Experimental therapies for Parkinson's disease are commonly validated in unilateral rat lesion models using simple tests of motor asymmetry such as rotation. However, the human disorder is considerably more complex than this, and alternative tests that yield a more relevant evaluation of the lesion-induced deficit could provide a powerful behavioural tool for analysis of novel therapies. This study evaluated the potential of a lateralised nose-poking task for detailed assessment of the deficit in rats with complete unilateral and partial nigrostriatal terminal lesions induced by unilateral injection of 6-hydroxydopamine (6-OHDA) into the medial forebrain bundle (MFB) or striatum, respectively. To maximise the task's potential for revealing chronic deficits, a version of the task that requires rats to nose-poke away from visual stimuli associated with food rewards was used. Because of its low stimulus-response compatibility, performance of this version is not mediated by the rats innate tendencies, and thus, is less likely to spontaneously recover after partial lesions. Two weeks and 4 months after lesion surgery, both groups of lesioned rats developed an ipsilateral responding bias with a corresponding drop in contralateral accuracy. Rats with complete (but not partial) lesions also developed a deficit in attempting trials, reacting to stimuli requiring a contralateral nose-poke and executing movements bilaterally. This experiment suggests that the version of the lateralised nose-poking task requiring an arbitrary stimulus-response association may be a powerful behavioural tool for assessment of complete or partial dopamine lesions, and any subsequent dopamine replacement strategy.