Association for Behavior Analysis International

Reading and writing behaviors are permeated by listening and speaking behaviors, and Equivalence-Based Instruction (EBI), has supported reading comprehension and improved speech accuracy in children with hard-of-hearing (HH) and cochlear implants. This work aimed to verify the effects of an EBI on oral word reading and picture naming in three hard-of-hearing children using bilateral hearing aids devices. In addition to behavioral reading and writing relationships, they were evaluated using standardized tests, and reading and vocabulary scores were demonstrated below expectations for their school year and chronological age. Participants received three ALEPP teaching units, EBI reading and writing software according to multiple baseline design. As a result, it was observed that naming pictures was much better (around 50% correct) than reading words (below 20% accurate) in the baseline. After teaching the respective units, everyone learned the relationships between taught stimuli, improved their naming of pictures and their reading of words emerged, achieving around 100% correct answers. Beyond being like those obtained by listeners, the results can support a proposal to complement regular education in the literacy process for children with hard of hearing and aid devices.

Human research demonstrates synchronous reinforcement is more reinforcing, effective, and preferred compared to asynchronous reinforcement, but there has been little research regarding this topic in non-humans. In prior human studies, the reinforcers were complex audiovisual stimuli (e.g., videos). A difficulty facing non-human research is identifying a continuous reinforcer suitable for studying both synchronous and asynchronous contingences, but sensory reinforcement may be a suitable paradigm with rodents according to the literature. Rats in a dark environment will press a lever to turn on a light, which has been shown to be a reinforcer. The current research uses this paradigm in the exploration of synchronous and asynchronous contingencies of reinforcement. During synchronous reinforcement, visual stimulus onsets with the depression of the lever and terminates with its release. During asynchronous reinforcement, depression of the lever starts a timer recording the response duration; release of the lever turns on the visual stimulus for a duration matched to the response. In the ongoing study, we have shown synchronous reinforcement maintains somewhat more responding than asynchronous reinforcement, similar to findings with humans. Ongoing phases of this research arrange choice between synchronous and asynchronous reinforcement to determine behavior allocation among different contingencies of sensory reinforcement.

The Odor Span Task (OST) is an incrementing non-matching-to-samples procedure to study remembering in rodents. Reinforcement is delivered for selection of odors when they are session-novel but not when presented later as a comparison. Testing occurs in an arena with multiple stimulus locations; rats remove a scented lid to obtain reinforcement. To examine whether rats track the sucrose pellet reinforcer rather than learn the OST, we presented four types of probe trials on separate days, mixed randomly with baseline trials, to rats that were performing accurately on the OST. On “no-bait” trials, no reinforcer was under either of the two stimulus lids. On “screens” trials, a pellet was in both stimulus cups; a plastic screen prevented access to the pellet in the S- cup. On “opposite” trials, the pellet was under the S- and no pellet was under the S+. On “all wrong” trials, two S- were presented on the trial with a pellet in one cup. Somewhat lower accuracy on probe (70%) compared to baseline trials (90%) was observed; on “all wrong” probes, performance was somewhat biased to the S- that contained a pellet. The screen condition may prove to be the most effective control.

We compared synchronous and asynchronous schedules on the acquisition and extinction of operant behavior reinforced by visual stimulation. Sixteen rats were exposed to a darkened operant chamber without illumination during a baseline period. Rats were randomly assigned to learn under one of two conditions. During Synchronous Reinforcement, the chamber light was illuminated whenever the response lever was held down. During Asynchronous Reinforcement, the light was illuminated after the lever press (the duration was matched to the duration of the press). The final condition was a return to the baseline condition to study extinction. After extinction, all rats repeated the study with the other contingency arranged on a different response lever. Lever pressing was acquired in both arrangements, and there was a slight tendency for increased pressing during asynchronous reinforcement. Press durations were noticeably increased during synchronous reinforcement compared with asynchronous reinforcement (see Figure), and this was the most dramatic effect observed. Responding declined during extinction, but was not affected by the conditions of acquisition. The findings suggest that the temporal alignment of reinforcement may shape different properties of behavior, e.g. holding the lever versus releasing it, laying the foundation for future study on further comparison between the two schedules.

