Content uploaded by Antigona Martinez
Author content
All content in this area was uploaded by Antigona Martinez on Dec 01, 2020
Content may be subject to copyright.
Comparison of Psychophysical, Electrophysiological, and fMRI
Assessment of Visual Contrast Responses in Patients with
Schizophrenia
Daniel J. Calderonea,b,c, Antígona Martineza,d, Vance Zemona,e, Matthew J. Hoptmana,b,c,
George Huf, Jade E. Watkinsa, Daniel C. Javitta,c,g, and Pamela D. Butlera,b,c
aNathan S. Kline Institute for Psychiatric Research, 140 Old Orangeburg Road, Orangeburg, NY
10962, USA
bDepartment of Psychiatry, NYU School of Medicine, 550 First Avenue, New York, NY 10016,
USA
cDepartment of Psychology, The Graduate Center, City University of New York, 365 Fifth Avenue,
New York, NY 10016, USA
dDepartment of Neurosciences, University of California San Diego, 9500 Gilman Drive, La Jolla,
California 92093, USA
eFerkauf Graduate School of Psychology, Rousso Building, Albert Einstein College of Medicine,
1165 Morris Park Avenue, Bronx, NY 10461, USA
fVerisci Corporation, Raritan, NJ 08869, USA
gDepartment of Psychiatry, Columbia University College of Physicians and Surgeons, 630 West
168th Street, New York, NY 10032, USA
Abstract
Perception has been identified by the NIMH-sponsored Cognitive Neuroscience Treatment
Research to Improve Cognition in Schizophrenia (CNTRICS) group as a useful domain for
assessing cognitive deficits in patients with schizophrenia. Specific measures of contrast gain
derived from recordings of steady-state visual evoked potentials (ssVEP) have demonstrated
neural deficits within the visual pathways of patients with schizophrenia. Psychophysical
measures of contrast sensitivity have also shown functional loss in these patients. In the current
study, functional magnetic resonance imaging (fMRI) was used in conjunction with ssVEP and
contrast sensitivity testing to elucidate the neural underpinnings of these deficits. During fMRI
scanning, participants viewed 1) the same low and higher spatial frequency stimuli used in the
psychophysical contrast sensitivity task, at both individual detection threshold contrast and at a
© 2012 Elsevier Inc. All rights reserved.
Corresponding Author: Daniel Calderone, dcalderone@nki.rfmh.org, 412-491-0527, 140 Old Orangeburg Road, Orangeburg, NY
10962.
Conflicts of Interest: The authors report the following conflicts of interest: Dr. Daniel Javitt holds intellectual property rights for use
of NMDA agonists, including glycine, D-serine, and glycine transport inhibitors in treatment of schizophrenia. Dr. Daniel Javitt is a
major shareholder in Glytech, Inc. and Amino Acids Solutions, Inc. Within the past year, Dr. Javitt has served as a paid consultant to
Sepracor, AstraZeneca, Pfizer, Cypress, Merck, Sunovion, Eli Lilly, and BMS. Drs. George Hu and Vance Zemon are major
shareholders in VeriSci Corp. Daniel Calderone, Antigona Martinez, Matthew Hoptman, Jade Watkins, and Pamela Butler have no
conflicts of interest in relation to the subject of this study.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our
customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of
the resulting proof before it is published in its final citable form. Please note that during the production process errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
NIH Public Access
Author Manuscript
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
Published in final edited form as:
Neuroimage
. 2013 February 15; 67: 153–162. doi:10.1016/j.neuroimage.2012.11.019.
$watermark-text $watermark-text $watermark-text
high contrast; and 2) the same stimuli used in the ssVEP paradigm, which were designed to be
biased toward either the magnocellular or parvocellular visual pathway. Patients showed
significant impairment in contrast sensitivity at both spatial frequencies in the psychophysical
task, but showed reduced occipital activation volume for low, but not higher, spatial frequency at
the low and high contrasts tested in the magnet. As expected, patients exhibited selective deficits
under the magnocellular-biased ssVEP condition. However, occipital lobe fMRI responses
demonstrated the same general pattern for magnocellular- and parvocellular-biased stimuli across
groups. These results indicate dissociation between the fMRI measures and the psychophysical/
ssVEP measures. These latter measures appear to have greater value for the functional assessment
of the contrast deficits explored here.
Keywords
contrast sensitivity; fMRI; gain control; magnocellular; schizophrenia; visual
1. Introduction
Over recent years it has become clear that patients with schizophrenia exhibit sensory
processing deficits in a number of modalities (Javitt, 2009, Koychev et al., 2011, Leitman et
al., 2011, Silverstein and Keane, 2011, Butler et al., 2012). Indeed, perception was chosen as
one of the key domains for development of measures that could be used in clinical trials in
schizophrenia by the NIH-sponsored Cognitive Neuroscience Treatment Research to
Improve Cognition in Schizophrenia (CNTRICS) initiative (Green et al., 2009, Butler et al.,
2012). In the visual system, behavioral, electrophysiological, and functional magnetic
resonance imaging (fMRI) studies have revealed early-stage sensory deficits, including
deficient processing of contrast (Slaghuis, 1998, Kéri et al., 2002, Kéri et al., 2004, Butler et
al., 2005, Butler et al., 2009, Green et al., 2009), motion (Chen et al., 2003b, Chen et al.,
2004, Kim et al., 2006), and spatial frequency information (O’Donnell et al., 2002, Martinez
et al., 2008, Martinez et al., 2012). These visual sensory processing deficits appear to
contribute to higher level dysfunction in reading (Revheim et al., 2006), object processing
and grouping (Doniger et al., 2002, Kurylo et al., 2007, Sehatpour et al., 2010, Calderone et
al., 2012), and emotion processing (Turetsky et al., 2007, Butler et al., 2009).
Within the domain of perception, the CNTRICS initiative included the neurophysiological
and psychophysical tasks that are the focus of the current study. The measures of interest
here are ones that quantify the gain and sensitivity of contrast responses and their underlying
mechanisms (Green et al., 2009, Butler et al., 2012). The neurophysiological measures are
based on the use of visual stimuli designed to emphasize either the magnocellular or
parvocellular contributions to visual processing (Zemon and Gordon, 2006). The subcortical
magnocellular pathway contains rapidly conducting neurons that project preferentially
through primary visual cortex (V1) to dorsal stream cortical areas while the parvocellular
pathway contains smaller, more slowly conducting neurons that project preferentially
through V1 to ventral stream areas, with extensive interaction between these pathways
following activation of V1 (Kaplan, 2003). While response properties of the two pathways
overlap, they can be preferentially activated by stimuli that differ in contrast, spatial, and
temporal frequency. With regard to contrast, magnocellular neurons have a nonlinear
response function with steep initial slope as contrast increases through the low contrast
region followed by decreasing slope (response compression) as contrast increases above
~12%. The steep initial slope reflects initial gain and is referred to as ‘contrast gain.’
Response compression which occurs with increases in contrast reflects a nonlinear inhibitory
mechanism and is a component of ‘contrast gain control’ (Shapley and Victor, 1979,
Ohzawa et al., 1982, 1985, Zemon et al., 1995, Carandini et al., 1997). The subcortical
Calderone et al. Page 2
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
parvocellular pathway and its recipient cortical neurons, on the other hand, do not respond
much at low contrast (<10%), and parvocellular response functions exhibit a shallow linear
slope in magnitude vs. contrast, i.e., low contrast gain (Kaplan and Shapley, 1982, 1986,
Tootell et al., 1988, Shapley, 1990, Benardete et al., 1992).
