INTRODUCTION This paper reports on a project carried out in 1991 by staff at Charles Sturt University - Mitchell. We believe this project addresses well the fact that basic research, practical application and policy need not be incompatible. Both practical concerns of classroom teachers involved in teaching students with severe intellectual disabilities and a background of research provided in the literature, have been combined to suggest a course of action. The result has been the development of a teacher skills based package addressing the area of communication programming for students with severe intellectual disability. Within the NSW Department of School Education, a student with severe intellectual disability is one who scores less than 35 on a standard intelligence test (Binet or WISC-R). In addition, there usually exists significant deficits in adaptive behaviours; the ability to maintain standards expected by society in personal independence and in social responsibility. Within the group identified as severely intellectually disabled there can be a range in ability levels, from those who require support for daily care for some of the time, to those who are almost totally dependent on others for their essential daily needs. It is also common to find multiple disabilities such as visual, hearing and physical impairments. Lack of communication skills is common. Education for this population addresses in each student the development of skills that will enable them to achieve maximum independence in home, school (or work) and community settings. Essential in each of these settings is the ability to communicate. Communication is a dynamic process and is defined by Dunst (1978, in Dunst & Lowe, 1986, p.12) as "...any overt conventional or non-conventional behaviour, whether used intentionally or not, that has the effect of arousing in an onlooker a belief that the signal producing organism is attempting to convey a message, make a demand, request etc to an onlooker." Such a definition has significant implications when discussing the population of interest. First, the behaviours need not be intentional to be communicative and the partner may infer the intent of the action. Secondly, the behaviours need not be conventional (eg. speech) to be communicative and in fact they may be very primitive in form (a facial movement) or even socially unacceptable (scratch). Thirdly, the emphasis is on the pragmatic quality of the message, indicating that the person concerned is communicating for a purpose. Finally, the process involves a partner to receive and understand the purpose of the behaviour. Traditionally, non verbal students have been trained to use an alternative to speech to augment or facilitate their communication abilities (eg. the use of signing or communication boards). Despite widespread and systematic training in such systems, reports indicate that students with severe intellectual disability continue to display significant deficits in communication abilities (Kaczmarek, 1990; Halle, 1987; Bryen & Joyce, 1985). When observing students closely, Houghton, Bronicki and Guess (1987) found more frequent use of body movement (eg. gestures) and social responses (eg. smile) than use of signing, communication boards or verbal behaviours. Recent research is therefore suggesting the recognition of existing informal systems and the need to build on these before training in the use of more conventional symbolic systems (Seigel-Causey & Guess, 1989). For students with severe intellectual disability it is important that they first discover the effectiveness of their behaviours in influencing the behaviour of others. From this basis, they can then further develop the ability to spontaneously use more conventional communicative structures or forms. The acceptance that communication can be achieved through multi-modal means, where any behaviour from the students can be interpreted as being communicative has significantly influenced our understanding of the behaviour of students with severe intellectual disabilities. Recognition is now given to the significant development in communicative abilities at the preverbal stage. Nonverbal means such as reaching and pointing can achieve the same purposes of communication possible through verbal means. Similarly, less socially accepted or indeed aberrant behaviours can also be seen to achieve functional ends (Burke, 1990; Carr & Durand, 1985). A pragmatic based system to study the communication behaviour of students with severe intellectual disability analyses the extent to which students are able to communicate to satisfy needs, wants and intentions (Halle, 1987; Linfoot, 1987; Cirrin & Rowland, 1985). Pragmatics refers to the function of a behaviour and the effect that this has in regulating the behaviours of a listener (Burke, 1990). Naturalistic observations of students in classroom settings (Linfoot, 1987) and observations in specific tasks structured to elicit communication (Cirrin & Rowland, 1985) have provided evidence that students with severe intellectual disability do in fact participate in a range of pragmatic functions. Regardless of the form of the communicative behaviour, attention to the skills of effective discourse is necessary. This includes aspects of initiating and maintaining communicative interactions and responding to communicative attempts with others. Concern exists that students with severe intellectual disability primarily respond and rarely initiate (Schwartz, Anderson & Halle, 1989; Halle, 1987; Kaczmarek, 1990). However, interesting evidence from classroom observations, shows that in fact students do initiate more than teachers recognise and acknowledge. Houghton, et al. (1987) observed students across 12 classrooms and found that students initiated communicative interactions approximately once each minute, while staff responded to these communications only 7 to 15% of the time. Finally, importance is attributed to the reciprocal nature of communication, where the development of skills in the communicative partner are seen to be as critical as the development of skills in the student. Partners need to recognise and respond to all communicative attempts by the student, to know how to create a communicative environment and to know when to intervene to generate increased interaction. As communication is a two way interactional process, we need therefore to place equal or greater emphasis on addressing how teachers and all who interact with the student, can be assisted to meet student needs, so that in fact, students with severe intellectual disability do become competent communicators. In the first instance it is paramount that teachers receive training in how to recognise student attempts to communicate (Houghton et al., 1987). By observing students, teachers will be able to assess more closely how students are currently communicating, what they are communicating about and what is motivating the students to be communicative. Secondly, the context of the instruction is important. In the past, students have been instructed using systematic stimulus control techniques at set times in the day (Houghton, et al, 1987; Goetz & Sailor, 1988). Such strategies have been powerful in advancing conceptual development and symbol recognition, but at the expense of improving the quality of the communicative interaction process. Such isolated training sessions have been associated with lack of generalisation and functional use of skills for communicative purposes (Bryen & Joyce, 1985). Teacher use of strictly stimulus controlled situations creates a dependency in students on their verbal cues, rather than natural cues, to occasion a response. Linfoot (1987) found, after observing students in three different classroom settings, that students primarily respond, and that the highest rate occurred in structured rather than unstructured sessions. On the other hand, unstructured sessions were more conducive to student initiated interactions. Improved results in effective use of communicative abilities have been reported to occur with teaching in natural settings, utilising activities the students usually perform, and enabling the naturally occurring cues of that setting to instigate communication (Mirenda, Iacono & Williams, 1990; Caro & Snell, 1989). Strategies have been documented in the literature that support such an approach, each having a particular strength. The mand model technique (Rogers-Warren & Warren, 1980) has been found to be effective in developing initial responding behaviours. The time delay strategy (Halle, Marshall & Spradlin, 1979) is then effective in transferring the ability to use such behaviours to initiate. Incidental teaching (Hart & Risley, 1982) provides a framework for extending existing forms. Interrupted behaviour chain (Goetz, Gee & Sailor, 1985) has been found to be effective in motivating a student who is reluctant to communicate and make social contact. Natural teaching strategies all share the following features. Opportunity exists to arrange the environment to increase interaction. While working within normal routines, specific communication objectives are targeted. Natural consequences provide reinforcement to communicative attempts while natural cues occasion the behaviour. Essential learning principles such as shaping, prompting, and modelling, are maintained and used following systematic and controlled procedures (Warren & Kaiser, 1986; Mirenda et al. 1990). While the use of such strategies to promote functional communication in everyday events by students with severe intellectual disability is supported by research (Caro & Snell, 1989; Warren & Kaiser, 1986) this information has been poorly disseminated to teachers (Schwartz et al., 1989). Teachers need training in how to structure naturally occurring situations throughout the day to occasion increased opportunities for communicative interaction. Evidence exists however to show that teachers can be trained to use such strategies and direct intervention to more functionally appropriate objectives. Haring, Neetz, Lovinger, Peck & Semmel (1987) were successful in training teachers to use four modified incidental techniques, including blocking access, providing for choices, placing materials out of reach, and offering students objects out of context. Teachers increased the number of opportunities for students to communicate, and students demonstrated higher responding rates. Schwartz et al. (1989) reported on successfully training teachers to utilize time delay in every day situations. As a result, students became more responsive and dependence on verbal prompts was lessened. By focusing on the reciprocal nature of communication, necessary attention is given to addressing both student and teacher needs. The goal of communication programming for students with severe intellectual disabilities is to "...increase the frequency of their communicative functions, and to promote their spontaneous and generalised use of communicative