Teaching New Symbols on AAC Devices:
Drill vs. Interactive Training
Jolyn Pate (NSSLHA Member)
University of New Mexico
536 Solano NE
Albuquerque, NM United States 87108
Janet L. Patterson (Member)
University of New Mexico
Department of Speech and Hearing Sciences
901 Vassar NE
Albuquerque, NM United States 87131
Mark A. McDaniel (Non-Member)
University of New Mexico
Department of Psychology
Logan Hall
Albuquerque, NM United States 87131-1161
Barbara L. Rodriguez (Member)
University of New Mexico
Department of Speech and Hearing Sciences
901 Vassar NE
Albuquerque, NM United States 87131
Abstract
This study compares 2 methods for teaching icon-referent associations to participants with impaired speech intelligibility and cognitive functioning using a DeltaTalker voice output communication device with Minspeak display. Recall and generalization measures are compared for drill versus interactive training conditions. Results and implications will be discussed.Narrative Summary
The most effective methods for teaching individuals new symbols on AAC devices have yet to be determined (Lloyd, Fuller, & Arvidson, 1997). In this study, drill and interactive dialogue methods for teaching icon-referent associations were compared.
Training using a drill format has been established as a useful method for teaching icon-referent associations (McDaniel, et al., 2001). This method is efficient in that it enables the student to learn the icon-referent associations and practice them repeatedly in a relatively short amount of time. The rehearsal and repetition that this training method supports could better facilitate the transfer of the new information to long-term memory, increasing the retention of the information. Thus, it is possible that the rehearsal and repetition inherent in a drill training method would support better recall of icon-referent associations.
Although recall of icon-referent associations might be increased with the use of a drill training method, without practical meaning attached to these symbols, generalization of the symbols to new contexts presents a problem for AAC users (Calculator, 1988). Language intervention literature shows that new language forms and new vocabulary are more effectively generalized when trained via a meaningful or naturalistic approach (Gillam, McFadden & van Kleeck, 1995; Norris & Damico, 1990; Scruggs, Mastropieri & Levin, 1985). Thus, it is also possible that using an interactive dialogue training method in which new icon-referent associations are attached to a meaningful context could better support the generalization of these icon-referent associations.
Research Question. This study sought to answer the following two part question: When using a Minspeak voice output communication aid, 1) does the performance of a group of 12 intellectually disabled students differ significantly on recall of icon-referent associations when trained via a drill-based approach versus when trained via an interactive dialogue approach, and 2) does this same group's performance on generalization of icon-referent associations to a story retell task differ significantly when trained via a drill-based approach versus when trained via a dialogue approach?Methods
Participants.
Participants will include twelve intellectually disabled 13- to 21-year-old students receiving special education services throughout the day. (Note: Data collection has been completed on four of the twelve participants to date.) Participants were identified by their teachers as having significantly impaired 1) speech intelligibility, and 2) cognitive functioning. TONI-3 (Brown, Sherbenou, & Johnsen, 1997) standard scores for the four subjects ranged from 65 to 74 and PPVT-III (Dunn & Dunn, 1997) standard scores ranged from 40 to 78.
Materials. Two 22-item vocabulary lists composed of eight nouns, eight verbs, four adjectives and two prepositions were used. The vocabulary items in these lists represent appropriate and common vocabulary in two separate everyday routines: dressing and eating. The DeltaTalker voice output communication device with Minspeak display was utilized. The DeltaTalker is a static display; the icons on the display board do not change. However, icons may represent more than one referent. For example, when the user selects the icon displaying a glass of juice plus the grammatical category, "noun," the device produces the word, "lemonade." However, when the user selects the juice icon plus, "adjective," the device produces the word, "thirsty." Thus, in order to communicate using the DeltaTalker, one must select a grammatical category in addition to a content icon to specify a referent for that icon.Procedures.
Each participant completed two training sessions. One training session included one set of vocabulary items learned using the drill training method, while the other training session included another set of vocabulary items learned using the dialogue training method. In "drill" training, the icon-referent associations and their rationale were presented. For example, while modeling the correct icon sequence on the device, the clinician instructs the student as follows: "To say, 'lemonade,' you press, 'juice,' because you drink juice, and 'noun,' because lemonade is a thing." In "dialogue" training, the clinician taught the student icon-referent associations by following a script and utilizing concrete contextual cues. The following excerpt from the eating routine script exemplifies this method: after making and eating popcorn with the student, the clinician says, "All that popcorn made me thirsty! To say, 'thirsty,' you press, 'juice,' because you drink juice when you're thirsty, and 'adjective,' because, 'thirsty' describes something." After modeling, the clinician instructs, "Now you try it," and the student has the opportunity to say, "thirsty" using the device.
Testing sessions followed each training session and included recall and generalization testing. Recall assessment was conducted by presenting 11 of the trained items in a drill format. Generalization was assessed utilizing a story retell task containing the remaining 11 items from the list. In order to counterbalance training method order, the participants were divided into two groups of six subjects.Results and Discussion
Data collection for the four subjects to date indicates little difference between training methods for recall and generalization of icon-referent associations. The mean scores for recall and generalization using a drill training method were 57% (SD=29.82) and 73% (SD=17.24) respectively. Using a dialogue training method, the mean recall and generalization scores were 61% (SD=20.99) and 77.5% (SD=13.5), respectively. Data collection continues and will be completed by July 2002, allowing for more conclusive results.
To date, a significant difference between drill and dialogue training methods is not apparent for recall and generalization of icon-referent associations. If this remains true following the completion of data collection, such findings would support either method when teaching new symbols on AAC devices and would provide clinicians with options to consider when selecting a training method for an individual client. For example, a student who not only needs to learn new symbols on his AAC device, but also requires considerable intervention to engage in social communication may benefit more from an interactive dialogue approach. In contrast, the drill training method, with its efficiency and relative simplicity, may be used with a client with good interaction skills who needs only to learn the icon-referent associations for his device.
