Assessing Stimulus Control in a Discrimination Task with Compound Stimuli: Evaluating Testing Procedures and Tracking Eye Fixations

Abstract

Experiments with pigeons have suggested that the way stimuli are arranged in tests affects the demonstration of the stimulus control established during training. The present study aimed to replicate these findings with humans exposed to a simultaneous-discrimination task with compound stimuli. Adults were exposed to a discrimination task, and their eye fixations were recorded. Two compound stimuli were used: a triangle and a red circle, and a square and a green circle. During Phase 1, responses to the first compound were rein- forced, and during Phase 2, these contingencies were reversed. Following training in each phase, the components of the stimulus compound were separated and presented across different tests to assess stimulus control by each stimulus component. Participants tended to choose the component on which their eyes had most frequently fixated during training. How- ever, the S+ component that was associated with fewer fixations also controlled participants’ choices. Results on tests replicate previous findings with pigeons. Possible effects of peripheral vision are discussed.

Full text

1 23
The Psychological Record
ISSN 0033-2933
Volume 65
Number 1
Psychol Rec (2015) 65:83-88
DOI 10.1007/s40732-014-0092-1
Assessing Stimulus Control in a
Discrimination Task with Compound
Stimuli: Evaluating Testing Procedures and
Tracking Eye Fixations
William F.Perez, Peter Endemann,
Candido V.B.B.Pessôa & Gerson
Y.Tomanari

1 23
Your article is protected by copyright and all
rights are held exclusively by Association for
Behavior Analysis International. This e-offprint
is for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

ORIGINAL ARTICLE
Assessing Stimulus Control in a Discrimination Task
with Compound Stimuli: Evaluating Testing Procedures
and Tracking Eye Fixations
William F. Perez &Peter Endemann &
Candido V. B. B. Pessôa &Gerson Y. Tomanari
Published online: 27 August 2014
#Association for Behavior Analysis International 2014
Abstract Experiments with pigeons have suggested that the
way stimuli are arranged in tests affects the demonstration of
the stimulus control established during training. The present
study aimed to replicate these findings with humans exposed
to a simultaneous-discrimination task with compound stimuli.
Adults were exposed to a discrimination task, and their eye
fixations were recorded. Two compound stimuli were used: a
triangle and a red circle, and a square and a green circle.
During Phase 1, responses to the first compound were rein-
forced, and during Phase 2, these contingencies were reversed.
Following training in each phase, the components of the
stimulus compound were separated and presented across dif-
ferent tests to assess stimulus control by each stimulus com-
ponent. Participants tended to choose the component on which
their eyes had most frequently fixated during training. How-
ever, the S+ component that was associated with fewer fixa-
tions also controlled participants’choices. Results on tests
replicate previous findings with pigeons. Possible effects of
peripheral vision are discussed.
Keywords Attention .Compound stimulus .Eye
movements .Humans .Simultaneous discrimination
Over 50 years ago, Reynolds (1961) conducted a heuristic
experiment that investigated “attention”in pigeons. In that
paper, he stated that “every part of the environment that is
present when a reinforced response occurs may not subse-
quently be an occasion for the emission of that response”(p.
208). Attention can be described as the controlling relation
between a specific part of the environment and the evoked
response (Skinner 1953). In Reynolds’original experiment,
two pigeons were trained on a discrimination task with com-
pound stimuli that had components that could be presented
separately. During subsequent tests, the stimuli were separated
and their components displayed individually to evaluate
which one had acquired control over pecking (Reynolds
1961). The research verified that both pigeons responded to
only one component of the stimulus related to reinforcement
(S+).
1
Subsequent experiments have replicated Reynolds (1961;
e.g., Birkimer 1969;BornandPeterson1969;Farthingand
Hearst 1970; Johnson and Cumming 1968; Kendall and Mills
1979; Reynolds and Limpo 1969; Wilkie and Masson 1976).