Patients with schizophrenia exhibit contrast response deficits in the visual system, which are
seen in electrophysiological (Butler et al., 2005, Green et al., 2009, Butler et al., 2012) as
well as behavioral studies (Slaghuis, 1998, Kéri et al., 2002, Kéri et al., 2004, Butler et al.,
2005, Butler et al., 2009, Green et al., 2009, Barch et al., 2012). An electrophysiological
technique that involves recording steady-state visual evoked potentials (ssVEP) to isolated-
check stimuli (Zemon et al., 1988, Zemon and Gordon, 2006) has previously been used to
demonstrate contrast gain deficits in schizophrenia (Butler et al., 2001, Butler et al., 2005,
Butler et al., 2008a). This technique can bias responses toward the magnocellular
contribution by keeping stimuli in the low contrast range, and can bias responses toward the
parvocellular contribution by modulating stimulus contrast around a high contrast “pedestal”
to keep stimuli within the contrast range at which magnocellular response saturation occurs
(Zemon and Gordon, 2006). Signal-to-noise ratios are obtained separately for
magnocellular- and parvocellular-biased responses over a range of increasing contrasts.
Schizophrenia patients have shown preferential deficits in the magnocellular-biased vs. the
parvocellular-biased contrast response function (Butler et al., 2001, Butler et al., 2005,
Butler et al., 2009). These deficits are thought to reflect a dysfunction in a nonlinear gain
mechanism. To better understand the neural underpinnings of these deficits, the current
study used the same stimuli from previous ssVEP work (Butler et al., 2005, Zemon and
Gordon, 2006) in an fMRI paradigm.
Schizophrenia patients also exhibit visual deficits in a psychophysical contrast sensitivity
task, in which contrast detection thresholds are found for grating stimuli of different spatial
frequencies. The magnocellular pathway responds preferentially to low contrasts (<10%) as
well as low spatial and high temporal frequencies, while the parvocellular pathway
preferentially responds to high spatial and low temporal frequencies (Tootell et al., 1988,
Shapley, 1990, Merigan and Maunsell, 1993, Wurtz and Kandel, 2000, Norman, 2002). For
contrast sensitivity tasks, shorter duration stimuli (i.e. higher temporal frequency), produce
the highest contrast sensitivities at low spatial frequencies, whereas longer duration stimuli
produce the highest contrast sensitivities at mid-range spatial frequencies (Tolhurst, 1975,
Legge, 1978). A number of studies show that patients with schizophrenia have higher
contrast thresholds (i.e., impaired contrast sensitivity) compared to healthy controls
(Slaghuis, 1998, Kéri et al., 2002, Chen et al., 2003a, Slaghuis, 2004, Butler et al., 2005,
Butler et al., 2008b, Norton et al., 2009, Dias et al., 2011). Selective deficits have been
found at low spatial frequencies in some studies (Butler et al., 2005, Butler et al., 2009),
though others found deficits across spatial frequencies (Slaghuis, 1998, Kéri et al., 2002) or
showed contradictory results of increased contrast sensitivity for first-episode schizophrenia
patients (Kiss et al., 2010).
The goal of the current study was to explore the cortical areas that underlie the deficits in
contrast responses in schizophrenia using stimuli from electrophysiological (Butler et al.,
2005, Zemon and Gordon, 2006) and psychophysical paradigms (Butler et al., 2001, Butler
et al., 2005, Butler et al., 2009). It is hoped that this work will assist in task development for
measures to be used in clinical trials aimed at assessing cognition in schizophrenia.
Calderone et al. Page 3
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
2. Methods
2.1. Participants
Fifteen patients who met DSM-IV criteria for schizophrenia and 15 healthy volunteers
participated. Patients were recruited through inpatient and outpatient facilities associated
with the Nathan Kline Institute for Psychiatric Research. Diagnoses were obtained using the
Structured Clinical Interview for DSM-IV (SCID) (First et al., 1997) and available clinical
information. Controls were recruited through the Volunteer Recruitment Program at the
Nathan Kline Institute. All participants provided informed consent and received cash
compensation for their time. The study was approved by the Nathan Kline Institutional
Review Board. Healthy volunteers with a history of SCID-defined Axis I psychiatric
disorders were excluded. Patients and controls were excluded if they had any neurological or
ophthalmological disorders, including glaucoma or cataracts, that might affect performance
or if they met criteria for alcohol or substance dependence within the last six months or
abuse within the last month. All participants had normal or corrected-to-normal visual acuity
of 20/32 or better on the Logarithmic Visual Acuity Chart (Precision Vision). All patients
were receiving antipsychotic medication at the time of testing. Chlorpromazine equivalents
were calculated as previously described (Woods, 2003, 2005, 2011). All data reported below
are means ± standard deviation.
Controls and patients did not differ in gender (patients: 13 males, 2 females; controls: 12
males, 3 females;
χ2
(1) = .240,
p =
.63) or age (patients: 40.40 ± 9.90; controls: 36.87 ±
10.01;
t
(28) = .972,
p =
.27). Patients had significantly lower socioeconomic status (SES) as
measured by the 4-factor Hollingshead Scale (patients: 23.31 ± 6.80; controls: 44.57 ± 9.88;
t
(25) = −6.463,
p <
.001), but parental SES did not differ between groups (patients: 39.92 ±
9.39; controls: 46.68 ± 14.05;
t
(14.17) = 1.260,
p =
.23). Patients had significantly reduced
IQ (patients: 97.46 ± 7.00; controls: 104.71 ± 8.65;
t
(25) = −2.38,
p =
.03) and education as
measured by highest grade achieved (patients: 11.54 ± 1.20; controls: 14.50 ± 1.99;
t
(25) =
−4.64,
p <
.001). Patients were ill for 14.58 ± 7.42 years, had an average Global Assessment
of Functioning (GAF) score of 48.67 ± 13.84, and were receiving antipsychotic doses
equivalent to an average of 783.33 ± 611.54 mg of chlorpromazine per day. Although
demographic data for some variables were unavailable for some participants, the overall
sample characteristics were similar to those in recent publications from our group (Dias et
al., 2011, Calderone et al., 2012, Martinez et al., 2012).
2.2. Psychophysical Contrast Sensitivity
Horizontal sine-wave gratings were presented on the left or right half of a computer screen
(VENUS system, Neuroscientific Corp., Farmingdale, NY), with the other side of the screen
blank. The mean luminance of each side of the display was 84 cd/m2. Participants indicated
on which side the grating pattern appeared in a two-alternative forced-choice paradigm
(Figure 1). Two sine-wave gratings of different spatial frequencies expressed in cycles per
degree of visual angle (c/deg) were used. The low spatial frequency (0.5 c/deg) stimuli were
shown for a short (32 ms) duration and the higher spatial frequency (4 c/deg) stimuli were
shown for a longer (500 ms) duration to bias stimuli toward eliciting responses from the
transient (magnocellular-like) and sustained (parvocellular-like) mechanism, respectively.
