Biology, 5, 679–685.
doi:10.1016/S0960-9822(95)00133-3
Alais, D., & Burr, D. (2003). The “flash-lag” effect occurs in audition and
cross-modally. Current Biology, 13, 59– 63. doi:10.1016/S0960-
9822(02)01402-1
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
332
HUBBARD
Anstis, S. (2007). The flash-lag effect during illusory chopstick motion.
Perception, 36, 1043–1048. doi:10.1068/p5653
Anstis, S. (2010). Illusions of time, space, and motion: Flash-lag meets
chopsticks and reversed phi. In R. Nijhawan & B. Khurana (Eds.),
Space and time in perception and action (pp. 408– 421). New York,
NY: Cambridge University Press. doi:10.1017/CBO9780511750540
.024
Arnold, D. H., Durant, S., & Johnston, A. (2003). Latency differences and
the flash-lag effect. Vision Research, 43, 1829 –1835. doi:10.1016/
S0042-6989(03)00281-5
Arnold, D. H., Ong, Y., & Roseboom, W. (2009). Simple differential
latencies modulate, but do not cause the flash-lag effect. Journal of
Vision, 9(5), 4. doi:10.1167/9.5.4
Arrighi, R., Alais, D., & Burr, D. (2005). Neural latencies do not explain
the auditory and audio-visual flash-lag effect. Vision Research, 45,
2917–2925. doi:10.1016/j.visres.2004.09.020
Bachmann, T. (2007). Binding binding: Departure points for a different
version of the perceptual retouch theory. Advances in Cognitive Psy-
chology, 3, 41–55. doi:10.2478/v10053-008-0013-4
Bachmann, T. (2010). Priming and retouch in flash-lag and other phe-
nomena of the streaming perceptual input. In R. Nijhawan & B.
Khurana (Eds.), Space and time in perception and action (pp. 536–557).
New York, NY: Cambridge University Press. doi:10.1017/
CBO9780511750540.030
Bachmann, T., & Kalev, K. (1997). Adjustment of collinearity of laterally
moving, vertically separated lines reveals compression of subjective
distance as a function of aperture size and speed of motion. Perception,
26, 119–120. doi:10.1068/v970218
Bachmann, T., Luiga, I., Põder, E., & Kalev, K. (2003). Perceptual accel-
eration of objects in stream: Evidence from flash-lag displays. Con-
sciousness and Cognition, 12, 279 –297. doi:10.1016/S1053-
8100(02)00067-3
Bachmann, T., & Põder, E. (2001). Change in feature space is not neces-
sary for the flash-lag effect. Vision Research, 41, 1103–1106. doi:
10.1016/S0042-6989(01)00003-7
Baldo, M. V. C., & Caticha, N. (2005). Computational neurobiology of the
flash-lag effect. Vision Research, 45, 2620–2630. doi:10.1016/j.visres
.2005.04.014
Baldo, M. V. C., Kihara, A. H., Namba, J., & Klein, S. A. (2002). Evidence
for an attentional component of the perceptual misalignment between
moving and flashing stimuli. Perception, 31, 17–30. doi:10.1068/p3302
Baldo, M. V. C., & Klein, S. A. (1995). Extrapolation or attention shift?
Nature, 378, 565–566. doi:10.1038/378565a0
Baldo, M. V. C., & Klein, S. A. (2010). Paying attention to the flash-lag
effect. In R. Nijhawan & B. Khurana (Eds.), Space and time in percep-
tion and action (pp. 396–407). New York, NY: Cambridge University
Press. doi:10.1017/CBO9780511750540.023
Baldo, M. V. C., & Namba, J. (2002). The attentional modulation of the
flash-lag effect. Brazilian Journal of Medical and Biological Research,
35, 969–972. doi:10.1590/S0100-879X2002000800014
Baldo, M. V. C., Ranvaud, R. D., & Morya, E. (2002). Flag errors in soccer
games: The flash-lag effect brought to real life. Perception, 31, 1205–
1210. doi:10.1068/p3422
Becker, S. I., Ansorge, U., & Turatto, M. (2009). Saccades reveal that
allocentric coding of the moving object causes mislocalization in the
flash-lag effect. Attention, Perception, & Psychophysics, 71, 1313–1324.
doi:10.3758/APP.71.6.1313
Blohm, G., Missal, M., & Lefèvre, P. (2003). Smooth anticipatory eye
movements alter memorized position of flashed targets. Journal of
Vision, 3(11), 10. doi:10.1167/3.11.10
Brenner, E., & Smeets, J. B. J. (2000). Motion extrapolation is not respon-
sible for the flash-lag effect. Vision Research, 40, 1645–1648. doi:
10.1016/S0042-6989(00)00067-5
Brenner, E., van Beers, R. J., Rotman, G., & Smeets, J. B. J. (2006). The
role of uncertainty in the systematic spatial mislocalization of moving
objects. Journal of Experimental Psychology: Human Perception and
Performance, 32, 811–825. doi:10.1037/0096-1523.32.4.811
Burr, D. (1980). Motion smear. Nature, 284, 164–165. doi:10.1038/
284164a0
Cai, R. H., Jacobson, K., Baloh, R., Schlag-Rey, M., & Schlag, J. (2000).
Vestibular signals can distort the perceived spatial relationship of retinal
stimuli. Experimental Brain Research, 135, 275–278. doi:10.1007/
s002210000549
Cai, R., & Schlag, J. (2001). A new form of illusory conjunction between
color and shape. Journal of Vision, 1(3), 127. doi:10.1167/1.3.127
Cantor, C. R. L., & Schor, C. M. (2007). Stimulus dependence of the
flash-lag effect. Vision Research, 47, 2841–2854. doi:10.1016/j.visres
.2007.06.023
Carbone, E., & Ansorge, U. (2008). Investigating the contribution of
metacontrast to the Fröhlich effect for size. Acta Psychologica, 128,
361–367. doi:10.1016/j.actpsy.2008.03.008
Catteeuw, P., Gilis, B., Jaspers, A., Wagemans, J., & Helsen, W. (2010).
