|
| Molar and Molecular Influences on Behaviour |
| Friday, November 30, 2001 |
| 10:00 AM–10:50 AM |
| Palladian Refectory Hall |
| Area: EAB; Domain: Applied Behavior Analysis |
| Chair: Phil Reed (University College London) |
| Abstract: These papers explore a range of issues connected with analysing whether behaviour is controlled by molar or molecular factors. |
| |
| Some Causes of Pauses in Operant Behavior |
| IVER H. IVERSEN (University of North Florida) |
| Abstract: When the behavioral microscope is applied to pauses between individual instances of operant responses a myriad of seemingly chaotic activities emerge. How are these activities related to the operant behavior? Is the operant superimposed on this background of behavioral noise, is the background generated by the reinforcement schedule, or does the background influence the operant? The research attempted to cast some light on these questions. RatsÆ lever pressing was reinforced by food pellets under fixed-ratio, fixed-interval, and variable-interval schedules. Some background activities such as water drinking, wheel running, contact with the feeder area, and general exploration and grooming activities were recorded concurrently with the operant. The results revealed that under some conditions the pauses in the operant behavior are directly related to these other activities. Such relationships were demonstrated in a functional analysis where some background activities were manipulated from moment to moment. For example, the duration of access to water was varied within sessions of fixed-ratio or fixed-interval food-reinforcement schedules for lever pressing among values of 0, 10, 20, 30, 40 and 60 s. The post-reinforcement pauses in operant lever pressing were systematically related to the duration of access to water; when rats drank for a long time the pause was long and when the rats drank little or no water the pause was short. Hence, at least under these circumstances, the operant pause was caused partly by the temporal extent of the background activity. Other experiments demonstrated similar relationships. With these manipulations one can demonstrate control of individual post-reinforcement pauses to the extent that the duration of a given pause can be predicted ahead of its occurrence with an accuracy error of just a few seconds. The more general conclusion from these experiments is that the moment-to-moment pattern of operant behavior is influenced both by the reinforcement schedule and by background activities. Hence, it is possible to identify some local |
| |
| The Effects of Concurrent Punishment on Rats' Sensitivity to the Response- Reinforcer Feedback Function |
| MARIANE SOH (University College London), Phil Reed (University College London) |
| Abstract: Two experiments examined the effects of punishment on rats' sensitivity to the molar and the molecular aspects of various types of reinforcement schedule. Punishment was produced by delivering a loud response-contingent tone on a schedule operating concurrently with the contingency under investigation. Experiment 1 investigated whether such a concurrent punishment procedure made rats more sensitive to the molar contingencies on variable ratio (VR) and variable- interval-plus-linear feedback (VI+) schedules. To the extent that the rats displayed sensitivity to the molar aspects of the schedules, performance on the VR and VI+ schedules should be similar. Experiment 2 compared the effects of tone- punishment on VI+ and variable interval (VI) schedules. To the extent that sensitivity to the molar aspects of the schedules was promoted, performance on the schedules would differ and responding would be greater on the VI+ schedule. The results are discussed in relation to the conditions under which rats' performance is governed by the molar or the molecular characteristics of the operative schedule of reinforcement. |
| |
| Choice and Temporally Extended Units: Evidence from People and Pigeons |
| TIMOTHY D. HACKENBERG (University of Florida), Cynthia J. Pietras (University of Florida), Eric A. Jacobs (University of Florida), Nicole R. Dorey (University of Florida) |
| Abstract: Fixed and progressive schedules have proven useful in studying choices in situations with contrasting short-term and longer-term consequences. The basic procedure involves recurrent choices between a fixed-interval (FI) schedule and a progressive-interval (PI) schedule that begins at 0 s and increases in fixed increments with each reinforcer delivered by that schedule. Selecting and completing the requirements of the FI schedule resets the PI schedule to its minimum value. Under these conditions, short-term consequences (i.e., delays to reinforcement on the upcoming trial) favor switching from the PI to the FI at the point of equality between the schedules, but longer-term consequences (i.e., overall reinforcement rate) favor switching well in advance of that point. Because switching prior to the equality point entails selecting an FI schedule over a currently-shorter PI schedule, long-term gains on this procedure are embedded within short-term costs. Despite these short-term costs, we have found that both pigeons and humans switch consistently prior to the equality point under a variety of conditions, indicating clear sensitivity to temporally remote consequences. Collapsed across procedural arrangements and species, switching from the PI to the FI has occurred prior to the equality point in 92% (133 of 144) of conditions conducted thus far. Taken as a whole, the results are in broad agreement with molar analyses of schedule preference and with versions of optimal foraging theory designed to address behavior in |
|
| |