The entire display subtended 6 x 6 degrees of visual angle, viewed from a distance of 190
cm. For each spatial frequency, an up-down transformed rule (UDTR) procedure (Wetherill
and Levitt, 1965) estimated the threshold contrast at which participants correctly identified
the location of the stimulus on 70.7% of the trials. Initially, contrast was changed in 6
decibel (dB) steps for each correct (−6 dB) or incorrect (+ 6 dB) responses. After two
incorrect responses, the UDTR was implemented with 3 dB steps. Following two correct
responses, contrast was decreased by 3 dB, whereas following one incorrect response,
Calderone et al. Page 4
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
contrast was increased by 3 dB. The threshold contrast was taken as the mean of ten contrast
reversals, and contrast sensitivity was calculated as the reciprocal of this threshold.
2.3. Steady-State Visual Evoked Potentials (ssVEP)
2.3.1. Apparatus—Stimulus presentation, ssVEP recording, and data analysis were
performed with a Neucodia system (VeriSci Corp., Raritan, NJ). For improved measurement
of responses, this system uses synchronized data collection: electroencephalographic (EEG)
signal sampling at integer multiples (4x) of the stimulus display’s frame rate (150 Hz). A
single channel of EEG recording was used (gain = 20,000, bandpass: 0.5–100 Hz) with one
(active) electrode at Oz, referenced to a second one at Cz with a floating ground at Pz, in
accordance with the 10–20 system (Jasper, 1958).
2.3.2. Stimuli and Procedure—Isolated dark checks, subtending 18.75 minutes of arc of
visual angle each, were shown in 16 × 16 check arrays subtending a total of 10 × 10° of
visual angle viewed from a distance of 114 cm. The background luminance was ~50 cd/m2.
Check luminance was modulated sinusoidally at 12.5 Hz. Seven depths of modulation
(DOM) (0, 1, 2, 4, 8, 16, and 32%) were presented for one second each in a seven-second
swept-parameter run. Ten such runs were obtained for M-biased and P-biased stimuli
separately, for each participant. For all runs, a standing check luminance (pedestal) was
used, with check luminance modulated above and below the pedestal according to the DOM.
In M-biased runs, the pedestal equaled the DOM, creating appearing and disappearing
stimuli (Figure 2). In P-biased runs, the pedestal was fixed at 48% Weber contrast, so that
stimuli never dropped below 16% contrast (Zemon and Gordon, 2006).
2.3.4. Analysis—A discrete Fourier transform was used to analyze the fundamental
frequency component of the ssVEP (response at the stimulus frequency) averaged for M-
and P-biased conditions separately. Signal-to-noise ratios (SNR) of the fundamental
frequency component computed for each set of 10 runs were used as the dependent measure
(Victor and Mast, 1991, Zemon et al., 1997) in a three-way ANOVA with group, DOM, and
bias condition as factors. A modified measure of initial gain was calculated as the slope of
the response function between 4 and 16% DOM, i.e. the change in SNR from 4 to 16%
DOM divided by the change in DOM (12%). A measure of maximal SNR was calculated as
the mean of the SNRs at 16 and 32% DOM.
2.4. Functional Magnetic Resonance Imaging (fMRI)
2.4.1. Apparatus—A 3T Siemens TIM Trio magnetic resonance scanner at the Nathan
Kline Institute was used for all functional and structural scans. Functional scans contained
34 axial slices, with TR=2000ms, TE=30ms, and voxel size=2.5×2.5×2.8 mm, with a 0.7
mm gap. High-resolution structural scans were performed with a 3-D magnetization
prepared rapid acquisition gradient echo (MPRAGE) sequence, having 192 sagital slices
with TR=2500ms, TE=3.5ms, FA=8°, and voxel size =1 mm3. Slice time correction, motion
correction, normalization to a value of 100, smoothing (8mm FWHM Gaussian kernel),
skull stripping, deconvolution of relevant time series, and first-order regression analyses
were performed using the AFNI (http://afni.nimh.nih.gov/; (Cox, 1996)). Functional and
structural scans were coregistered and transformed into a common Talairach space using the
Automatic Registration Toolbox (Ardekani et al., 2004, Klein et al., 2009).
2.4.2. Stimuli and Procedure—Similar isolated-check stimuli as those used in the
ssVEP paradigm were presented to create M-biased and P-biased fMRI scanning runs.
Check size was slightly larger, with each isolated check subtending 24 minutes of arc of
visual angle. The background luminance was ~100 cd/m2. Check luminance was modulated
sinusoidally at 12 Hz. Five DOMs (2, 4, 8, 16, and 32%) were presented in a block design
Calderone et al. Page 5
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
for both M-biased and P-biased conditions. For each DOM, 12 seconds of pattern
presentation was followed by 12 seconds of blank background luminance, and this cycle was
repeated four times. To maintain attention to the stimuli, participants pressed a response
button when a fixation cross in the center of the screen changed into a dot for 300ms. The
dot appeared randomly during half of the stimulus presentations.
Contrast sensitivity tasks were performed during fMRI scanning for the same two sinusoidal
grating stimulus conditions used in the psychophysical task: 0.5 c/deg displayed for 32 ms,
and 4 c/deg displayed for 500 ms. The mean luminance of each half of the display was 68.6
cd/m2. During structural scans, participants performed the task starting at 50% contrast. A
UDTR procedure similar to the one described above in section 2.2 that estimated an
accuracy of 84.09% changed stimulus contrast by 6 dB until the first incorrect response, and
then changed stimulus contrast by 3 dB afterwards. Following four correct responses,
contrast was decreased by 3 dB, whereas following 1 incorrect response, contrast was
increased by 3 dB. When five contrast reversals were obtained, the task paused until
structural scanning was complete. The threshold contrast was taken as the mean of the two
contrasts comprising the fifth contrast reversal. During a subsequent functional scan, the
task resumed at the contrast level reached at the end of the fifth reversal. Participants
continued performing the task at threshold contrast for 45 TRs, and then performed the task
at an unchanging high contrast of 71% for an additional 45 TRs. This entire procedure was
completed for each spatial frequency separately.
Display equipment used for the fMRI paradigm was limited to 256 gray levels, in contrast to
the 4096 gray levels utilized by the VENUS system for the psychophysical contrast
sensitivity task. Thus, the lowest contrast that could be displayed in the fMRI paradigm was
~1%. To partially compensate for this, a neutral density filter was used to reduce the
luminance of the display by 1 log unit during the contrast sensitivity paradigm only,
resulting in a mean luminance of 6.8 cd/m2. Contrast sensitivities, particularly to higher
spatial frequencies, are known to be reduced under conditions of lower luminance (Patel,
1966, Sperling, 1970, Peli, 1990). The fMRI paradigm utilized lower luminance in order to
bias threshold contrasts to be higher, though the display’s inability to show contrasts below
1% remained a limitation.
All stimuli were viewed through a mirror system mounted on the head coil that reflected a
projection screen behind the scanner. The luminance of the projection display was obtained
for the complete range of grayscale values by using a photometer (Photoresearch, Inc.
Spectrascan Model 650). This information was used to accurately calculate contrast when
designing stimuli.
