Printer Friendly

Prompting Sidman Avoidance Behavior in Betta splendens.

Avoidance behavior prevents or postpones the presentation of an aversive stimulus. Two types of avoidance discussed in this paper are avoidance with an exteroceptive warning stimulus, referred to as "standard avoidance" or "signaled avoidance," and avoidance without an exteroceptive warning stimulus, referred to as "Sidman avoidance" or "free-operant avoidance" (Sidman, 1953). Signaled avoidance is reinforced by the termination of a warning stimulus which also results in the nonoccurrence or avoidance of the primary (or backup) aversive stimulus, thereby avoiding the aversive stimulus (Bolles & Popp, 1964; Pear, 2016). No change in stimulus is involved during Sidman avoidance. Since it is unclear how the absence of a stimulus can act as a reinforcer, a number of theories have been proposed to explain how animals can learn to avoid the primary aversive stimulus in a Sidman avoidance schedule (e.g., Anger, 1963; Benedict, 1975; Bolles & Popp, 1964; Rescorla, 1968). Moreover, research indicates that Sidman avoidance is not readily established in some species, such as Betta splendens (Siamese fighting fish), Carassius auratus (goldfish), and Stegastes leucostictus (beau Gregory; Hurtado-Parrado, 2015; Otis & Cerf, 1963; Sidman, 1953; Wodinsky, Behrend, & Bitterman, 1962).

Animal research on conditioning Sidman avoidance may have practical value for humans. A review by Higgins and Morris (1984) examined the generality of avoidance conditioning in animals and humans. They found some parallels among variables of avoidance behavior for both human and nonhuman species, such as maintaining avoidance behavior, acquisition of alternative responses, and the effects of the exteroceptive stimuli. Studying Sidman avoidance is also important as it relates to teaching individuals with anxiety disorders to stop avoiding unharmful stimuli, such as occurs in obsessive compulsive disorders like excessive hand washing. Further investigation into how avoidance is learned is important for creation of effective treatments. It is possible that teaching clients to predict and avoid aversive events such as social disapproval could be an important therapeutic goal. There are many situations in which learning avoidance behavior is important (Baron & Perone, 2001). For example, it is sometimes important for clinicians to teach their clients to refrain from talking to strangers, hugging people they do not know, wearing inappropriate clothing, overeating, abusing alcohol and other drugs, and gambling (Baron & Perone, 2001). These considerations suggest that animal research on the conditioning of Sidman avoidance using behavioral procedures may have practical value.

In addition to being studied with mammals and birds (e.g., Davidson, 1971; Griffin, Medearis, & Hughes, 1973; Sidman, 1953; Smith & Keller, 1970), Sidman avoidance has been observed with at least one species of fish, namely, goldfish (Behrend & Bitterman, 1963; Otis & Cerf, 1963). By contrast, experiments on Sidman avoidance with Betta splendens have been inconclusive or negative (Hurtado-Parrado, 2015; Otis & Cerf, 1963). Otis and Cerf (1963) hypothesized that due to their aggressive nature, Betta splendens would learn avoidance more readily than goldfish. Their study involved an experimental tank that required subjects to follow a moving light within the tank. A shock administered through the water was delivered if the subject did not crossover to the other side of the tank within 15s of the presentation of the moving light. Of the 12 goldfish in their study, 10 met the criterion of avoidance responses set by the authors of the study, compared to only four of the 12 Betta splendens that did. Otis and Cerf suggested that this might be due to inherent differences between the two species. For example, goldfish are bottom feeders, whereas Betta splendens are surface swimmers. In other words, the presence of light in the tank may have been more salient to the goldfish because they swim at the bottom of bodies of water where it is dark. Therefore, light is less accessible and more novel to goldfish compared to Betta splendens. This may have resulted in rapid learning.

