Variability

Abstract

After experiments with shaping new behavior in a publicly displayed porpoise, Karen Pryor and colleagues suggested that creativity and variability could be reinforced directly. Other researchers have followed up this claim in the laboratory, studying different species under different contingencies, and elaborating over several related issues. A major point on the theoretical side is the suggestion that variability be labelled an operant in its own right. The following dispute over this, with referal to what constitutes an operant and whether variability is explained better by already existing basic principles of behavior analysis, is adressed. A few other minor discussions concerning incongruences in experimental results are also treated, among them the basis for the present study in the experimental part of the thesis. An experiment was conducted to examine if lag schedules depending on responses or response classes yield different results. At the same time, the study facilitated two strains of rats to maintain if there would be differences in the variability registered. The hypothesis for both issues was that a difference would be evident in either changeover from baseline, between experimental phases, between groups or between the two strains used, or possibly on all these measures. Eight non-naive rats of two strains (SHR and WKY) were subjected to lag 3 schedules; four rats were required to respond to any of four operanda, the other four were required to respond to all of four operanda. The groups were switched after 16 sessions and run for a further 8 sessions. The overall finding was that there is no difference between lag schedules depending on either responses or response classes, but that higher levels of activity influence variability in genetically predisposed strains.