2.4.3. Analysis—For each participant, first-order regression analyses isolated fMRI
activity related to specific task conditions, generating brain maps (termed “beta maps”) in
which each voxel contained a beta coefficient. The square of these beta coefficients
represent the amount of variance explained by the activity in that voxel during a specific
task condition. These analyses were restricted to the occipital lobe, based on the a priori
assumption that deficits in schizophrenia to these particular tasks occur in early visual
processing areas. These beta maps were used as input for higher-order group analyses and
averages. All
p
values were corrected for multiple comparisons using AlphaSim, such that
only clusters of 48 voxels or more were considered significant. For each task, two measures
were assessed. Volume of activation in milliliters was obtained for each individual based on
beta maps for each condition thresholded at
p
= .001. Strength of activation was measured as
the proportion of total variance explained by each condition obtained by squaring the beta
values and calculating the mean squared beta value over occipital cortex for values surviving
p
= .001.
Calderone et al. Page 6
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Data from the isolated-check scans included activity related to each DOM (2, 4, 8, 16, and
32%) separately for M-biased and P-biased conditions. Measures of activation volume and
strength were used as dependent variables in two separate three-way ANOVAs with group
as a between-subjects factor and DOM and bias condition as within-subjects factors. In
addition, location and direction of activation were determined by group averages of beta
coefficient maps thresholded at
p
= .001. These were displayed on a flattened anatomical
map of the occipital cortex.
Contrast sensitivity scans included four task conditions: 0.5 and 4 c/deg for near threshold
contrast and for high contrast conditions separately. Because this task showed stimuli on the
left or right visual field, first-order regression analysis yielded separate beta maps for left vs.
right hemispheric activation for each of the four task conditions. Volume of activation and
strength of activation were obtained for each task condition as described above, for
combined left-stimuli and right-stimuli beta maps.
3. Results
3.1. Psychophysical Contrast Sensitivity
A main effect of group (
F
(1,56) = 11.769,
p
< .005) showed that patients had lower contrast
sensitivities than did healthy controls to both the 0.5 and 4 c/deg spatial frequency
conditions (Figure 3). Further, a two-way Group × Condition interaction (
F
(1,56) = 4.632,
p
< .05) indicated that this deficit in contrast sensitivity was greater for the higher spatial
frequency as compared to the low spatial frequency condition.
3.2. Steady-State Visual Evoked Potentials
Signal-to-noise ratios (SNR) of the evoked potential response were used as the dependent
measure (Victor and Mast, 1991, Zemon et al., 1997). A significant three-way interaction for
Group × Condition × DOM (
F
(6,392) = 7.001,
p <
.001) suggested greater deficits for
patients in the M-biased than the P-biased condition, which was confirmed by a significant
two-way interaction for Group × Condition (
F
(1,56) = 14.517,
p
= .001). Post-hoc t-tests
showed that these deficits occurred at 4, 8, 16, and 32% DOM, where controls had signals
greater than the noise (Figure 4).
For controls, the M-biased condition generated an initial steep rise in response indicative of
strong contrast gain. There was response compression and the response function exhibited
the greatest SNR at 16% DOM. Patients had decreased initial gain (i.e., decreased slope
from 4 to 16% DOM) (
t
(28) = 2.587,
p <
.05) and lower maximal SNR (i.e., mean SNR for
16 and 32% DOM) (
t
(28) = 3.335,
p <
.005) compared to controls. For both controls and
patients, the P-biased condition generated a steadily rising response function. No group
differences in SNR were found for any DOM where the signal was greater than the noise.
Initial gain (
t
(28) = 1.091,
p >
.28) and maximal SNR (
t
(28) = 1.815,
p >
.08) measures did
not significantly differ between groups.
3.3. Functional Magnetic Resonance Imaging (fMRI): Contrast Sensitivity Task
Contrast sensitivity tasks were performed during fMRI scanning at 0.5 and 4 c/deg, at near
threshold levels for each observer and at a fixed 71% contrast. Five controls were outliers
with near threshold contrast levels above 50% and one patient was an outlier with an error in
blood oxygenation level dependent (BOLD) signal acquisition, and they were removed from
all analyses. Controls and patients had similar contrast sensitivities for 0.5 c/deg at 32 ms (M
± SD contrast sensitivity: controls: 54.69 ± 17.03; patients: 50.70 ± 19.15) as well as for 4 c/
deg at 500 ms (M ± SD contrast sensitivity: controls: 41.43 ± 24.17; patients: 23.52 ± 21.68)
(data not graphed).
Calderone et al. Page 7
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
For volume of activation, no interactions were found, but a main effect of group indicated
greater occipital recruitment for controls than for patients (
F
(1,88) = 17.583,
p <
.001). This
difference was significant for the 0.5 c/deg condition at both near threshold contrast and at
high contrast, but not for 4 c/deg condition at either contrast level (Figure 5A). For strength
of activation, no interactions or main effects were found, indicating equivalent activation
strength for all task conditions and both groups (Figure 5B).
3.4. Functional Magnetic Resonance Imaging (fMRI): Isolated Check Task
Volume and strength of activation were assessed for each bias condition and DOM (Figure
6). In addition, location of activation is represented in group-averages displayed on flattened
anatomical images of the occipital cortex, for each bias condition and DOM (Figure 7). For
both volume and strength of activation measures, there was no significant main effect of
group and no significant interactions containing group in a three-way ANOVA with group
as a between subjects factor and condition (M- vs. P-biased) and DOM as within subjects
factors. Collapsed across groups, there was a significant Condition × DOM interaction
(
F
(4,290) = 3.192,
p <
.05) for activation volume. While both the M- and P-biased
conditions showed a steep increase in volume of occipital activation from 2 to 8% DOM,
and a decrease in activation volume from 8 to 16% DOM across groups, the P-biased
condition yielded an increase in activation volume from 16 to 32% DOM whereas the M-
biased condition did not (Table 1). For strength of activation, a main effect of DOM
(
F
(4,295) = 17.280,
p <
.001) was found. The pattern of activation strength as DOM
increased was strikingly different from the pattern of activation volume. Activation strength
remained low over the 2 to 8% DOM range, and only increased at 16 and 32% DOM (Table
1).
Group-averages displayed on flattened maps of occipital cortex (corrected
p =
.001) showed
differences in the location and direction of activation for different DOMs (Figure 7). For
both bias conditions, the activation maps showed that the foveal representation was activated
at all DOMs except 2%, where there was little activation. In addition, parafoveal areas
showed positive activation (i.e. increased activation relative to rest) at 4 and 8% DOM, but
negative activation (i.e. decreased activation relative to rest) at 16 and 32% DOM.
4. Discussion
This study investigated the cortical regions associated with contrast response deficits in the
visual system in schizophrenia. An ssVEP paradigm utilizing contrast stimuli (Zemon et al.,
1988, Zemon and Gordon, 2006) has previously revealed such deficits (Butler et al., 2001,
Butler et al., 2005, Butler et al., 2008a, Butler et al., 2009), as have psychophysical contrast
sensitivity tasks (Slaghuis, 1998, Kéri et al., 2002, Butler et al., 2005, Butler et al., 2009,
Dias et al., 2011), but no study has localized these processes in the visual cortex of controls
or patients. The current study utilized similar stimuli as those used in these
electrophysiological and behavioral tasks in an fMRI paradigm in order to elucidate the
neural substrates involved in visual contrast processing in healthy controls and patients with
schizophrenia.
4.1. Psychophysical Contrast Sensitivity
Psychophysical contrast sensitivity results obtained with the VENUS system indicated that
schizophrenia patients had contrast sensitivity deficits to both the shorter duration low
spatial frequency and longer duration higher spatial frequency conditions, indicating that
they were unable to detect low contrast stimuli as well as controls for both conditions.