For male Betta splendens, the sight of its mirror image elicits aggression (e.g., a darkening of body color, gill erection that makes their head appear larger to an opponent and swimming into the mirror making biting movements; Meliska, Meliska, & Peeke, 1980). Further, there is evidence that a mirror image is reinforcing to male Betta splendens (Bols & Hogan, 1979; Lattal & Metzger, 1994). These fish will learn to emit an arbitrary response to produce their mirror image. For example, Lattal and Metzger (1994) demonstrated that a mirror--even when presentation is delayed--is an effective reinforcer when teaching the fish to swim through a ring.

Although previous research has demonstrated that some Betta splendens may learn Sidman avoidance when shock is administered as the aversive stimulus (e.g., Otis & Cerfs 1963), research by Hurtado-Parrado (2015) has indicated that Betta splendens do not learn Sidman avoidance when turbulent water disturbance is delivered as the aversive stimulus, even following an extensive numbers of trials. Hurtado-Parrado demonstrated that turbulent water flow is aversive to Betta splendens, possibly because their natural habitat is a still-water environment. However, he was not able to establish Sidman avoidance when turbulent water disturbance was used as the aversive stimulus. Learning more about Sidman avoidance and the conditions under which it can be explicitly taught--as opposed to simply being learned due to exposure to a Sidman avoidance schedule (the typical way it has been "taught")--could have practical value. The present study sought to determine whether Sidman avoidance can be taught to Betta splendens using behavioral prompting and reinforcement strategies.

Method

Subjects

Three short-tailed male adult Betta splendens served in this study. The fish were purchased from a local pet store, their ages were unknown, and they had not previously been exposed to other experiments. Males were chosen for their high level of aggression and their availability. A fourth Betta splendens was discontinued due to a health condition.

Apparatus

A 33-L (41 x 41 x 20 cm) experimental fish tank, separated in the middle by a white plastic partition, was filled with 18 L of water (see Fig. 1). There was a 2.5-cm wide gap in the center of the partition that was large enough for a fish to swim from one side of the tank to the other. The experimental tank also contained a compartment inaccessible to the fish that housed two AquaClear 50[R] water pumps. The sides and bottom of the experimental tank were made of clear glass, the top was open, and the tank sat on a table with a white top. The compartment housing the pumps was covered with a white plastic sheet. The pumps produced turbulent water flow of approximately 3600 mL per minute. The water in both the experimental and home tanks was dechlorinated due to chlorine's harmful effects on fish. After each session, the water in the experimental tank was emptied using a Shop-Vac[R] and dried with paper towels to minimize pheromones and harmful microorganisms.

A computer-automated video-tracking system (VTS), similar to the one used by Pear and Legris (1987), recorded the three-dimensional position of the fish ten times per second during experimental sessions. Given that the fish was the only nonwhite object in view of the system, the VTS used darkness contrast to track the fish. The VTS consisted of two video cameras, a circuit board connected to an IBM XT computer, a television monitor, and a Windows 2000 computer containing software for collecting and analyzing data received by the IBM XT computer from the VTS. The video cameras were mounted on an extension arm of a sealed wooden scaffold so that they looked directly down on the experimental tank. The cameras were 31 cm apart, angled toward each other at a combined angle of 14[degrees], and were connected to the television monitor and the VTS circuit board. The VTS combined the input from the two cameras by tracking the location of the first dark region three pixels wide that it detected when analyzing the images from the left to right and back (the side farthest from the door of the experimental room) to front (the side closest to the door of the experimental room), and then presented the center of this location in three coordinates (three dimensions) ten times per second. In addition to spatio-temporal recording of the data, sessions were also observed in real time (as explained in the procedure section below) and video recorded. The two computers and the television monitor were located in a room adjacent to the room containing the experimental tank. Because certain aspects of the procedure could not be accurately monitored by the VTS, data were also collected through visual observation using an application called ABC Data Pro on an iPod (see http://cbtaonline.com/drupal/ abcdatapro). This application was chosen for its multi-button functions that allowed collection of multiple target behaviors.

