Association for Behavior Analysis International

A number of nonoverlap methods to calculate treatment effect estimates have been developed for single-subject experimental research studies. Although extant literature contains several studies comparing some of these methods with each other and visual analysis, there is no study in the literature investigating most of the available nonoverlap methods in a single study using same graphs and visual analysts. Purpose of our study was to compare 12 commonly-used nonoverlap methods with each other and visual analysis. Data were obtained from 25 studies focused on embedded instruction and schema-based instruction and included a total of 101 graphs. Treatment effect estimates using 12 nonoverlap methods were calculated for each graph by hand or using an online calculator. Five experts conducted visual analysis of each graph. Results showed that strong agreements existed between visual analysis and Percentage of Nonoverlapping Data (PND), TauNOVLAP, and Tau-U when raw data were analyzed, and PND, Percentage Of Non-Overlapping Corrected Data (PNCD), and Percentage of Data Exceeding a Median Trend (PEM-T) when categorized data were analyzed. Among 12 methods investigated, PND had the highest agreement rate with visual analysis, followed by PEM-T, Percentage of All Nonoverlapping Data (PAND), PNCD, Improvement Rate Difference, Nonoverlap of All Pairs, and TauNOVLAP. Overall, visual analysis appeared to be more conservative as most nonoverlap methods overestimated treatment effect. Additional research is needed to replicate findings of the present study.

Schizophrenia involves debilitating and pervasive symptoms that can cause the persons manifesting them significant functional impairments in nearly every aspect of their lives. Traditionally, schizophrenia is considered a primarily biological presentation, which is reflected in the use of antipsychotics as the most preferred therapeutic intervention. Alternative to the biomedical model, “psychotic” behavior patterns can be conceptualized as the presence of behavioral excesses and/or deficits that are maintained by social positive, negative, and/or automatic-sensory reinforcement rather than relevant neurochemical, neuroanatomical or gene abnormalities. By contextualizing an individual’s verbal or motoric bizarre behavior, oppositional responses, perseverative or hallucinatory speech, and stereotypies employing functional analytic methodology, behavior analysts can destigmatize the persons behavior and develop effective interventions, such as social skills training, that improve the impairments while simultaneously addressing limitations of the pharmacotherapeutic treatments. Functionally-based treatments support the idea that the primary goal should not be reducing symptomatology, but rather increasing the frequency of alternative more functional behaviors. Most importantly, they reinforce the view that schizophrenia can be sufficiently understood when considering the purpose behaviors serve in the person’s environment.

The Journal of Applied Behavior Analysis (JABA) periodically publishes articles meant to summarize an aspect of the journal as it relates to the broad mission of “publishing research about applications of the experimental analysis of behavior to problems of social importance.” Journal editors review and shape the work of authors, and in so doing, influence the journal’s direction and quality. They also serve as authors themselves contributing to the body of knowledge within a field. Mathews (1997) illuminated the role of editors of the Journal of Applied Behavior Analysis by analyzing the annual publication rates of 55 editors and associate editors, which showed an increasing trend of postdoctoral experience and greater number of publications prior to appointment as editors. The current paper extends the previous analysis over the subsequent two decades, including 44 editors, adds editor alma mater and gender, and allows multiple comparisons between the original work and current project.

Behavior analysts employ a function-based approach to the analysis of behavior. Responses are defined not by their topography, but by the relations that exist between individual responses and environmental variables. This kind of functional categorization has been fundamental to the development of an effective science of behavior. Although behavior analysts have established a conceptual system that allows for the categorization of many behavioral responses, some classes of behavior require further examination. Of specific interest is the notion of generalized operant classes. Typically defined by the observation of responding in untrained contexts, it appears that this term has been used in a variety of ways. A review of published literature shows that generalized response classes are invoked in the explanation of complex behavior (imitation, match-to-sample, language) for which first-order environmental variables may provide a more precise account. To begin, this poster will analyze the functional use of the terms “generalized” and “higher-order” response class. In addition, the poster will provide alternative accounts based on the concepts of atomic behavioral repertoires and abstracted stimulus dimensions.

One of the eminent standards set by Sidman’s (1960) Tactics of Scientific Research for behavioral research was to avoid inferential statistics in the form of null hypothesis statistical testing (NHST). Foster, Jarema, and Poling (1999) documented a growing trend in reporting of such inferential statistics in the Journal of the Experimental Analysis of Behavior (JEAB) through 1995. Zimmermann, Watkins, and Poling (2015) confirmed the trends into the 2010s. The present study examined articles appearing in JEAB volumes from 2011 to 2018 for use and reporting of NHST, species, and experimental design. Results show continued high reporting of between-subject designs (8-38% across the years), experiments with human participants (17-50%), and increasingly high reporting of NHST in Small-N design studies (24-60%). Whereas the former findings are consistent with previous predictions by Foster et al. and Zimmermann et al., the latter finding has not been reported previously. As we did not examine their appropriateness, it remains unclear what is behind this new trend of combining NHST with Small-N designs.

Skinner discussed a type of controlling response that has an effect on the probability response that follows. In his work The Technology of Teaching (1968), he termed this response as precurrent behavior, which he describes as a type of behavior that functions to make subsequent behavior more effective. Polson & Parsons (1994) further emphasized that this type of behavior does not produce the reinforcer itself but produces stimuli that affect the likelihood of behavior that follows, and therefore increases the likelihood of behavior resulting in reinforcement. Though most research about this operant pertains to precurrent behavior in the acquisition of problem solving skills, it can also play an important role in the acquisition and maintenance of precursor behavior that precedes problem behavior. In this conceptual analysis, we will review empirical literature on the topic as well as define the core features of precurrent behavior, discuss the altering effects precurrent behavior has on the current operant being observed, and discuss how precursor behavior that precedes problem behavior may be maintained by a precurrent operant contingency.

Drosophila melanogaster (fruit fly) has long been a model organism in genetics research. Behavioral genetics research with these organisms has relied on the use of neural activation or negative reinforcement by use of aversive stimulation (heat) to analyze behavioral contingencies and the effects of organisms’ phylogenetic endowments. Additionally existent research has required the constraint of the organism in a flight chamber. Optical neural activation of reward pathways has been demonstrated. However both of these paradigms prevent evaluation of naturalistic responding in the context of positive reinforcement contingencies. Drosophila melanogaster’s status as model organism in other fields like genetics means the ability to demonstrate positive reinforcement and schedule control with this organism will help to contextualize modern behavior analytic models alongside her sister sciences. Presented will be this laboratories initial designs, successes and failures at designing an operant chamber which permits free response to positive reinforcement. While the organism’s size and lifespan and present problems for traditional operant chamber design we hope that our designs open avenues for positive reinforcement research.

A large constraint placed on operant research with nonhuman organisms is the existing apparatuses which investigators use. These constraints may be purposeful, as with Skinner’s boxes, which were originally meant to control for confounding variables as best as possible. But many experiments are ran with prototypical Skinner boxes simply out of convention, instead of out of sensitivity towards the potential influence of confounding variables. An unfortunate side-effect of this is that many researchers—at least those fortunate enough to still have animal laboratories—feel obligated to spend a large amount of money on pre-made chambers to ask any number of research questions. The time and effort it would take to create one’s own chamber—or variations on it—may seem immense and overwhelming. However, work within the NRRL laboratory at the University of North Texas over the past year has indicated that it is feasible to build and program one’s own chamber cheaply. Additionally, our lab has created a “supercage” capable of housing at least a dozen rats that allows for social enrichment and the examination of nonlinear contingencies.