Predicting Cognitive Status of Older Adults by Using Directional Accuracy in Explicit Timing Tasks

The early effects of age-related cognitive decline on explicit timing have been widely reported. However, it is not clear to what extent measures from the internal clock and working memory are predictive of cognitive function in older adults. During this study, we examined three target intervals (i.e., 2, 6, 10s) using Production and Reproduction tasks to assess the performance measures representative of the functioning of an internal clock and working memory, respectively. Participants were 36 older adults (19 females) with mean age of 68.4 +/- 5.1year and mean Montreal Cognitive Assessment (MoCA) scores of 27.9 +/- 2.0. During the Reproduction task, the participants were asked to reproduce the duration of a previously seen stimulus generated by a virtual reality program, whereas in the Production task the participants had to produce a target interval explicitly requested by the program. Each task consisted of nine trials with three repetitions of each target interval in a pseudo-random sequence. The average relative signed error for each target interval was calculated based on the participants' estimations. The six calculated signed errors, as well as the participant's age and gender were used as predictors for a linear regression model to determine the best predictors for the participants' MoCA scores using a backward method. The final regression model was found to be significant (F(2, 33)=5.06, p<.01, R-2=0.24) using only the two predictors of 6- and 10-s reproduction intervals. Our results show that measures of working memory provide a more reliable account of the variation of older adults' cognitive scores compared to age and the measures of the internal clock.