Association for Behavior Analysis International

Four groups of young adults were exposed to an arbitrary matching-to-sample training as well as to symetry, transitivity, and equivalence test-trials. During training, for two groups correct and incorrect matching responses produced the corresponding feedback according to continuous and intermittent schedules, respectively. Correct responses produced feedback and incorrect responses produced blanks and vice versa for other two groups, respectively. Intermitent feedback obstructed the aquisition of the conditional discrimination and produced the lowest percentages of correct responses in test-trials. Providing feedback for errors and blanks for correct responses allowed the acquisition of the conditional discrimination, but high percentage of correct responses in simetry test-trials only. Continious feedback as well as providing feedback for correct responses and blanks for errors both allowed acquisition of the conditional discrimination and high percentage of correct responses in all test trials. These results support postulates of interbehavioral psychology and are disccused in the context of such theory.

The demand for effective and efficient instruction in education is clear; multiple statistics demonstrate that research at the experimental level is needed in order to extend our understanding of the process of learning. For instance, The Nation’s Report Card (2017) demonstrates this claim through data collected on average 4th-grade students’ test scores. Over the past 30 years, the average math score has been below the proficient level (defined by procedural, conceptual, and problem-solving skills); furthermore, the last 10 years have shown stability- exhibiting a lack of progress toward proficiency. The purpose of the current study is to extend our understanding of learning by manipulating the number of stimuli presented in an array during an arbitrary match-to-sample task. Preliminary data shows that 5 stimuli in an array results in faster acquisition, as defined by a frequency aim, than 7 stimuli. The study will also assess acquisition in relation to application of the skills learned which can more directly inform educational and teaching aims outside of the laboratory.

This experiment evaluated whether the outcomes of RFT studies on the frames of sameness and opposition are instances of contextual control over equivalence and nonequivalence and exclusion-based responding. Four college students participated. In Phase 1, we trained SAME as a contextual cue for selecting comparisons physically identical to the samples, and OPPOSITE as a contextual cue for selecting the comparisons most physically dissimilar to the samples. In Phase 2, SAME and OPPOSITE controlled the maintenance and reversal of equivalence classes, respectively. In Phase 3, we trained SAME-A1B1, SAME-A1C1, OPPOSITE-A1B2, and OPPOSITE-A1C2. The four participants derived SAME-B1C1, SAME-B2C2, OPPOSITE-B1C2, and OPPOSITE-B2C1. SAME-B2C2 could be explained by the exclusion of C1, a stimulus nonequivalent to B2, or by the combinatorial entailment of the opposition relations A1B2 and A1C2, as RFT proposes. In Phase 4, we trained SAME-A1B1, SAME-A1C1, SAME-A2B2, and OPPOSITE-A1C2, with a new set of stimuli. The four participants derived the same relations as in Phase 3. SAME-B2C2 can be explained by the exclusion of C1, but not by the combinatorial entailment of opposition relations because OPPOSITE-A1B2 was not trained. In summary, the outcomes of RFT studies on the frames of sameness and opposition are probably explained by alternative processes.

A common phenomenon associated with extinction-based treatments for problem behavior is the occurrence of a transient increase in the rate, magnitude, or duration of the targeted behavior. We sought to replicate and extend the findings of Lerman and Iwata (1995, 1999) by examining the magnitude and persistence of this phenomenon. Specifically, we reviewed data from 108 participants with autism spectrum disorder who had been admitted to an intensive outpatient clinic for the assessment and treatment of problem behavior and extinction was a component of the treatment. We found a prevalence in our sample of 19.1% bursts observed out of 335 exposures, which was similar to the prevalence of 15% previously reported in Lerman and Iwata (1999). The extinction bursts were most frequently classified as large in magnitude, accounting for over half of the bursts observed in this sample. Additionally, we observed a decreasing trend across ten sessions that approached levels prior to the implementation of extinction.

Obesity is a public health problem in Mexico. More than 36.3% of adolescents have obesity. Although the literature mentions selection by type of food and physical activity are the main factors for the development of obesity, the multidimensional effect of decision making is still unclear. Delay discounting studies show that adults with obesity discount more steeply immediate rewards than the controls. However, mixed findings are with probability discounting, and the effort discounting has not been explored as a possible additional factor to describe the choices for poor physical activity. This study aimed to compare the loss of the subjective value of different rewards in delay, probability, and effort discounting tasks with hypothetical outcomes in adolescents with obesity. The choice trials were presented with the adjusting amount procedure. We use a mixed design. Preliminary results show the hyperbolic model had moderate adjustments for the obesity group and higher adjustments for the controls. Findings were replicated for delay discounting with money. A probability and steeply effort discounting were found with money for obese participants than for the controls. Findings suggest the variable delay and effort seem to describe the "impulsive" choices for the obesity group, unlike the pattern of risk-seeking.