Training of perceptual–cognitive skills in offside decision making. Jour-
nal of Sport & Exercise Psychology, 32, 845–861.
Catteeuw, P., Gilis, B., Wagemans, J., & Helsen, W. (2010). Perceptual-
cognitive skills in offside decision making: Expertise and training ef-
fects. Journal of Sport & Exercise Psychology, 32, 828– 844.
Catteeuw, P., Helsen, W., Gilis, B., van Roie, E., & Wagemans, J. (2009).
Visual scan patterns and decision-making skills of expert assistant
referees in offside situations. Journal of Sport & Exercise Psychology,
31, 786–797.
Chappell, M., & Hine, T. J. (2004). Events before the flash do influence the
flash-lag magnitude. Vision Research, 44, 235–239. doi:10.1016/j.visres
.2003.09.021
Chappell, M., Hine, T. J., Acworth, C., & Hardwick, D. R. (2006).
Attention “capture” by the flash-lag flash. Vision Research, 46, 3205–
3213. doi:10.1016/j.visres.2006.04.017
Chappell, M., & Mullen, K. T. (2010). The magnocellular visual pathway
and the flash-lag illusion. Journal of Vision, 10(11), 24. doi:10.1167/10
.11.24
Collyer, C. E. (1977). Discrimination of spatial and temporal intervals
defined by three light flashes: Effect of spacing on temporal judgments
and of timing on spatial judgments. Perception & Psychophysics, 21,
357–364. doi:10.3758/BF03199487
Cravo, A. M., & Baldo, M. V. C. (2008). A psychophysical and compu-
tational analysis of the spatio-temporal mechanisms underlying the
flash-lag effect. Perception, 37, 1850 –1866. doi:10.1068/p6053
De Valois, R. L., & De Valois, K. K. (1991). Vernier acuity with stationary
moving Gabors. Vision Research, 31, 1619–1626. doi:10.1016/0042-
6989(91)90138-U
Durant, S., & Johnston, A. (2004). Temporal dependence of local motion
induced shifts in perceived position. Vision Research, 44, 357–366.
doi:10.1016/j.visres.2003.09.022
Eagleman, D. M., & Sejnowski, T. J. (2000a). Flash-lag effect: Differential
latency, not postdiction: Response. Science, 290, 1051a. doi:10.1126/
science.290.5494.1051a
Eagleman, D. M., & Sejnowski, T. J. (2000b). Motion integration and
postdiction in visual awareness. Science, 287, 2036 –2038. doi:10.1126/
science.287.5460.2036
Eagleman, D. M., & Sejnowski, T. J. (2000c). The position of moving
objects: Response. Science, 289, 1107a. doi:10.1126/science.289.5482
.1107a
Eagleman, D. M., & Sejnowski, T. J. (2002). Untangling spatial from
temporal illusions. Trends in Neurosciences, 25, 293. doi:10.1016/
S0166-2236(02)02179-3
Eagleman, D. M., & Sejnowski, T. J. (2007). Motion signals bias local-
ization judgments: A unified explanation for the flash-lag, flash-drag,
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
333
FLASH-LAG EFFECT
flash-jump, and Fröhlich illusions. Journal of Vision, 7(4), 3. doi:
10.1167/7.4.3
Efron, R. (1970). The minimum duration of a perception. Neuropsycholo-
gia, 8, 57–63. doi:10.1016/0028-3932(70)90025-4
Erlhagen, W. (2003). Internal models for visual perception. Biological
Cybernetics, 88, 409–417. doi:10.1007/s00422-002-0387-1
Erlhagen, W., & Jancke, D. (2004). The role of action plans and other
cognitive factors in motion extrapolation: A modeling study. Visual
Cognition, 11, 315–340. doi:10.1080/13506280344000293
Finke, R. A., Freyd, J. J., & Shyi, G. C. W. (1986). Implied velocity and
acceleration induce transformations of visual memory. Journal of Ex-
perimental Psychology: General, 115, 175–188. doi:10.1037/0096-3445
.115.2.175
Fouriezos, G., Capstick, G., Monette, F., Bellemare, C., Parkinson, M., &
Dumoulin, A. (2007). Judgments of synchrony between auditory and
moving or still visual stimuli. Canadian Journal of Experimental Psy-
chology, 61, 277–292. doi:10.1037/cjep2007028
Freyd, J. J. (1987). Dynamic mental representations. Psychological Re-
view, 94, 427–438. doi:10.1037/0033-295X.94.4.427
Freyd, J. J., & Finke, R. A. (1984). Representational momentum. Journal
of Experimental Psychology: Learning, Memory, and Cognition, 10,
126–132. doi:10.1037/0278-7393.10.1.126
Freyd, J. J., & Johnson, J. Q. (1987). Probing the time course of
representational momentum. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 13, 259 –268. doi:10.1037/0278-
7393.13.2.259
Fröhlich, F. W. (1923). U
¨
ber die messing der empfindungszeit [On the
measurement of sensation time]. Zeitschrift für Sinnesphysiologie, 54,
58–78.
Fu, Y. X., Shen, Y., & Dan, Y. (2001). Motion-induced perceptual
extrapolation of blurred visual targets. Journal of Neuroscience, 21,
RC172.
