EP 124. Dual-task processing in young and older adults

Abstract

Introduction A substantial body of research has indicated impairments in dual-task performance with advanced age. Aim To corroborate previously reported age effects on dual-task performance and to examine potential moderating factors. In particular, we examined associations between dual-tasking and visuomotor coordination, cognitive flexibility, divided attention and task-switching. Methods The sample consisted of 20 young (mean age: 25.7 yrs.) and 20 elderly (mean age: 57.4 yrs.) adults. Dual-task performance was assessed using the well-established psychological refractory period (PRP) paradigm, in which two speeded choice-reaction tasks need to be performed either simultaneously or successively. Task 1 required a discrimination of high and low tones; Task 2 required a discrimination of letters “a” and “e”. A variable stimulus onset asynchrony (SOA; 50, 100, 350 or 800 ms) separated the two tasks. In addition, several tests were conducted to assess potential influencing factors including the Trail-Making Test (TMT) for visuomotor coordination and cognitive flexibility, a cross-modal divided-attention test, and an alternating-runs task-switching paradigm producing global and local switch costs. Results We observed typical dual-task costs for response speed (i.e., a significant PRP effect on reaction time, RT) but neither found a significant main effect of age on RT, nor a significant interaction between age and dual-task costs for RT in either Task. In the letter discrimination, however, there was a significant interaction between age and dual-task costs for accuracy driven by a more steeply increasing error rate with increasing task overlap in the older (vs. young) group. Neither visuomotor coordination nor cognitive flexibility co-varied with dual-task costs in RT and accuracy, but divided-attention and task-switching performance did. Difficulties in dividing attention and larger local switch costs were associated with a larger PRP effect on RT but not on accuracy. Conclusion When two speeded reaction tasks need to be performed in parallel, healthy aging may lead to deterioration in accuracy, rather than speed, relative to serial task processing. With regard to speed, however, both the ability to efficiently divide attention across sensory modalities and the ability to shift to a given task right after performing another one significantly predict low dual-tasking costs. As to accuracy, the pattern appears to be reversed, with age, rather than divided-attention or task-switching abilities, compounding dual-tasking costs. This suggests that doing two things at once is not per se compromised in old age but may depend more strongly on the integrity of other executive functions such as attentional control and task-set activation. Also, differential speed-accuracy tradeoffs need to be considered when studying age effects on performance.