Analysis of Mobile Typing Characteristics in the Light of Cognition

Abstract

Cognitive decline is associated with a variety of neurological disorders. Assessment of cognitive domains beyond the clinical environment can support the detection of short- and long-term changes. It is particularly relevant in the early diagnosis of neurocognitive diseases and gaining insights into treatment progress. In this context, the most commonly used feature of mobile phones, the keyboard, provides a rich source to measure specific dimensions of cognition. The objective of this work involves revealing patterns of typing behavior among a population of healthy subjects and evaluating the applied methodology concerning a prospective clinical study on the determination of digital biomarkers for neurocognitive diseases. Therefore, this work introduces a modified version of the iOS default keyboard to measure typing speed and variation in character usage. A study is conducted on eleven healthy subjects to collect typing metrics for one week. The core results of the data analysis yield a positive-skewed distribution for typing speed and homogeneity in typing behavior among the population. Due to the similar statistical properties in typing behavior among healthy people, further studies surrounding subjects with neurocognitive impairment and diverse demographics are encouraged.