Procedure

Between sessions, each fish was housed individually in a rectangular-shaped 11.35-L tank, filled with dechlorinated water and decorated with aquarium gravel and two plastic aquarium plants. The home tank water temperature was maintained at 27 [+ or -] 1 [degrees]C. The fish were visually isolated from each other in order to minimize potential changes in behavior as a result of seeing a conspecific. The home tanks were located in a room with a 12-h light/dark cycle. The animal facility manager monitored the health and feeding of the fish.

One research assistant was involved in carrying out the procedure in each session, accompanied by the principal investigator. The principal investigator implemented the experimental procedures and the research assistant collected the data. A second research assistant collected interobserver agreement (IOA) data for one-third of the sessions, following the same data collection procedures.

All sessions were 20-min long and occurred four times a week. Start times varied by the day of the week due to scheduling constraints (e.g., every Monday at 5:00 pm and every Tuesday at 11:30 am). This schedule remained consistent throughout the study. The study consisted of four conditions. The first condition, baseline 1, examined whether the fish engaged in Sidman avoidance without special training. The second and third conditions, phases 1 and 2, were two intervention conditions. The fourth condition, baseline 2, was a return to baseline.

Baseline 1

This phase involved 10-s presentations of water disturbances or flows produced by activations of both pumps. The fish was on a Sidman avoidance schedule for water disturbances with a flow-flow interval of 30 s and a response-flow interval of 30 s. When the fish had been on one side of the tank continuously for 30 s, the water pump on that side produced a steady 10-s water flow (flow-flow interval). To escape the water flow, fish could swim to the other side of the divided tank, which terminated the water flow. To avoid water flows completely, the subject could repeatedly swim from one side of the tank to the other side before the end of each 30-s interval.

The principal investigator and assistants visually tracked (1) the number of crossovers that fish made as an escape response while the pump was active and (2) the number of crossovers made prior to pump activation that thereby prevented the water flow. If the fish swam across the midline (i.e., through the gap in the partition separating the two sides of the tank such that any part of its body was detected by the VTS or by visual observation as being on the other side of the tank) in less than 30 s (response-flow interval) after the previous water flow or response, the pumps were not activated. The number of crossovers made prior to the next scheduled pump activation indicated whether the fish showed avoidance. During baseline conditions, the VTS responsible for water disturbance activation also automatically recorded the total time and frequency that the pumps were activated on each side, as well as the response rates (i.e., rate of crossovers). Data on the frequency of escapes and avoidances were also collected by real-time visual inspection. To account for any residual water flow, avoidances were considered only after 3 or more seconds following the cessation of water activation; otherwise, they were considered escapes.

Some tracking errors (i.e., failure of the VTS to accurately detect the fish) occurred but did not seriously affect the accuracy of the data. As mentioned above, the VTS tracked ten responses per second. Therefore, within a 20-min session, there was an opportunity for 720,000 errors. Tracking errors ranged between 6 and 1094 errors per session, with an average of 486 errors per session. To ensure integrity of the VTS, the system was tested before each session by using a "dummy fish," which was simply a black bolt on a stick.

During baseline 1, the fish were exposed to periodic water disturbances that they did not avoid. It is possible that elicited behavior, such as freezing, may have competed with avoidance learning. In effect, due to the absence of a contingency for avoidance responding, the fish may have learned not to emit avoidance responses. Therefore, in order to minimize this possibility, only as many sessions as were necessary to determine stability of responding were conducted during baseline 1. Stable responding was defined as 3-5 data points with no apparent changing increasing or decreasing trend. Once stability was observed the subject was moved to phase 1.

