Instructions can override the influence of programmed schedules of reinforcement. Although this finding has been interpreted as a limitation of reinforcement schedule control in humans, an alternative approach considers instructional control, itself, as a phenomenon determined by subjects' reinforcement histories. This approach was supported in a series of experiments that studied instructional and schedule control when instructions either did or did not accord with the schedule of reinforcement. Experiment I demonstrated that accurate instructions control discriminative performances on multiple avoidance schedules, and that such control persists in a novel discrimination. Experiments II and III showed that elimination of instruction-following occurs when inaccurate instructions cause subjects to contact a monetary loss contingency. Experiment IV demonstrated the reinforcing properties of accurate instructions. Skinner's view of rule-governed behavior is consistent with these findings, and can be extended to account for many aspects of instructional control of human operant behavior.

Categorization and concept learning encompass some of the most important aspects of behavior, but historically they have not been central topics in the experimental analysis of behavior. To introduce this special issue of the Journal of the Experimental Analysis of Behavior (JEAB), we define key terms; distinguish between the study of concepts and the study of concept learning; describe three types of concept learning characterized by the stimulus classes they yield; and briefly identify several other themes (e.g., quantitative modeling and ties to language) that appear in the literature. As the special issue demonstrates, a surprising amount and diversity of work is being conducted that either represents a behavior‐analytic perspective or can inform or constructively challenge this perspective.

Michael (1975) reviewed efforts to classify reinforcing events in terms of whether stimuli are added (positive reinforcement) or removed (negative reinforcement). He concluded that distinctions in these terms are confusing and ambiguous. Of necessity, adding a stimulus requires its previous absence and removing a stimulus its previous presence. Moreover, there is no good basis, either behavioral or physiological, that indicates the involvement of distinctly different processes, and on these grounds he proposed that the distinction be abandoned. Despite the cogency of Michael's analysis, the distinction between positive and negative reinforcement is still being taught. In this paper, we reconsider the issue from the perspective of 30 years. However, we could not find new evidence in contemporary research and theory that allows reliable classification of an event as a positive rather than a negative reinforcer. We conclude by reiterating Michael's admonitions about the conceptual confusion created by such a distinction.

Following the emergence of two three-member equivalence classes (AlBlCl and A2B2C2), 5 college students were exposed to one or more changes in the reinforcement contingencies controlling baseline conditional discriminations. AC relations were either reversed (i.e., C2 was reinforced and Cl punished when Al was the sample; Cl was reinforced and C2 punished when A2 was the sample) or arranged randomly (i.e., C2 and Cl were reinforced and punished equally often in the presence of Al and A2). In a third condition, the original AB and AC relations were reversed. Results showed that although baseline conditional discrimination performances were under the control of reinforcement contingencies, and performances on symmetry trials varied with baseline responding for 3 of 4 subjects when contingencies were reversed, performances on transitivity probes remained consistent with the initial equivalence class. These inconsistencies between probe and baseline performances were striking because conditional discriminations are thought to be the determinants of equivalence class performance. Similarly, the contrast between performances on symmetry and transitivity probes was of theoretical interest because equivalence classes are defined by congruent patterns of responding on probe trials.

Rationale A number of tasks are used to assess working memory in rodents, but the Odor Span task (OST) is unique in studying performance as a function of the number of stimuli to remember. Objectives The purpose of the present study was to better characterize the behavioral pharmacology of the OST by exploring the effects of several amnestic agents including an NMDA antagonist (dizocilpine), a positive GABA-A modulator (chlordiazepoxide), an anticholinergic compound (scopolamine) and as a negative control, an opiate receptor agonist (morphine). Methods Rats were trained to perform on the OST which is a non-match-to-sample procedure with an incrementing number of sample odors to remember as the session progresses. Trials with a simple odor discrimination task (SD) were interspersed to provide a control for effects unrelated to memory load. Results All four drugs disrupted performances on the OST task in a dose-dependent fashion, but only the NMDA antagonist dizocilpine produced impairments that were clearly dependent on the number of stimuli to remember. Dizocilpine impaired OST performance at a dose (0.1 mg/kg) that did not affect SD and that impairment depended on memory-load. Chlordiazepoxide (3.0 mg/kg) also produced amnestic effects that were manifest by shorter memory spans and runs of correct responding. In contrast, morphine and scopolamine impaired OST accuracy only at doses that also disrupted SD (18.0 and 0.3 mg/kg, respectively). Conclusions These results provide evidence of NMDA and benzodiazepine modulation of working memory as assessed by the OST.