CAPS: Context Aware Prompting System for Persons with Cognitive Disabilities

Project Team

Project Co-Directors: Cathy Bodine, PhD, CCC-SLP; Michael Melonis, PE
Project Team: K. Hogue, MS, J. Sanford M.Arch., C. Lewis, U of Colorado-Boulder
Project Partners: RERC-ACT, AbleLink Technologies, Inc.
Advisor: P. Nishman, NISH

Summary / Outcome Goals

The primary objective of this collaboration is to develop, test, and refine a context aware prompting system designed to act as a job coach for adults with cognitive disabilities working in assembly line jobs in competitive employment settings. A secondary objective is to conduct a randomized clinical trial comparing the impact of three types of work-related prompting systems: pen and paper, electronic aid (e.g., personal digital assistants (PDAs)); and CAPS.

The specific aims of this project are to: 1) characterize specific work tasks found in assembly line jobs; 2) customize paper-and-pencil (e.g., written instructions, symbolic representations) and electronic PDA (i.e., PocketCoach by AbleLink Technologies) prompting technologies for the specific tasks identified; 3) develop, refine, and test a context aware prompting system for the same work tasks; and 4) complete a randomized controlled trial to compare the impact of these three types of prompting systems on work outcomes by persons with cognitive disabilities.


Current Work

As part of the Work RERC, the engineering team at Assistive Technology Partners, University of Colorado, Anschutz Medical Campus, has been exploring the use of task-prompting systems that could aid an individual with a disability to successfully perform assembly line work. A linear and a non-linear prompting system are being compared to determine if one is more effective than the other. In both systems, there is a sequence of tasks that need to be performed. In the linear prompting system, the individual must determine if the task has been performed and then hit a button to check it off the list and proceed to the next task. In the non-linear system, we have developed a tool that automatically determines if the task has been performed. The tool uses environmental sensors to know what steps have been completed. If the task has been completed, the next step is automatically presented. In addition to task advancement, the automated prompting systems can recognize basic error scenarios.

The job that was chosen to test the prompting system involves packaging novelty Chocolate First Aid Kits. The assembly process consists of two jobs. The first job is to construct a box filled with four different kinds of chocolate bars. The second job involves placing two of the candy bar boxes, plus two bottles of chocolates, in a case that resembles a first aid kit. The position of the bottles and boxes are critical in the final assembly. Once the contents have been correctly placed, the lid is snapped shut and is sent off to be shipped.

Final Assembly Workstation A Sample of the Final Assembly Images

Our hypothesis is that the non-linear system will enable persons with cognitive disabilities to perform more effectively (fewer errors) and more efficiently (decreased time to complete a task). To date, 25 of 40 participants have completed the research protocol. Preliminary results suggest:


NIDRR

CATEA