The aim of this study is to compare the motor-output of two groups of Drosophila fruitflies in the flight simulator which both are trained to avoid a flight direction towards a given pattern in their environment. One group is trained operantly to perform the task, the other classically, according to the definitions in 1.2 and 1.3. In such an experiment, it is plausible that the conditioned responses might deviate as an effect of the different associations made during the different training procedures. More specifically, it can be expected that the operantly trained flies acquire 'new' behavioral strategies the classically trained flies lack or that the flies selectively activate and inactivate certain behaviors from a range of motor-programs while the classical group still uses the whole range. To investigate this, the assessment techniques measuring the performance of the behavior have to be identical in both the classical and the operant experiment. The experimental setup has to allow for exquisite control of stimulus presentation and response generation in order to 1) avoid complications with stimuli unintentionally connected with the experiment, 2) assure that the two experiments differ only in training and 3) detect differences in response generation with sufficiently high accuracy. The flight simulator provides the means to achieve this goal. Moreover, the concept of studying 'microbehavior' (Heisenberg and Wolf, 1984) enables the student to pose his questions more specifically than in some setups often used by psychologists, where gross behavior is analyzed. In such preparations the expression of learning in gross behavior is most likely to be the product of a rather large amount of post-acquisition processing, complicating the interpretation of the data in terms of what has been learned.