A major finding throughout these studies has been that differ-
ent results may be observed depending on the kind of test
used. For example, by varying the stimulus arrangements (see
below), either of the two S+ components to some degree could
control test responses. In Kendall and Mills (1979), response
rates evoked by the S+ decreased as the apparently noncon-
trolling S+ component was withdrawn. Wilkie and Masson
(1976) showed that the noncontrolling S+ component, as
determined by Reynolds’test, acquired control over
responding faster than any S- component in a resistance-to-
reinforcement test. Furthermore, Farthing and Hearst (1970)
found that recombining S+ and S- components during tests
may reveal control by both S+ components, whereas testing
with individually presented components (cf. Reynolds 1961)
may not. According to this set of results, whether stimulus
control is evident after discrimination training is partly
1
The notation S+ and S- hereafter will be used in reference to the
programmed contingency. Thus, even if a programmed S+ component
does not control responses during tests, it will be referred to as an S+
component.
W. F. Pe rez :P. Endemann :C. V. B. B. Pessôa :G. Y. Tomanari
University of São Paulo, São Paulo, Brazil
Present Address:
W. F. Pe rez (*):C. V. B. B. Pessôa
Núcleo Paradigma de Análise do Comportamento, Rua Vanderley,
611, Perdizes, São Paulo, SP CEP: 05011-001, Brazil
e-mail: will.f.perez@gmail.com
Psychol Rec (2015) 65:83–88
DOI 10.1007/s40732-014-0092-1
Author's personal copy

influenced by how that control is assessed; in other words, the
absence of control in one setting does not necessarily indicate
that control will be absent in another setting. Therefore, of
major importance to the study of “attention”is how to mea-
sure stimulus control over behavior, particularly control by the
S+ components of compound stimuli.
All of the experiments described above used pigeons as
subjects. They developed a range of procedures to assess the
stimulus control established by typical discrimination tasks
with compound stimuli. This study aimed to replicate these
findings by exposing human participants to discrimination
training with compound stimuli and to evaluate the results
obtained from different stimulus-control assessment proce-
dures. First, human participants were exposed to a simulta-
neous discrimination task. Once discrimination training end-
ed, several tests were used to assess stimulus control. During
these tests, the components of the compound stimuli were
separated and/or recombined to form novel stimuli, and the
participants’choices were measured.
In addition to measures of responding during tests, a sec-
ond (but not less important) approach to inferring stimulus
control was used for this study. Previous research has sug-
gested that the frequency and duration of participants’eye
fixations to discriminative stimuli during training can be used
as a measure of stimulus-control relations (Dube et al. 1999,
2003,2006,2010;Magnusson2002;Perez2008;Pessôaetal.
2009; Schroeder 1969,1970,1997). In studies like Dube et al.
(2010), subjects that presented with overselectivity (i.e., re-
stricted stimulus control) also failed to observe all of the
relevant stimuli in the discrimination task. Schroeder (1969,
1970; see also Pessôa et al. 2009) found that participants in a
discrimination task with two programmed S+ stimuli eventu-
ally moved their eyes toward and responded to only one S+.
Magnusson (2002)andPerez(2008) showed that, in a
matching-to-sample task, correct matching responses at times
occur after fixating on only the sample and S- comparison.
Considering these findings, this study also tracked partici-
pants’eye fixations during discrimination training in order
to evaluate the coherence between eye-fixation measures and
participants’choices during tests.
Method
Participants
Three undergraduate students were personally recruited for
this study. Each student read and signed an informed consent
document that described general information about the exper-
iment. The participants were naïve with respect to the task and
the apparatus and declared having normal vision. Once the
experimental session was over, the participants were debriefed
and the researchers clarified all their questions about the
experiment.
Setting, Equipment, and Stimuli
Sessions were held in a 2 m×3 m room. The room was divided
in two by a partition wall. On one side of the partition, there
was a chair anda table with a computer monitor and keyboard.
