University of Wuerzburg Biocenter Genetics D-97074 Wuerzburg Germany,
A simple apparatus has been constructed for operant conditioning of freely walking single flies. A fly can be conditioned to avoid one side of a small test chamber if the chamber is heated whenever the fly enters this side. In a subsequent memory test without heat wildtype Drosophila keeps avoiding the heat associated side. The test apparatus is a transparent rectangular chamber (40 x 4 x 2.5mm). The upper and lower surfaces consist of Peltier-elements by which the temperature is adjustable in the chamber. A control circuit and a thermo-sensor in the chamber keep the Peltier-element at a permessive „COLD“ (25°C /-2°C) or non-permessive „HOT“ (37°C /-2°C) temperature. The chamber is virtually subdivided perpendicular to its long axis into two halves („LEFT“ and „RIGHT“) by a directionally selective light gate. The experiment normally consists of a preference test, training and a memory test. A computer controls the experiment in the following manner: During all 3 phases it continuously monitors the time and direction of transitions at the light gate. During training, in addition, it turns the heat on when the fly enters one side and turns it off when the fly leaves it again. Note that the Peltier-elements cover the whole length of the container. This ensures that the fly does not experience a spatial temperature gradient after-effect. During the preference and memory tests the temperature is fixed at 25°C ( /-2°C), independent of the position of the fly. After training, the memory test does not start until the fly changes sides, therefore inactive flies are excluded. The performance of individual flies is calculated as Performance Index PI=(A-B)/(A B) with A indicating the time the fly spends on the side that during training is associated with low temperature („COLD“) and B indicating the time on the „HOT“ side. The PI can vary between -1 and 1. A PI of zero indicates that the fly spends 50% of the time on the „HOT“ side. Half of the flies are conditioned to avoid „LEFT“ and the other half to avoid „RIGHT“. In this way the effects of spontaneous preferences for one side and of slight asymmetries in the apparatus are largely eliminated from the data. Flies from the wildtype strains Canton-S and Berlin avoid the heat associated side even after heat is switched off (memory test). The mutations dunce and rutabaga avoid the heat too, but failed in the subsequent memory test. The positive aftereffect of the wildtype flies in the memory test is interpreted as learning in the previous training period. As a proof of associated learning, flies were trained in one chamber and testet (memory test) in another similar one. In this situation the flies show a positive aftereffect as well. In a screen for learning mutants in the apparatus about 480 mutagenized 2nd chromosomes (vg bw marked) have been tested and two dominant and two recessive mutations have been found.