Fukiage, T., & Murakami, I. (2010). The tilt aftereffect occurs indepen-
dently of the flash-lag effect. Vision Research, 50, 1949–1956. doi:
10.1016/j.visres.2010.07.002
Fukiage, T., Whitney, D., & Murakami, I. (2011). A flash-drag effect in
random motion reveals involvement of preattentive motion processing.
Journal of Vision, 11(13), 12. doi:10.1167/11.13.12
Gauch, A., & Kerzel, D. (2008a). Comparison of flashed and moving
probes in the flash-lag effect: Evidence for misbinding of abrupt and
continuous changes. Vision Research, 48, 1584 –1591. doi:10.1016/j
.visres.2008.04.025
Gauch, A., & Kerzel, D. (2008b). Perceptual asynchronies between
color and motion at the onset of motion and along the motion
trajectory. Perception & Psychophysics, 70, 1092–1103. doi:10.3758/
PP.70.6.1092
Gauch, A., & Kerzel, D. (2009). Contributions of visible persistence and
perceptual set to the flash-lag effect: Focusing on flash onset abolishes
the illusion. Vision Research, 49, 2983–2991. doi:10.1016/j.visres.2009
.09.018
Getzmann, S. (2005). Shifting the onset of a moving sound source: A
Fröhlich effect in spatial hearing. Hearing Research, 210, 104 –111.
doi:10.1016/j.heares.2005.08.003
Gilis, B., Helsen, W., Catteeuw, P., & Wagemans, J. (2008). Offside
decisions by expert assistant referees in association football: Perception
and recall of spatial positions in complex dynamic events. Journal of
Experimental Psychology: Applied, 14, 21–35. doi:10.1037/1076-898X
.14.1.21
Gollisch, T., & Meister, M. (2010). Eye smarter than scientists believed:
Neural computations in circuits of the retina. Neuron, 65, 150 –164.
doi:10.1016/j.neuron.2009.12.009
Halpern, A. R., & Kelly, M. H. (1993). Memory biases in left versus right
implied motion. Journal of Experimental Psychology: Learning, Mem-
ory, and Cognition, 19, 471– 484. doi:10.1037/0278-7393.19.2.471
Harris, L. R., Duke, P. A., & Kopinska, A. (2006). Flash lag in depth.
Vision Research, 46, 2735–2742. doi:10.1016/j.visres.2006.01.001
Hayes, A. E., & Freyd, J. J. (2002). Representational momentum when
attention is divided. Visual Cognition, 9, 8–27. doi:10.1080/
13506280143000296
Hazelhoff, F., & Wiersma, H. (1924). Die Wahrnehmungszeit [The sen-
sation time]. Zeitschrift für Psychologie, 96, 171–188.
Hine, T. J., White, A. M. V., & Chappell, M. (2003). Is there an auditory–
visual flash-lag effect? Clinical and Experimental Ophthalmology, 31,
254–257. doi:10.1046/j.1442-9071.2003.00640.x
Hogben, J. H., & Di Lollo, V. (1974). Perceptual integration and perceptual
segregation of brief visual stimuli. Vision Research, 14, 1059–1069.
doi:10.1016/0042-6989(74)90202-8
Hommuk, K., Bachmann, T., & Oja, A. (2008). Precuing an isolated
stimulus temporarily outweighs in-stream stimulus facilitation. Journal
of General Psychology, 135, 167–181. doi:10.3200/GENP.135.2.167-
182
Hubbard, T. L. (1990). Cognitive representation of linear motion: Possible
direction and gravity effects in judged displacement. Memory & Cog-
nition, 18, 299–309. doi:10.3758/BF03213883
Hubbard, T. L. (1995). Environmental invariants in the representation of
motion: Implied dynamics and representational momentum, gravity,
friction, and centripetal force. Psychonomic Bulletin & Review, 2, 322–
338. doi:10.3758/BF03210971
Hubbard, T. L. (1997). Target size and displacement along the axis of
implied gravitational attraction: Effects of implied weight and evidence
of representational gravity. Journal of Experimental Psychology: Learn-
ing, Memory, and Cognition, 23, 1484 –1493. doi:10.1037/0278-7393
.23.6.1484
Hubbard, T. L. (2005). Representational momentum and related displace-
ment in spatial memory: A review of the findings. Psychonomic Bulletin
& Review, 12, 822–851. doi:10.3758/BF03196775
Hubbard, T. L. (2006a). Bridging the gap: Possible roles and contributions
of representational momentum. Psicológica, 27, 1–34.
Hubbard, T. L. (2006b). Computational theory and cognition in represen-
tational momentum and related types of displacement: A reply to Kerzel.
Psychonomic Bulletin & Review, 13, 174–177. doi:10.3758/
BF03193830
Hubbard, T. L. (2008). Representational momentum contributes to motion
induced mislocalization of stationary objects. Visual Cognition, 16,
44– 67. doi:10.1080/13506280601155468
Hubbard, T. L. (2010). Approaches to representational momentum: Theo-
ries and models. In R. Nijhawan & B. Khurana (Eds.), Space and time
in perception and action (pp. 338–365). New York, NY: Cambridge
University Press. doi:10.1017/CBO9780511750540.020
Hubbard, T. L. (2013). Launching, entraining, and representational mo-
mentum: Evidence consistent with an impetus heuristic in perception of
causality. Axiomathes. Advance online publication. doi:10.1007/s10516-
012-9186-z
Hubbard, T. L., & Bharucha, J. J. (1988). Judged displacement in apparent
vertical and horizontal motion. Perception & Psychophysics, 44, 211–
221. doi:10.3758/BF03206290
Hubbard, T. L., Blessum, J. A., & Ruppel, S. E. (2001). Representational
momentum and Michotte’s (1946/1963) “launching effect” paradigm.