A Macintosh Performa 5215CD using MTS v.11.6.7 software
(Dube and Hiris 1999) conducted the stimulus presentations,
delivered the consequences, and recorded the participant’s
choices. On the other side of the partition was the equipment
used by the experimenter. The apparatus used for recording
eye movements was a RK-826PCI Pupil/Corneal Reflection
Tracking Hardware System with a precision of .3 deg in a
visual field of 20 deg×20 deg. This equipment permitted free
movements of the head. The study also used a RK-630 Auto-
Calibration System, installed on a PC platform with ISCAN
Raw Movement Data Acquisition software. Images were cap-
tured in another PC platform with Pinnacle Studio Plus 9®
software and analyzed frame by frame (30 Hz) with Video
Frame Coder software (Abilities Software, Sudbury, MA,
USA).
A minimum viewing distance of 60 cm was maintained
between the participant’s eye and the monitor. As displayed in
Fig. 1, the stimulus components used were a triangle, a red
circle, a square, and a green circle. The maximum stimulus-
component size was .5 deg×5 deg (1 deg×1 deg for Test 6).
During training, the distance between components displayed
in the same presentation area was 7.6 deg. During training and
tests, the distance center-to-center for the presentation areas
was 20.8 deg. The maximum size of each presentation area
was 9.5 deg×9.5 deg.
Procedure
The experiment took place in two phases (Phase 1 and Phase
2) immediately following one another. Each phase involved a
discrimination-training task and six different tests. During
Phase 1 discrimination training, the S+ was a white triangle
and a red circle, and the S- was a white square and a green
circle. During Phase 2 discrimination training, these contin-
gencies were reversed. The compound stimuli used during
training in both phases are presented in Fig. 1.
Before starting discrimination training, the partici-
pants were told that correct choices would be followed
by a beep, and incorrect choices would be followed by
a dissonant sound. Thus, their aim was to produce as
many beeps as they could. Each trial began with a
simultaneous presentation of the S+ and S-, one in each
upper corner of the presentation area. A mouse click on
the S+ produced a short “beep”sound, and a mouse
click on the S- produced a dissonant sound. After either
84 Psychol Rec (2015) 65:83–88
Author's personal copy

sound, the stimuli were withdrawn and a .5 s intertrial
interval (ITI) was initiated. Training ended when partic-
ipants emitted 12 responses in the programmed S+ in a
block of 12 trials. If this criterion was not achieved,
another 12-trial block was run. Tests started immediately
after training and consisted of 12 trials each.
Before the tests, the participants were told that no
feedback would follow their responses, but the computer
would keep recording their choices as correct or incor-
rect. Each test was comprised of 12 trials. Across trials,
mouse clicks were followed only by the withdrawal of
the stimuli and the ITI. During tests, stimuli were pre-
sented separately (Tests 1 to 5) or recombined (Test 6;
see the x-axis of Figs. 2,3and 4). In Test 1, all com-
ponents were separated and presented simultaneously.
Test 1 assessed which component the participants would
choose more frequently. In Tests 2–5, one of the two S+
components was simultaneously presented against one of
the two S- components. Tests 2–5 evaluated if choices
would occur to both S+ components. In Test 6, each S+
component was combined with an S- component to form
two novel compound stimuli: a green triangle (see the
lower right panel of Fig. 2) and a red square. This test
assessed which S+ component the participants would
chose more frequently when combined with an S-
component.
During Tests 2–6, components were presented on the
upper corners of the screen. The exception was Test 1, in
which one of the four components appeared in each
corner. When a component was presented alone, it was
displayed on the center of the presentation area, as shown
in the right panel of Fig. 1. During training and testing,
stimuli were all presented an equal number of times in
each presentation area, but they were not presented in the
same presentation area for more than three times
consecutively.
Analysis of Eye-Tracking Data
Two independent observers analyzed eye-tracking data. A
fixation was coded when the compound stimuli were
displayed and the point-of-gaze cursor generated by the
eye-tracking device remained still on any part of any
component of the compound stimuli for more than 1/30
of a second (i.e., a fixation was never recorded when the
eye was just moving from one component to another).
