Free operant avoidance in humans

Abstract

The purpose of the current study was to develop a method of studying human avoidance learning that provided data comparable to that obtained from non-human animals. Modifications were made to a typical human avoidance paradigm to make it more analogous to the type of paradigm commonly used to investigate non-human animal avoidance responding. Specifically, a consumable, immediate reinforcer was used, participants were exposed to repeated sessions in each condition, and verbal mediation was minimised. In the current study, five adult human participants learned to make avoidance responses using a trackball in order to avoid the interruption of access to a computer game. Audio and visual "blackouts" were scheduled using a free operant avoidance procedure. Responding shared characteristics of that observed in previous human studies of negative reinforcement. Participants typically over responded and did not exhibit signs of sensitivity to changes in conditions. Steps to eliminate over responding using the current paradigm were discussed including restricting the range of possible responses, punishing ineffective responding and increasing feedback. Options for increasing schedule sensitivity include the use of increased feedback or by putting in place measures to punish ineffective responding.