Journal of Experimental Psychology: Learning, Memory, and Cogni-
tion, 27, 294–301. doi:10.1037/0278-7393.27.1.294
Hubbard, T. L., Kumar, A. M., & Carp, C. L. (2009). Effects of spatial
cueing on representational momentum. Journal of Experimental Psy-
chology: Learning, Memory, and Cognition, 35, 666 –677. doi:10.1037/
a0014870
Hubbard, T. L., & Ruppel, S. E. (1999). Representational momentum and
the landmark attraction effect. Canadian Journal of Experimental Psy-
chology, 53, 242–256. doi:10.1037/h0087313
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
334
HUBBARD
Hubbard, T. L., & Ruppel, S. E. (2013). A Fröhlich effect and represen-
tational gravity in memory for auditory pitch. Journal of Experimental
Psychology: Human Perception and Performance. Advance online pub-
lication. doi:10.1037/a0031103
Ichikawa, M., & Masakura, Y. (2006). Manual control of the visual
stimulus reduces the flash-lag effect. Vision Research, 46, 2192–2203.
doi:10.1016/j.visres.2005.12.021
Ichikawa, M., & Masakura, Y. (2010). Reduction of the flash-lag effect in
terms of active observation. Attention, Perception, & Psychophysics, 72,
1032–1044. doi:10.3758/APP.72.4.1032
Ishii, M., Seekkuarachchi, H., Tamura, H., & Tang, Z. (2004). 3D flash lag
illusion. Vision Research, 44, 1981–1984. doi:10.1016/j.visres.2004.03
.026
Jancke, D., & Erlhagen, W. (2010). Bridging the gap: A model of common
neural mechanisms underlying the Fröhlich effect, the flash-lag effect,
and the representational momentum effect. In R. Nijhawan & B.
Khurana (Eds.), Space and time in perception and action (pp. 422–440).
New York, NY: Cambridge University Press. doi:10.1017/
CBO9780511750540.025
Jancke, D., Erlhagen, W., Schöner, G., & Dinse, H. R. (2004). Shorter
latencies for motion trajectories than for flashes in population responses
of cat primary visual cortex. Journal of Physiology, 556, 971–982.
doi:10.1113/jphysiol.2003.058941
Johnston, H. M., & Jones, M. R. (2006). Higher order pattern structure
influences auditory representational momentum. Journal of Experimen-
tal Psychology: Human Perception and Performance, 32, 2–17. doi:
10.1037/0096-1523.32.1.2
Jones, B., & Huang, Y. L. (1982). Space–time dependencies in psycho-
physical judgment of extent and duration: Algebraic models of the tau
and kappa effects. Psychological Bulletin, 91, 128–142. doi:10.1037/
0033-2909.91.1.128
Joordens, S., Spalek, T. M., Razmy, S., & van Duijn, M. (2004). A
Clockwork Orange: Compensation opposing momentum in memory
for location. Memory & Cognition, 32, 39–50. doi:10.3758/
BF03195819
Jordan, J. S., & Knoblich, G. (2004). Spatial perception and control.
Psychonomic Bulletin & Review, 11, 54–59. doi:10.3758/
BF03206460
Kafaligönül, H., Patel, S. S., Ög
˘
men, H., Bedell, H. E., & Purushothaman,
G. (2010). Perceptual asynchronies and the dual-channel differential
latency hypothesis. In R. Nijhawan & B. Khurana (Eds.), Space and time
in perception and action (pp. 379–395). New York, NY: Cambridge
University Press. doi:10.1017/CBO9780511750540.022
Kanai, R., Sheth, B. R., & Shimojo, S. (2004). Stopping the motion and
sleuthing the flash-lag effect: Spatial uncertainty is the key to perceptual
mislocalization. Vision Research, 44, 2605–2619. doi:10.1016/j.visres
.2003.10.028
Kanai, R., & Verstraten, F. A. J. (2006). Visual transients reveal the
veridical position of a moving object. Perception, 35, 453– 460. doi:
10.1068/p5443
Kelly, M. H., & Freyd, J. J. (1987). Explorations of representational
momentum. Cognitive Psychology, 19, 369– 401. doi:10.1016/0010-
0285(87)90009-0
Kerzel, D. (2000). Eye movements and visible persistence explain the
mislocalization of the final position of a moving target. Vision Research,
40, 3703–3715. doi:10.1016/S0042-6989(00)00226-1
Kerzel, D. (2003). Asynchronous perception of motion and luminance
change. Psychological Research, 67, 233–239. doi:10.1007/s00426-002-
0121-6
Kerzel, D. (2010). The Fröhlich effect: Past and present. In R. Nijhawan &
B. Khurana (Eds.), Space and time in perception and action (pp. 321–
337). New York, NY: Cambridge University Press. doi:10.1017/
CBO9780511750540.019
Kessler, K., Gordon, L., Cessford, K., & Lages, M. (2010). Characteristics
of motor resonance produce the pattern of flash-lag effects for biological
motion. PLoS ONE, 5, e8258. doi:10.1371/journal.pone.0008258
Khurana, B., Carter, R. M., Watanabe, K., & Nijhawan, R. (2006). Flash-
lag chimeras: The role of perceived alignment in the composite face
effect. Vision Research, 46, 2757–2772. doi:10.1016/j.visres.2006.02
.001
Khurana, B., & Nijhawan, R. (1995). Extrapolation or attention shift?