We exposed 34 children to the adjusting-amount procedure (standard amount = 16 prizes) presenting by five delays (5, 10, 20, 30, and 60 seconds), odds against (0.111, 0.333, 1, 3, and 9) and effort levels (0.2, 0.4, 06, 0.8, and 1) in three Experiential Discounting Tasks (EDTs). We found high values for the adjustment to the hyperboloid model in the three tasks (R2 = .92). We observed a steep probability discounting curve, a moderately delay discounting curve and a shallow effort discounting curve. Statistically significant differences were found in the degree of discounting between delay and effort discounting, and probability and effort discounting. The Principal component analysis showed four components that explain 67.46% of the variance. The findings of this study suggested different discounting patterns based on the parameters manipulated and supported the internal validity of the tasks. The EDTs contributed to solving practical issues by using real contingencies with children.

Previous research showed that delay, but not social discounting is significantly correlated with Japanese students’ using smartphones while walking (USWW) rates. In addition, there is mixed evidence for a relationship between delay discounting and texting while driving (TWD). There are no published studies examining relationships between social discounting and TWD. The current study explored relationships between two types of discounting (delay and social) and two risky behaviors involving smartphones (TWD and USWW). A total of 456 US college students completed demographic, discounting, and self-reported risky smartphone measures. Figure 1 shows social discounting box-plots for the bottom 25%, middle 50% and top 25% TWD participants. As TWD increased, social discounting rate significantly increased. However, the same pattern was not shown with delay discounting and TWD. Figure 2 shows that low USWW rates were correlated with lower social discounting rates. There was no correlation between delay discounting and USWW. TWD and USWW were significantly correlated (r = 0.40), whereas delay and social discounting were not significantly correlated (r = 0.03). The results are interpreted in regards to the absolute differences in self-reported USWW rates between the US and Japan, and how converging evidence suggests that delay and social discounting are separate processes.

Implications of research on the sign effect may indicate how to deal with decision-making challenges. The sign effect is the steeper discounting of gains compared to losses. In this experiment, we compared discounting of gains and losses in order to explore when and to what extent human adults discount. Thirty-one participants went through a computer-based choice-task procedure of hypothetical monetary gains and losses. The results show clear qualitative differences. Whereas gains mostly involve systematic devaluations, losses often correspond to zero discounting. These results replicate earlier studies suggesting that discounting of gains and losses involve different reinforcing contingencies. A paired samples t-test indicated that the mean AUCgain is significantly smaller than the mean AUCloss, t (30) = -8.22, p = .00, r = .83, confirming the sign effect. Correlations between AUCgain and AUCloss values show that AUCgain was weak to moderately, and not significantly, correlated to AUCloss (r = .223; p = .228). The weakness of the relationship between these variables strengthen the argument that discounting of gains and losses are not part of the same process.

Research shows that prolonged training in the impulsive task determines impulsive choice in nonhuman animals. We extended the generality of this finding to the impulsive choices of Spontaneously Hypertensive (SHR) and Lewis (LEW) rats responding to a novel concurrent-chains procedure. The initial-link arranged choices between smaller-sooner (SS) and larger-later (LL) foods. Choice was measured on two levers concurrently available in the initial link, where two non-independent random interval schedules arranged entries to two terminal links. One terminal link delayed the delivery of the LL food (4-pellets) six times, and the other terminal link delivered the SS food (1-pellet) immediately. Five models of intertemporal choice and the Generalized Matching Law (GML) fitted the data from the SHRs and LEWs well. Discounting rate (k) and sensitivity to immediacy of reinforcement (s) increased with prolonged training; estimates of k were positively correlated with estimates of s, suggesting compatibilities between models of intertemporal choice and the GML. It is proposed that the behavior pattern labeled impulsivity changes with the rats’ experience in the choice situation.

