Optimizing musical variability

Variability of responses is an essential skill, and deficits in this area can contribute to pathologies such as autism and depression. It has been widely accepted that response contingencies can increase variability, but research suggests that verbal instructions are ineffective in increasing response variability. However, this research has only considered either verbal instructions or response contingencies, and has not looked at the combination of rules and contingencies. The aim of this study was to determine whether verbal instructions, response contingencies, or a combination, are more effective in increasing response variability. Sixty psychology students were quasirandomly assigned to one of four groups, and required to create seven note musical sequences on a virtual piano keyboard. Two groups received instructions that varying sequences would earn points, and two groups received no instructions. Two groups earned points contingent upon composing varied sequences, and two groups earned points on a variable ratio schedule of reinforcement. Results from a multi-factorial ANOVA found the proportion of novel sequences composed was higher when reinforcement was contingent upon varying responses. Providing rules did not have an effect on any measure of response variability. In addition, combining rules and contingent reinforcement did not have a statistical impact on response variability, despite the group which received rules and contingent reinforcement having the highest proportion of novel sequences. Keywords: Variability, Rule-Governed Behavior, Lag Schedules