Nature, 378, 566. doi:10.1038/378566a0
Khurana, B., Watanabe, K., & Nijhawan, R. (2000). The role of attention
in motion extrapolation: Are moving objects “corrected” or flashed
objects attentionally delayed? Perception, 29, 675– 692. doi:10.1068/
p3066
Kirschfeld, K. (2006). Stopping motion and the flash-lag effect. Vision
Research, 46, 1547–1551. doi:10.1016/j.visres.2005.07.010
Kirschfeld, K., & Kammer, T. (1999). The Fröhlich effect: A consequence
of the interaction of visual focal attention and metacontrast. Vision
Research, 39, 3702–3709. doi:10.1016/S0042-6989(99)00089-9
Klein, S. (2002). Libet’s research on the timing of conscious intention to
act: A commentary. Consciousness and Cognition, 11, 273–279. doi:
10.1006/ccog.2002.0557
Kourtzi, Z., & Kanwisher, N. (2000). Activation in human MT/MST for
static images with implied motion. Journal of Cognitive Neuroscience,
12, 48–55. doi:10.1162/08989290051137594
Kreegipuu, K., & Allik, J. (2003). Perceived onset time and position of a
moving stimulus. Vision Research, 43, 1625–1635. doi:10.1016/S0042-
6989(03)00165-2
Kreegipuu, K., & Allik, J. (2004). Confusion of space and time in the
flash-lag effect. Perception, 33, 293–306. doi:10.1068/p5043
Krekelberg, B. (2001). The persistence of position. Vision Research, 41,
529–539. doi:10.1016/S0042-6989(00)00281-9
Krekelberg, B. (2003). Sound and vision. Trends in Cognitive Sciences, 7,
277–279. doi:10.1016/S1364-6613(03)00133-5
Krekelberg, B., & Lappe, M. (1999). Temporal recruitment along the
trajectory of moving objects and the perception of position. Vision
Research, 39, 2669–2679. doi:10.1016/S0042-6989(98)00287-9
Krekelberg, B., & Lappe, M. (2000a). A model of the perceived relative
positions of moving objects based upon a slow averaging process. Vision
Research, 40, 201–215. doi:10.1016/S0042-6989(99)00168-6
Krekelberg, B., & Lappe, M. (2000b). The position of moving objects.
Science, 289, 1107a. doi:10.1126/science.289.5482.1107a
Krekelberg, B., & Lappe, M. (2001). Neuronal latencies and the position of
moving objects. Trends in Neurosciences, 24, 335–339. doi:10.1016/
S0166-2236(00)01795-1
Krekelberg, B., & Lappe, M. (2002). Response: Untangling spatial from
temporal illusions. Trends in Neurosciences, 25, 294. doi:10.1016/
S0166-2236(02)02180-X
Lankheet, M. J. M., & van de Grind, W. A. (2010). Simultaneity versus
asynchrony of visual motion and luminance changes. In R. Nijhawan &
B. Khurana (Eds.), Space and time in perception and action (pp. 301–
317). New York, NY: Cambridge University Press. doi:10.1017/
CBO9780511750540.018
Lappe, M., & Krekelberg, B. (1998). The position of moving objects.
Perception, 27, 1437–1449. doi:10.1068/p271437
Lee, T. C. P., Khuu, S. K., Li, W., & Hayes, A. (2008). Distortion in
perceived image size accompanies flash lag in depth. Journal of Vision,
8(11), 20. doi:10.1167/8.11.20
Libet, B. (1985). Unconscious cerebral initiative and the role of conscious
will in voluntary action. Behavioral and Brain Sciences, 8, 529 –566.
doi:10.1017/S0140525X00044903
Libet, B., Gleason, C. A., Wright, E. W., & Pearl, D. K. (1983). Time of
conscious intention to act in relation to onset of cerebral activity (read-
iness-potential): The unconscious initiation of a freely voluntary act.
Brain, 106, 623–642. doi:10.1093/brain/106.3.623
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
335
FLASH-LAG EFFECT
Libet, B., Wright, E. W., Jr., & Gleason, C. A. (1982). Readiness-potentials
preceding unrestricted “spontaneous” vs. pre-planned voluntary acts.
Electroencephalography and Clinical Neurophysiology, 54, 322–335.
doi:10.1016/0013-4694(82)90181-X
Linares, D., Holcombe, A. O., & White, A. L. (2009). Where is the
moving object now? Judgments of instantaneous position show poor
temporal precision (SD ⫽ 70 ms). Journal of Vision, 9(13), 9.
doi:10.1167/9.13.9
Linares, D., & López-Moliner, J. (2007). Absence of flash-lag when
judging global shape from local positions. Vision Research, 47, 357–
362. doi:10.1016/j.visres.2006.10.013
Linares, D., López-Moliner, J., & Johnston, A. (2007). Motion signal and
the perceived positions of moving objects. Journal of Vision, 7(7), 1.
doi:10.1167/7.7.1
López-Moliner, J., & Linares, D. (2006). The flash-lag is reduced when the
flash is perceived as a sensory consequence of our action. Vision
Research, 46, 2122–2129. doi:10.1016/j.visres.2005.11.016
Mach, E. (1897). Contributions to the analysis of sensations (C. M.
Williams, Trans.). Chicago, IL: Open Court. (Original work published
1885)
MacKay, D. M. (1958). Perceptual stability of a stroboscopically lit visual
field containing self-luminous objects. Nature, 181, 507–508. doi:
10.1038/181507a0
Maiche, A., Budelli, R., & Gómez-Sena, L. (2007). Spatial facilitation is
involved in flash-lag effect. Vision Research, 47, 1655–1661. doi:
10.1016/j.visres.2007.02.008
Mateeff, S., & Hohnsbein, J. (1988). Perceptual latencies are shorter for
motion towards the fovea than for motion away. Vision Research, 28,
711–719. doi:10.1016/0042-6989(88)90050-8
Mateeff, S., Yakimoff, N., Hohnsbein, J., Ehrenstein, W. H., Bohdanecky,
Z., & Radil, T. (1991). Selective directional sensitivity in visual motion
perception. Vision Research, 31, 131–138. doi:10.1016/0042-
6989(91)90080-O
Maus, G. W., Khurana, B., & Nijhawan, R. (2010). History and theory of
flash-lag: Past, present, and future. In R. Nijhawan & B. Khurana (Eds.),
Space and time in perception and action (pp. 477–499). New York, NY:
Cambridge University Press. doi:10.1017/CBO9780511750540.027
Maus, G. W., & Nijhawan, R. (2006). Forward displacement of fading
objects in motion: The role of transient signals in perceiving position.