Phase 1

Phases 1 and 2 consisted of prompting to test behavioral strategies to train the fish to engage in avoidance behavior. During these intervention phases, the pumps were not activated. The fish experienced a prompting sequence that facilitated crossing over at 30-s intervals. Avoidances during these interventions were considered independent crossovers (i.e., simulating avoidance in the intervention phase) if the fish crossed from one side to the other within a 30-s interval without prompting. Prompting consisted of gently guiding the fish through the gap in the partition separating the two sides of the experimental tank by using a 10 x 10-cm plastic panel. The fish were blocked from swimming away from the divider gap. Once a fish had crossed the midline through the gap, it immediately received 15 s of access to a mirror that was placed in the tank directly in front of the fish. Reinforcement delivered via the mirror was administered on fixed-ratio (FR) 1 schedule (i.e., continuous reinforcement: reinforcement was provided after each crossover response, whether prompted or independent). The duration required to prompt the fish before they crossed over was recorded to compare prompting times between fish. The next 30-s trial then took place, in which the fish were not prompted but could receive reinforcement if it crossed over to the other side of the tank. This length of time was chosen to be consistent with the previously mentioned procedure in the avoidance condition, in which the pumps activated when the fish stayed on one side of the tank for 30 s. The pumps were not activated in the intervention conditions, i.e. phases 1 and 2. The interval schedules would have been less controlled if the tracking/pump activations had occurred during the intervention conditions because the mirror and research assistant's hands would have interfered with the accuracy of the VTS.

The principal investigator recorded whether the fish showed gill erection or flaring following the presentation of the mirror. This information was useful in determining the reinforcing strength of the mirror. During phase 1, the mirror was submerged immediately into the tank and placed directly in front of the fish for 15 s each time the fish crossed from one side of the tank to the other. This prompting and reinforcement sequence was administered after each 30-s interval. In addition, mirror presentation was immediately administered if the fish crossed over before the 30 s had elapsed. A research assistant collected the following data using the ABC Data Pro on the iPod: duration and frequency of prompts and reinforcement, independent responses, and flaring (i.e., gill erection). The data were then exported to be analyzed. Phase 1 continued until a stable number of independent responses occurred. Stable in this context meant a leveling off of the increasing crossovers.

Phase 2

Phase 2 involved continuing the aforementioned prompting and reinforcement sequence. However, the schedule of reinforcement was changed to a variable-ratio (VR) 2 schedule. That is, on average the subject received a mirror presentation after two responses, varying randomly between one, two, and three responses.

Baseline 2

The final phase of the study involved removing both prompting and reinforcement procedures and a return to the initial baseline condition. This phase was not initiated until the fish appeared to be consistently swimming back-and-forth between the two sides of the experimental tank (i.e., 3-5 stable data points of responding). During this condition, the number of crossovers in a 20-min trial was recorded to see whether the intervention phases resulted in an increase in avoidance behavior, whether the behavior was maintained, and, if the behavior was not maintained, the rate of decrease in the behavior.

Interobserver Agreement and Procedural Integrity

All study personnel completed a fish ethics online course, offered by the Office of Research Ethics and Compliance of the University of Manitoba, before participating in any lab activities. After completing this course, assistants learned proper handling of the fish and how to accurately collect data on the fish's behavior, so that all phases were conducted as closely as possible across research assistants. Interobserver agreement (IOA) and procedural integrity (PI) indices were calculated as agreement scores, by dividing the number of agreements by the number of agreements plus disagreements and multiplying the result by 100 to obtain a percentage (Harris & Lahey, 1978; Iwata, DeLeon & Roscoe, 2013; Kazdin, 2011). IOA for frequency of independent crossovers was 97.25%. IOA for whether or not the fish flared was 93%. PI for prompting procedures was 100% except for one outlier that lowered PI to 86%. Finally, PI for reinforcement procedures was 99%.

Results

The data were analyzed through visual inspection with special attention to the following for each subject: (1) changes in the means, (2) changes in levels of performance (i.e., variability and stability), (3) trends (i.e., increasing, decreasing, and zero trend), (4) latency of behavior change (i.e., duration before changes in responding), and (5) nonoverlapping data points between the phases (Kazdin, 2011).