Observers evaluated the total frequency and duration of
fixation on each component of the compound stimuli. The
lowest value was divided by the highest. The intercoder
agreement ranged from 0.93 to 1.00.
Fig. 1 On the upper left panel, the components of the compound stimuli
used during training are displayed. The lower left panel shows an exam-
ple of a training trial with compound stimuli displayed on the computer
screen. On the right panel, two examples of test trials are depicted, one
with four components presented separately (Test 1) and another with two
components (Tests 2–5 and Test 6, with components recombined as a red
square and a green triangle). The size of each component and presentation
area was actually smaller then depicted on this figure (see text for actual
measures and angular distance)
Fig. 2 Results of participant A. On the upper panel, the duration and
frequency of eye fixations for each compound component (triangle, red,
square, and green) presented during training for Phases 1 and 2; on the
lower panel, choices (using mouse click) during tests after Phases 1 and 2
in which components presented during training were separated or
recombined
Psychol Rec (2015) 65:83–88 85
Author's personal copy

Results
Participants made few mistakes on both experimental phases
during discrimination training. After a correct response to the
S+, they continued to respond correctly on the following trials.
In the discrimination training of Phase 1, Participants C and J
first responded to the S+ on the second trial, and Participant A
on the third trial. In the discrimination training of Phase 2, all
the participants responded to the S+ on the second trial.
The analysis of eye fixations during training included only
fixations that occurred after the discrimination had been
established (i.e., after the second consecutive trial with an S+
selection). Figures 2,3and 4show eye-fixations during training
and participants’choices during tests. The upper panels show the
total duration of fixation and frequency to each component of the
compound stimuli during training in Phases 1 and 2. The lower
panel shows the results of Tests 1–6 during Phases 1 and 2.
In both phases, all the participants displayed a higher
duration and frequency of fixation to one component of the
S+ when compared to the other S+ component and to the two
S- components. Two participants (A and J) fixated more on
shapes than colors (the triangle in Phase 1 and the square in
Phase 2). The other participant (C) fixated more on colors than
shapes (red in Phase 1 and green in Phase 2). For all partici-
pants, in both phases, the second highest number of fixations
and duration of fixations to the S- component corresponded to
the same dimension as the S+ component with the highest
number and duration of fixations (i.e., shape for A and J and
color for C). Repetitive fixations on shapes or colors during
training are translated into horizontal saccades followed by
eye fixations distributed on the upper or lower portion of the
compounds, respectively (see Fig. 1).
Across trials in each test, participants responded by
clicking on only one component. During Test 1, in which all
four components were presented simultaneously and separate-
ly (see the horizontal axis on the lower panel of Figs. 2,3and
4), the participants clicked on the component with the highest
number and duration of fixations during training. The only
exception to this trend was participant J in Phase 1, who
Fig. 3 Results of participant C. On the upper panel, the duration and
frequency of eye fixations for each compound component (triangle, red,
square, and green) presented during training for Phases 1 and 2; on the lower
panel, choices (using a mouse click) during tests after Phases 1 and 2 in which
components presented during training were separated or recombined
Fig. 4 Results of participant J. On the upper panel, the duration and
frequency of eye fixations for each compound component (triangle, red,
square and green) presented during training for Phases 1 and 2; on the
lower panel, choices (using a mouse click) during tests after Phases 1 and
2, in which components presented during training were separated or
recombined
86 Psychol Rec (2015) 65:83–88
Author's personal copy

fixated on the triangle most often during training, but chose
red on the test. Across Tests 2–5, in which one S+ component
was presented with an S- component, participants always
chose the S+, even when the S+ component displayed was
not the one with the most fixations during training. In Test 6,
in which components were recombined, participants chose the
component with the most fixations during training, except for
participant J, who chose red instead of the triangle in Phase 1.