Vision Research, 46, 4375–4381. doi:10.1016/j.visres.2006.08.028
Maus, G. W., & Nijhawan, R. (2008). Motion extrapolation into the blind
spot. Psychological Science, 19, 1087–1091. doi:10.1111/j.1467-9280
.2008.02205.x
Maus, G. W., & Nijhawan, R. (2009). Going, going, gone: Localizing
abrupt offsets of moving objects. Journal of Experimental Psychology:
Human Perception and Performance, 35, 611–626. doi:10.1037/
a0012317
Maus, G. W., Ward, J., Nijhawan, R., & Whitney, D. (2013). The perceived
position of moving objects: Transcranial magnetic stimulation of area
MT⫹ reduces the flash-lag effect. Cerebral Cortex, 23, 241–247. doi:
10.1093/cercor/bhs021
Metzger, W. (1932). Versuch einer gemeinamen Theorie der Phänomene
Fröhlichs und Hazeloffs und Kritikihrer Verfahren zur Messung der
Empfindungszeit [An attempt toward a common theory of the phenom-
ena of Fröhlich and Hazelhoff and a criticism of their methods to
measure sensation time]. Psychologische Forschung, 16, 176 –200. doi:
10.1007/BF00409732
Mitrani, L., & Dimitrov, G. (1982). Retinal location and visual localization
during pursuit eye movements. Vision Research, 22, 1047–1051. doi:
10.1016/0042-6989(82)90041-4
Moore, C. M., & Enns, J. T. (2004). Object updating and the flash-lag
effect. Psychological Science, 15, 866 –871. doi:10.1111/j.0956-7976
.2004.00768.x
Munger, M. P., & Owens, T. R. (2004). Representational momentum and
the flash-lag effect. Visual Cognition, 11, 81–103. doi:10.1080/
13506280344000257
Murai, Y., & Murakami, I. (2012). The flash-lag effect is observed some-
what before, but never after, the display period of a moving stimulus.
Journal of Vision, 12(9), 1232. doi:10.1167/12.9.1232
Murakami, I. (2001a). The flash-lag effect as a spatiotemporal correlation
structure. Journal of Vision, 1(2), 6. doi:10.1167/1.2.6
Murakami, I. (2001b). A flash-lag effect in random motion. Vision Re-
search, 41, 3101–3119. doi:10.1016/S0042-6989(01)00193-6
Müsseler, J., & Aschersleben, G. (1998). Localizing the first position of a
moving stimulus: The Fröhlich effect and an attention-shifting explana-
tion. Perception & Psychophysics, 60, 683–695. doi:10.3758/
BF03206055
Müsseler, J., Stork, S., & Kerzel, D. (2002). Comparing mislocalizations
with moving stimuli: The Fröhlich effect, the flash-lag, and representa-
tional momentum. Visual Cognition, 9, 120–138. doi:10.1080/
13506280143000359
Nagai, M., Suganuma, M., Nijhawan, R., Freyd, J. J., Miller, G., &
Watanabe, K. (2010). Conceptual influence on the flash-lag effect and
representational momentum. In R. Nijhawan & B. Khurana (Eds.), Space
and time in perception and action (pp. 366 –378). New York, NY:
Cambridge University Press. doi:10.1017/CBO9780511750540.021
Namba, J., & Baldo, M. V. C. (2004). The modulation of the flash-lag
effect by voluntary attention. Perception, 33, 621–631. doi:10.1068/
p5212
Nieman, D., Nijhawan, R., Khurana, B., & Shimojo, S. (2006). Cyclopean
flash-lag illusion. Vision Research, 46, 3909 –3914. doi:10.1016/j.visres
.2006.06.003
Nijhawan, R. (1994). Motion extrapolation in catching. Nature, 370, 256–
257. doi:10.1038/370256b0
Nijhawan, R. (1997). Visual decomposition of colour through motion
extrapolation. Nature, 386, 66–69. doi:10.1038/386066a0
Nijhawan, R. (2001). The flash-lag phenomenon: Object motion and eye
movements. Perception, 30, 263–282. doi:10.1068/p3172
Nijhawan, R. (2002). Neural delays, visual motion, and the flash-lag effect.
Trends in Cognitive Sciences, 6, 387–393. doi:10.1016/S1364-
6613(02)01963-0
Nijhawan, R. (2008). Visual prediction: Psychophysics and neurophysiol-
ogy of compensation for time delays. Behavioral and Brain Sciences, 31,
179–198. doi:10.1017/S0140525X08003804
Nijhawan, R., Khurana, B., Kamitani, Y., Watanabe, K., & Shimojo, S.
(1998). Eye-movement based extrapolation leads to decomposition of
color. Investigative Ophthalmology and Visual Science, 39, 1045.
