Perspectives on Behavior Science最新文献

For behavior analysts, social validity should be a standard and evolving component of our own self-evaluation process. Our field has known that for decades, but we have not always been good about implementing such self-evaluations. Recent criticism of our field should be viewed as setting the occasion for self-reflection and improvement of our practices. To accomplish those aims (self-relection and improvement), we need to identify specific practices to extinguish, modify, or (at least) explain better. We will address this challenge in three contexts. First, we provide discussion on what is "meant" by "ABA." Second, we distinguish between ABA as a discipline and ABA as an intervention (we contend it is the former and not the latter). Third, we acknowledge and identify nuances of social validation, often by drawing parallels and comparisons to medical fields.

Although associative learning research has been conducted for more than a century, little is known about learning processes when subjects are not alone, but in a group-a phenomenon termed collective learning. In collective learning situations, the behavior of conspecifics may serve as an associative cue for learning, like any other stimulus during individual learning. Two experiments investigated how individual versus collective training affects associative learning. Experiment 1 utilized a simultaneous discrimination task, whereas Experiment 2 implemented a serial go/no-go discrimination task. In both experiments, rats were trained either individually or collectively, exposing them to two distinct stimuli with only one of them signaling the availability of food reinforcement. Following acquisition training, all rats were tested both individually and collectively. Contingencies were then reversed: the previously nonreinforced stimulus now signaled the availability of food, and the previously reinforced stimulus now signaled the absence of food. Following reversal training, the rats were again tested individually and collectively. Results from both experiments suggest that the training condition (individual or collective) had little effect on learning the cue-outcome association. However, individual training negatively affected test performance in a collective context. These results suggest that collective training may have a facilitative effect on learning and points out key methodological considerations for more in-depth examination of this effect.

Supplementary information: The online version contains supplementary material available at 10.1007/s40614-025-00450-8.

The concepts of reinforcement and punishment arose in two disparate scientific domains of psychology and artificial intelligence (AI). Behavior scientists study how biological organisms do behave as a function of their environment, whereas AI focuses on how artificial agents should behave to maximize reward or minimize punishment. This article describes the broad characteristics of AI-based reinforcement learning (RL), how those differ from operant research, and how combining insights from each might advance research in both domains. To demonstrate this mutual utility, 12 artificial organisms (AOs) were built for six participants to predict the next response they emitted. Each AO used one of six combinations of feature sets informed by operant research, with or without punishing incorrect predictions. A 13^th predictive approach, termed "human choice modeled by Q-learning," uses the mechanism of Q-learning to update context-response-outcome values following each response and to choose the next response. This approach achieved the highest average predictive accuracy of 95% (range 90%-99%). The next highest accuracy, averaging 89% (range: 85%-93%), required molecular and molar information and punishment contingencies. Predictions based only on molar or molecular information and with punishment contingencies averaged 71%-72% accuracy. Without punishment, prediction accuracy dropped to 47%-54%, regardless of the feature set. This work highlights how AI-based RL techniques, combined with operant and respondent domain knowledge, can enhance behavior scientists' ability to predict the behavior of organisms. These techniques also allow researchers to address theoretical questions about important topics such as multiscale models of behavior and the role of punishment in learning.

Delay discounting is a behavioral phenomenon in which the subjective value of a reinforcer decreases as the reinforcer becomes more delayed. Two procedures are commonly used to assess how the value of a reinforcer changes as a function of delay: adjusting-delay and adjusting-amount. The evolutionary theory of behavior dynamics (ETBD) is a complex systems theory that uses an algorithm based on Darwinian principles of natural selection to animate artificial organisms. The behavior of artificial organisms animated by the theory are its predictions, and the theory has been shown to make accurate predictions about how living organisms behave in a variety of experimental arrangements. In the present article, we generated predictions with the ETBD for adjusting-delay and adjusting-amount procedures and evaluated whether these predictions align with live-organism delay discounting. The predictions were generated using modified procedures that could be conducted with continuous choice arrangements rather than discrete trials; however, despite these procedural differences, the ETBD's predictions were generally consistent with equations known to describe live-organism delay discounting well. This suggests that the ETBD might be used to generate other predictions that could expand our understanding of delay discounting.