Discussion
Previous studies with pigeons have suggested that measuring
stimulus control in discrimination tasks with compound stimuli
might require further testing to evaluate whether one or more
individual components of the compound stimulus acquired con-
trol over responses. According to this literature, some stimulus-
control relations might only be revealed when different types of
tests are carried out (Farthing and Hearst 1970; Kendall and Mills
1979; Reynolds 1961; Wilkie and Masson 1976). The present
study replicated these findings by exposing humans to a simul-
taneous discrimination task with compound stimuli while record-
ing their eye fixations. Participants rapidly learned to respond
accurately during discrimination training and were then exposed
to a series of different tests to assess which stimulus compo-
nent(s) (color or shape) had acquired control over their
responding. The results suggest that (a) the controlling compo-
nent was idiosyncratic across participants, with some participants
responding during tests on the basis of color and some
responding on the basis of shape; (b) the participants tended to
choose the component to which they had more fixations during
training; and (c) the S+ component that was associated with
fewer fixations during training also controlled participants’
choices on some of the tests.
Tests 1 and 6 aimed to reveal the component that acquired
the most control over participants’choices: S+ color or shape.
In Test 1, all components were separated and simultaneously
presented (cf. Reynolds 1961). In Test 6, the individual com-
ponents of the compound stimulus were recombined (cf. Born
and Peterson 1969). In both tests, the two S+ components
were presented on the same trial and only one of them could
be chosen. Participants consistently chose the same one,
which suggests that only color or shape acquired control over
responding (cf. Pessôa et al. 2009; Saunders et al. 1988;
Schroeder 1969,1970). On the other hand, Tests 2–5present-
ed each of the S+ components separately and pitted them
against an S- component. The results from Tests 2–5suggest
that both S+ components (color and shape) acquired control
over participants’choices. Similar to previous experiments
using pigeons as subjects, the current study shows that pre-
senting the S+ components of a compound stimulus separated
on the same test trial might reveal control by the stronger
component but not by the weaker one (cf. Wilkie and Masson
1976). To evaluate the degree of control acquired by the
weaker component, this component should be separately pre-
sented along with an S- component (Farthing and Hearst
1970;KendallandMills1979; Wilkie and Masson 1976).
Another aim of this study was to evaluate the coherence
between participants’responses during tests and eye fixations
during training. Eye fixations during training occurred more
frequently and with higher duration to the S+ component that
controlled responding in Tests 1 and 6. However, the analysis
of eye fixations did not readily reveal control by the second S+
component that was shown separately during Tests 2–5.
In Tests 1 and 6, in five out of six cases, participants chose
the S+ component that had the highest number and duration of
fixations during training. These results confirm previous find-
ings suggesting that tracking eye fixations is a possible way to
assess stimulus control relations (cf. Dube et al. 1999,2003,
2006,2010;Magnusson2002;Perez2008; Pessôa et al. 2009;
Schroeder 1969,1970). However, eye-fixation measures did
not reveal any control by the weaker S+ component. In other
words, the second highest number and duration of fixations
did not occur to the other S+ component. It instead occurred to
the S- component from the same dimension as the stronger S+
component that was displayed on the same portion of the
presentation area. Shapes were presented on the upper portion
of the compound and colors on the bottom (see Fig. 1). Since a
greater number and duration of fixations were made to one S+
component, eye movements occurred in either the upper or
lower portion of the compound stimulus, depending on the
portion of the presentation area where the stronger S+ com-
ponent was displayed. After a correct choice was made, fixa-
tions on the same portion of the compound might have been
intermittently reinforced by the presentation of the S+ in the
same presentation area. If this were correct, the presentation of
an S- would set the occasion for participants to move their
eyes toward the same portion of the other presentation area,
where inevitably the stronger S+ component would be locat-
ed. Since this is speculation, future studies should investigate
this issue by varying the portion of the presentation area in
which stimulus components are displayed.
