Homelessness has been associated with neuropsychological (NP) impairment, but few studies have adequately controlled for factors known to affect NP performance. The authors performed brief NP testing that examined learning, recall, processing speed, executive functioning, and verbal fluency in 50 ever- and 22 never-homeless persons entering outpatient psychiatric treatment. Groups were matched a priori on key demographic, substance use, psychiatric, and premorbid intelligence quotient characteristics. Rates of NP impairment were high among both groups (46%-54%). There were no significant differences in global NP impairment. There were trends toward better levels of processing speed and executive functioning among never-homeless relative to ever-homeless. Among the ever-homeless group, NP test performance was unrelated to number of homelessness episodes (median 3). Findings confirm high prevalence of NP impairment among homeless individuals but provide little evidence for broad NP differences between ever- and never-homeless persons matched for coexisting conditions that have confounded interpretation of previous results in the literature.