Nijhawan, R., & Kirschfeld, K. (2003). Analogous mechanisms compen-
sate for neural delays in the sensory and the motor pathways; evidence
from motor flash-lag. Current Biology, 13, 749 –753. doi:10.1016/
S0960-9822(03)00248-3
Nijhawan, R., Watanabe, K., Khurana, B., & Shimojo, S. (2004). Com-
pensation of neural delays in visual–motor behavior: No evidence for
shorter afferent delays for visual motion. Visual Cognition, 11, 275–298.
doi:10.1080/13506280344000347
Noguchi, Y., & Kakigi, R. (2008). Knowledge-based correction of flash-
lag illusion. Journal of Cognitive Neuroscience, 20, 513–525. doi:
10.1162/jocn.2008.20033
Ög
˘
men, H., Patel, S. S., Bedell, H. E., & Camuz, K. (2004). Differential
latencies and the dynamics of the position computation process for
moving targets, assessed with the flash-lag effect. Vision Research, 44,
2109–2128. doi:10.1016/j.visres.2004.04.003
Orban, G. A., Hoffman, K. P., & Duysens, J. (1985). Velocity selectivity
in the cat visual system. I. Response of LGN cells to moving bar stimuli:
A comparison with cortical areas 17 and 18. Journal of Neurophysiol-
ogy, 54, 1026–1049.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
336
HUBBARD
Palmer, E. M., & Kellman, P. J. (2002). Underestimation of velocity after
occlusion causes the aperture-capture illusion. Journal of Vision, 2(7),
477. doi:10.1167/2.7.477
Palmer, E. M., & Kellman, P. J. (2003). (Mis)perception of motion and
form after occlusion: Anorthoscopic perception revisited. Journal of
Vision, 3(9), 251. doi:10.1167/3.9.251
Patel, S. S., Ög
˘
men, H., Bedell, H. E., & Sampath, V. (2000). Flash-lag
effect: Differential latency, not postdiction. Science, 290, 1051a. doi:
10.1126/science.290.5494.1051a
Purushothaman, G., Patel, S. S. Bedell, H. E., & Ög
˘
men, H. (1998).
Moving ahead through differential visual latency. Nature, 396, 424. doi:
10.1038/24766
Raiguel, S. E., Lagae, L., Gulya
`
s, B., & Orban, G. A. (1989). Response
latencies of visual cells in macaque areas V1, V2 and V5. Brain
Research, 493, 155–159. doi:10.1016/0006-8993(89)91010-X
Rao, R. P. N., Eagleman, D. M., & Sejnowski, T. J. (2001). Optimal
smoothing in visual motion perception. Neural Computation, 13, 1243–
1253. doi:10.1162/08997660152002843
Reed, C. L., & Vinson, N. G. (1996). Conceptual effects on representa-
tional momentum. Journal of Experimental Psychology: Human Per-
ception and Performance, 22, 839 –850. doi:10.1037/0096-1523.22.4
.839
Rizk, J. K., Chappell, M., & Hine, T. J. (2009). Effect of motion smooth-
ness on the flash-lag illusion. Vision Research, 49, 2201–2208. doi:
10.1016/j.visres.2009.06.010
Rotman, G., Brenner, E., & Smeets, J. B. J. (2002). Spatial but not temporal
cueing influences the mislocalization of a target flashed during smooth
pursuit. Perception, 31, 1195–1203. doi:10.1068/p3411
Rotman, G., Brenner, E., & Smeets, J. B. J. (2004). Mislocalization of
targets flashed during smooth pursuit depends on the change in gaze
direction after the flash. Journal of Vision, 4(7), 4. doi:10.1167/4.7.4
Rotman, G., Brenner, E., & Smeets, J. B. J. (2005). Flashes are localized
as if they were moving with the eyes. Vision Research, 45, 355–364.
doi:10.1016/j.visres.2004.08.014
Roulston, B. W., Self, M. W., & Zeki, S. (2006). Perceptual compression
of space through position integration. Proceedings of the Royal Society:
Series B. Biological Sciences, 273, 2507–2512. doi:10.1098/rspb.2006
.3616
Sarich, D., Chappell, M., & Burgess, C. (2007). Dividing attention in the
flash-lag illusion. Vision Research, 47, 544 –547. doi:10.1016/j.visres
.2006.09.029
Schlag, J., Cai, R. H., Dorfman, A., Mohempour, A., & Schlag-Rey, M.
(2000). Extrapolating movement without retinal motion. Nature, 403,
38–39. doi:10.1038/47402
Schlag, J., & Schlag-Rey, M. (2002). Through the eye, slowly: Delays and
localization errors in the visual system. Nature Reviews Neuroscience, 3,
191–200. doi:10.1038/nrn750
Schwartz, O., Sejnowski, T. J., & Dayan, P. (2009). Perceptual orga-
nization in the tilt illusion. Journal of Vision, 9(4), 19. doi:10.1167/
9.4.19
Scocchia, L., Actis-Grosso, R., de’Sperati, C., Stucchi, N., & Baud-Bovy,
G. (2009). Observer’s control of the moving stimulus increases the
flash-lag effect. Vision Research, 49, 2363–2370. doi:10.1016/j.visres
.2009.06.023
Senior, C., Barnes, J., Giampietro, V., Simmons, A., Bullmore, E. T.,
Brammer, M., & David, A. S. (2000). The functional neuroanatomy of
implicit-motion perception or “representational momentum”. Current
Biology, 10, 16–22. doi:10.1016/S0960-9822(99)00259-6
Senior, C., Ward, J., & David, A. S. (2002). Representational momentum
and the brain: An investigation into the functional necessity of V5/MT.
Visual Cognition, 9, 81–92. doi:10.1080/13506280143000331
Shen, M., Zhou, J., Gao, T., Liang, J., & Shui, R. (2007). The perceived
position of a moving object is not the result of position integration.
