Fall Research Expo 2023

Developing An Algorithm for Runtime Gameplay Adjustments in A Mobile Cognitive Assessment

The mCAPP study, led by Principal Investigator Dr. Dawn Mechanic-Hamilton, is a mobile Cognitive App Performance Platform delivering cognitive assessments to those at risk for Alzheimer’s Disease and Related Dementias (ADRDs) in order to detect associated cognitive changes. The mCAPP mobile app contains three minigames aimed to assess memory and executive functioning skills, one of which is the Brick Drop Game (Mechanic-Hamilton).

 

Brick Drop is a processing speed and executive functioning test "modeled after the Stroop Test, which was developed by John Ridley Stroop in 1935 to measure cognitive processing speed and selective attention, as well as the ability to inhibit automatic responses. In essence, the Stroop Test measures how quickly and accurately a person can name the colors of words while ignoring the actual words themselves."

-ChatGPT

 

Brick Drop focuses on processing speed and executive functioning because these are elements of cognition that change in aging and can be further impacted by neurodegenerative diseases.  Therefore, by measuring and tracking these elements, the influence of ADRDs on cognition can be identified.

The goal throughout Summer 2023 was to update Brick Drop such that game difficulty (i.e. speed) adjusted to user interaction in real time. The design decision for the implementation of this update was that difficulty would depend on the user’s average reaction time. Thus, the hypothesis was that updating the queue speed property of the bricks based on an average reaction time calculated over 4 clicks will be the least noticeable whereas updating it based on an average reaction time calculated over 6 or more clicks will be the most noticeable.

The number of clicks (time stamps of mouse clicks) is important because the more clicks the algorithm needs to collect for the average reaction time calculation, the longer it takes to update game difficulty.  For instance, if the game waited until 100 click time stamps were collected before updating the game play, this would increase the noticeability of any gameplay change. 

Therefore, the independent variable is the number of clicks used to calculate average reaction time and the dependent variables are noticeability and difficulty ratings, which are determined by a questionnaire delivered after each user test of the game.

As of August 2023, a total of two testers have completed UX testing via a laptop computer.  Future steps for this project would be to conduct additional UX tests via mobile phone with at least six to ten more testers in order to reduce the impact of outliers and confounding variables influencing reaction time, such as mouse pad movement difficulties.  Additionally, it would be important to evaluate the efficacy of the updated Brick Drop version by collecting UX and performance data from mCAPP study participants.

 

References:

Mechanic-Hamilton D., et al. Remote App-Based Assessment of Memory and Executive Functioning in Aging and Pre-Clinical Alzheimer’s Disease.

PRESENTED BY
PURM - Penn Undergraduate Research Mentoring Program
Engineering & Applied Sciences 2026
Advised By
Dawn Mechanic-Hamilton
Assistant Professor, Department of Neurology & Chief, Division of Neuropsychology at the Hospital of the University of Pennsylvania
Jeffrey Vadala
Director and Lead Programmer of Penn Neurology VR Laboratory
PRESENTED BY
PURM - Penn Undergraduate Research Mentoring Program
Engineering & Applied Sciences 2026
Advised By
Dawn Mechanic-Hamilton
Assistant Professor, Department of Neurology & Chief, Division of Neuropsychology at the Hospital of the University of Pennsylvania
Jeffrey Vadala
Director and Lead Programmer of Penn Neurology VR Laboratory

Comments