Private equity (PE) refers to a financial investment model that involves investment firms pooling funds from multiple investors to acquire, manage, and resell companies for profit. PE firms have identified applied behavior analysis (ABA) service providers as a lucrative investment opportunity, leading to a surge of involvement within the industry. Although concern has been expressed about PE involvement in ABA, there are few resources available to behavior analysts interested in understanding its potential impacts. In this article, I will provide an overview of the risks and appeals of PE ownership and offer my reckoning of what may come of PE involvement in ABA.

Many recent studies have investigated rats' choice between drug and nondrug reinforcers to model variables influencing drug taking in humans. As research using this model accumulates, the complexity of factors affecting drug choice has become increasingly apparent. This review applies a behavioral economic perspective to research that has used this model. The focus is on experiments that have manipulated behavioral economic variables in studies of rats' choice between drugs like cocaine or heroin and nondrug reinforcers like saccharin or social interaction. Price effects, reinforcer interactions (i.e., as substitutes or complements), economy type, and income effects are described. Results of experiments testing the impact of these variables on rats' choice are presented and analyzed. Although rats' behavior in this model often conforms well with behavioral economic principles, there have also been instances where further explanation is required. By appreciating the behavioral economic context in which rats' choice between drug and nondrug reinforcers occurs, and by recognizing that both consequences and antecedents can play important roles in this behavior, our understanding of the complexity of factors involved in drug choice can be increased.

Research on the role of time in associative learning has changed our understanding of what an association is. It is a measurable fact about the distribution of events in time, not an altered activation-conducting connection in a mind, brain or net. Associative learning is the process of perceiving temporal associations and deciding to act on them. Informativeness- the ratio of a conditional rate to an unconditional rate-is the essential empirical variable, not the probability of reinforcement. The communicated information between temporally associated behavioral and reinforcing events is the log of informativeness. Because the time units in the rate estimates cancel, associative-learning is time-scale invariant: Perceivably associated events may be arbitrarily widely separated. There are no windows of associability nor decaying eligibility traces. The learning rate-operationally defined as the reciprocal of reinforcements prior to the appearance of a conditioned response-is an almost scalar function of relative temporal separation, as measured by informativeness. The central role of informativeness unites our understanding of Pavlovian and operant/instrumental phenomena, revealing unexpected quantitative and conceptual communalities.

Researchers have identified questionable research practices that compromise replicability and validity of conclusions. However, this concept of questionable research practices has not been widely applied to single-case experimental designs (SCED). Moreover, to date researchers have focused little attention on improved research practices as alternatives to questionable practices. This article describes initial steps toward identifying questionable and improved research practices in SCED. Participants were 63 SCED researcher experts with varying backgrounds and expertise. They attended a 1-day virtual microconference with focus groups to solicit examples of questionable and improved research practices at different stages of the research process. A qualitative analysis of over 2,000 notes from the participants yielded shared perspectives, resulting in 64 pairs of questionable and improved research practices in SCED. Our results highlight the need for further evaluation and efforts to disseminate improved research practices as alternatives to questionable practices.

Although applied behavior analysis (ABA) is regarded as providing the gold standard for interventions designed to meet the needs of autistic individuals in the United States, elsewhere this is not the case. In Northern Ireland, for example, successive governments have portrayed ABA simply as one of a number of commercially available interventions for autism. In this article, I argue that this view arises directly from the practice of behavior analysts who have courted the development of branded versions of ABA at the expense of promoting ABA directly. Because clinicians who advise government ministers are not trained in ABA, it is understandable that a discrimination issue arises whereby ministers are then encouraged not to invest in only "one of the commercially available interventions." To address this problem, the article ends with a suggestion in how a specially designed ethical code of practice might hold behavior analysts accountable for the discrimination problems that could arise as a consequence of their actions in countries struggling to promote the uptake of ABA.

Recent attacks on applied behavior analysis (ABA) by neurodiversity advocates share a common theme with opposition to other supports, such as subminimum wage vocational programs and congregate residential settings: the intact mind assumption, which maintains that even profoundly autistic people have typical intelligence, even if they present as severely cognitively impaired. This article examines the history of the intact mind assumption, which was largely shaped by psychoanalytic theory in the mid-20^th century, as well as its impact on contemporary disability policy and practice.