The purpose of this study is to describe possible differences in cognitive functioning between smoking and non-smoking patients with first-episode psychosis and to determine whether there is a better cognitive profile associated with smoking. We assessed 61 first-episode psychosis patients with a neuropsychological battery that included computerized measurements of attention, working memory, and executive functioning. Patients were grouped into two categories: non-smokers (0 cigarettes/day; n = 30) and smokers (>/=20 cigarettes/day; n = 31). No significant differences were detected in sociodemographic and clinical data between the two groups. For attention tasks, smokers exhibited shorter reaction times in the sustained attention test than non-smokers (P = 0.039) and needed less time to complete the Stroop interference test (P = 0.013). In the working memory task, smokers exhibited shorter reaction times (P = 0.029) and presented a significantly lower percentage of omission (P = 0.002) and commission errors (P = 0.020) than non-smokers. For executive functioning, no differences were detected between groups in performance on the Wisconsin Card Sorting Test. Results indicate that first-episode psychosis patients who are nicotine users have better cognitive functioning in the areas of attention and working memory than patients who are not nicotine users. This study supports the cognitive approach to the self-medication hypothesis, to explain the high rates of cigarette smoking among psychosis patients. These results may be relevant for developing new strategies involving nicotinic receptors for cognitive enhancement in psychosis.