Vision Research, 47, 3088–3095. doi:10.1016/j.visres.2007.08.012
Sheth, B. R., Nijhawan, R., & Shimojo, S. (2000). Changing objects lead
briefly flashed ones. Nature Neuroscience, 3, 489 –495. doi:10.1038/
74865
Shi, Z., & de’Sperati, C. (2008). Motion-induced positional biases in the
flash-lag configuration. Cognitive Neuropsychology, 25, 1027–1038.
doi:10.1080/02643290701866051
Shi, Z., & Nijhawan, R. (2008). Behavioral significance of motion direc-
tion causes anisotropic flash-lag, flash-drag, flash-repulsion, and
movement-mislocalization effects. Journal of Vision, 8(7), 24. doi:
10.1167/8.7.24
Shioiri, S., Yamamoto, K., Oshida, H., Matsubara, K., & Yaguchi, H.
(2010). Measuring attention using flash-lag effect. Journal of Vision,
10(10), 10. doi:10.1167/10.10.10
Stekelenburg, J. J., & Vroomen, J. (2005). An event-related potential
investigation of the time-course of temporal ventriloquism. NeuroRe-
port, 16, 641–644. doi:10.1097/00001756-200504250-00025
Sundberg, K. A., Fallah, M., & Reynolds, J. H. (2006). A motion-
dependent distortion of retinotopy in area V4. Neuron, 49, 447–457.
doi:10.1016/j.neuron.2005.12.023
Thornton, I. M., & Hayes, A. E. (2004). Anticipating action in complex scenes.
Visual Cognition, 11, 341–370. doi:10.1080/13506280344000374
van Beers, R. J., Wolpert, D. M., & Haggard, P. (2001). Sensorimotor
integration compensates for visual localization errors during smooth
pursuit eye movements. Journal of Neurophysiology, 85, 1914–1922.
van de Grind, W. (2002). Physical, neural, and mental timing. Conscious-
ness and Cognition, 11, 241–264. doi:10.1006/ccog.2002.0560
van de Grind, W. (2008). Motion as a reference for positions. Behavioral
and Brain Sciences, 31, 218 –219. doi:10.1017/S0140525X08004020
Vinson, N. G., & Reed, C. L. (2002). Sources of object-specific effects in
representational momentum. Visual Cognition, 9, 41–65. doi:10.1080/
13506280143000313
Vreven, D., & Verghese, P. (2005). Predictability and the dynamics of
position processing in the flash-lag effect. Perception, 34, 31– 44. doi:
10.1068/p5371
Vroomen, J., & de Gelder, B. (2004). Temporal ventriloquism: Sound
modulates the flash-lag effect. Journal of Experimental Psychology:
Human Perception and Performance, 30, 513–518. doi:10.1037/0096-
1523.30.3.513
Walker, J. T., & Irion, A. L. (1982). Apparent displacement of moving and
stationary strobe flashes. Human Factors, 24, 213–224. doi:10.1177/
001872088202400207
Watanabe, K. (2004). Visual grouping by motion precedes the relative
localization between moving and flashed stimuli. Journal of Experimen-
tal Psychology: Human Perception and Performance, 30, 504 –512.
doi:10.1037/0096-1523.30.3.504
Watanabe, K., Nijhawan, R., Khurana, B., & Shimojo, S. (2001). Percep-
tual organization of moving stimuli modulates the flash-lag effect.
Journal of Experimental Psychology: Human Perception and Perfor-
mance, 27, 879–894. doi:10.1037/0096-1523.27.4.879
Watanabe, K., & Yokoi, K. (2006). Object-based anisotropies in the
flash-lag effect. Psychological Science, 17, 728–735. doi:10.1111/j
.1467-9280.2006.01773.x
Watanabe, K., & Yokoi, K. (2007). Object-based anisotropic mislocaliza-
tion by retinotopic motion signals. Vision Research, 47, 1662–1667.
doi:10.1016/j.visres.2007.03.006
White, H., Minor, S. W., Merrell, J., & Smith, T. (1993). Representational-
momentum effects in the cerebral hemispheres. Brain and Cognition, 22,
161–170. doi:10.1006/brcg.1993.1031
Whitney, D. (2002). The influence of visual motion on perceived position.
Trends in Cognitive Sciences, 6, 211–216. doi:10.1016/S1364-
6613(02)01887-9
Whitney, D., & Cavanagh, P. (2000). Motion distorts visual space: Shifting
the perceived position of remote stationary objects. Nature Neurosci-
ence, 3, 954–959. doi:10.1038/78878
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
337
FLASH-LAG EFFECT
Whitney, D., Cavanagh, P., & Murakami, I. (2000). Temporal facilitation
for moving stimuli is independent of changes in direction. Vision Re-
search, 40, 3829–3839. doi:10.1016/S0042-6989(00)00225-X
Whitney, D., & Murakami, I. (1998). Latency difference, not spatial
extrapolation. Nature Neuroscience, 1, 656 –657. doi:10.1038/3659
Whitney, D., Murakami, I., & Cavanagh, P. (2000). Illusory spatial offset
of a flash relative to a moving stimulus is caused by differential latencies
for moving and flashed stimuli. Vision Research, 40, 137–149. doi:
10.1016/S0042-6989(99)00166-2
Wojtach, W. T., Sung, K., Truong, S., & Purves, D. (2008). An empirical
explanation of the flash-lag effect. Proceedings of the National Academy
of Sciences of the United States of America, 105, 16338–16343. doi:
10.1073/pnas.0808916105
Young, A. W., Hellawell, D., & Hay, D. C. (1987). Configural information
in face perception. Perception, 16, 747–759. doi:10.1068/p160747
Received August 24, 2012
Revision received March 17, 2013
Accepted March 20, 2013 䡲
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
338
HUBBARD
A preview of this full-text is provided by American Psychological Association.
Content available from Psychological Bulletin
This content is subject to copyright. Terms and conditions apply.
