The assessment of limb apraxia: an investigation of task effects and their cause.

To test the effects of variations in the structure of tasks used to assess limb apraxia, eight tasks--differing in their modes of elicitation (tactile, verbal, visual, imitation) and/or types of movement elicited (transitive, intransitive, meaningless)--were administered to 25 carefully selected subjects with idiomotor limb apraxia. When standardized scores (based on the performance of 10 nonneurologically impaired control subjects) were used to measure the differences between tasks, no clinically significant task effects were found. Additional investigation was made of two causal hypotheses regarding the processes underlying limb apractic performance. The results of factor analyses and a search for double dissociations among individual subjects were consistent with the (severity) hypothesis that differences in task performances (raw scores) reflect degrees of task difficulty; and they failed to support the (disconnection) hypothesis that posits separate and independent neural/mental processes underlying task performances.