Association for Behavior Analysis International

Decision about how much time to dedicate to certain behaviors can depend on both the cost for engaging in the behavior and its associated benefits. Past research has demonstrated the feasibility of hypothetical purchase tasks in measuring the relative demand for reinforcers such as alcohol and prescription drugs. However, the application of hypothetical purchase tasks to understand behavior related to scholarships has not been previously explored. In this study, hypothetical purchase tasks were used to examine the relative reinforcing efficacy of scholarships of varying amounts anchored to either academics or athletics. Specifically, students indicated how much time they would study/train per day if they were offered 100%, 90%, 80%, 70%,...0% academic/athletic scholarships. We received 86 valid academic scholarship responses and 43 valid athletic scholarship responses. All indices of demand were higher for academic scholarships than athletic scholarships. The specific fit of demand indices with this dataset and whether purchases in terms of time dedicated toward specific activities can be usefully incorporated into standard behavioral economic models are discussed.

While many established delay discounting models fit observed indifference point data closely, more attention is needed to assess whether models are sufficiently specified such that data generated from these models closely resembles human data. Nonlinear least squares (NLS) is the most widely used approach to fit indifference points to models because it adheres closely to observed data. The common strategy of considering data with normally distributed residuals is flawed here, because such models can produce invalid indifference points above the larger later amount or below zero. This indicates that the models we rely on are not fully specified in an adequate manner so as to describe human-like patterns of behavior. Thus, the class of Monte Carlo simulation techniques (useful for everything from sample size calculations to comparative assessment of completing analytical techniques) is not fully reliable since data produced via simulations do not resemble human data adequately. We introduce a class of nonlinear beta regression models that provide excellent fit to discounting data and also improve simulation based approaches due to their ability to honor bounds on indifference point data.

Basic studies within the field of stimulus equivalence often arrange training phases with gradual thinning of programmed consequences before test for emerged relations. One reason for this is to prepare the participant for responding in extinction during test. At the same time, a gradual training procedure increases the minimum number of trials and time spent in conditional discrimination training. Sidman (1994) suggested that procedures in the applied setting need not be as austere as in the laboratory. 30 adult participants, 17 women and 13 men, were randomly assigned to (a) gradual thinning of programmed consequences, (b) no thinning phases before test, and (c) overtraining. Overtraining was arranged to control for differences in minimum trials required in the two other conditions. Results showed that most participants formed equivalence classes in gradual thinning and overtraining, but no significant results were obtained. Furthermore, baseline relations were marginally better maintained in overtraining than in gradual thinning indicating that number of trials in baseline training was a core variable for class formation in these participants. Further research on the role of gradual thinning of consequences and number of trials is needed before firm conclusions can be drawn.

Choice bundling (when a single choice produces repeating consequences over time) has been shown to increase valuation of larger later (LL) rewards in cigarette smokers. However, no prior work has investigated the effects of choice bundling in individuals in recovery from substance use disorders (SUDs). In this replication of Stein et al (2022), 194 participants from the International Quit & Recovery Registry, were randomized to complete delay discounting tasks for either monetary gains or losses. Participants completed a six-trial task to establish Effective Delay 50 (ED50; the delay required for an outcome to lose half of its value) and three adjusting-amount tasks with bundle sizes of 1, 3, and 9 rewards per choice (order counterbalanced). Differences in discounting gains and losses were controlled using ED50 individual values to set the LL amount. ED50 values were not significantly different between losses and gains (p=0.526). Choice bundling significantly increased valuation of LL gains and losses (p<0.001), with significantly greater effects in ascending losses (p<0.01) and descending gains (p<0.05) compared to ascending gains. Future research examining the role of remission status on choices may elucidate the potential clinical utility of choice bundling in SUD treatment.

An observing response (OR) is a response that produces access to discriminative conditions. In typical matching-to-sample arrangements, ORs are made to the sample stimulus, which then produce the comparison array. Previous research suggests that ORs may facilitate or be necessary to produce or maintain matching performances. The current study sought to extend prior research on ORs in a gamified human operant arrangement hosted in Minecraft Education Edition. In this study, undergraduate students completed identity matching-to-sample trials in which the 12-stimulus comparison array was immediately visible. An OR was possible, but not required, which produced the sample. The OR requirement varied across sessions (FR 1, FR 5, or FR 10), as well as the number of duplicate comparison stimuli. Given the sample, the probability of selecting the target comparison changed from 1.0 to .08 across trials as the number of unique comparison stimuli decreased. Participant’s total number of ORs per trial was measured. The OR requirement and number of unique comparison stimuli were shown to be functionally related to the participant’s emission of the OR. Implications for ORs in matching-to-sample arrangements in applied settings will be discussed.