behaviour in everyday life" (Caro & Snell, 1989, p.67). This can only be addressed by attaching importance to the training of skills in those who act as the interactive partner for the student. In summary, the research reported in the literature highlights the fact that students with severe intellectual disability are not effective users of formal communication systems, but can achieve communicative power if recognition is given to their existing abilities and the functions achieved by those actions. Importance is placed on the reciprocal nature of communication. This therefore has significant implications for increasing skills in the communicative partner. Formal structured teaching sessions do not meet the needs of teaching interactive and effective communication. Research supports the use of naturalistic teaching strategies. Teachers have been successfully trained to use these strategies, yet dissemination of this information from research to practice has not been adequately achieved. The following section describes the results of a survey sent to teachers of students with severe and moderate intellectual disability in N.S.W. It will be shown that the findings support the trends reported in the literature, with interesting directions indicated in terms of the effect of training received by teachers. A package addressing the assessment of student abilities and detailing instructional strategies, has been developed at Charles Sturt University - Mitchell, to respond to needs expressed in the survey and the supporting evidence found in the literature. METHOD Subjects A comprehensive survey* involving open-ended and rated items was sent to all teachers of students with moderate or severe intellectual disabilities in government schools across N.S.W. during the first half of 1991. In broad terms, targeted teachers were either working in support classes in regular public schools, or schools for specific purposes (S.S.P's). A sample of this magnitude could reasonably be expected to reflect a wide variation in levels of teacher education and experience in the field. Given the vagaries of flexible staffing, it was impossible to establish exact teacher numbers in any given region. Accordingly, initial estimates were inflated by approximately 10% when introductory letters, survey forms and reply-paid envelopes were mailed out to schools. Similarly, although the research team was specifically interested in the needs of teachers working with students with severe intellectual disabilities, the heterogeneous nature of class groupings for students with moderate (IO) or severe (IS) intellectual disabilities meant that ALL classroom teachers working with these populations were potential respondents. Both of the above factors may have important implications for the resultant return rate. Procedure 676 survey forms coded for demographic but not personal details, were sent to a total of 208 schools across the ten Departmental regions of N.S.W., with appropriate covering letters to Principals and teachers. A suggested final return date was indicated in this information, and when this date had passed, schools with a poor response rate were systematically phoned twice over a two week period to remind staff about the survey. The instrument required respondents to detail the communication skills of students in their class, outline the wider processes of service delivery and ecological factors at work in these students' lives, and express their personal skills and concerns in this area. Student detail was necessary to allow the research team to build up a picture of the communicative levels of individuals and groups, and match this information to teachers' perceived needs. The information provided by respondents was computed either as a simple frequency count or, in the case of open-ended questions, categorised on the basis of groupings identified across the data set. In a number of instances, chi-square analyses were then applied as a measure of association. In the following sections, a number of important outcomes will be presented and discussed. RESULTS Response rate Overall, a mean school response rate of 56% was achieved statewide (Regional Range 34-76%). This is, however, rather too broad a measure of teacher involvement, given that the dependent measure for school response was the return of one (1) or more surveys. More specifically, the mean teacher response rate was 36% statewide (n=241), with a regional range of 20-62%. In an effort to examine reasons for this response rate, a short follow-up survey has been sent to schools for distribution to non-respondents, and may confirm the distribution and identification difficulties identified earlier in the paper. As expected, the respondents represented a range of teacher training backgrounds, with 69% (n=167) having had some special education training (from Masters' level to preservice course components). Experience in special education teaching varied considerably, with the largest respondent group (42%) indicating service in the 0-5 year range. (Note: All percentages rounded to nearest whole number.) Differentiating the needs of teachers of students with moderate (IO) intellectual disabilities from those of teachers of students with severe (IS) intellectual disabilities. To distinguish between teachers of the two populations, a set of decision rules was introduced into the analysis to allow the identification of the lowest functioning student profiled in each survey. The communicative issues of