|
| Matching to Sample 2 |
| Tuesday, June 1, 2004 |
| 12:00 PM–12:50 PM |
| Beacon D |
| Area: EAB |
| Chair: Joseph L. Cermak (University of North Texas) |
| |
| The Effects of Discriminability of Sample and Comparison Stimuli on the Acquisition of Conditional Discriminations in Adult Humans |
| Domain: Applied Research |
| YUSUKE HAYASHI (University of North Texas), Manish Vaidya (University of North Texas) |
| |
| Abstract: This study attempted to investigate the effects of the discriminability of sample and comparison stimuli on the acquisition of conditional discriminations in adult humans. The task involved learning four different types of conditional relations based on combinations of simple and complex stimuli serving as samples and comparisons. Simple stimuli consisted of one Japanese character and complex stimuli consisted of two Japanese characters. Four university students learned four instances of simple-simple, simple-complex, complex-simple and complex-complex conditional relations (where the first term describes the type of sample (simple or complex) and the second term describes the type of comparison stimuli used). The steepness of the learning curve in the four different conditional discrimination tasks was compared within individual subjects. Other measures included the latencies to observe sample stimuli and select comparison stimuli. Preliminary results suggest that stimulus discriminability plays a functional role in conditional discriminations and that discrimination of sample stimuli (successive discrimination) may be more important than discrimination of comparison stimuli (simultaneous discrimination) in the acquisition of conditional relations in adult humans with the matching-to-sample procedure. Future studies will attempt to relate these findings to the likelihood of stimulus-function interchangeability characteristic of outcomes in stimulus equivalence. |
| |
| Pigeons’ and Children’s Performance on an Identity Match-to-Sample Task under Fixed Ratio Schedules |
| Domain: Applied Research |
| JOSEPH L. CERMAK (University of North Texas), Manish Vaidya (University of North Texas) |
| |
| Abstract: Prior research on the effects of intermittent reinforcement on identity matching tasks has shown that accuracy is a linear, increasing function of ordinal position in fixed ratio schedules (Ferster, 1960; Nevin, Cumming & Berryman, 1963; Mintz, Mourer & Weinberg, 1966). Nevin (1967), using a simple discrimination procedure, intermittent reinforcement, and a limited hold, found that accuracy remained constant throughout the ratio. Instead, the probability of responding was shown to be a linearly increasing function of ordinal position in the fixed ratio. The current experiment sought to systematically replicate Nevin’s (1967) study with a conditional discrimination procedure. Four naïve pigeons were first exposed to an ascending series of schedule requirements (FR1, 2, 4, and 8). In the second phase, an x-second limited hold was placed on the selection of comparison stimulus. Results suggest that the addition of the limited hold contingency substantially attenuated the relation between likelihood of error and ordinal position in the programmed schedule. These data, together with the results of earlier studies, suggest that the decreases in accuracy seen under intermittent reinforcement is a schedule-typical effect. They also suggest that “incorrect” responses in these procedures may be likened to the pauses seen with intermittent reinforcement of simple keypecking. |
| |
|
| |