Generating Different Version of Game Based Identification using Procedural Content Generation and Parallelism for Identifying Adult with Dyslexia

Muhammad Maulana Ramadhan, Ari Moesriami Barmawi, Ph.D, Kristiantini Dewi, Sp.A

Telkom University, Indonesia
Dyslexia Association of Indonesia, Indonesia

Abstract

Identifying adult with dyslexia using game based test is a tough endeavor. This is due to two reason. First, unlike children, adult dyslexic no longer has problem on reading and writing but on managing their daily life. This entails that using language domain for the game is not a viable strategy. The second lies in their general ability which is higher than children. The higher abilities imply that the game mechanic used to measure them should be hard to master.

In order to solve the problems, this research proposes a game based test for identifying adult with dyslexia along with a method to generate different version of the game. The former utilizes executive function which is an aspect that still persists throughout adulthood in people with dyslexia. The latter is designed to store, retrieve and generate a different version of the test each time it is played. The method combines procedural content generation algorithm typically found in commercial games with parallelism. The procedural content generation ensures that different game can be generated, whereas the parallelism ensures that each generated test will have similar psychometric properties.

Keywords: Game Based Identification; dyslexia; parallel test; procedural content generation.