Selective attention can be measured through analysis of errors and reaction time (RT) for trials in which targets are presented alone compared with trials in which targets and distractors are presented. This study investigated selective attention using a reaching task, in which subjects made rapid reaches to targets. Thirty-seven patients with lesions of the prefrontal cortex (PFC) were compared with 19 healthy age- and IQ-matched volunteers and 18 patients with early-stage Huntington’s disease (HD), a neurodegenerative disease primarily affecting the basal ganglia. It was hypothesised that, if fronto-striatal circuits as a whole support selection-for-action, then the pattern of behavioural performance of both patient groups would be similar. Alternatively, if the functional roles of PFC and basal ganglia in selection-for-action are dissociable, then two different patterns would emerge. It was found that that both HD and frontal groups were significantly more distractible than controls for RT, but they had a different pattern of errors. Frontal patients made significantly more touches of the distractor location itself than did controls, while this was not the case for HD. It is argued that a reactive- inhibition mechanism, required in the circumstance of strong distractor activation, is affected by frontal damage, while a lateral- inhibition mechanism, invoked during the recruitment of selective attention, is affected in HD. Additionally, there were significant correlations between the degree of distractibility for RT and the extent of lateral PFC damage, and between cue-generated preparation and lateral PFC damage, thus highlighting the critical importance of lateral, rather than orbital or medial, PFC for attention to action.