Killeen and Jacobs (2017) discussed James Gibson’s concept of affordance and its place within the analysis of behavior. According to Gibson (1979), affordances are possibilities for behavior scaled to the action capabilities of the organism. Previous research has found that when asked to lift and move objects that vary only by their lengths, participants make a transition from one- to two-hands or two- to one-hand at predictable affordance boundaries defined as a ratio of hand-span to object length (Lopresti-Goodman et al. 2009; Isenhower et al. 2010). In order to further explore the relationship between affordances and behavior analysis, the current study manipulated object properties in a similar fashion to the aforementioned studies but treated the choice (one vs. two hands) as a concurrent operant arrangement. Reinforcement was delivered in the form of points on concurrent VR-VR schedules. The participants’ goal was to maximize the number of points they earned. During baseline conditions no reinforcement was delivered regardless of how participants chose to move the objects. During the 1H condition, participants contacted reinforcement on a VR2 schedule for moving objects with one hand and a VR5 schedule for moving objects with two hands. During the 2H condition, the concurrent schedules were reversed such that participants contacted reinforcement on a VR2 schedule for moving the objects with two hands and a VR5 schedule for moving objects with one hand. Preliminary results indicate that the affordance boundary moves relative to baseline for the 2H condition but not the 1H condition. Implications for incorporating affordances into the analysis of behavior will be discussed.

Reinforcement-based procedures are key tools for implementing effective behavior change. However, one of the main criticisms of reinforcement-centered programs is the overjustification hypothesis, which posits that when “extrinsic” rewards are delivered contingent on the occurrence of behavior, the rate of that behavior will fall to below baseline levels when these rewards are removed. The current study is a replication of a 1971 study by Edward Deci evaluating the overjustitication hypothesis. The subjects in this replication were ten undergraduate students who were asked to complete puzzle configurations across three phases. These subjects were divided into a treatment group, who received rewards for puzzle completion, and a control group, who received no extrinsic rewards. Participants in the treatment group received no “extrinsic” rewards in phase one, a monetary reward contingent on puzzle completion in phase two, and a return to no monetary rewards in phase three. The results were highly variable but the majority of participants in both the treatment and control group showed a decline in engagement across the three phases. The findings of this replication suggest that satiation, rather than overjustification, may decrease subject puzzle interaction.

Metacontingencies have not yet been implemented in conjunction with the Good Behavior Game to facilitate its maintenance. Although the literature reports strong effects from the Good Behavior Game, diffusion of the Brazilian culturally adapted version Elos Program – Communities Coalitions was suspended by the federal government. Even though national available data should be considered, designs with higher levels of experimental control are necessary to estimate the intervention effects. Through reversal and multiple baseline designs across three municipalities, this study aims to test the effects of Elos Program – Communities Coalitions with the addition of support metacontingencies, subsidized in the program manuals available for public access. Dependent variables include on-task, cooperative, disruptive and aggressive behavior from pupils; nurturing practices from teachers; and interlocked behavioral contingencies of mutual support between practitioners. 320 participants are expected from six public schools. Operant and cultural consequences will be manipulated in three phases: Operant selection, Cultural selection and Cultural maintenance. Conditions differ regarding individual or group feedback delivered to professionals as well as to reinforcement schemes contingent to the occurrence of interlocked behavioral contingencies. The first phase is running and preliminary results from one third-grade class can be discussed. Disruptive behavior rates markedly decreased while on-task behavior rates increased to levels far from baseline conditions since intervention started. If repeated across classrooms, this result shall increase external validity of Good Behavior Game variations to Brazilian educational system. Data from longitudinal studies using randomized control trial methods identified correlations between participating in the intervention and lower rates of drug abuse at adult age. Although its measurements are limited to few points over time, long-term research literature regarding the Good Behavior Game highlights its potential outcomes that should be investigated with reliable experimental control.

Group contingency (GC) is a behavior management strategy in which a consequence is contingent upon the behavior of all or selected portion of a group of people. However, studies comparing the different types of GC are limited, especially those aimed at enhancing physical activity. The purpose of this study was to examine the effectiveness and cost-efficiency of GC for promoting walking behavior of college students. 72 college students were grouped as a team of three based on their step counts and were randomly assigned to three conditions for 66 days of intervention. In independent condition, participants earned points each time they met the predetermined goal, regardless of their teammates’ performance. In interdependent condition, participants earned points each time all teammates met the goal. In random dependent condition, participants earned points each time the randomly selected teammate met the goal. In terms of effectiveness, random dependent condition showed a significant increase in step counts during intervention than the others. In terms of cost-efficiency, when the same amount of points was provided, increase in step counts of interdependent condition was the highest. The results suggest that random dependent GC is effective, whereas interdependent GC is cost-efficient in promoting walking behavior.

