Association for Behavior Analysis International

Transferability of resistance to change was assessed. On Experiment 1, five pigeons were exposed to a control and experimental conditions. Control condition consisted of pigeons trained on a multiple schedule, with a rich component (VI 30s) and a lean component (VI 120s), signaled by different key lights. Pigeons were then exposed to the presentation of a pair of neutral stimuli, with no programmed consequences. Extinction and prefeeding probe sessions were done afterwards with the presentation of the neutral and discriminative stimuli. Experimental condition consisted on the same manipulations, however, pigeons were exposed to a second order conditioning procedure, using the neutral stimuli as second order stimuli and discriminative stimuli of the multiple schedule as first order stimuli. Probes were done afterwards. Results are interpreted according to Behavioral Momentum Theory, assessing transfer of function. Experiment 2, assessed transfer of persistence with five pigeons without a control condition. Pigeons were exposed directly to experimental conditions, as mentioned above, in order to avoid effects of history. Extinction probes were done in order to assess persistence to both neutral and discriminative stimuli.

When confronted with a choice between a smaller-sooner reinforcer (SSR) and a larger-later reinforcer (LLR), selecting the SSR over the LLR is considered impulsive. These two options differ in both magnitude and delay. The present study sought to develop an experimental arrangement to determine how each of these reinforcer dimensions contribute to impulsive choice. Rats responded under a concurrent-chains procedure. The magnitude and delay for one option was constant across the experiment, whereas the magnitude and delay for the other option changed across session blocks. Baseline data indicated that choice was controlled by the combined effects of reinforcement magnitude and delay. Using the generalized matching equation, sensitivity to magnitude was, on average, slightly higher than sensitivity to delay (.83 vs. .77, respectively) but this varied somewhat across rats. These data suggest this may be a practical and efficient method for quantifying the impact of the individual reinforcer dimensions when choice involves combinations of reinforcement dimensions (e.g., impulsive choice) in rats. Such a method may provide a sensitive baseline to examine the effects of neurobiological manipulations (e.g., drug administration) on impulsive choice.

Based on the assumption that meaningful stimuli can be members of pre-experimentally established equivalence classes, this ongoing research aims to assess the reliability of online sorting tests for verifying pre-experimental relations. Ten participants were asked to sort 40 pictures based on what they considered as correct and then, to name the sorted sets. These pictures were taken from seven likely pre-experimental classes; Birds (A1B1C1D1E1), Vessels (A2B2C2D2), Gardening (A3B3C3D3), Dentistry (A4B4C4D4E4F4), Baking (A5B5C5D5E5F5), Car Mechanics (A6B6C6D6E6F6), and Abstract items (A7B7C7E7F7G7H7I7). Each class comprised physically similar and dissimilar pictures in order to assess non-arbitrary and arbitrary relations. Five participants sorted stimuli according to the expected classes. Three participants mixed classes and named them according to their mixed features (e.g., P101 blended Birds and Gardening classes and named them as one single class, “Nature”). Two participants sorted Abstract items into different classes. Partial results support sorting as a reliable strategy for assessing stimulus relations. Although participants have mostly sorted the pictures as expected, differences were exhibited when they clustered pre-experimentally defined classes. Naming requirements may help to identify discriminative controls for relational responses. Future stages will expand the size of each class and analyze the consistency of relations across participants.

Previous research has found that choice in and of itself can be a reinforcer; however, research has also found there are limits to this, such as when there are too many choices. The current study, replicated and extended behavior-economic research using behavior-analytic methods by (a) determining the prevalence of preference for choice in a large number of adults, (b) evaluating large numbers of options on preference for choice and satisfaction, and (c) evaluating the effects of time constraints on an individual’s preference for choice and satisfaction with those choices. Results showed that many participants preferred choice over no-choice contexts. One third of the participants reached a breaking point during the choice overload phase and time constraint phase. Additionally, results indicated that it was more difficult to choose when there were more choices and when there was less time. These findings demonstrate that offering extensive amounts of options may have negative effects but suggest further research be conducted on more substantial contexts.

Conditioned reinforcement is widely used in applied behavior analysis, but variables influencing its efficacy are not well understood. Pavlovian learning underlies conditioned reinforcement and larger C:t ratios facilitate that learning (C is the inter-reinforcer interval and t is the conditioned stimulus (CS) ---> reinforcer interval). Following Pavlovian training with nonhumans, the CS evokes sign-tracking (CS approach) and/or goal-tracking (feeder entry). Prior studies have shown that larger C:t values promote sign-tracking, and separate studies report that the CS is more reinforcing in rats’ that naturally sign-track. In this rat study, we combined these separate findings by manipulating C:t values across groups (1.75, 3.5, and 12). Sign-tracking was measured during training and CS reinforcer efficacy was measured in sessions in which an operant nosepoke produced the CS alone. With data-collection ongoing, more sign-tracking occurs in the C:t 12 than in the 1.75 group (Figure 1). The amount of sign-tracking was a significant predictor of operant responding for the CS on testing day 1 (Figure 2). Finally, C:t 12 rats responded more for the CS than C:t 3.5 rats; the difference between C:t 1.75 and 12, thus far, only approaches significance. These findings suggest C:t ratios and sign-tracking play a role in conditioned reinforcer efficacy.