Probing spatial working memory with the Corsi Blocks task.

The Corsi Blocks task (CBT) measures spatial memory span, but methodological differences across previous studies prevent a clear appreciation of perceptual, memory, and motor contributions to spatial span. CBT performance in a standard condition was compared to performance with longer encoding intervals (3 vs 1 s), longer memory intervals (9 vs 1 s), or fewer response alternatives (all nine vs only the relevant positions). All experiments also contrasted ascending vs descending order of item difficulty and recorded response-initiation times. Performance improved with longer encoding and maintenance intervals and with fewer response alternatives. Item order had no reliable effects. Response-initiation times were a useful additional measure of spatial span. Implications of these findings for research using the CBT are discussed.