Using Behavior Measurement to Estimate Cognitive Function Based on Computational Models

Abstract

The measurement of cognitive performance is important in diagnosing and monitoring interventions for a wide variety of neurological conditions, such as dementias (Alzheimer's, vascular, etc.), multiple sclerosis, Parkinson's disease, and stroke recovery. These risk factors for cognitive decline are further aggravated with advancing age. The encouraging news is that recent research has shown that there is significant neuroplasticity in the adult brain, and that even the elderly are capable of achieving measurable changes in brain organization and function. Maximizing effectiveness of such interventions requires continuous (or at least very frequent), unobtrusive assessment of cognitive functions. This chapter describes how new behavioral informatics with computational models can be used to assess various cognitive functions in the wild and over time using new behavioral metrics, including walking speed, computer interactions and embedded measures in cognitive computer games. This new approach to cognitive monitoring offers substantial improvements over conventional infrequent assessments performed in the clinic. Namely, repeated measures in the home environment offer the ability to measure within-subject trends and potentially detect cognitive problems and intervene at an earlier stage.