This study investigated the cognitive and linguistic processing demands of message-encoding techniques used to retrieve prestored messages from computer-based augmentative communication systems. Twelve physically disabled adults with severe speech impairments participated in six counterbalanced experimental conditions, consisting of three encoding techniques (salient letter, letter category, and iconic codes), each in a personalized condition in which subjects selected their own codes and in a nonpersonalized condition in which codes were preselected by a clinician. In each of the experimental conditions, the subjects participated in five learning and testing sessions. During these sessions, the subjects learned the codes for lists of 80 messages, half of which involved concrete referents and half abstract concepts. Results indicated that the subjects were more accurate in recalling the codes to retrieve preprogrammed messages when using the two letter encoding techniques than when using the iconic technique. No reliable differences were found between recall performances with personalized codes and with nonpersonalized ones for any of the three encoding techniques. Code recall improved consistently across the series of learning sessions; there were no significant differences in the rates of learning across the encoding techniques. Subjects were found to be more accurate at recalling the codes to retrieve concrete messages than those to retrieve abstract messages.
|Original language||English (US)|
|Number of pages||12|
|Journal||Journal of speech and hearing research|
|State||Published - Dec 1 1992|
All Science Journal Classification (ASJC) codes