The present study examines the potentially harmful effects of prenatal cocaine exposure on later visuospatial memory functions. A novel neuropsychological measure of immediate- and short-term memory for visuospatial information was administered to 40 children, who were identified as cocaine-exposed, and 11 age and socioeconomic status matched controls (all children were 8-9 years old). The Groton Maze Learning Test is a computer-based hidden maze learning test that consists of a "timed chase test" (a simple measure of visuomotor speed), 5 learning trials on a hidden maze, followed by a delayed recall trial after an 8min delay. The specific test parameters are chosen based on the age cohort of the subjects. The cocaine-exposed group performed significantly slower on the "timed chase test," the last 3 learning trials, and the delayed recall trial (p is less than or equal to 0.05 for all comparisons). Although there was a modest trend for the cocaine-exposed group to make more errors throughout the learning trials, there were no significant group differences. These results suggest that children who were exposed in utero to cocaine exhibit slowed visuomotor speed, a possible impairment in procedural learning, and diminished efficiency in accessing and using the internal spatial map that subjects create to master the maze.