I. Classical conditioning at the flight simulator

In classical conditioning, the stimulus is made contingent upon the reinforcer in a way that whenever the stimulus is presented, the reinforcement will be applied. This reinforcement leads to a behavioral response, often a reflex. There is no feedback between the behavior of the animal and its stimulus situation, hence the above diagram only shows feed-forward relations (solid arrows) from the CS (stimulus) to the US (reinforcer) to the behavior.

Classical conditioning at the Drosophila flight simulator is brought about by turning the arena in open loop (i.e. there is no feedback between the fly's control maneuvers and the rotational speed of the arena) with a constant rotational speed of 30°/s. The reinforcement regime is as follows: whenever one of the two patterns enters the frontal 90° sector of its visual field, the fly is punished by the heat beam. The shutter closes as soon as the next pattern is brought to the front. Thus, pattern presentation feeds forward to shutter closure/opening. Classical color learning is conducted accordingly: the patterns are replaced by four identical vertical stripes and the illumination of the arena is changed whenever a new stripe enters the frontal visual field of the fly. After the open loop training the fly tested in closed loop for pattern (color) preference.

I. Classical conditioning
II. Pure operant conditioning
III. Parallel operant conditioning
IV. Facilitating operant conditioning

--- back to parent document ---

All content on brembs.net is copyright © Björn Brembs, 2001.