Reaction Time Decomposition as a Tool to Study Subcortical Ischemic Vascular Cognitive Impairment

Abstract

Background: The study of reaction time (RT) and its intraindividual variability (IIV) in aging, cognitive impairment, and dementia typically fails to investigate the processing stages that contribute to an overall response. Applying "mental chronometry" techniques makes it possible to separately assess the role of processing components during environmental interaction.

Objective: To determine whether RT and IIV-decomposition techniques can shed light on the nature of underlying deficits in subcortical ischemic vascular cognitive impairment (VCI). Using a novel iPad task, we examined whether VCI deficits occur during both initiation and movement phases of a response, and whether they are equally reflected in both RT and IIV.

Methods: Touch cancellation RT and its IIV were measured in a group of younger adults (n = 22), cognitively healthy older adults (n = 21), and patients with VCI (n = 21) using an iPad task.

Results: Whereas cognitively healthy aging affected the speed (RT) of response initiation and movement but not its variability (IIV), VCI resulted in both slowed RT and increased IIV for both response phases. Furthermore, there were group differences with respect to response phase.

Conclusion: These results indicate that IIV can be more sensitive than absolute RT in separating VCI from normal aging. Furthermore, compared to cognitively healthy aging, VCI was characterized by significant deficits in planning/initiating action as well as performing movements. Such deficits have important implications for real life actions such as driving safety, employment, and falls risk.

Keywords: Aging; attention; cerebral small vessel disease; dementia; intra-individual variability; mental chronometry; reaction time; stimulus response/movement initiation & control; vascular cognitive impairment.

Fig. 1

A) The initial state of the response screen before the index finger is placed on the red home button. B) When the index finger (not shown) is placed on the home button, it turns green, and after a random interval one of the orange targets turns blue and begins to blink. The participants’ task is to move from the home button and touch the target as quickly as possible.

Fig. 2

RT and IQR patterns for both the lift and move measures, averaged across participants as a function of group. Horizontal lines show the median and large circle the mean of each distribution, with notches indicating 95%confidence intervals. Boxes identify the upper (75%) and lower (25%) quartiles, with the whiskers indicating maximum and minimum values, excluding outlier values. Individual participant median values are plotted as small black dots.

Fig. 3

Touch precision in terms of Euclidean distance (pixels) from centre of target objects averaged across participants as a function of group and object position. Note that object positions are numbered sequentially from left to right as they would appear on screen. Error bars indicated the standard error of the mean.