Modern applications of operant conditioning through the training of a beaching behaviour with bottlenose dolphins (Tursiops truncatus)

Abstract

Operant conditioning techniques, such as positive reinforcement and shaping through successive approximation, are used in applied settings to train a variety of species to complete behaviours voluntarily. The reinforcement and shaping procedures implemented by trainers can vary widely and are generally guided by qualitative assessment rather than quantitative data. The aim of this study was to complete a preliminary investigation of several factors that may affect the success of training a new behaviour in order to present information on current techniques and determine focus areas for future studies employing experimental designs. Data were collected during training sessions involving six bottlenose dolphins Tursiops truncatus at the US Navy Marine Mammal Program in San Diego, CA. The focal behaviour required the dolphin to slide up and out of the water and beach onto a foam transport mat. The study discusses three components of operant conditioning: 1) How often should a behaviour be shaped? 2) When and how often should the behaviour be reinforced? 3) How are criteria changes or previous beaches related to future success? Based on these data, limiting the number of distance building approximations within a session may play a role in acquisition. Providing small reinforcement magnitudes for attempting did not clearly promote successful beaches but also did not seem to hinder success. The duration of hand station avoidance was related to the outcome of the previous trial but did not impact the following trial.