Analysis of the automated delayed nonmatching-to-position task: The effects of changing contiguity between stimulus, response and reinforcement, and of providing a salient spatial cue within the apparatus

Abstract

The present report examines two factors that may affect the acquisition and performance of a delayed nonmatching to position (DNMP) task. Both of these concern the design and layout of the apparatus. The first is the contiguity between stimulus, response and reinforcement while the second concerns the impact of asymmetry within the behavioural apparatus. In the standard version of the task, food rewards are dispensed at a central well, equidistant from the stimuli to be discriminated and from the site of the animals' responses. The present experiment demonstrated that acquisition of the task is significantly quicker when there is greater continguity between the stimuli and rewards i.e. when rats are reinforced in the same location as where they respond. We also studied the potential impact of the fact that the standard two-lever operant chamber is asymmetrical as the transparent access door is typically at the side of the box i.e. adjacent to one lever. By comparing test chambers with or without single Plexiglas walls we found clear evidence that the door/wall provides a salient cue that aids DNMP performance. This casts light on the nature of the actual memory demands in the DNMP task.