Video-recording equipment is potentially useful for research within Behavior Analysis. Whether for human or nonhuman animals, recording an organism’s behavior, particularly when not otherwise observable, can provide valuable insight into answering experimental questions. Detailed within this poster is a low-cost method for acquiring, building, and constructing cameras using Raspberry Pi components. Raspberry Pi makes highly customizable micro-computers that can be used to fulfill a variety of needs. The cameras described within this poster were built to record sessions within standard two-lever operant-conditioning chambers and to develop sign- and goal-tracking indices for rats. Additionally, these devices can be operated remotely and can automatically transfer video files to other computers or storage systems over a local network. The cameras remain customizable after their construction as parts like the lens can be changed as needed to satisfy the demands of the current research project. A list of parts used to build the cameras used in our lab is provided, along with suggestions for replacement parts to satisfy other research purposes. We share this information and detailed plan to aid other researchers in collecting supplemental data in a relatively easy and affordable way.

Renewal occurs when extinguished behavior reemerges following a change in the extinction context. This study investigated the effects of conducting extinction sessions in multiple contexts, known as Multiple Context Training (MCT), on renewal with rats. During baseline, rats pressed a lever on a variable interval (VI) 30-s schedule to receive food pellets. Rats were then assigned to four groups. The two experimental groups underwent extinction of lever pressing in 6 and 12 different contexts, respectively, while the two control groups experienced extinction in a single context (one for 6 sessions and the other for 12 sessions). Finally, each group was exposed to the training context to test for renewal. Although prior studies suggested that extinction across multiple contexts may mitigate renewal, our findings did not support this hypothesis. No differences in mean response rates were observed between the 6-context and 12-context groups compared to their respective control groups during renewal testing. These results suggest that merely increasing the number of extinction contexts may not attenuate renewal. This highlights the complexity of context effects on relapse and underscores the need for further investigation into the factors that influence extinction generalization.

There is growing interest in testing interventions that reduce delay discounting (DD). Delay exposure (DE) training has shown promising results in rodents. It consists of rats pressing a lever to produce a cue stimulus for a 17.5 s interval, at the end of which a reinforcer is delivered (Renda et al., 2021). The present study aimed to address two limitations in DE research: extensive number of training sessions required (60 or more) and exclusive use of group designs. Using a within-subjects design, a respondent variation of DE and its standard control condition (Immediate Exposure-IE) were evaluated in DD of rats tested in adolescence and adulthood. All DE and IE standard aspects remained the same except that the presentations of delay cues and reinforcers were response independent due to equipment limitations that prevented using a third operandum. Eighteen male Sprague-Dawley rats were daily exposed to a delay-discounting task (DDT, Liley et al., 2019) until steady responding was obtained at late adolescence (starting postnatal day – PND98, Ghasemi et al., 2021) and early adulthood (week 25 = PND175). Eight of the 18 rats continued DD assessment plus different experimental manipulations added to the DDT across six conditions. The remaining 10 rats were kept as a waitlist control group (no DD assessment/treatment until PND 300). The first manipulation added to the DDT controlled for the additional access to pellets entailed in subsequent DE/IE conditions. This 60Pellets+DDT condition consisted of 1-hr pre-DDT response-independent access to the same amount of reinforcement delivered during DE/IE manipulations (i.e., 60 pellets). The third and fifth conditions consisted of DDT+DE and DDT+IE (counterbalanced across half of the animals) with 60Pellets+DDT reversals interspersed between them (fourth and sixth conditions). Based on subjects’ performance during the first six conditions, a second control manipulation was introduced. This Wait+DDT seventh condition entailed rats waiting in the operant chamber during the 60-min before the daily DDT session without operanda, delay cue, or pellets. All 18 rats, including waitlist, were exposed to a DDT retest starting on PND 300 (i.e., middle adulthood). Waitlist rats were then exposed to 60Pellets+DDT and DE conditions starting on PND 315. We found that exposure to 60Pellets+DDT condition during early adulthood unexpectedly produced important reductions in DD of five out of eight animals, with an average of 16 sessions needed to produce steady low DD. These rats maintained low DD levels throughout the remaining six conditions of the experiment, regardless the manipulation, including DE/IE. The DD of the three remaining animals also dramatically decreased during subsequent conditions and, when reached, remained low until the end of the experiment. DD of waitlist rats did not change between early- and middle-adulthood tests separated by ~20 weeks, and only slightly decreased under 60Pellets+DDT and DE conditions. These preliminary and promising findings suggest early- but not late-adulthood response-independent access to reinforcement reduces impulsive choice long-term with a relatively brief intervention. The contribution of respondent DE/IE, repeated DD assessment, and maturation on this effect is unclear. Further efforts aimed at reproducing these effects and identify the processes involved are warranted.

