Dokument

Summary:
The study of sequential behaviour which relies among others on dopamine mechanisms and basal ganglia networks, is particularly relevant in Parkinsonian patients. Sequential behaviour can be ex- tensively studied through the use of a standard test known as the Serial Reaction Time Task (SRTT) in humans and non-human pri- mates. Although a rodent model of such a test would be very useful to investigate the underlying brain mechanisms of this type of learn- ing, there is no standardised rodent test. The aim of the three studies presented in this work was to characterise sequential behaviour in the intact rat as an analogy to the human standard test. The aim of the first study was to implement a rat model of the hu- man standard SRTT. The designed task required the rats to poke fast with their nose (motor answer) into lit holes (visual stimulus, one of four locations) and to perform a series of such nosepokes in order to get a food-reward, according to a fixed ratio schedule of reinforce- ment (FR). The location of the light was displayed in either random or sequential order and sequential learning was inferred from the dif- ference in performance between the two conditions within-session. We found that the rats performed better in the sequential condition, in terms of speed, accuracy and number of rewards earned. Details of the test were improved in the course of the studies to ensure that the better performance in sequential condition could only be attributed to the learning of the serial order information and no other general skill. Rats were finally tested on a repeated sequence of twelve ordered locations under a FR13. The length of the FR13 series was intention- ally longer than the length of the repeated sequence to dissociate the sequence locations from the FR schedule positions. The sequence structure was cautiously generated according to statistical rules (e.g. locations frequency, transitions frequency). These features provided a level of sequence difficulty comparable to the human one. This test was used in the second study to investigate the role of dopamine in this task in general and in the sequential performance of well-trained rats in particular. As this SRTT was planned to be applied in dopamine-depleted rats, the effects of the blockade of the dopaminergic transmission were first studied. A D1 and a D2 se- lective antagonists were used and injected systemically. We found that both antagonists produced dramatic disruption of responding, decreased response rate and increased the number of omissions. Only the D1 antagonist increased accuracy to a small extent. These effects were independent of the condition and dose-dependent. The D1 antagonist specifically impaired initial reaction times (within the first halves) of the series, whereas the D2 antagonist affected the whole pattern. Under D1 antagonist treatment, reaction times did not improve in sequential condition compared to random condition, which would reflect a specific effect of the D1 receptor in sequential performance. The third study aimed at investigating to which extent well-trained rats in the SRTT developed a habit. Rats were trained in sequential con- dition and were then confronted during a test with randomly inserted unique sequence violations. A detailed analysis of the performance yielded that rats showed indices of habit but also that attention was still playing a role. At the position of the violation, either the rats dis- played lengthened reaction times for correct pokes or poked fast into the hole where the light should have appeared according to the se- quential order (“expected” light location). This fast answer was how- ever now incorrect because of the sequence violation. Repetition of this test in a bigger group of rats proved the reliability of these results. In this repeated experiment, the apparatus and details of the task (but not of the sequence) were modified to suit application in dopamine- depleted animals for which motor requirements for example, have to be minimized. The rat SRTT with food-reinforcement described here shows high face-validity with the standard human SRTT. It has been effective for the biopsychological characterisation in intact rats of sequential performance, which in many aspects resembled the human one. The designed SRTT with food-reinforcement will probably be of value as a rodent model for the study of sequential behaviour in dopamine- depleted animals as a model for Parkinson disease.

Das Dokument ist im Internet frei zugänglich - Hinweise zu den Nutzungsrechten