Does Time-on-Task effect vary as a function of age? A distribution analysis in young, middle-aged, and older people by fitting the ex-Gaussian parameters to the response time distributions.

Introduction.
Young people experiment mental fatigue after long-lasting cognitive tasks (Time-on-Task effect), evidenced by changes in response time (RT) distribution (Wang et al., 2014). As RT distribution are often positively skewed, the ex-Gaussian distribution (exponential-Gaussian convolution) fits RT data. It comprises three parameters: mu & sigma are mean & SD of the Gaussian while tau describes the exponential (the increase in extreme RT). In Young mu and sigma are not fatigue sensitive while tau suffers from Time-on-Task. Aim. To replicate results in Young (Wang et al., 2014) and to test Time-on-Task in Middle-aged and Older by fitting ex-Gaussian parameters to RT distributions.
Method.
Twenty-one Young (MAge=21.29), 16 Middle-aged (MAge=49.94), and 17 Older (MAge=65.06) performed a 160min Stroop task. It comprised facilitator (FA), interfering (I), and neutral (NE) items, these latter ones systematically appearing one item out of two. The duration was divided into four blocks of 40min each and we fitted the ex-Gaussian parameters to the RT distributions in those four blocks, separately for each item types. 3(Group) X 4(Block) repeated measure ANOVAs were carried out on the ex-Gaussian parameters for each item types.
Results.
Mu for FA and NE (both p<.01) were globally greater in Middle-aged and in Older as compared to Young while Middle-aged and Older did not differ from each other. Regarding I, mu was greater in Older as compared to Young. No significant effect found for Sigma. Tau in Young showed an increase for NE between the 1st and the 4th blocks (p=.02). Similarly, Middle-aged tended to show an increased tau for NE between the 1st and the 4th block (p=.05) and showed a significant increased tau between the 2nd and the 4th block (p=0.02). However, Older did not show any increased tau as a function of time block.
Discussion.
Only NE items were sensitive to mental fatigue. Given the predictability of NE, changes in tau with Time-on-Task could be due to alterations in top-down control necessary to counteract task-disengagement associated with task-monotony. Mental fatigue is differently expressed across lifespan. Young evidence an increase in extreme RT with Time-on-Task, particularly for predictive items. Older did not suffer from Time-on-task. However, they show very slow performance from the beginning. This floor effect possibly prevents Older from showing extreme RT, hiding fatigue installation. Interestingly, Middle-aged are relatively
similar to Older regarding their global slowness but, like Young, they suffer from increased
extreme RT on predictive items with Time-on-Task.