Developing a psychological test battery to measure cognition in daily life.

Abstract

Cognition is not a static process but is subject to substantial and meaningful variation within individuals over time. This has led to an increase in studies that aim to describe cognition in daily life by sampling participants repeatedly and remotely. Such studies, which we call high-frequency cognitive assessment or "HFCA", tend to use a limited number of brief cognitive tests. This focus on a small number of tests leaves open questions concerning the psychometric characteristics of a wide array of cognitive tasks useful for HFCA that can guide researchers on appropriate task selection. We developed the Cognitive Variability Battery (CVB), a series of nine cognitive tests clustered into three distinct cognitive domains: attentional control, processing speed, and episodic memory. CVB was administered to participants three times per day for 3 weeks. Cognitive tests were rotated to reduce the length of any single testing session. Each test was administered up to 60 times. We provide detailed descriptions of the performance of each task, including variability, skew, and reliability statistics, using intraclass correlations. We examine the sensitivity of each test to several contextual factors including stress, affect, and social interactions. Finally, we provide power analyses on each cognitive test to determine how many assessments and participants are needed to detect effects of interest. These analyses will be essential to anyone seeking to implement HFCA testing in their own research programs, by providing guidance on which cognitive tests to select and how many participants and observations may be needed.