Figure 2 shows the frequency of independent crossovers between the two sides of the experimental tank throughout the study for all fish. Note that all three fish showed an increase in avoidance responding from baseline 1 to phase 1 and a further increase from phase 1 to phase 2. This was followed by a decrease in avoidance responding during baseline 2. Note that while avoidance responding decreased to baseline 1 levels for fishes 2 and 3, avoidance responding remained above baseline levels for fish 1, suggesting that this fish had learned avoidance responding.

Figure 3 depicts the duration of time required to prompt the fish to cross over sides of the tank during intervention. Note that the time needed to prompt fishes 1 and 2 decreased from phase 1 to phase 2, although this was not the case for fish 3. This suggests that some learning to respond to the prompts had occurred for fishes 1 and 2. However, the duration of the prompts needed for fish 3 were relatively short, ranging between 0 and 23 s. The average percentage of time each fish flared when the mirror was presented during intervention was 0.2, 88, and 94% for fishes 1, 2, and 3, respectively. Based on the small number of fish, the results of this study should be considered with caution.

Discussion

The major findings and their implications from this study were as follows. First, for each fish, the interventions (phases 1 and 2) increased the rates of responding (i.e., independent crossovers) compared to both the first and second baseline conditions. This is important because it indicates that the interventions were responsible for the changes in behavior. It also replicates previous research showing that Betta splendens can learn simple tasks (i.e., crossing over between two sides of a tank) by implementation of reinforcement schedules using a mirror as a reinforcer (Bols & Hogan, 1979).

Second, the crossover and flaring data indicated that administration of the mirror can be considered a strong reinforcer, which complements previous research in this area. The mirror image elicited an aggressive response from fishes 2 and 3 almost every time it was presented, confirming that aggression-eliciting stimuli can serve as reinforcers. However, fish 1 rarely made an aggressive response toward the mirror, although the frequency of this fish's independent crossovers remained high. This raises the question of whether maintenance did not occur in fish 2 and fish 3 because their aggressive responsiveness toward the mirror impeded their learning to avoid the water disturbance by competing with avoidance responses.

Third, the frequency and duration of prompts decreased from the initial sessions in which prompts were used. This indicates that the prompting procedure was effective in teaching the fish to cross from one of the experimental tanks to the other without the use of an aversive stimulus. Fourth, Sidman avoidance was not demonstrated during the initial baseline, suggesting that Betta splendens do not engage in Sidman avoidance in response to water disturbances without an intervention to develop that behavior.

The most important result was that for one of the fish (fish 1) levels of responding (i.e., crossovers) remained above initial baseline levels once the intervention was removed. This is important because it indicates that for fish 1, some level of avoidance of the water disturbance occurred. This suggests that Sidman avoidance can be taught with prompting and reinforcement.

A strength of this study includes its contribution confirming that mirrors can reinforce relatively complex behavior in Betta splendens. The study also found that habituation of the mirror did not occur throughout the course of the intervention, as indicated by stable increases in responding during the intervention phases. This is consistent with previous research that found that habituation to the mirror occurs very slowly (Clayton & Hinde, 1967).

The results of this study during baseline conditions were consistent with those of Hurtado-Parrado (2015), which suggested that Betta splendens do not learn Sidman avoidance without explicit teaching when water disturbance is the aversive stimulus. The results are also consistent with Otis and Cerf's (1963) finding that Betta splendens do not readily learn Sidman avoidance even when shock is the aversive stimulus.

For future studies, additional conditions for strengthening maintenance of the independent responses may be considered. A possible way to increase maintenance could include additional phases of increasingly leaner VR schedules to systematically fade out the frequency of prompting. Using higher VR schedules could increase responding, which may increase the probability of the maintenance of the crossovers. However, there is the possibility of habituation and overtraining with increasing the amount of training trials (Dickinson, 1985). Although studies have demonstrated that habituation to a mirror by Betta splendens takes a long time, Clayton and Hinde (1967) found that even after 10 days of various schedules of interaction with the mirror, the aggressive response of Betta splendens weakened over time but never ceased altogether.

