The linguistic development and academic achievement of severely and profoundly hearing-impaired students is significantly delayed in comparison to their hearing peers. A major contributing factor to these delays has been the difficulty of identifying communication methods that facilitate development of students' linguistic and academic competencies (Moores, 1982; Quigley & Paul, 1984). Although there has been emotional debate concerning the effectiveness of particular methods of communication since the sixteenth century, it is only in the last thirty years that there have been empirical studies of this important issue.
The results of these empirical studies have been interpreted as supporting the use of simultaneous combinations of oral and manual methods of communication with severely and profoundly hearing-impaired students. This approach has been implemented in a majority of schools and is most frequently described as Total Communication or "simultaneous communication" (Schlesinger, 1986; Johnson & Kadunc, 1980).
In addition to studies which examined the methods of communication used with hearing-impaired students, another area of relevant research has involved studies of the information processing consequences of hearing impairment. The findings of studies related to communication methods and to information processing abilities have been inconsistent. The linguistic and cognitive abilities of subjects, the difficulty of tasks, and the nature of stimuli have not always been considered in these studies.
The present study attempted to control the effects of these variables. The 30 subjects were the total metropolitan population of severely and profoundly hearing-impaired students between the ages of ten and sixteen years who met certain criteria for the study. These criteria were intelligence within the average range, no additional disability, similar communication experience, linguistic competence, and instructional and social background. The communication experience consisted of exposure to Total Communication (including the use of Australian Signed English) for at least five years. Linguistic competence involved being able to understand and use simple sentences and instructional background involved at least three years attendance at the student's present school.
Control of other experimental variables related to task difficulty was attempted by using a low-demand, four-alternative recognition task with planned linguistic content. The nature of the stimuli for each presentation was controlled by using (a) a consistent signal-to-noise ratio that replicated the acoustic conditions measured in classrooms, (b) videotaped presentation of all communication conditions for re-test reliability, (c) a large number of trials per subject per condition to achieve data stability, and (d) different testing sequences for various groups of subjects to protect against serial learning effects.
Another variable that may have influenced the results of previous studies was the subjects' degree of hearing loss. Although differences between severely and profoundly hearing-impaired subjects have been demonstrated for listening and for lipreading (Erber, 1974c), the possibility of differences between the two groups with other communication methods has not been examined. Most studies have described their subjects as "severely to profoundly hearing-impaired" and their results may have been influenced by the proportions of severely versus profoundly hearing-impaired students involved. For this reason, the present study allocated the students to either the "Severely Hearing-Impaired Group" or to the "Profoundly Hearing-Impaired Group" based on their better-ear-average, unaided hearing loss.
Eleven communication conditions were used to examine the receptive communication abilities of the severely versus the profoundly hearing-impaired students. The communication conditions involved the presentation of videotaped sentences unimodally, bimodally and multimodally. The unimodal conditions were audition, lipreading, fingerspelling, and signs (Signed English). The bimodal conditions were all possible bimodal combinations of these four communication methods. The multimodal conditions presented information through the two combinations most frequently used in classrooms: either through audition, lipreading and signs or through audition, lipreading and fingerspelling.
In addition, a questionnaire was used to obtain teacher ratings of the students' communication abilities. The questionnaire was adapted from that used by Conrad (1979) and sought information on the students' receptive communication abilities and their speech intelligibility. It was found that the association between these ratings and the study communication measures was only moderate, with the rating for fingerspelling the only one to reach significance.
The main results indicated that (i) there was no evidence that the most common communication practice used with Total Communication (simultaneous audition, lipreading and signing) was more effective than the use of signing alone for either group or the use of the auditory/lipreading combination for the severe group, (ii) the communication abilities of the severe group were significantly better than the profound group on all communication conditions not involving signing, and (iii) there were significant negative correlations between the severely hearing-impaired group's ability to listen and their ability to receive signing or fingerspelling.
It was also noted that the addition of speech to signed information did not produce an enhancement of students' scores. It was suggested that this may have been caused by the slower production rate of Signed English disturbing the prosodies of speech. This could have made the use of lipreading and listening more difficult. An alternative explanation for this result may have been due to students' facility in encoding signed information, rather than speech information, in short-term memory.
Several other findings of the study, when taken in conjunction with each other, indicated that the simultaneous use of audition, lipreading and signs may not necessarily produce benefits in both the oral and manual skills of students. These findings include (i) the significant negative correlations between the severely hearing-impaired groups' listening and "manual" abilities (fingerspelling and signing), (ii) the poor performance of the profound group with "oral" communication conditions (audition, lipreading and the auditory/lipreading combination), (iii) the students' low speech intelligibility ratings.
To develop the listening, lipreading and speaking skills of hearing-impaired students may require separate attention to the relevant components and a more "active" use of oral communication strategies in Total Communication programs. Based on information gathered during the study, improvement of the signal-to-noise ratios in classrooms would also seem to be relevant to acquisition of these skills.
Large standard deviations were obtained for both groups for fingerspelling and also for the profound group for "oral" communication conditions. In addition, certain students performed in a manner that was discordant with the range of group measures. These variations indicated that there are individual differences in communication competencies. It was suggested that these should be taken into account by using the methods of Total Communication in a flexible and discriminating manner which caters for individual differences. The practical implications of the poor performance of the profound group with fingerspelling were discussed in relation to the relatively small size of the Signed English lexicon.
The results of the study have shown that there are group and individual differences in the communication abilities of students and only a moderate correlation between the ratings of teachers and the study measures of communication abilities. These results have important implications for the implementation of Total Communication and the evaluation and remediation of the communication abilities of hearing-impaired students.