Association for Behavior Analysis International

The Association for Behavior Analysis International® (ABAI) is a nonprofit membership organization with the mission to contribute to the well-being of society by developing, enhancing, and supporting the growth and vitality of the science of behavior analysis through research, education, and practice.


44th Annual Convention; San Diego, CA; 2018

Event Details

Previous Page


Symposium #342
Punishment: Don't Throw the Baby Out With the Bathwater!
Sunday, May 27, 2018
5:00 PM–6:50 PM
Marriott Marquis, San Diego Ballroom C
Area: EAB; Domain: Basic Research
Chair: Toshikazu Kuroda (Aichi Bunkyo University)
Discussant: Cynthia J. Pietras (Western Michigan University)

Reinforcement and punishment are the two pillars of operant conditioning. Yet punishment has not been investigated as extensively as reinforcement, sometimes leading to throw-the-baby-out-with-the-bathwater claims against its use with limited understanding of basic processes and effects. This symposium presents a series of laboratory experiments on punishment with a variety of experimental arrangements and species, investigating more about procedures and variables that lead to behavior reduction (e.g., the use of non-harmful punishers). Kuroda et al. investigated the punishment effects of predator videos on the sensor-approach behavior of zebrafish. Bland et al. assessed the punishment effects of discriminative stimuli correlated with the absence of reinforcers (S-) on pigeons' key-pecking for food and on humans' key-pressing for money. Whirtley and Perone experimentally analyzed how the rate of reinforcement affects the punishment effects of timeout from positive reinforcement. Fontes et al. found that resurgence—the recurrence of a previously extinguished target response, usually upon the termination of reinforcement for an alternative response—occurs when electric shock is delivered as a punisher for the alternative response. With Dr. Cynthia Pietras as a discussant, we will discuss what punishment is, what it does, and what it does not do in relation to its negative image.

Instruction Level: Intermediate
Keyword(s): discriminative stimulus, punishment, resurgence, timeout

Punishment of Zebrafish (Danio rerio) Approach Behavior With the Motion Picture of Their Predator

TOSHIKAZU KURODA (Aichi Bunkyo University), Yuto Mizutani (Aichi Gakuin University), Carlos Renato Xavier Cançado (Universidade de Brasilia, Brazil), Christopher A. Podlesnik (Florida Institute of Technology)

Zebrafish (Danio rerio) are a widely used animal model in biological research for having such features as fully sequenced genome and transparent embryo. A number of studies demonstrate operant behavior in zebrafish. The present study investigated whether a motion picture of their predator (Indian Leaf fish; Nandus nandus) functions as a punisher of operant behavior. The predator image served as an ecologically relevant alternative to a more common punisher, namely electric shock, which can directly stimulate the brain under water and alter behavior. We established approach to a sensor reinforced with decapsulated brine shrimp eggs with a variable-interval (VI) schedule of reinforcement (Experiment 1) or in a concurrent VI VI schedule (Experiment 2). Next, the predator picture was presented on a monitor according to a VI schedule and moved in one of five different directions across presentations. Responding decreased relative to the absence of the predator picture in both experiments with little change in the unpunished response. Therefore, the motion picture of a predator functioned as an ecologically relevant punisher of zebrafish operant behavior.


Using a Negative Discriminative Stimulus as a Punishing Consequence in Pigeons and Humans

Vikki J. Bland (University of Auckland), Sarah Cowie (University of Auckland), Jason Landon (Auckland University of Technology), STEPHANIE GOMES-NG (University of Auckland), Douglas Elliffe (University of Auckland), Ellen Michie (Auckland University of Technology), Christopher A. Podlesnik (Florida Institute of Technology)

The use of punishment in behavioral treatments is restrained by ethical concerns. However, there remains a need to reduce harmful behaviour unable to be reduced by alternative methods. We used two studies to investigate whether a negative discriminative stimulus signaling an absence of reinforcers would function as a punishing consequence for pigeons' key-pecking for food, and humans key-pressing for money. Subjects were trained to discriminate between a positive- (S+) and negative discriminative stimulus (S-). After training, every five responses on average during S+ resulted in S- being super-imposed on S+ for 1.5-s. In addition, any response could produce reinforcers. Log proportion of S+ response rate was negative in most tests for all five pigeons, whether the corresponding log proportion of S+ reinforcers was positive or negative. Therefore, S+ response rate was relatively suppressed in punishment tests. Log proportion of S+ response rate was negative, or suppressed relative to S+ reinforcer rate, for 7 of 11 humans. We also used pigeons to show that a S- contingent stimulus will shift choice from a punished- to an unpunished alternative, despite a 1:1 reinforcer ratio. These findings provide a foundation for continued investigation of a negative discriminative stimulus as a punishing consequence.


Effects of Reinforcement Rate on the Aversive Function of Timeout From Positive Reinforcement

CORY WHIRTLEY (West Virginia University), Michael Perone (West Virginia University)

Although use of timeout is widespread and has largely been shown to be an effective punisher, the factors responsible for its effectiveness are not fully understood. Our research program seeks to identify these factors. Of present interest is the rate of reinforcement underway during periods of time-in. Our first experiment underscores the complexity of this factor as it affects two countervailing forces: the aversive function of timeout (which would increase its response-reducing function) and the strength of the responding that the timeout is supposed to punish (which would resist reduction in responding by timeout). A second experiment, currently underway, continues the study of these forces. Rats' lever pressing is maintained on variable-interval schedules of food reinforcement in a multiple schedule with two components. Across conditions, the reinforcement rates in both components range from 0.5 to 6 pellets per min. In one component, some presses are followed by a 30-s timeout during which a tone sounds, the component light is extinguished, and the food schedule is suspended. These manipulations are designed to clarify the relations among reinforcement rate, response strength, and the aversiveness of timeout.


Punishing an Alternative Behavior Generates Resurgence of a Previously Extinguished Target Behavior

RAFAELA FONTES (Universidade de Brasilia), João Claudio Todorov (Universidade de Brasilia), Timothy A. Shahan (Utah State University)

Resurgence is the recurrence of a previously extinguished target response when a more recently reinforced alternative response is also extinguished. Although resurgence is typically observed during extinction of an alternative response, other means of devaluation may also lead to resurgence. The present studies investigated the effect of punishment of the alternative response on the recurrence of target response in rats. In Experiment 1, lever presses (target) were reinforced on a VR 20 during Phase 1, extinguished while nose pokes (alternative) were reinforced on a VI 15 s during Phase 2, and during Phase 3 a punishment contingency (0.05 s 0.6mA shocks with a 0.5 probability for each response) was superimposed for the alternative response, while target remained on extinction. Resurgence was observed for three out of six rats. In Experiment 2, the same contingencies as Experiment 1 were programmed, except that lever presses (target) were reinforced on a VI 15 s during Phase 1 and the shock intensity during Phase 3 was escalated from 0.25 mA to 0.75mA across sessions. Resurgence was observed in all rats and was replicated in two subsequent exposures to the sequence of conditions.




Back to Top
Modifed by Eddie Soh