The cerebellum and cognition: cerebellar lesions impair sequence learning but not conditional visuomotor learning in monkeys

Abstract

Claims that the cerebellum contributes to cognitive processing in humans have arisen from both functional neuroimaging and patient studies. These claims challenge traditional theories of cerebellar function that ascribe motor functions to this structure. We trained monkeys to perform both a visuomotor conditional associative learning task and a visually guided sequence task, and studied the effects of bilateral excitotoxic lesions in the lateral cerebellar nuclei. In the first experiment three operated monkeys showed a small impairment in post-operative retention of a visuomotor associative task (A) but were then not impaired in learning a new task (B). However, the impairment on A could have been due to a problem in making the movements themselves. In a second experiment we therefore gave the three control animals a further pre-operative retest on both A and B and then tested after surgery on retention of both tasks. Though again the animals showed motor problems on task A, they reached criterion, and at this stage could clearly make both movements satisfactorily. The critical test was then retention of task B, and they were not impaired. In the final experiment (serial reaction time task) the monkeys response times on a repeating visuomotor sequence were compared with those for a pseudo-random control sequence. After bilateral nuclei lesions they were slow to execute the pre-operatively learned sequence but were still faster on this than on the control task. However, when they were then given a new repeating sequence to learn, they never performed the sequence as quickly as they had on retention of the first sequence. We conclude that the cerebellum is not essential for the learning or recall of stimulus-response associations but that it is crucially involved in the process by which motor sequences become automatic with extended practice.

Introduction

The present decade has seen a striking increase in the number of studies that suggest that the cerebellum contributes to cognitive performance [1], [14], [23], [40], [41], [42]. These claims have been based almost entirely on evidence from human subjects, that is on reports of cognitive impairments in patients with cerebellar lesions and on neuroimaging studies of normal subjects reporting cerebellar activations even when controls are introduced for overt motor responses. These reports range from studies of language [15], [37], [38], [45], attentional processes [1], [2], [3], associative learning [6], [8], [14], [51], sequence learning [12], [13], [20], [34], [44], and executive functions [15], [17], [22]. Other studies have failed to corroborate these reports [9], [10].

One problem with patient studies is that it is never possible to be certain that the pathology is confined to the cerebellum. One strategy is therefore to place lesions in the cerebellum in non-human primates and to study the behavioural effects. In a related paper we have reported that monkeys with bilateral lesions in the lateral cerebellar nuclei are not impaired on spatial working memory and can learn visual discriminations normally [33]. The present paper considers associative learning. There have been several reports that patients with cerebellar pathology are poor at associative learning tasks on which they must learn to give one response to stimulus A, a different response to stimulus B, and so on [6], [8], [51]. There have also been several reports that patients with cerebellar pathology are slow to learn the serial reaction time task [12], [13], [34]. Here the subject responds to targets in turn and learns, either implicitly or explicitly, that there is a sequence such that presentation of target A is followed in time by the presentation of target B and so on [31]. On both these tasks the response is determined by an associative rule. The correct response is either conditional on a stimulus or on the last response.

The first experiment investigates whether the cerebellum is essential for the normal learning and retention of visual-motor associations. Monkeys were taught to move a joystick in one direction in response to one visual cue and in another direction in response to a second cue. As in the studies of patients [51] the monkeys learned the associations by trial and error.

The second experiment investigates whether the cerebellum is essential for the normal learning and retention of a motor sequence. The task was modelled on the serial reaction time task [31]. Four targets lit up in sequence, and the measure of learning was the response times. As the animal came to learn the sequence, these response times became faster than for performance of a control task in which the sequence was pseudo-random. The serial reaction time task is commonly given to human subjects as an ‘implicit’ learning task; that is the subjects show a decrease in response times while remaining unaware that there is a sequence. In the case of animal subjects the distinction between ‘implicit’ and ‘explicit’ learning cannot be drawn.

In an earlier pilot study [32], monkeys were taught a visuomotor conditional motor task and then retested after bilateral lesions had been made in the dentate nuclei. However, the lesions were made using a radiofrequency probe, and the lesion included white matter as well as the grey matter of the dentate. Two of the animals had small lesions and were unimpaired on retention, but a third monkey had a more complete lesion and though this animal showed evidence of relearning, it did not reach the criterion again within 2000 trials.

In this animal, however, there was complete retrograde degeneration in the principal division of the inferior olive due to damage to the white matter that involved fibres passing from the inferior olive to the lateral cerebellar cortex. In the present study we therefore made lesions in the lateral cerebellar nuclei using an excitotoxin so as to avoid damage to the white matter [25], [53]. The aim was to make complete lesions in the dentate, but this could not be achieved without the excitotoxin also affecting the interpositus nucleus.

To assess associative learning or retention it was necessary to distinguish between an impairment that was due to a problem in making the relevant movements from an impairment in selecting the appropriate movement. For example, in the pilot experiment [32] the animal that was slow to relearn the visual conditional task also had considerable difficulty at first in making one of the movements; and thus it was possible that the impairment in relearning was partly or wholly due to this motor problem. In assessing the retention of visuomotor learning in the present experiment we therefore taught two tasks (A and B) before surgery, and first retraining the animals on task A after surgery. Once they had reached criterion on retention of this task, it could no longer be argued that any residual motor problem made it difficult for them to make the appropriate movements. The crucial test was therefore subsequent retention of task B.

On the sequence task the same issue can be raised. The animals with cerebellar lesions might be slow to perform a sequence because they had problems in reaching for the targets. We found that these animals were slow to learn a novel sequence (sequence 2). However, the animals had already been retested on another sequence of the same length (sequence 1), and the response times on this sequence could be used as a measure of the times that the animals ought to be able to achieve in learning sequence 2. The evidence that they were genuinely impaired in learning sequence 2 is that they did not achieve this criterion within the limits of testing.