An important contribution of this study is that it provides a new methodology for studying Sidman avoidance. It may be useful for studying avoidance in other animals that do not readily learn Sidman avoidance, even with species that typically do learn it. For example, although most rats learn Sidman avoidance, many do not--at least with standard Sidman avoidance conditioning procedures (M. Perone, personal communication, January 23, 2015).

Moreover, the results of this study provide an animal model for studying a particular type of coping behavior under aversive stimulation. This could lead to further understanding of the mechanisms of avoidance in humans as well as other animals. Animal models for this type of research are important because this type of research may not be possible to conduct with humans for ethical reasons. The results of this study suggest that reinforcement can be used to teach avoidance of aversive stimuli. This could be beneficial for teaching humans to avoid dangerous or aversive situations and to refrain from unhealthy or problem behaviors.

Funding A Psychology Graduate Fellowship for financial support for educational programs from the University of Manitoba was given to the first author.

https://doi.org/10.1007/s40732-018-0285-0

Compliance with Ethical Standards

Conflict of Interest The authors declare that there are no conflicts of interests.

Statement on the Welfare of Animals This paper involved animal subjects and was approved by the Animal Ethics Research Board of the University of Manitoba. All personnel handling the fish in this research were required to take the Animal Users' Training Course offered by the Office of Research Ethics and Compliance of the University of Manitoba.

References

Anger, D. (1963). The role of temporal discriminations in the reinforcement of Sidman avoidance behavior. Journal of the Experimental Analysis of Behavior, 6, 477-506. https://doi.org/10.1901/jeab. 1963.6-s477

Baron, A., & Perone, M. (2001). Explaining avoidance: two factors are still better than one. Journal of the Experimental Analysis of Behavior, 75, 357-361. https://doi.org/10.1901/jeab.2001.75-357

Behrend, E. R., & Bitterman, M. E. (1963). Sidman avoidance in the fish. Journal of the Experimental Analysis of Behavior, 6, 47-52. https:// doi.org/10.1901/jeab.1963.6-47

Benedict, J. O. (1975). Response-shock delay as a reinforcer in avoidance behavior. Journal of the Experimental Analysis of Behavior, 24, 323-332. https://doi.org/10.1901/jeab.1975.24-323

Bolles, R. C., & Popp, R. J. (1964). Parameters affecting the acquisition of Sidman avoidance. Journal of the Experimental Analysis of Behavior, 7, 315-321. https://doi.org/10.1901/jeab.1964.7-315

Bols, R. J., & Hogan, J. A. (1979). Runway behavior of Siamese fighting fish, Betta splendens, for aggressive display and food reinforcement. Animal Learning and Behavior, 7, 537-542. https://doi.org/10.3758/ BF03209715

Clayton, F. L., & Hinde, R. A. (1967). The habituation and recovery of aggressive display in Betta splendens. Behavior, 30, 96-105. https:// doi.org/10.1163/156853968X00207

Davidson, R. S. (1971). Effects of food deprivation on free-operant avoidance. Psychological Reports, 29, 1019-1022. https://doi.org/10. 2466/pr0.1971.29.3.1019

Dickinson, A. (1985). Actions and habits: the development of behavioural autonomy. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 308(1135), 67-78.

Griffin, P., Medearis, H. D., & Hughes, W. R. (1973). Food deprivation and free-operant avoidance in the pigeon. Bulletin of the Psychonomic Society, 1, 393-394. https://doi.org/10.3758/ BF03209715

Harris, F. C., & Lahey, B. B. (1978). A method for combining occurrence and nonoccurrence interobserver agreement scores. Journal of Applied Behavior Analysis, 11, 523-527. https://doi.org/10.1901/ jaba.1978.11-523

Higgins, S. T., & Morris, E. K. (1984). Generality of free-operant avoidance conditioning to human behavior. Psychological Bulletin, 96, 247-272. https://doi.org/10.1037/0033-2909.96.2.247

Hurtado-Parrado, C. (2015). Effects of different aversive contingencies involving water disturbances on the behavior of Betta splendens: performance in a shuttle tank during signaled and un-signaled free-operant avoidance with and without differential consequences for escape (unpublished doctoral dissertation). University of Manitoba, Manitoba.