system type, language functions, social interaction skills and initiation/response factors, were used as the basis for this process. Once a particular individual was identified within a class as the lowest functioning student, the teachers' designated placement of that student (IO/IS) became the guide to the identification of that person as a teacher of students with moderate or severe intellectual disabilities. Of the 241 respondents, 69 (29%) were classified as teachers of students with moderate intellectual disabilities (IO), while 169 (71%) taught students with severe intellectual disabilities. There were 3 respondents who did not indicate placement of students. Reliability of the decision rules As a test of reliability for this identification procedure, the decision rules were independently and separately followed with a subset of 20 randomly selected survey forms by two members of the research team involved in the coding process. Using the formula - Agreements __________________________________ x 100 Agreements plus Disagreements (Alberto and Troutman, 1990, p.129) reliability on the exercise was calculated at 90%. Student use of communication systems Tables 1 and 2 (below) suggest that students at the lower end of the intellectual scale are not effective users of formal systems of communication, such as speech, or augmentative systems. Almost 38% of the 241 lowest functioning students profiled were not reported as using an augmentative system, although 22% used a combination of such forms. As expected, use of speech by students with moderate or severe intellectual disabilities was significantly different from the level expected by chance, as indicated by the chi-square analysis reported in Table 2. TABLE 1 Augmentative systems used by the lowest functioning students Frequency % Compic 48 20 Manual signs 29 12 Combination 54 22 Miscellaneous 19 8 Not reported 91 38 TOTAL 241 100 TABLE 2 Use of speech by lowest functioning students (IO and IS) Students (IO) Students (IS) Consistent Use 47 19 Inconsistent Use 16 24 Not reported 6 126 TOTAL 69 169 (X p <.0001) However, low functioning students may be better users of informal modes of communication, a point suggested in the current literature (Houghton, et al., 1987; Siegel-Causey & Guess, 1989) and supported by data collected for the present study. Table 3 indicates that although many of the students with severe intellectual disabilities did not communicate a request for an object in any way, an important number did do so in an informal manner. The presence of an informal mode of communication amongst these students is consistently evident in the wider data pertaining to the various pragmatic components profiled in the survey. TABLE 3 Mode of request for an object by students. Students (IO) Students (IS) Formal System 39 (57%) 22 (13%) Informal Mode 13(19%) 63(37%) Not at all 10(14%) 72(43%) (X p < .0001) TABLE 4 Instructional strategies used by teachers of students with severe intellectual disabilities (IS) Arranging Learning Environment to Enhance Communi- cative Interactions Social Skills Training Teaching for Generalisation Reported 8 10 2 Not reported 161 159 167 TOTAL 169 169 169 Teacher Familiarity with, or Use of Techniques in Communication, Facilitation and Enhancement. The frequency figures presented in Table 4 derive from an open ended question in which teachers were asked to 'list any techniques in communication enhancement and/or facilitation that they were familiar with or regularly used with their students.' Although current research on best practice (Goetz & Sailor, 1988) suggests that techniques such as arranging the learning environment are important, even crucial to communication development, most teachers surveyed for the study do not appear to be either familiar with or using them to any significant level. In contrast, the data presented in Table 5, also summarised from responses to the above question, suggests that teachers place heavy emphasis on the communicative forms used by their students such as symbols (p <.0001) or manual signs (p <.003). TABLE 5 Augmentative forms reported as enhancement strategies by teachers Use of symbol systems Use of signing systems Reported Not Reported Reported Not Reported Teachers (IO) 17(25%) 52(75%) 23(33%) 46(67%) Teachers (IS) 104(62%) 65(38%) 92(54%) 77(46%) Teachers of the target population, therefore, appear to be focused on the communicative forms used by their students, but perhaps significantly less able to identify or use ways in which communication, as a wider process, can or should be taught. How to teach use of oneÕs communicative opportunities and enhance the power of that communication for the student, is a central theme of much of the current literature, discussed earlier, and is underlined as a vital professional need of the teachers who responded in this study. As if to further confirm this point, 137 of the 241 respondent teachers (57%) answered yes when asked the question - ÒDoes your school have a preferred system of communication augmentation which it uses with most of its students?