6-Hydroxydopamine Lesions in the Rat Neostriatum Impair Sequential Learning in a Serial Reaction Time Task

Sequential behavior has been intensively investigated in humans using so-called serial reaction time tasks (SRTT), in which visual stimuli are either presented in a random or sequential order. Typically, when the stimulus presentation follows a previously learned sequential order, reaction times are decreased compared to random stimulus presentation and become partly automated. A vast amount of SRTT findings indicates that sequential learning and performance seem to be mediated amongst others by the basal ganglia-especially the striatum-and the neurotransmitter dopamine therein. In this study we used an operant rat version of the human four choice SRTT to investigate the effect of bilateral neostriatal dopamine lesions induced by 6-hydroxydopamine on sequential learning. The rats' task was to respond rapidly to illuminated holes by nose-poking into them. During extensive training, the position of the illuminated hole followed a 12-item sequence. The outcome of this sequential training was also investigated in two tests, namely an interference test, where stimulus presentation switched between this sequential and a pseudo random order every five minutes, and a violation test, in which only one sequence item was eventually skipped. The neurotoxic lesions, which was placed before the start of training, led to the expected sub-total dopamine depletions (i.e. residual levels around 34-56% of controls), especially in the medial neostriatum. These lesions did not lead to general motor deficits in a catalepsy task, but moderate deficits in locomotion in an activity box, which largely recovered with time after lesion. In the SRTT, rats with lesions showed impaired learning, that is, less response accuracy and slower reaction times than the control group. During a subsequent test with alternating phases of sequential and random stimulus presentations, reaction times and accuracy of the control group were superior during sequential as compared to random stimulus phases. In the lesion group, only a moderate advantage in accuracy was observed. In the violation test, another outcome measure, the control group showed an expected increase in reaction times on the violated positions. By contrast, the lesion group showed no such increase, which indicates less automation of sequential behavior in these animals. For one, these findings support previous evidence in showing that neostriatal dopamine plays an important role for instrumental behavior, in general. Furthermore, and most importantly, they suggest that dopaminergic-striatal networks also play an important role in sequential behavior, especially its acquisition.