Contrast sensitivity for controls was greater for the higher spatial frequency condition, and
the deficit for patients compared to controls was also larger for this condition. These results
Calderone et al. Page 8
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
are consistent with studies showing contrast sensitivity deficits across spatial frequencies
(Slaghuis, 1998, Kéri et al., 2002), and suggest a robust and basic deficit in schizophrenia
for the perception of low contrasts.
4.2. Functional Magnetic Resonance Imaging (fMRI): Contrast Sensitivity Task
In the fMRI paradigm, contrast sensitivity tasks were performed at low spatial frequency
(0.5 c/deg) with short stimulus duration (32 ms) and at a higher spatial frequency (4 c/deg)
with long stimulus duration (500 ms). For the 0.5 c/deg condition, mean threshold contrasts
from the VENUS psychophysical paradigm (controls: 0.88%, patients: 1.22%) were close to
the lowest contrast possible in the scanner (1%), and to mean thresholds obtained during
fMRI scanning (controls: 1.83%, patients: 1.97%). On the other hand, for the 4 c/deg
condition, mean threshold contrasts from the VENUS psychophysical paradigm (controls:
0.34%, patients: 0.57%) were well below 1%. However, a lower luminance was used for the
scanner display (6.8 cd/m2) compared to the VENUS display (84 cd/m2) which increases
threshold contrasts, particularly for higher spatial frequencies (Patel, 1966, Sperling, 1970,
Peli, 1990). Indeed, the fMRI paradigm produced higher mean threshold contrasts for the 4
c/deg condition (controls: 2.41%, patients: 4.25%) than the 0.5 c/deg condition, though the
inability to obtain thresholds below 1% contrast remained a limitation.
Schizophrenia patients had lower volume of activation measures for the low spatial
frequency stimuli, at both threshold and high contrast (71%), but no significant differences
were seen between groups for the higher spatial frequency stimuli at either threshold or high
contrast. However, strength of activation measures were equivalent between groups and
across all stimulus conditions. This indicates a selective deficit in processing low spatial
frequency information for schizophrenia patients reflective of reduced volume of occipital
activation, rather than reduced activation strength. The psychophysical and fMRI measures
thus showed different deficits for schizophrenia patients. While psychophysical measures
showed a deficit in perception of low contrast across spatial frequencies, fMRI measures
showed a lack of cortical recruitment to low spatial frequencies across contrast. The loss of
cortical recruitment for the 0.5 c/deg condition at threshold in schizophrenia patients may
not be related to a deficit in psychophysical contrast sensitivity, as the same loss of
recruitment was found at high contrast. However, the deficit in volume of activation for low
spatial frequency is consistent with previous findings described below.
A recent study by Martinez and colleagues (Martinez et al., 2008) also found reduced
volume of activation to low spatial frequency (0.2–1.4 c/deg) stimuli in schizophrenia at
high (100%) and low (12%) contrast, in retinotopically defined V1 and V2. As in the current
results, this study also did not find deficits to higher spatial frequencies (3.5–4.9 c/deg) at
either contrast level. The current results extend this finding of a selective deficit in activation
volume for low spatial frequencies to even lower contrasts (1–2%). Martinez and colleagues
presented stimuli centrally and had participants press a button when a central fixation cross
dimmed. The current results show that this deficit is also present during the frequently used
psychophysical contrast sensitivity task. An even more recent study by Martinez and
colleagues (Martinez et al., 2012) found reduced fMRI activation to an attended low spatial
frequency (0.8 c/deg) compared to an attended high spatial frequency (5 c/deg) grating in
schizophrenia patients. However, no group differences were found in areas known to be
involved in feature-guided attention, suggesting that sensory processing of low spatial
frequencies is impaired in schizophrenia independent of attentional deficits. Additionally,
Calderone et al. (2012) recently found deficits in a network of cortical areas including
occipital cortex to low spatial frequency object stimuli (~6 cycles per image) in
schizophrenia. This finding indicates that low spatial frequency processing deficits are not
limited to simple grating stimuli, but also occur with complex images. These previous
Calderone et al. Page 9
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
findings and the current results indicate a robust deficit in fMRI activation to low spatial
frequency stimuli in schizophrenia, across stimulus contrast and complexity.
Neither controls nor patients showed differences in activation volume or strength between
threshold and high contrast conditions for either spatial frequency. This is in contrast to the
isolated-check paradigm, which showed dramatic changes in these ssVEP measures as
contrast increased. Isolated check stimuli were passively viewed, while the contrast
sensitivity task required responses based on stimulus location. Equivalent activation at
threshold and high contrast may indicate that the active detection of a stimulus in this task
recruits a specific volume and strength of occipital activation regardless of stimulus contrast.
4.3. Steady-State Visual Evoked Potentials
Consistent with our previous findings, schizophrenia patients showed a selective deficit in
the magnocellular- vs. the parvocellular-biased condition (Butler et al., 2001, Butler et al.,
2005, Butler et al., 2009). For the magnocellular-biased condition, healthy controls showed a
steep initial increase in response as contrast increased over the low contrast range, followed
by a peak SNR when contrast reached 16%. Patients showed a less steep initial increase in
response, indicative of reduced signal amplification to low contrasts. In addition, maximal
SNRs were lower for patients. For the parvocellular-biased condition, both groups
demonstrated a linear increase in response with a shallow slope over the full range of
contrasts, with no group differences for responses out of the noise (SNR > 1). The shape of
the curves is consistent with nonlinear gain in the magnocellular-biased condition and
supports previous studies (Butler et al., 2001, Butler et al., 2005, Zemon and Gordon, 2006,
Butler et al., 2007, Butler et al., 2008a, Green et al., 2009) that show schizophrenia is
associated with specific deficits in contrast gain.
Our previous ssVEP results (Butler et al., 2001, Butler et al., 2005, Butler et al., 2009) were
obtained utilizing a VENUS system (Neuroscientific Corp., Farmingdale, NY) for stimulus
presentation, VEP recording, and analysis, which is no longer manufactured. The current
results were obtained utilizing a recently developed Neucodia system (VeriSci Corp.,
Raritan, NJ) which includes the feature of synchronized data collection, utilizes modern
equipment, and provides ease of use. Thus, this deficit in contrast gain seen in the ssVEP
paradigm is robust across assessment systems and cohorts of patients.
4.4. Functional Magnetic Resonance Imaging: Isolated-Check Task
Similar stimuli to those used in the ssVEP paradigm were shown in the fMRI task in order to
localize the ssVEP response to specific visual cortical areas. Across groups and for both
magnocellular- and parvocellular-biased stimuli, a general pattern emerged in which
increasingly greater volumes of occipital cortex were recruited as contrast increased through
the low contrast range, while higher contrasts showed reduced volume of activation (Figure
6A). Conversely, strength of activation remained low over the low contrast range, and
increased dramatically at higher contrasts (Figure 6B). These fMRI results conflict with the
ssVEP results, since they show similar patterns for magnocellular- and parvocellular-biased
conditions, as well as for controls and patients. Likewise, the lack of increase in fMRI
activation strength over the low contrast region conflicts with previous single-cell work,
which has shown increases in the firing rates of individual cells as contrast increases
(Kaplan and Shapley, 1986, Shapley, 1990). This indicates a dissociation between ssVEP
and fMRI for this task, and suggests that patients may recruit the same occipital areas with
the same amount of metabolic energy as controls, but are not able to utilize these areas for
enhanced contrast gain.