Ó. Of the 137 who answered yes, 35% identified compics (computer pictures) as the school preferred system, with 21% indicating manual signs and 39% indicating a combination of systems. This begs the question: how well are we seeking to match individual student needs to relevant modes and functions of communication? Could it be that communication programming is system (or forms) driven, rather than having it's focus on the existing and unique skills and needs of the individual? Teacher training and instructional techniques Finally, it is important to consider the impact of special education training on teacher use of appropriate techniques in communication instruction, enhancement and facilitation. Table 6, below, indicates that specialised training has not significantly influencd teacher use of relevant strategies, although this conclusion is limited by the small report levels for the strategies overall. TABLE 6 Teacher training in special education and use of instructional techniques Use of structured instruction ______________ Arranging the Environment to Enhance Communication Teaching for Generalisation _______________ Reported Not Rpt. Reported Not Rpt. Reported Not Rpt. YES 7(4%) 160(96%) 8(5%) 159(95%) 1(1%) 166(99%) NO 1(1%) 73(99%) 2(3%) 72(97%) 1(1%) 73(99%) These data, and those which precede it, have a number of important implications both for policy directions in the area and practice in the classroom. These will now be explored. IMPLICATIONS FOR POLICY AND PRACTICE Data from the teacher survey are limited to some extent by the response rate and by the number of respondents who addressed questions relating to instructional methods in communication. Nevertheless, these data do point to some important issues for consideration by those involved in curriculum planning and in the provision of in-service education, consultancy services and other forms of support for teachers of students with severe intellectual disability. The literature review and our own experiences, indicate that many students with severe intellectual disabilities lack an effective oral communication system and frequently have difficulty in achieving the prerequisites for learning to use (formal) augmentative communicative systems such as the many electronic computer-assisted, symbol or manual signing systems (Arthur, 1989). Indeed, for those students who manage to master the motor tasks, the cognitive (symbol use) and the memory processes involved in using formal augmentative or alternative communication systems, there still remains the problem of learning the pragmatic or socially appropriate aspects of communication highlighted earlier in this paper. These are often very difficult to incorporate with the learning of a formal communication system (Molloy, 1987). Recent findings that many students have developed their own informal, if idiosyncratic means for acknowledging, for initiating interactions and for engaging in other interactive behaviours, demonstrate the interest of researchers in understanding student's success with the functions of communication, and not merely with mastery in using traditional forms (Bloom & Lahey, 1978). While communication using comprehensible forms of communication (like speech, signing or symbol use) is clearly an important task for students with severe disabilities and one for which their teachersÕ input will be very important, many researchers and writers on communication development are clear now that this emphasis on forms is not a sufficient educational or developmental goal. Students need assistance in incorporating repertoires for everyday social opportunities into their interactive behaviours which are sensitive to the knowledge and feelings or expectations of their communicative partners. These pragmatic considerations in training are probably more important to communication learning for this population than the acquisition of augmentative communication forms. It is interesting that responses to the survey showed a consistently greater concern by teachers for providing for students with severe disabilities by enhancing the available forms or modes of communication. For example, the data reported in Table 5 show that teachers of students with severe disabilities (IS) reported significantly greater use of symbol systems (p <.0001), and of manual signing (p < .003), than teachers of students with moderate disabilities (IO). By contrast, most teachers of students with severe intellectual disabilities (Table 4) did not report use of strategies which would promote pragmatic, or functional aspects of communication learning. These include such strategies as 'arranging the learning environment to enhance communicative interactions', 'social skills training' and 'teaching for generalisation.' Clearly, the perception of teachers in providing for students, is that the outward sign of the studentÕs communicative difficulty, i.e. the form of communication, is the first step for trying to develop a constructive intervention. Perhaps disappointingly, these orientations appear to be consistent across all groups of teachers irrespective of level of training in special education. The impact of recent research on communicative behaviour of students with severe intellectual disabilities has either not found its way to special education courses or has not transferred to the attitudes and beliefs of teachers completing them. This paper concludes with a number of statements which appear to be crucial for those involved in supporting the service delivery arrangements for students with severe intellectual disabilities. Firstly, there needs to be an appreciation by all individuals concerned with this population that these students can and do communicate, if we are attuned to observing and responding to efforts made. Secondly this view introduces the idea that teaching communication systems involving signing, symbol use or other systems is not an