DESCRIBING COGNITIVE DECLINE THROUGH A PICTURE DESCRIPTION WRITING TASK

Abstract

To describe cognitive decline, a large variety of neuropsychological tests are available. Linguistic processing is an important factor in some of these tests because language disturbances might serve as a strong indicator for an underlying neurodegenerative disorder, such as Alzheimer's disease. However, the current instruments for language assessment mainly focus on product measures, ignoring the importance of the process that leads to written or spoken language production. In this study, we aim to describe and test writing differences between healthy and cognitively impaired elderly on the basis of a selection of product and process variables. The latter are mainly related to pause times, because the number, length and location of pauses characterize writing fluency and flow. As a consequence, they are proven to reveal traces of the complexity of underlying cognitive processes. We matched fifteen cognitively impaired patients (10 patients with mild cognitive impairment and 5 patients with probable Alzheimer's disease) with fifteen cognitively healthy elderly for gender and age. Our research involved two copy and two picture description tasks. Both tasks were logged with Inputlog, a keystroke logging tool that allowed us to log and time stamp keystroke activity to reconstruct and describe text production processes. Preliminary analysis showed some promising results for cognitive effort related to word production in a written task. We used mixed effects models that included participants as a random effect and word categories as a fixed effect. Our presentation will focus on both product and process measures. Results associated with the process leading to a final text will mainly focus on pause times. For instance, cognitively impaired patients pause longer than healthy controls before, within and after words. Neuropsychological tests that assess language production based on patients' writing processes could be used for assessing cognitive decline in the future. As will be illustrated in our presentation, analyses already revealed a number of key variables that could distinguish cognitively impaired from healthy elderly. Hence, this points to a set of predefined and strictly controlled product and process aspects as important criteria for improving the current task design.