Tes ts 2–5 aimed to investigate if the weaker S+ component
would also control participants’choices during tests even
though there were no responses to that component during Test
1. The results of these tests suggest that both S+ components
acquired control over mouse-clinking responses. However,
the eye-fixation results highlight an alternative possibility:
reject control (e.g., Johnson and Sidman 1993) might have
occurred on tests that presented the weaker S+ component and
the S- that had the second highest number of fixations (e.g., in
Phase 1 for Participant A: the red circle vs. the square). In
other words, the choice of the weaker S+ component might
have been controlled by the S- component (e.g., Magnusson
2002; Perez 2008; Stromer and Osborne 1982). Additional
tests are needed to rule out the possibility of reject control.
Psychol Rec (2015) 65:83–88 87
Author's personal copy

One alternative testing method is to change one of these
components to a novel stimulus. Presenting the S- component
with the second highest number of fixations along with a
novel stimulus would reveal reject control if the novel stimu-
lus were chosen. On the other hand, presenting the weaker S+
component along with a novel stimulus would reveal control
by the first stimulus if it were consistently chosen (e.g.,
Stromer and Osborne 1982; see Goulart et al. 2005). Future
studies should consider using such probe trials.
Another methodological issue concerns the formal proper-
ties of the stimulus components. In the present study, the
stimulus components were dissimilar in shape, color, and
luminance. Thus, a peripheral vision effect was to be expected
from the very beginning of training (cf. Schroeder 1969,
1970). The results of Tests 2–5 in Phase 2 suggest that the
participants were “paying attention”to both S+ components
despite the fact that they only fixed their eyes on one compo-
nent. To decrease peripheral-vision effects and to increase
prediction of stimulus control relations by eye fixations, future
studies should present components that are similar in lumi-
nance (cf. Pessôa et al. 2009) by changing the size of the
stimuli or the angular distance in relation to the participant’s
eyes (Perez 2008).
Author’s note During the preparation of the manuscript, the
authors were supported by the following grants: William Ferreira
Perez by FAPESP (doctoral fellowship, Grant # 2011/19125-2),
Peter Endemann by CNPq (doctoral fellowship, Grant # 140636/
2009-9), Candido V. B. B. Pessôa by FAPESP (post-doc fellow-
ship, Grant # 2011/19125-2), and Gerson Y. Tomanari by CNPq.
Data collection and preparation of the manuscript were supported
by CNPq (Grant # 573972/2008-7) and FAPESP (Grant # 08/
57705-8), both at the Instituto Nacional de Ciência e Tecnologia
sobre Comportamento, Cognição e Ensino (INCT-ECCE), coordi-
nated by Dr. Deisy G. de Souza (UFSCar).
We are grateful to Saulo Velasco, Erik Arntzen, the editor and an
anonymous reviewer for helpful comments on early versions of the
manuscript.
References
Birkimer, J. C. (1969). Control of responding by the elements of a
compound discriminative stimulus and by the elements as individual
discriminative stimuli. Journal of the Experimental Analysis of
Behavior, 12,431–436.
Born, D. G., & Peterson, J. L. (1969). Stimulus control acquired by
components of two color-form compound stimuli. Journal of the
Experimental Analysis of Behavior, 12,437–442.
Dube, W. V., & Hiris, J. (1999). Match to sample program (Version
11.6.7) [Computer software]. Walthan, MA: E. K. Shriver Center
of Mental Retardation.
Dube, W. V., Lombard, K. M., Farren, K. M., Flusser, D. S., Balsamo, L.
M., & Fowler, T. R. (1999). Eye tracking assessment of stimulus
overselectivity in individuals with mental retardation. Experimental
Analysis of Human Behavior Bulletin, 17,8–14.
Dube, W. V., Lombard, K. M., Farren, K. M., Flusser, D. S., Balsamo, L.
M., Fowler, T. R., & Tomanari, G. Y. (2003). Stimulus
overselectivity and observing behavior in individuals with mental
retardation. In S. Soraci & K. Murata-Soraci (Eds.), Visual informa-
tion processing (pp. 109–123). London: Proeger.