Negative reinforcement has been programmed using a variety of arrangements with humans and nonhumans. However, to our knowledge, no previous experiments have evaluated human responding on variable cycle schedules of avoidance, which may be more representative of avoidance schedules maintaining responding in daily life. In a human-operant arrangement, we evaluated acquisition and extinction of responding reinforced by variable-cycle schedules of money-loss avoidance. We compared responding maintained by a denser (10 s) and leaner (40 s) variable cycle avoidance schedule using a multielement design. Using a computer program, we presented six participants with two buttons, each corresponding to one condition. During acquisition, participants began each session with $20 for each condition and lost $0.10 to the variable cycle schedule. During extinction, no losses occurred. The leaner reinforcement schedule tended to lead to higher response rates and slower extinction than the denser schedule. The results also demonstrate a method of evaluating responding maintained by negative reinforcement in a practical, in-person, single-session format.

Eating disorders (EDs) are pathological conditions characterized by alterations in behaviors related to eating, significantly impacting physical health and psychosocial functioning. Among the functional aspects underlying the diagnoses of EDs, guilt emerges as a central symbolic context in establishing verbal responses essential to the development and maintenance of these disorders, enabling the transformation of aversive functions related to eating. Grounded in the behavioral analysis framework, this study explores how covert verbal behaviors, such as guilt, influence the establishment and maintenance of EDs. Using the Implicit Relational Assessment Procedure (IRAP), 34 university students were assessed on their relational responses involving healthy and fast-food items paired with phrases indicating guilt or indifference. Participants also completed the Sick Control Stone Fat Food Questionnaire (SCOFF) and the Eating Disorder Examination Questionnaire (EDE-Q) to measure ED-related symptoms. Preliminary results revealed a statistically significant differentiation between participants with and without ED indicators, especially in guilt-related fast-food responses (Group 2: t=3.073, p<0.01). These findings highlight the sensitivity of IRAP in identifying potential ED risks, offering insights into the role of verbal relations in ED pathology.

Students play a video game consisting of Multiple Concurrent Fixed Interval and Fixed Ratio schedules and told to find as many hidden objects as possible. Previous research in this lab has explored the effects of schedule parameter changes, such as schedule values and incentives, on sensitivity to the contingencies and strategies developed. Instructions in these studies has been minimal. The current study specifically evaluated the effects of instructions. For some groups the instructions were consistent with the contingencies, such as “some games you have played required hitting a troll multiple times” (considered appropriate instructions for FR contingencies) or “some games you have played required timing your responses” (considered appropriate for FI contingencies). Another group was given instructions opposite to the schedule contingency. Another group was given instructions for both FI and FR contingencies. Strategies used to hunt invisible items was recorded three times during each of the three sessions. Results suggest that sensitivity to FR contingencies appear faster than sensitivity to FI contingencies, and that instructions may have a greater effect on strategies developed under an FR than an FI schedule.opposite instructions, and another group was given instructions for both FI and FR schedules. Strategies used to hunt invisible items was recorded three times for each of the three sessions. Results suggest that sensitivity to FR contingencies appear faster than sensitivity to FI contingencies, and that instructions may have a greater effect on strategies used under an FR than an FI schedule.