Calderone et al. Page 10
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Several studies have shown correspondence between negative VEP deflections and positive
BOLD activity under certain stimulus conditions (Whittingstall et al., 2007, Whittingstall et
al., 2008, Yesilyurt et al., 2010). Other studies, however, have shown VEP and BOLD
responses to be unrelated for particular stimuli (Janz et al., 2001), and that pharmacological
treatment can reduce BOLD activity without affecting VEP responses (Seaquist et al., 2007).
Di Russo and colleagues (Di Russo et al., 2007) recently demonstrated both correspondence
and dissociation between ssVEP and BOLD responses, such that some areas of occipital
BOLD activation were shown to contribute to ssVEP response, while other occipital BOLD
activation did not. The current results showing an overall pattern of similar BOLD responses
for magnocellular- and parvocellular-biased conditions across groups suggests that the
neural processes measured by ssVEPs are divergent from the broader class of processes
measured by fMRI, and that patient deficits to these particular stimuli occur at the neural
level measured by ssVEP.
4.5. Conclusions
Utilizing the CNTRICS domain of perception, this study examined the cortical
underpinnings of two tasks that utilized contrast responses and that have previously been
used to demonstrate deficits in early-stage visual perception in schizophrenia. Both
psychophysical contrast sensitivity and ssVEP measures demonstrated dysfunctional
processing of low contrasts for schizophrenia patients, while fMRI revealed only a deficit in
cortical recruitment for low spatial frequencies. The similarity between magnocellular- and
parvocellular-biased fMRI responses to the isolated-check paradigm, as well as equivalent
fMRI activation to high and low contrast in the contrast sensitivity paradigm, indicate that
these ssVEP and psychophysical techniques are of greater assessment value for the contrast
deficits explored. Further work is required to investigate the neural correlates of contrast
response deficits in schizophrenia.
References
Ardekani BA, Bachman AH, Strother SC, Fujibayashi Y, Yonekura Y. Impact of inter-subject image
registration on group analysis of fMRI data. International Congress Series. 2004; 1265:49–59.
Barch DM, Carter CS, Dakin SC, Gold J, Luck SJ, Macdonald A 3rd, Ragland JD, Silverstein S,
Strauss ME. The clinical translation of a measure of gain control: the contrast-contrast effect task.
Schizophrenia Bulletin. 2012; 38:135–143. [PubMed: 22101963]
Benardete EA, Kaplan E, Knight BW. Contrast gain control in the primate retina: P cells are not X-
like, some M cells are. Visual Neuroscience. 1992; 8:483–486. [PubMed: 1586649]
Butler PD, Abeles IY, Weiskopf NG, Tambini A, Jalbrzikowski M, Legatt ME, Zemon V, Loughead J,
Gur RC, Javitt DC. Sensory contributions to impaired emotion processing in schizophrenia.
Schizophrenia Bulletin. 2009; 35:1095–1107. [PubMed: 19793797]
Butler PD, Chen Y, Ford JM, Geyer MA, Silverstein SM, Green MF. Perceptual measurement in
schizophrenia: promising electrophysiology and neuroimaging paradigms from CNTRICS.
Schizophrenia Bulletin. 2012; 38:81–91. [PubMed: 21890745]
Butler PD, Martinez A, Foxe JJ, Kim D, Zemon V, Silipo G, Mahoney J, Shpaner M, Jalbrzikowski M,
Javitt DC. Subcortical visual dysfunction in schizophrenia drives secondary cortical impairments.
Brain. 2007; 130:417–430. [PubMed: 16984902]
Butler PD, Schechter I, Zemon V, Schwartz SG, Greenstein VC, Gordon J, Schroeder CE, Javitt DC.
Dysfunction of early-stage visual processing in schizophrenia. American Journal of Psychiatry.
2001; 158:1126–1133. [PubMed: 11431235]
Butler PD, Silverstein SM, Dakin SC. Visual perception and its impairment in schizophrenia.
Biological Psychiatry. 2008a; 64:40–47. [PubMed: 18549875]
Butler PD, Tambini A, Yovel G, Jalbrzikowski M, Ziwich R, Silipo G, Kanwisher N, Javitt DC.
What’s in a face? Effects of stimulus duration and inversion on face processing in schizophrenia.
Schizophrenia Research. 2008b; 103:283–292. [PubMed: 18450426]
Calderone et al. Page 11
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Butler PD, Zemon V, Schechter I, Saperstein AM, Hoptman MJ, Lim KO, Revheim N, Silipo G, Javitt
DC. Early-stage visual processing and cortical amplification deficits in schizophrenia. Archives of
General Psychiatry. 2005; 62:495–504. [PubMed: 15867102]
Calderone DJ, Hoptman MJ, Martinez A, Nair-Collins S, Mauro CJ, Bar M, Javitt DC, Butler PD.
Contributions of Low and High Spatial Frequency Processing to Impaired Object Recognition
Circuitry in Schizophrenia. Cerebral Cortex. 2012 In Press.
Carandini M, Heeger DJ, Movshon JA. Linearity and normalization in simple cells of the macaque
primary visual cortex. J Neurosci. 1997; 17:8621–8644. [PubMed: 9334433]
Chen Y, Levy DL, Sheremata S, Holzman PS. Compromised late-stage motion processing in
schizophrenia. Biological Psychiatry. 2004:55.
Chen Y, Levy DL, Sheremata S, Nakayama K, Matthysse S, Holzman PS. Effects of typical, atypical,
and no antipsychotic drugs on visual contrast detection in schizophrenia. The American Journal of
Psychiatry. 2003a; 160:1795–1801. [PubMed: 14514493]
Chen Y, Nakayama K, Levy D, Matthysse S, Holzman P. Processing of global, but not local, motion
direction is deficient in schizophrenia. Schizophrenia Research. 2003b; 61:215–227. [PubMed:
12729873]
Cox RW. AFNI: software for analysis and visualization of functional magnetic resonance
neuroimages. Comput Biomed Res. 1996; 29:162–173. [PubMed: 8812068]
Di Russo F, Pitzalis S, Aprile T, Spitoni G, Patria F, Stella A, Spinelli D, Hillyard SA. Spatiotemporal
analysis of the cortical sources of the steady-state visual evoked potential. Hum Brain Mapp. 2007;
28:323–334. [PubMed: 16779799]
Dias EC, Butler PD, Hoptman MJ, Javitt DC. Early sensory contributions to contextual encoding
deficits in schizophrenia. Archives of General Psychiatry. 2011; 68:654–664. [PubMed:
21383251]
Doniger GM, Foxe JJ, Murray MM, Higgins BA, Javitt DC. Impaired visual object recognition and
dorsal/ventral stream interaction in schizophrenia. Archives of General Psychiatry. 2002; 59:1011–
1020. [PubMed: 12418934]
First, MB.; Spitzer, RL.; Gibbon, M.; Williams, JBW. Structured Clinical Interview for DSM-IV Axis
I Disorders. New York: New York State Psychiatric Institute; 1997.