Dube, W. V., Balsamo, L. M., Fowler, T. R., Dickson, C. A., Lombard, K.
M., & Tomanari, G. Y. (2006). Observing behavior topography in
delayed matching to multiple samples. The Psychological Record,
56,233–244.
Dube,W.V.,Dickson,C.A.,Balsamo,L.M.,O’Donnell, K. L.,
Tomanari, G. Y., Farren, K. M.,...McIlvane,W.J.(2010).
Observing behavior and atypically restricted stimulus control.
Journal of the Experimental Analysis of Behavior,94,297–313.
Farthing,G. W., & Hearst, E. (1970). Attention in the pigeon: testing with
compounds of elements. Learning and Motivation, 1,65–78.
Goulart, P. R. K., Mendonça, M. B., Barros, R. S., Galvão, O. F., &
McIlvane, W. J. (2005). A note on select- and reject-controlling
relations in the simple discrimination of capuchin monkeys (Cebus
apella). Behavioural Processes, 69,295–302.
Johnson, D. F., & Cumming, W. W. (1968). Some determiners of atten-
tion. Journal of the Experimental Analysis of Behavior, 11,157–166.
Johnson, C., & Sidman, M. (1993). Conditional discrimination and
equivalence relations: control by negative stimuli. Journal of the
Experimental Analysis of Behavior, 59,333–347.
Kendall, S. B., & Mills, W. A. (1979). Attention in the pigeon: testing for
excitatory and inhibitory control by the weak elements. Journal of
the Experimental Analysis of Behavior, 31,421–431.
Magnusson, A. (2002). Topography of eye movements under select and
rejec t control. Unpublished doctoral dissertation, Northeastern
University, Boston.
Perez, W. F. (2008). Movimentos dos olhos e topografias de controle de
estímulos em treino de discriminação condicional e testes de
equivalência (Master’s thesis, Programa de Pós-graduação
Psicologia Experimental, Universidade de São Paulo, São Paulo).
Retrieved from http://www.teses.usp.br.
Pessôa, C., Huziwara, E., Perez, W. F., Endemann, P., & Tomanari, G. Y.
(2009). Eye fixations to figures in a four-choice situation with
luminance balanced areas: Evaluating practice effects. Journal of
Eye Movement Research,2(5), 3, 1–6. Retrieved from http://www.
jemr.org/online/2/5/3.
Reynolds, G. S. (1961). Attention in the pigeon. Journal of the
Experimental Analysis of Behavior, 4,203–208.
Reynolds, G. S., & Limpo, A. J. (1969). Attention and generalization
during a conditional discrimination. Journal of the Experimental
Analysis of Behavior, 12,911–916.
Saunders, R. R., Saunders, K. J., Kirby, K. C., & Spradlin, J. E. (1988).
The merger and development of equivalence classes by unreinforced
conditional selection of comparison stimuli. Journal of the
Experimental Analysis of Behavior, 50,145–162.
Schroeder, S. R. (1969). Effects of cue factors on selective eye move-
ments and choices during successive discriminations. Perceptual
and Motor Skills, 29,991–998.
Schroeder, S. R. (1970). Selective eye movements to simultaneously pre-
sented stimuli during discrimination. Perception & Psychophysics, 7,
121–124.
Schroeder, S. R. (1997). Selective eye fixations during transfer of dis-
criminative stimulus control. In D. M. Baer & E. M. Pinkston (Eds.),
Environment and behavior (pp. 97–110). Boulder: Westview.
Skinner, B. F. (1953). Science and human behavior. New York: The Free
Press.
Stromer, R., & Osborne, J. G. (1982). Control by adolescent’s arbitrary
matching-to-sample by positive and negative stimulus relations.
Journal of the Experimental Analysis of Behavior, 37,329–348.
Wilkie, D. M., & Masson, M. E. (1976). Attention in the pigeon: a re-
evalation. Journal of the Experimental Analysis of Behavior, 26,
207–212.
88 Psychol Rec (2015) 65:83–88
Author's personal copy