Integrative Graduate
Education and
Research Traineeship

Highlight

Predicting Activity Transitions to Improve Prompt Timing

Experimenter monitoring study participant

Projects:

Integrative Training in Health-Assistive Smart Environments

Institution:

Washington State University

Primary Strategic Goal:

Discovery: Foster research that will advance the frontiers of knowledge, emphasizing areas of greatest opportunity and potential benefit and establishing the Nation as a global leader in fundamental transformational science and engineering.

Secondary Strategic Goal:

Learning: Build the Nation's research capability through critical investments in advanced instrumentation, facilities, cyber infrastructure, and experimental tools.

Achievement/Results

Prompting technologies have gained popularity because of their effectiveness to increase successful activity completion for people with cognitive impairments. However, little is known about when prompts should be delivered. Early prompting literature focused on time-based prompts, which were shown to be useful; however, they do not consider the individuals’ environment, which can be problematic. To address this limitation, our IGERT trainees are designing technologies for context-aware activity prompting. The goal of context-aware prompting is to recognize what activity the individual is engaged in and then assess when would be the most effective time to prompt. Our IGERT trainees have developed a machine-learning algorithm that detects transition periods between activities. They are performing a study with healthy participants to assess whether context-aware transition period prompting will be more effective than traditional time-based prompting. Prompts are delivered via Android tablets or smart phones using an interface designed by the IGERT trainees.

Address Goals

Currently, context-aware prompting determines which activities to prompt using by a set of assumptions designed by the experimenters. However, these assumptions may not always be accurate. Prompting during activity transition, when a person is not engaged in anything, may be a more effective prompting time. We hypothesize that prompting during transitions between activities will lead to better task performance because there will not be a requirement to divide attention between the task at hand and the task directly following. We also hypothesize that individuals will be less likely to ignore prompts if they do not interrupt current activities. This project represents a unique multi-disciplinary collaboration between students in Computer Science, Experimental Psychology, and Clinical Psychology. The students have not worked together before or worked in multi-disciplinary settings. In addition to gaining unique perspectives on their research they are learning to communicate about their specialties to a broader audience.