Green MF, Butler PD, Chen Y, Geyer MA, Silverstein S, Wynn JK, Yoon JH, Zemon V. Perception
measurement in clinical trials of schizophrenia: Promising paradigms from CNTRICS.
Schizophrenia Bulletin. 2009; 35:163–181. [PubMed: 19023123]
Janz C, Heinrich SP, Kornmayer J, Bach M, Hennig J. Coupling of neural activity and BOLD fMRI
response: new insights by combination of fMRI and VEP experiments in transition from single
events to continuous stimulation. Magn Reson Med. 2001; 46:482–486. [PubMed: 11550239]
Jasper HH. The ten twenty electrode system of the International Federation. Electroencephalography
Journal. 1958; 10:371–375.
Javitt DC. When doors of perception close: Bottom-up models of disrupted cognition in schizophrenia.
The Annual Review of Clinical Psychology. 2009; 5:249–275.
Kaplan, E. The M,P, and K pathways of the primate visual system. In: Chalupa, L.; Werner, J., editors.
The Visual Neurosciences. Cambridge, Mass: MIT Press; 2003.
Kaplan E, Shapley RM. X and Y cells in the lateral geniculate nucleus of macaque monkeys. The
Journal of Physiology. 1982; 330:125–143. [PubMed: 7175738]
Kaplan E, Shapley RM. The primate retina contains two types of ganglion cells, with high and low
contrast sensitivity. Proceedings of the National Academy of Sciences USA. 1986; 83:2755–2757.
Kéri S, Antal A, Szekeres G, Benedek G, Janka Z. Spatiotemporal visual processing in schizophrenia.
Journal of Clinical Neuroscience. 2002; 14:190–196.
Kéri S, Kelemen O, Benedek G, Janka Z. Vernier threshold in patients with schizophrenia and in their
unaffected siblings. Neuropsychology. 2004; 18:537–542. [PubMed: 15291731]
Kim D, Wylie G, Pasternak R, Butler PD, Javitt DC. Magnocellular contributions to impaired motion
processing in schizophrenia. Schizophrenia Research. 2006; 82:1–8. [PubMed: 16325377]
Kiss I, Fabian A, Benedek G, Keri S. When doors of perception open: visual contrast sensitivity in
never-medicated, first-episode schizophrenia. Journal of Abnormal Psychology. 2010; 119:586–
593. [PubMed: 20677847]
Calderone et al. Page 12
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Klein A, Andersson J, Ardekani BA, Ashburner J, Avants B, Chiang MC, Christensen GE, Collins DL,
Gee J, Hellier P, Song JH, Jenkinson M, Lepage C, Rueckert D, Thompson P, Vercauteren T,
Woods RP, Mann JJ, Parsey RV. Evaluation of 14 nonlinear deformation algorithms applied to
human brain MRI registration. Neuroimage. 2009; 46:786–802. [PubMed: 19195496]
Koychev I, El-Deredy W, Deakin JF. New visual information processing abnormality biomarker for
the diagnosis of Schizophrenia. Expert Opin Med Diagn. 2011; 5:357–368. [PubMed: 22003364]
Kurylo DD, Pasternak R, Silipo G, Javitt DC, Butler PD. Perceptual organization by proximity and
similarity in schizophrenia. Schizophrenia Research. 2007; 95:205–214. [PubMed: 17681736]
Legge GE. Sustained and transient mechanisms in human vision: temporal and spatial properties.
Vision Research. 1978; 18:69–81. [PubMed: 664278]
Leitman DI, Wolf DH, Laukka P, Ragland JD, Valdez JN, Turetsky BI, Gur RE, Gur RC. Not pitch
perfect: sensory contributions to affective communication impairment in schizophrenia. Biological
Psychiatry. 2011; 70:611–618. [PubMed: 21762876]
Martinez A, Hillyard SA, Bickel S, Dias EC, Butler PD, Javitt DC. Consequences of magnocellular
dysfunction on processing attended information in schizophrenia. Cerebral Cortex. 2012; 22:1282–
1293. [PubMed: 21840846]
Martinez A, Hillyard SA, Dias EC, Hagler DJ, Butler PD, Guilfoyle DN, Jalbrzikowski M, Silipo G,
Javitt DC. Magnocellular pathway impairment in schizophrenia: Evidence from functional
magnetic resonance imaging. The Journal of Neuroscience. 2008; 28:7492–7500. [PubMed:
18650327]
Merigan WH, Maunsell JHR. How parallel are the primate visual pathways? Annu Rev Neurosci.
1993; 16:369–402. [PubMed: 8460898]
Norman J. Two visual systems and two theories of perception: An attempt to reconcile the
constructivist and ecological approaches. Behavioral and Brain Sciences. 2002; 25:73–144.
[PubMed: 12625088]
Norton D, McBain R, Holt DJ, Ongur D, Chen Y. Association of impaired facial affect recognition
with basic facial and visual processing deficits in schizophrenia. Biological Psychiatry. 2009;
65:1094–1098. [PubMed: 19268917]
O’Donnell BF, Potts GF, Nestor PG, Stylianopoulos KC, Shenton ME, McCarley RW. Spatial
frequency discrimination in schizophrenia. Jounral of Abnormal Psychology. 2002; 111:620–625.
Ohzawa I, Sclar G, Freeman RD. Contrast gain control in the cat visual cortex. Nature. 1982; 298:266–
268. [PubMed: 7088176]
Ohzawa I, Sclar G, Freeman RD. Contrast gain control in the cat’s visual system. Journal of
Neurophysiology. 1985; 54:651–667. [PubMed: 4045542]
Patel AS. Spatial resolution by the human visual system. The effect of mean retinal illuminance. J Opt
Soc Am. 1966; 56:689–694. [PubMed: 5963523]
Peli E. Contrast in complex images. J Opt Soc Am A. 1990; 7:2032–2040. [PubMed: 2231113]
Revheim N, Butler PD, Schechter I, Jalbrzikowski M, Silipo G, Javitt DC. Reading impairment and
visual processing deficits in schizophrenia. Schizophrenia Research. 2006; 87:238–245. [PubMed:
16890409]
Seaquist ER, Chen W, Benedict LE, Ugurbil K, Kwag JH, Zhu XH, Nelson CA. Insulin reduces the
BOLD response but is without effect on the VEP during presentation of a visual task in humans. J
Cereb Blood Flow Metab. 2007; 27:154–160. [PubMed: 16639425]
Sehatpour P, Dias EC, Butler PD, Revheim N, Guilfoyle DN, Foxe JJ, Javitt DC. Impaired visual
object processing across an occipital- frontal-hippocampal brain network in schizophrenia: An
integrated neuroimaging study. Archives of General Psychiatry. 2010; 67:772–782. [PubMed:
20679585]
Shapley R. Visual sensitivity and parallel retinocortical channels. The Annual Review of Psychology.
1990; 41:635–658.
Shapley R, Victor JD. The contrast gain control of the cat retina. Vision Research. 1979; 19:431–434.
[PubMed: 473613]
Silverstein SM, Keane BP. Perceptual organization impairment in schizophrenia and associated brain
mechanisms: review of research from 2005 to 2010. Schizophrenia Bulletin. 2011; 37:690–699.
