Qualitative performance features and extended-time effects on the Block Design subtest of the WAIS–R were examined in 145 healthy Ss ages 50–95. Raw scores were calculated at standard and extended time limits; error types and starting points were recorded for a subgroup of Ss. In Ss age 60 and older, there was a consistent decline with age in overall scores at both time limits. Extra time allotments resulted in modest but significant increases in scores. Older and younger Ss benefited equally from the extra time, indicating that the age-related decline on this task cannot be accounted for by general age-related psychomotor slowing. Single- and multiblock rotation errors were relatively common, whereas stimulus boundedness and broken configurations were rare. Despite some potential limitations in generalizability, the results suggest that the incorporation of qualitative scoring procedures in the assessment of visuoconstructional skills may enhance understanding of normal and abnormal brain–behavior relationships.