The assumptions for this study were as follows: in operant conditioning a response-reinforcer association is formed, whereas in classical conditioning the association is stimulus-response in nature. Using the Drosophila flight simulator, the motor-output of two groups of Drosophila flies - one presented with stationary stimulus-reinforcer pairings (open loop training), one enabled to control the presentation of the pairings (closed loop training) - was compared. It was assumed that the open loop group was classically, the closed loop group operantly conditioned. If this were correct, there might be a difference in the behavioral strategies an operantly trained subject might acquire compared to a subject classically trained to perform the same task. It can be concluded from the gathered data that one of the observed behavioral strategies was significantly enhanced in the operantly but not in the classically trained flies: choosing a flight direction as far away from the reinforced stimulus as possible.
It appears, however, that an individual during conditioning is not confined to making singular associations but is rather evaluating complex stimulus situations. This leads to the transfer of reinforcer properties to the unconditioned stimulus even in the operant paradigm. An experiment in which the subject uses one behavioral repertoire (e.g. walking behavior) to control the reinforcer and a sensory stimulus during training and another (e.g. flight control behavior) to control the conditioned stimulus during test, deserves special interest in resolving this issue.