In this experiment, we report a novel drug-free behavioural test of lateralised neglect which is sensitive to unilateral dopamine-denervating lesions and subsequent graft-derived striatal dopamine replacement. For the task, white plastic lids containing sugar pellets were placed along the left and right sides of the floor of a long narrow corridor at regular intervals. Hungry female Sprague-Dawley rats were placed individually into the corridor where they were allowed to make up to 20 pellet retrievals. The number of retrievals each rat made from its left and right sides was counted. Complete mesencephalic or partial nigrostriatal lesions were induced by injection of 6-hydroxydopamine into the medial forebrain bundle or striatum, respectively. Both lesions induced a pronounced ipsilateral retrieval bias in the task. Five weeks after lesion surgery, half of the rats from each lesion group were given E14 ventral mesencephalic cell suspension transplants into the denervated striatum, and were then re-tested in the Corridor Task 5 and 10 weeks later. There was no amelioration of the side bias in rats with medial forebrain bundle lesions. In contrast, in nigrostriatal-lesioned rats, the graft significantly reduced the lesion-induced ipsilateral bias. We conclude that the Corridor Task is a sensitive test of lateralised sensorimotor response selection, and is suitable for assessing deficits and recovery associated with lesions and grafts within the nigrostriatal system.