[PubMed: 21700589]
Calderone et al. Page 13
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Slaghuis WL. Contrast sensitivity for stationary and drifting spatial frequency gratings in positive- and
negative-symptom schizophrenia. Journal of Abnormal Psychology. 1998; 107:49–62. [PubMed:
9505038]
Slaghuis WL. Spatio-temporal luminance contrast sensitivity and visual backward masking in
schizophrenia. Experimental Brain Research. 2004; 156:196–211.
Sperling G. Model of visual adaptation and contrast detection. Perception & Psychophysics. 1970;
8:143–157.
Tolhurst DJ. Reaction times in the detection of gratings by human observers: a probabilistic
mechanism. Vision Research. 1975; 15:1143–1149. [PubMed: 1166615]
Tootell RBH, Hamilton SL, Switkes E. Functional anatomy of macaque striate cortex. IV. Contrast
and magno-parvo streams. The Journal of Neuroscience. 1988; 8:1594–1609. [PubMed: 3367212]
Turetsky BI, Kohler CG, Indersmitten T, Bhati MT, Charbonnier D, Gur RC. Facial emotion
recognition in schizophrenia: when and why does it go awry? Schizophrenia Research. 2007;
94:253–263. [PubMed: 17583481]
Victor JD, Mast J. A new statistic for steady-state evoked potentials. Electroencephalogr Clin
Neurophysiol. 1991; 78:378–388. [PubMed: 1711456]
Wetherill GB, Levitt H. Sequential Estimation of Points on a Psychometric Function. Br J Math Stat
Psychol. 1965; 18:1–10. [PubMed: 14324842]
Whittingstall K, Stroink G, Schmidt M. Evaluating the spatial relationship of event-related potential
and functional MRI sources in the primary visual cortex. Hum Brain Mapp. 2007; 28:134–142.
[PubMed: 16761265]
Whittingstall K, Wilson D, Schmidt M, Stroink G. Correspondence of visual evoked potentials with
FMRI signals in human visual cortex. Brain Topogr. 2008; 21:86–92. [PubMed: 18841455]
Woods SW. Chlorpromazine equivalent doses for the newer atypical antipsychotics. J Clin Psychiatry.
2003; 64:663–667. [PubMed: 12823080]
Woods SW. Calculation of CPZ Equivalents. 2005; 2012
Woods SW. Chlorpromazine Equivalent Doses for the Newer Atypical Antipsychotics. 2011; 2012
Wurtz, RH.; Kandel, ER. Central Visual Pathways. In: Kandel, ER., et al., editors. Principles of Neural
Science. New York, NY: McGraw-Hill; 2000. p. 523-545.
Yesilyurt B, Whittingstall K, Ugurbil K, Logothetis NK, Uludag K. Relationship of the BOLD signal
with VEP for ultrashort duration visual stimuli (0.1 to 5 ms) in humans. J Cereb Blood Flow
Metab. 2010; 30:449–458. [PubMed: 19844243]
Zemon V, Eisner W, Gordon J, Grose-Fifer J, Tenedios F, Shoup H. Contrast-dependent responses in
the human visual system: childhood through adulthood. Int J Neurosci. 1995; 80:181–201.
[PubMed: 7775048]
Zemon V, Gordon J. Luminance-contrast mechanisms in humans: Visual evoked potentials and a
nonlinear model. Vision Research. 2006; 46:4163–4180. [PubMed: 16997347]
Zemon V, Gordon J, Welch J. Asymmetries in ON and OFF visual pathways of humans revealed using
contrast-evoked cortical potentials. Visual Neuroscience. 1988; 1:145–150. [PubMed: 3154786]
Zemon V, Hartmann EE, Gordon J, Prunte-Glowazki A. An electrophysiological technique for
assessment of the development of spatial vision. Optom Vis Sci. 1997; 74:708–716. [PubMed:
9380368]
Calderone et al. Page 14
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Highlights
Previous findings of psychophysical contrast sensitivity deficits were replicated.
MRI deficits to only some of the contrast sensitivity stimuli were found.
Deficits in electrophysiological measures of visual gain control were replicated.
Normal recruitment of visual cortical areas was seen in a corresponding fMRI task.
Psychophysical, electrophysiological, and fMRI measures were dissociated.
Calderone et al. Page 15
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Figure 1.
Contrast sensitivity task used in the fMRI paradigm. Participants indicated whether stimuli
appeared on the left or right side of the screen. A. Low spatial frequency condition: stimuli
were presented at 0.5 c/deg for approximately 32 ms. B. High spatial frequency condition:
stimuli were presented at 4 c/deg for approximately 500 ms.
Calderone et al. Page 16
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Figure 2.
Sinusoidal modulation of isolated check stimuli used in the ssVEP and fMRI paradigms.
Stimuli were sinusoidally modulated at ~12 Hz, such that each cycle through contrast levels
shown in A and B occurred ~12 times per second. A. Magnocellular-biased condition:
pedestal around which contrast was modulated equaled the depth of modulation, resulting in
appearance/disappearance stimuli. B. Parvocellular-biased condition: pedestal around which
contrast was modulated equaled 48%, so that contrast remained high during modulation.
Calderone et al. Page 17
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Figure 3.
Psychophysical contrast sensitivities obtained with the VENUS system, displayed on a log
base 10 scale. Error bars show standard error. *
p <
.05, **
p <
.01.
Calderone et al. Page 18
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Figure 4.
Contrast response functions for the ssVEP paradigm. Error bars show standard error. *
p <
.
05.
Calderone et al. Page 19
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Figure 5.
fMRI measures obtained during the contrast sensitivity task for 0.5 and 4 c/deg at either near
threshold contrast or at high contrast. A. Volume of activation measured in milliliters. B.
Strength of activation measured as first-order regression betas squared (percent of total
variance). Error bars show standard error. *
p <
.05.
Calderone et al. Page 20
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Figure 6.
fMRI measures for isolated check stimuli calculated for occipital cortex. A. Volume of
activation measured in microliters. B. Strength of activation measured as first-order
regression beta squared (variance). Error bars show standard error.
Calderone et al. Page 21
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
Figure 7.
Flattened anatomical maps of occipital cortex showing direction and location of fMRI
activation to isolated check stimuli. Each map shows a group-average of first-order
regression beta values thresholded at
p =
.001. No differences between hemispheres were
found for any condition at
p =
.001, and thus right hemisphere maps are shown here as
representative of all occipital activation.
Calderone et al. Page 22
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
$watermark-text $watermark-text $watermark-text
$watermark-text $watermark-text $watermark-text
Calderone et al. Page 23
Table 1
fMRI Isolated check task: Differences in occipital activation volume and activation strength across groups as depth of modulation increases in the M-
biased and P-biased conditions.
Volume of Activation Activation Strength
Condition Depth of Modulation t(29) p t(29) p
M-biased 2% vs. 4% −4.033 <0.001* 4.028 <0.001*
M-biased 4% vs. 8% −3.135 <0.005* −1.653 0.109
M-biased 8% vs. 16% 5.639 <0.001* −5.862 <0.001*
M-biased 16% vs. 32% −0.393 0.697 −1.751 0.091
P-biased 2% vs. 4% −6.603 <0.001* 1.798 0.083
P-biased 4% vs. 8% −2.217 <0.050* 0.854 0.400
P-biased 8% vs. 16% 8.050 <0.001* −4.868 <0.001*
P-biased 16% vs. 32% −3.611 <0.005* −2.396 <0.050*
Neuroimage
. Author manuscript; available in PMC 2014 February 15.