end in itself but rather a tool for expressing communicative intentions in ways which are more easily understood than student's own often idiosyncratic behaviours. Thirdly, in-service training is needed which will promote knowledge about the assessment of studentsÕ communicative behaviours and about a variety of teaching methods for enhancing communication opportunities. Finally, it is critical that parents, teachers and other professionals focus on identifying and developing a range of communicative functions in contexts relevant to the student. The dissemination of these skills in communication programming for students with severe intellectual disabilities would appear to be a high priority in the light of both current literature and the study reported here. The teacher skills package currently in its final stages of preparation at Charles Sturt University - Mitchell will go some way toward meeting these needs. *Note: Copies of the survey instrument are available from the first author, School of Teacher Education, CSU-Mitchell, Bathurst, NSW 2795. BIBLIOGRAPHY Alberto, P.A., & Troutman, A.C. (1990). Applied behavior analysis for teachers (3rd ed.) Columbus: Merrill. Arthur, M. (1989). Augmentative communication systems for learners with severe disability: Towards effective assessment and placement practices. Australian and New Zealand Journal of Developmental Disabilities, 15 (2), 119-125. Bloom, L., & Lahey, M. (1978). Language development and language disorders, New York: John Wiley & Sons. Bryen, D.N., & Joyce, D.E. (1985). Language intervention with the severely handicapped: A decade of research. The Journal of Special Education, 19 (1), 7-39. Burke, G.M. (1990). Unconventional behavior: A communicative interpretation in individuals with severe disabilities. Topics in Language Disorders, 10 (4), 75-85. Caro, P. ,& Snell, M.E. (1989). Characteristics of teaching communication to people with moderate and severe disabilities. Education and Training of the Mentally Retarded , March, 63-77. Carr, E.G., & Durand, V.M. (1985). Reducing behavior problems through functional communication training. Journal of Applied Behavior Analysis, 2. Cirrin, F.M., & Rowland, C.M. (1985). Communicative assessment of nonverbal youths with severe/profound mental retardation. Mental Retardation, 23, 52-62. Dunst, C.J., & Lowe, L.W. (1986). From reflex to symbol: Describing, explaining, and fostering communicative competence. Augmentative and Alternative Communication, 11-18. Goetz,L., Gee, K., & Sailor, W. (1985). Using a behavior chain interruption strategy to teach communication skills to students with severe disabilities. Journal of The Association for Persons with Severe Handicaps, 10, 21-30 Goetz, L., & Sailor, W. (1988). New directions: Communication development in persons with severe disabilities. Topics in Language Disorders, 8 (4), 41-54. Halle, J.W. (1987). Teaching language in the natural environment: An analysis of spontaneity. Journal of The Association for Persons with Severe Handicaps, 12 (1), 28-37. Halle, J.W., Marshall, A.M., & Spradlin, J. (1979). Time delay: A technique to increase language use and facilitate generalization in retarded children. Journal of Applied Behavior Analysis, 1,109-120. Haring, T.G., Neetz, J.A., Lovinger, L., Peck, C., & Semmel, M.I. (1987). Effects of four modified incidental teaching procedures to create opportunities for communication. Journal of The Association for Persons with Severe Handicaps, 12 (3), 218-226. Hart, B.M., & Risley, T.R. (1980). How to use incidental teaching for elaborating language. Lawrence KS: H & H Enterprises. Houghton, J., Bronicki, B., & Guess, D. (1987). Opportunities to express preferences and make choices among students with severe disabilities in classroom settings. Journal of The Association for Persons with Severe Handicaps, 12 (1), 18-27. Kaczmarek, L.A. (1990) Teaching spontaneous language to individuals with severe handicaps: A matrix model. Journal of The Association for Persons with Severe Handicaps, 15 (3), 160-169. Linfoot, K.W. (1987). Communicative behaviour in children with severe developmental delay, Unpublished doctoral dissertation, Macquarie University, Sydney. Mirenda, P., Iacono, T., & Williams, R. (1990). Communication options for persons with severe and profound disabilities: State of the art and future directions. Journal of The Association for Persons with Severe Handicaps, 15 (1), 3-21. Molloy, S. (1987). An investigation into the introduction of the vocriss, a synthesized voice output communicator, with a non-verbal child who has cerebral palsy. Unpublished graduate diploma dissertation, Mitchell College of Advanced Education, Bathurst. Rogers-Warren, A., & Warren, S. (1980). Mands for verbalization: Facilitating the display of newly trained language in children. Behavior Modification, 4, 361-382. Russel, M. (1984). Assessment and intervention issues with the nonspeaking child. Exceptional Children, 51, (1), 64-71. Schwartz,I.S., Anderson, S.R., & Halle, J.W. (1989). Training teachers to use naturalistic time delay: Effects on teacher behavior and on the language use of students. Journal of The Association for Persons with Severe Handicaps, 14 (1), 48-57. Siegel-Causey, E., & Guess, D. (1989). Enhancing nonsymbolic communication interactions among learners with severe disabilities. Baltimore: Paul H Brookes. Warren, S.F., & Kaiser, A.P. (1986). Incidental language teaching: A critical review. Journal of Speech and Hearing Disorders, 51, 291-299.