Coding and recall of categorized material in aphasics

The aim of the present study was to examine whether the often reported impairments of aphasics in different short-term memory tasks could be the result of a failure to use the facilitating strategy of clustering. Aphasics (n = 60), right hemisphere brain-damaged (n = 36) and normal (n = 10) controls were tested with a modification of tasks developed by Petrides and Milner (1982). They were presented with sets of 16 cards with a random arrangement of the same 16 stimuli. On each card subjects had to point to one of the stimuli, trying not to point to the same stimulus in different cards. In some of the tasks pictures were selected to suggest a clustering into four equally sized subsets; in others the stimuli were highly heterogeneous with respect to perceptual and semantic characteristics. Prior to each task with items easy to categorize, half of the subjects were requested to sort the pictures into subsets, while the other half were given the pictures without any specific instruction, having been requested to do the sortings only after the main task. Aphasics (1) showed less clustering, i.e., sequential pointing according to the predefined categories, and (2) made significantly more errors than RH-controls especially in the tasks easy to categorize. While RH-controls benefited from the preceding sorting of the pictures aphasics made more errors when first introduced to the categorization task.