Many homeless people may have been exposed to events and situations mat could potentially produce neuropsychological impairments. In me current study, 80 homeless participants underwent a battery of tests designed to 1) estimate long-standing established memory and intelligence, which was assumed to indicate prehomeless function and 2) measure current memory and intelligence function. Mental health screening and substance misuse data were also obtained. Results indicated that current memory and IQ were significantly lower tiian die estimated normal population means and also their prehomeless estimates. The memory score change was from 100.5 to 90.3 (p < .001) and IQ change from 98.8 to 95.6 (p = .038). The interaction between task type (memory or IQ) and measure (prehomeless or current) was also significant (p = .003), signifying that there was a greater change in the domain of memory function than in IQ. Many participants reported substance misuse and clinically significant mental health concerns. We conclude that die homeless individuals in our sample appear to have suffered a reduction in cognitive function, which may have occurred either during homelessness or prior to it.