Resurgence is the recurrence of previously established behavior (e.g., problem behavior) when conditions of reinforcement for current behavior (e.g., alternative behavior) worsen (Lattal, 2017). Understanding resurgence is important for designing durable clinical intervention. In a recent review, Podlesnik (2023) found a steep increase in the publication rate in the past decade; however, studies varied in terms of design, procedural manipulations, definitions of resurgence, and types of analyses. Typically, models of resurgence consist of three phases (training, elimination, and testing). LeCompte (2021) compared the three-phase model to a four-phase model with rats in response to mixed findings regarding the influence of exposure to extinction on resurgence. Results showed greater levels of resurgence for rats exposed to the three-phase procedure as compared to the four-phase procedure. The purpose of the current study is to replicate and extend LeCompte (2021) by comparing three- and four-phase resurgence models with young children in an analogue clinical setting. Preliminary results suggest both models can capture resurgence effects, but differences in resurgence magnitude may be more a function of the order of exposure (i.e., reductions during second exposure to extinction relative to first exposure). Results discussed in terms of clinical application of the basic resurgence assessment models.

Social Discounting is a quantitative description of how the subjective reinforcing value of an outcome decreases as social distance from the respondent increases (Jones & Rachlin, 2006). An individual may perceive relatives as socially closer than non-relatives and, therefore, be more altruistic with respect to relatives than non-relatives (Rachlin & Jones, 2008). Reciprocation history, defined by Osinski (2009) as “recipient’s loyalty”, has been shown to change the subjective value of an outcome decreases as a person’s loyalty decreases. The present study aims to evaluate how different social distance prompts, manipulating reciprocation and relatedness, impact social discounting. The present study exposed participants to three social distance prompts, each of which will prompt participants to order their social distance list by a different variable: (1) relatedness, (2) reciprocation history, and (3) closeness. After participants complete a social discounting task, rates of altruism and fit to the hyperbolic function as a function of social distance prompts will be evaluated.

Under the premise that chaining might occur between proximate responses independently of an explicit chaining contingency (“superstitious” chaining), the present study assessed whether the transition to response classes that occur later in the inter-reinforcement interval might reinforce responses that tend to occur earlier , such as adjunctive behaviors, regardless of the presence of an explicit chaining contingency. In the first experiment, rats were exposed to a fixed interval of lever pressing for food while water was concurrently available, until adjunctive drinking (SID) was stable and the typical bitonic pattern of drinking developed in the first segment of the inter-reinforcement interval. In the second experiment, rats were exposed to a fixed interval chained program in which wheel running produced the extension of a lever. In both experiments, test sessions were conducted where we measured the effects of removing the lever on SID and wheel running, while maintaining food delivery constant. In both experiments, we observed a decrease in the occurrence of the first behavior when the second behavior was absent and in close proximity with slight pressing. The similarity of the effects in both experiments and the relevance of the proximity between response distributions are consistent with a “superstitious” chaining hypothesis.

Two studies assessed the effects of asymmetric or symmetric resource distribution in prosocial interactions in same-sex or different-sex dyads. The experimental situation involved solving shared virtual puzzles. Participants could take their or their partner's pieces to complete the puzzle and obtain points only for themselves, or they could give pieces to their partner to work and gain points. In Experiment 1, sixteen university students were randomly distributed into two experimental groups of four dyads, each group exposed to different sequences of asymmetric (75%-25%) and symmetric (50%-50%) resource distribution. Participants worked in different experimental cubicles, visually and verbally isolated from each other. In Experiment 2, twenty-four university students were assigned into three groups of four dyads, exposed to the same resource distribution sequence (symmetric-asymmetric-symmetric), and differentiated by the participants' sex (man or/and woman). Participants worked in the same cubicle and could establish verbal exchanges. Findings suggest that asymmetric resource distribution affects prosocial interactions, favoring taking away resources from others over giving resources to another. However, effects differ by initial exposure to symmetric or symmetric distribution and dyad composition. Results are discussed in terms of the relation between institutional contingencies and prosocial interactions.