Effective note-taking is a critical skill that may enhance learning outcomes for students across age and disability status. Note-taking is also a form of mediation which may be directly observed. Frampton et al. (in press) found that the inclusion of a stimulus fading package to teach note-taking in the form of a graphic organizer (GO) resulted in moderately high equivalence yields with young adults. The present study utilized a pre-training treatment package to teach young adults to construct a GO during matching-to-sample baseline relations training (MTS-BRT). The package included video illustration, voice-over instructions, and feedback when practiced with familiar stimuli. Following pre-training, a non-concurrent multiple probe design across participants was used to evaluate the effects of MTS-BRT with arbitrary stimuli in three five-member classes. The participants engaged in GO construction during MTS-BRT and efficiently acquired the trained relations. Both participants passed the posttest on the first attempt, though neither drew a GO with integrity during the posttest. These results extend prior findings suggesting that effective note-taking may facilitate learning of the baseline relations and result in high yields. The voluntary demonstration of notetaking by both participants during MTS-BRT suggests this may be a beneficial addition to equivalence-based instruction.

Laboratory research has a long history of contributing to applied psychology. Behavior analytic gambling research has been largely experimental and translational, informing applied work. Previous data from the Rider Gambling Laboratory has shown that venue and laboratory data are comparable in behavioral factors of number of bets and time spent gambling. The data suggested that, generally, people gamble more and longer at the venue than in the laboratory. The previous data were collected from people already at the venue and people who participated in research form a university population. In this study, to further examine the setting effect, leisure gamblers gambled in the laboratory setting and in the venue setting in different sessions in a single case experimental design ABAB fashion. Consistent with the observational data, generally participants gambled more and longer at the venue than the laboratory. Participants also were given a self-directed gambling intervention, after which gambling generally reduced across the settings, though gambling reduction in the laboratory was less pronounced.

Positive practice overcorrection (PPOC) is a commonly used punishment procedure within a variety of settings. Early research on PPOC was conducted within academic, clinical, and home settings (e.g., Doleys et al., 1976; Carey & Bucher, 1983; Cole et al., 2000); however, it seems necessary and important to make the connection between applied and basic realms to fully understand the mechanisms underlying PPOC’s effectiveness. This study is designed to examine the mechanisms underlying the effects of PPOC procedures using an animal model (e.g., subjects are pigeons). First, it is unclear if the work requirement or timeout (TO) from positive reinforcement is responsible for behavior suppression. Second, the baseline through which we compare punishment effects is a crucial consideration when designing an experiment that not only is in line with previous basic punishment research, but also has translational outcomes. In Experiment 1 (single reinforcement key), there is only one source of reinforcement; the same response produces reinforcement and PPOC/TO. In Experiment 2 (concurrent), there is an alternative source of reinforcement without PPOC/TO. In Experiments 1 and 2, the average duration to complete the terminal PPOC ratio is yoked to a TO to disentangle the confounded components of PPOC.

Gender role refers the set of behaviors and expectations that should govern the way women and men are, feel and act (González, 1999). In this way, the goal of this study is to assess if gender roles influence the Pavlovian conditioning of behaviors that occur in situations of street harassment. Forty participants (18-30 years old) completed the Gender Attitude Scale (GRAS), and were later exposed to a Pavlovian experimental task. In the Training Phase (30 trials), vignettes of neutral interaction between a man and a woman, and street harassment with a geometric figure were presented, each of these paired with appetitive or aversive stimuli. In the following phase (Test Phase, 3 trials), participants were asked about the association of the previously-presented vignettes. Our results showed some evidence that people with lower scores on the GRAS tended to have greater difficulty in appetitive conditioning street harassment and greater ease in aversive conditioning; whereas, people with high scores tended to have the opposite results. Altogether, gender roles might play an important function in people's ability to identify risky behaviors for themselves.

There have been numerous studies that have assessed the effectiveness of self-management strategies for on-task behaviors. However, many of these studies focus on increasing on-task behaviors in younger populations. This current study was conducted to increase on-task behaviors in an applied behavior analysis graduate intern (myself), while completing internship assignments in a shared office space. A changing criteria design was used to increase time spent on-task and momentary sampling was used to assess if on-task behaviors occurred during each interval. The criteria for on-task behavior increased, until the final goal time of 25 minutes was met. Based on the results, it was concluded that this intervention effectively increased on-task behaviors. Additionally, during the maintenance trials, on-task behavior remained consistent with the final goal criteria. This shows that the effects of the treatment remained consistent 3 months after conclusion of the intervention.

Temporal discounting is an important measure of impulsivity associated with a variety of behavioural problems. Sosa and Santos (2018) proposed that impulsivity demonstrated in a temporal discounting task may reflect the overlap between the primary and secondary reinforcers for the immediate and delayed choice. If so, they argued that less discounting should be observed if the immediate and delayed reinforcers were different to each other. University students (n = 320) completed an online discounting task in which the immediate and delayed reinforcers were either the same or different to each other in a 2x2 fully randomised between-groups design. The two reinforcers employed were a subscription for varying lengths of time to either a music streaming (music) or food-delivery service (meal). When the delayed reinforcer was a meal, shallower discounting was obtained when music was the immediate reinforcer as predicted by Sosa and Santos. However, there was no difference between the two conditions in which music was the delayed reinforcer. These results provide some support for Sosa and Santos, but a more detailed consideration of cues associated with reinforcement in humans may be required before this account can be fully evaluated.

