Association for Behavior Analysis International

This symposium presents two experimental studies and one conceptual paper discussing a behavioral account of problem solving where the emission of a precurrent response increases the probability of a response that is scheduled for reinforcement. In both studies, mediating verbal behavior may have allowed access to already-established repertoires. The investigation by Esch and Esch used a verbal rehearsal procedure during a delay-to-respond to increase correct stimulus selection when shown an array of numerals in a problem-solving joint control task. Preliminary results show that participants correctly selected numerals when instructed to rehearse them during a delay. Sundberg et al. present the results of a study comparing performances of individuals defined as either high-verbal or low-verbal on a matching-to-sample task. Results indicated that performance deteriorated when rehearsal (i.e., covert verbal behavior) was disrupted, thus preventing the establishment of joint control over the selection response. Hall presents an analysis of abstract control in terms of its role in verbal problem solving and discusses how such control might be established via strategic teaching. The discussant, Dr. Mark Sundberg, will offer remarks regarding these papers.

Much complex behavior requires probing established verbal and nonverbal repertoires to present (access?) stimuli that will evoke a response that currently is unable to be emitted in the absence of those stimuli (Skinner, 1957). This study follows previous research findings (Esch et al., 2010) showing that, for some individuals, a self-echoic response after a correct echoic response is weakened when a delay-to-respond is imposed. Such discrepancies between echoic and self-echoic behavior may be predictive of deficits in verbal problem solving. Following echoic and self-echoic assessments, participants were asked to find the correct numeric sequence when shown 4 quadrants of various numeric sequences on a problem-solving joint control task (Lowenkron, 1998). In an ABCBC experimental design, participants were asked to find the correct sequence under 3 conditions: (a) instruction and sequence presented simultaneously, (b) sequence presentation delayed by 5 s, and (c) instructed echoic and self-echoic rehearsal provided during a 5-s delay to presentation. We discuss the results of this study as it relates to teaching selection-based problem-solving skills and a descriptive autoclitic as a form of automatic reinforcement.

There is a sizeable body of basic behavioral research that employs matchingto-sample (MTS) preparations with verbal adults as participants. The goal of much of this research involves the study of complex behavioral relations (e.g., problem solving, emerging behavior, equivalence, relational frames). The current empirical investigation examines the role of multiple control and covert verbal behavior in high-verbal participants (college students) and low-verbal participants (adults with severe to moderate developmental disabilities). The data demonstrate that additional independent variables are responsible for the comparison stimulus selection behavior for the high-verbal participants. These variables primarily involve covert verbal behavior that occurs for verbal participants between the presentation of sample stimuli and the selection of a comparison stimulus. The covert verbal behavior then provides multiple sources of stimulus control in the form of joint control over selection behavior (Lowenkron, 1992). When the covert verbal behavior is disrupted, joint multiple control becomes hard to establish and performance clearly deteriorates. The implications of these results on current experimental practice and claims regarding emerging behavior are discussed.

The study of complex human behavior including transfer versus functional independence between verbal relations, and emergent problem solving continues to intrigue behavior analysts. Such work serves to counter the arguments of linguists and others who may claim that behavior analysis cannot account for behavior, which occurs for the first time without direct training, and is very important in the area of education. Children with and without language delays can greatly benefit from strategic teaching procedures designed to establish "general patterns of responding that can in turn be used to produce effective responses in an infinite variety of situations" (Alessi, 1987). To successfully generate complex emergent relations, it appears critical to conduct a detailed analysis of the relations to be trained and tested, then strategically program for transformation of stimulus function and recombination of repertoires. Engelmann and Carnine (1982) have described such an approach. Abstract verbal responding may range in complexity from simple abstract tacting to verbal problem solving. Regardless of the complexity of the performance, when abstract control is present, an "observed sameness" in a set of examples controls a consistent pattern of responding. The present paper will analyze the abstract control that appears to be present in transfer between certain verbal operants and speaker and listener relations, and verbal problem solving, and discuss how such control might be established via strategic teaching.