Variables facilitating “cognitive defusion” are still incompletely understood in laboratory settings. The present study focused on the persistent contextual control of the rock-paper-scissors (RPS) game, where responding based on the circular relation is reinforced and feedback in accordance with an alternative rule is deactivated. The alternative context was introduced, where responding based on the linear relation of the number of fingers (NOF), is reinforced. As a general procedure, a participant chose the winning/losing one of two handsigns on a computer screen. Trainings provided a feedback on “correct/incorrect” per trial, then its effect was assessed in the subsequent tests with no feedback. In Experiment 1, 12 undergraduates’ data were obtained. Two contextual cues were introduced, indicating either NOF or RPS responding is correct. In the trainings where the handsigns not appearing in the RPS game were added, nine participants showed over 75% correct responses in the NOF context. In Experiment 2, eight undergraduates’ data were obtained. After the training where rock, paper, and scissors were presented in the same trial, six participants showed over 75% NOF responses. In conclusion, situations that conflict with the regular RPS game could facilitate defusion from its persistent contextual control.

We have developed a specialized apparatus for single-subject research with rat colonies. This “One Rat Door” (ORD) sits between an operant chamber and group housing. Only one rat may pass from the group housing, through the door, to the operant chamber. Once a rat has passed, a locking mechanism is engaged that prevents additional rats from entering the ORD. Only when the initial rat leaves the operant chamber is the door unlocked, allowing another rat to enter. The design of the ORD is cheap, simple, fully mechanical, and can accommodate different rat sizes. The mechanical design reduces risk of injury from motorized moving parts and reduces transition disruption. It also enables automated experimentation; it can be run 24/7 without requiring experimenter time and effort. Used in conjunction with RFID technology, the ORD allows for both single-subject data and individualized contingencies in the operant chamber. This apparatus is currently being used to conduct single-subject research with rats relearning a distal forelimb reach task following ischemic stroke. Because of its simple and enabling design, the ORD may be used to measure self-initiated behavior within a social context and to study behavioral economics and nonlinear contingencies with high external validity.

In Experiment 1, each rat in the 0-minute group moved freely in a chamber where a wall blocked access to a restraint tube. After 10 minutes, the wall was removed, permitting 15 minutes of chamber access and tube entry. The other two groups were locked in the tube for 10 and 20 minutes respectively before release into the chamber for 15 minutes. Across sessions, rats locked up for 10 and 20 minutes entered the tube more frequently than rats in the 0-minute group, and during the first two sessions rats in the 20-minute group stayed in the tube longer than the other groups. Over sessions this difference disappeared. However, for all groups and sessions the mean percentage of session time in the tube exceeded chance expectations. This result suggests tube occupation was reinforcing. In Experiment 2’s Phase 1, rats could enter an open tube. On exiting, the tube door closed. A lever press opened the door for the rest of the one-hour session. In Phase 2, these rats were locked in the tube for 10 minutes before the door opened. Upon exiting, the door closed. To return to the tube they pressed a lever, opening the door for the rest of the session. The latency between pressing and tube entry decreased over sessions, indicating tube entry reinforced lever pressing. These results are difficult to reconcile with accounts of rat empathy based on the thesis that tube restraint distresses occupants.

Schedule- Induced drinking (SID) occurs when food-deprived rats are exposed to intermittent schedules of food reinforcement. The explanation of SID in terms of operant behavior reinforced by food has remained elusive. SID may be reinterpreted as operant behavior reinforced directly by water. Food deprivation and intermittent food delivery may function as establishing operations that increase the reinforcing value of water. The purpose of the study was to determine the effects of varying levels of food deprivation on water reinforcer efficacy during SID sessions. In successive conditions, four rats were food deprived at 100, 90, 80 and 75% of their free -feeding weights. During each condition, rats were exposed to daily SID sessions in which food was delivered according to a tandem fixed- time 176- s, differential-reinforcement-of-other-behavior 4-s. Each lever press resulted in water delivery. Holding constant the food schedule, water was then delivered according to a progressive- ratio 5 schedule, in which the response requirement increased for successive reinforcers. In general, decreasing body weight resulted in higher response rates and higher break points, suggesting that food deprivation increases the reinforcing value of water. The results complement earlier findings showing that SID is operant behavior reinforced directly by water.