The purpose of this study was to assess changes in visual scanning behavior as a function of stimulus equivalence training. Four adult participants completed three stimulus-association training sessions and a follow-up stimulus equivalence test. Eight stimuli were employed. Each stimulus contained a Girl Face, a Boy Face, a Man Face, or a Woman Face paired with three fish shapes differing in color. Participants were trained to correctly match the color of the fish with a given face. Each training session consisted of 12 trials. After the training sessions a 6-trial equivalence test was given. Across training trials, there was a significant decrease in the number of fixations and fixation duration to the stimulus components; and visual response latencies became faster to the correct stimulus association. Participants’ attending behavior became more selective, suggesting greater volitional control, and less influenced by stimulus salience features that guide perception. In summary, visual scanning measures should provide a superior direct assessment of an individual’s attending behavior to stimulus features (over-selectivity or under-selectivity) and the rate of stimulus comparisons, which should better serve the behavior analyst in the designing and implementation of appropriate intervention schemes for higher order learning in at-risk and non-verbal populations.

The purpose of the current investigation was to evaluate whether within-session exposure to high-probability of winning during a gambling game produces tolerance-like behaviors, e.g. progressively increased bet size, and if such behavior is maintained when exposed to low-probability win conditions. Participants were assigned to either an experimental or control group. Each group played 10 practice hands of blackjack. The experimental group played 25 hands of blackjack where winning outcomes occurred on 80% of hands, followed by 25 hands in which the winning outcomes occurred on 20% of hands. The control group played 25 hands consisting of only the 20% winning condition and each participant was started with a bankroll equal to an experimental participant bankroll prior to the low-winning condition. To evaluate tolerance development, a dependent T-Test was used to compare the experimental group's average bet size in the high-winning condition with the low-winning condition. A one-way ANOVA compared the experimental group's average bet size to the control group’s in the low-winning condition, i.e. establishing whether or not decreases in bet size are a function of the prior winning history. Results of these analysis provide implications for the establishment of within-session tolerance and the development of potential exposure treatment exercises.

The COVID pandemic forced an abrupt transition to distance learning, but the initial diagnoses show that strategy was not very successful. In Mexico, the Human Capital Project identifies that not only hundreds of students abandoned their studies, but those who stayed lost knowledge equivalent to two years of schooling. Despite these discouraging results, we cannot abandon distance learning as it may still a useful tool in today's world. One way distance learning can be improved is by using collaborative tasks because in them students engage more quickly in effective social networks. In this context, we tested a remote and synchronized collaborative task in which complexity level served as independent variable. Six teams of 5 members of fourth-semester Psychology students participated. Each member worked remotely, but simultaneously on two shared documents (google docs and google slides, respectively). The task consisted of five exercises, each one developed on a different complexity (i.e repetition vs inference). The dependent variables, frequency and duration of completed trials, showed differences by complexity level. The possibilities for structuring remote tasks that allow complex and structured learning that go beyond memorization and repetition tasks are discussed.

For individuals with automatically maintained self-injurious behavior (SIB), it is best practice to conduct a competing stimulus assessment to identify stimuli associated with reductions in SIB. Notably, individuals with Subtype 2 and 3 SIB are more likely to be treatment resistant to common interventions established as effective for individuals with Subtype 1 SIB, and thus often require additional tactics to reduce SIB. Hagopian et al. (2020) found that augmenting the CSA (A-CSA) by incorporating response promotion and disruption tactics can assist with identifying high competition stimuli (HCS) to reduce SIB. In this study, we replicated procedures from Hagopian et al. for an 8-year-old male diagnosed with autism who engaged in head-directed SIB (H-SIB) classified as Subtype 2. Various stimuli were assessed across Free Access (FA) and Response Promotion and Response Disruption (RP+RD) conditions until multiple HCS were identified. The largest number of HCS for reducing SIB, 8, were only identified after RP+RD tactics were included; 6 of these 8 HCS were associated with greater than 50% engagement. During the Repeated Free Access condition, when disruption tactics were removed, only 1 stimulus met the criteria to be a HCS. Clinical recommendations for practitioners and implications for future research will be discussed.

For most studies, assessing the recurrence of socially-compatible or -incompatible behavior requires the training of a target, and an alternative response while the former is under extinction. Once both behaviors are under extinction conditions, target responding increases (i.e., resurgence is observed). Training more than one alternative response could reduce the probability of resurgence once all responding is placed in extinction. Furthermore, alternative-response recurrence could be observed as a result training primacy or recency. Using PORTL, undergrad students were exposed to a modified-resurgence procedure. Spinning a toy was established as target responding for all students. During the second phase, students had to shake, slide, pile and turn-over different toys; spinning was not reinforced. During the third phase, target and all alternative responses were under extinction. For most of the students, as alternative responses were trained, target and previously-trained alternative responding decreased to near-zero responses. During the third phase, target and all previously-trained alternative responding increased. However, target responding was never greater than any of the alternative responses. These results suggest that resurgence can be prevented by training more than a single alternative response.