Iwata, B. A., DeLeon, I. G., & Roscoe, E. M. (2013). Reliability and validity of the functional analysis screening tool. Journal of Applied Behavior Analysis, 46, 271-284. https://doi.org/10.1002/ jaba.31

Kazdin, A. E. (2011). Single-case research designs: Methods for clinical and applied settings. New York: Oxford University Press.

Lattal, K. A., & Metzger, B. (1994). Response acquisition by Siamese fighting fish (Betta splendens) with delayed visual reinforcement. Journal of the Experimental Analysis of Behavior, 61, 35-44. https:// doi.org/10.1901/jeab.1994.61-35

Meliska, C. J., Meliska, J. A., & Peeke, H. V. (1980). Threat displays and combat aggression in Betta splendens following visual exposure to conspecifics and one-way mirrors. Behavioral and Neural Biology, 28, 473-486. https://doi.org/10.1016/S0163-1047(80)91842-7

Otis, L. S., & Cerf, J. E. (1963). Conditioned avoidance learning in two fish species. Psychological Reports, 12, 679-682. https://doi.org/10. 2466/pr0.1963.12.3.679

Pear, J. J. (2016). The science of learning (Second ed.). New York: Taylor & Francis.

Pear, J. J., & Legris, J. A. (1987). Shaping by automated tracking of an arbitrary operant response. Journal of the Experimental Analysis of Behavior, 47, 241-247. https://doi.org/10.1901/jeab.1987.47-241

Rescorla, R. A. (1968). Pavlovian conditioned fear in Sidman avoidance learning. Journal of Comparative and Physiological Psychology, 65, 55-60. https://doi.org/10.1037/h0025412

Sidman, M. (1953). Avoidance conditioning with brief shock and no exteroceptive warning stimulus. Science, 118, 157-158. https://doi. org/10.1126/science.118.3058.157

Smith, R. F., & Keller, F. R. (1970). Free-operant avoidance in the pigeon using a treadle response. Journal of the Experimental Analysis of Behavior, 13, 211-214. https://doi.org/10.1901/jeab.1970.13-211

Wodinsky, J., Behrend, E. R., & Bitterman, M. E. (1962). Avoidance-conditioning in two species of fish. Animal Behavior, 10, 76-78. https://doi.org/10.1016/0003-3472(62)90134-3

Lisa Hunter (1) * Karli Pedreira (1) * Joseph Pear (1)

Published online: 7 May 2018

([mail]) Lisa Hunter

umhunt45@myumanitoba.ca

Karli Pedreira

umpedrek@myumanitoba.ca

Joseph Pear

Joseph.Pear@umanitoba.ca

(1) Department of Psychology, University of Manitoba, 190 Dysart Road, Winnipeg, Manitoba R3T 2N2, Canada

Caption: Fig. 1 Photo of the 33-L experimental fish tank separated in the middle by a plastic partition. Two water pumps are placed in the long compartment, divided from the main tank

Caption: Fig. 2 Frequency of independent crossovers between the two sides of the experimental tank during baseline 1, phase 1, phase 2, and baseline 2 for all fishes

Caption: Fig. 3 Time, in seconds, spent in prompts during phases 1 and 2 for all fishes
COPYRIGHT 2018 Springer
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:ORIGINAL ARTICLE
Author:Hunter, Lisa; Pedreira, Karli; Pear, Joseph
Publication:The Psychological Record
Article Type:Case study
Date:Jun 1, 2018
Words:4925
Previous Article:An Examination of the Transdiagnostic Role of Delay Discounting in Psychological Inflexibility and Mental Health Problems.
Next Article:Acquisition of Complex Conditional Discriminations in a Child with Autism Spectrum Disorder.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters