The formative research that investigated the underlying methodologies applied in Visual Assistant began in 1993.  Since then, AbleLink has completed an initial research and development project (9/99-3/00) funded by the US Department of Education's National Institute on Disability Rehabilitation Research (NIDRR) that successfully demonstrated the feasibility of using a multimedia training program on a Handheld PC to effectively enhance independence for adults with mental retardation in performing community-based vocational tasks.  This six month project involved a process of determining the functional and technical requirements for the system, developing the software prototype and then testing it in a pilot study.

The pilot study that concluded the research plan was designed to assess the effectiveness of the portable visual/audio-based training system for training specific community-based vocational skills to individuals with mental retardation.  The basic hypothesis was that the Visual Assistant system would improve vocational skill acquisition and reduce the amount of staff time required to train individuals on new community-based tasks. The hypothesis was that 1) there would be significant improvement in the ability for an individual to perform the tasks independently, and 2) that the amount of errors in completing the task would be less for subjects when using the portable Visual Assistant prototype system.

The study consisted of a comparison of 1) the portable visual/audio training approach provided with Visual Assistant and 2) the traditional training method of task demonstration. Two tasks representative of community-based vocational tasks were selected after consulting with counselors at local adult services agencies and school district transition offices.  The Pizza Hut organization was generous enough to provide supplies for the first task, pizza box assembly.  The second task that was selected was software packaging, another task common to community-based vocational settings.  The independent variable in the pilot study was the type of training and task assistance provided to study participants.  The two levels of the independent variable were 1) live demonstration and use of the Visual Assistant prototype and 2) live demonstration only.  The two tasks were analyzed to document the steps required in performing the task.  Ten individuals (eight males and two females) participated in the study, ranging in age from 18 to 70 with an average age of 41.9.  WAIS-r scores of study participants ranged from 39 to 72 with an average score of 54.8.

The order in which the subject completed the task with the Visual Assistant prototype was randomized within constraints to control for order effects. Data were collected on the both dependent variables: 1) independence measured by the number of prompts required for each step in a task; and 2) accuracy measured by the number of errors made for each vocational task.  The results of this data demonstrated strong support for the contention that using a multimedia training approach with a Palmtop PC can effectively enhance independence for adults with mental retardation in performing community-based vocational tasks.   In summary, use of the Visual Assistant prototype resulted in improved task accuracy and increased independence for ten adults that were trained on two different vocational tasks as compared to a traditional training approach.  The average errors per task when individuals used Visual Assistant was 0.75, while the average errors per task when individuals did not use the Visual Assistant prototype was 2.25.  The observed mean difference for errors per task was statistically significant (p<.006).   Additionally, the average prompts per task when individuals used Visual Assistant was 1.05, compared to the average prompts per task of 2.40 when individuals did not use the Visual Assistant prototype.  The observed mean difference for prompts required per task was statistically significant (p<.032). These results are detailed below in Table 4.

Table 4:  Results of Visual Assistant Pilot Study

  * Significant difference at p < .05

In summary, the results of this pilot study demonstrated the effectiveness of the Visual Assistant prototype for improving task accuracy and increasing independence for ten adults with mental retardation on two different vocational tasks as compared to a traditional training approach.  These results must be considered preliminary as the research scope and time available for the evaluation were limited.  However, the pilot study results were strong enough to result in NIDRR's approval of a two year continuation project to: 1) complete the Visual Assistant multimedia software program, 2) build a companion software application for the PC that helps manage the setup of the Visual Assistant training tasks, and 3) perform expanded field testing of the system to evaluate its applicability for a wide range of activities of daily living and levels of disability.  This project, which will run from September 1, 2000 through August 31, 2002, will result in a commercial software application, Visual Assistant, that will be available for a wide variety of portable PC’s running the Windows CE operating system.  By capitalizing on commercially produced portable PCs, future Visual  Assistant users will be able to choose from multiple supported living and training products that will  be developed by AbleLink and other rehabilitation technology companies.  This provides a much more cost-effective approach to providing portable independent living software because a single hardware purchase will be able to operate a variety of Windows CE software titles.