Age differences in the longitudinal associations of leisure-time physical activity and depressive symptoms with cognitive decline in older Taiwanese.

Objectives: The associations of physical activity and depressive symptoms with cognitive decline are likely age dependent. The study was therefore aimed to investigate the age differences in the associations of leisure-time physical activity (LTPA) and depressive symptoms with cognitive decline in older Taiwanese.Method: A total of 3545 participants, age ≥50, from 2003 (baseline) and 2007 (end-point) survey datasets were analyzed. Cognitive decline was defined as a decrease of two or more points on the Short Portable Mental Status Questionnaire. The long-term LTPA and depressive symptoms were the change of status between the two time-points. Multivariable logistic regression was conducted to evaluate the longitudinal associations, stratified by age.Results: Significant associations between all levels of long-term LTPA status and cognitive decline were observed for the older group (age ≥70), but not any for the younger group (age 50-69). Significant associations between long-term depressive symptoms status and cognitive decline were also found in all levels for the older group, but only in one level for the younger group. The two odds ratios of cognitive decline between the age groups were further found to be significantly different for those with active LTPA in both time-points (p=.005) and for those with depressive symptoms decreased (p=.014).Conclusion: The findings suggest that there are age differences in the longitudinal associations of LTPA and depressive symptoms with cognitive decline in older Taiwanese. Understanding the pattern of age differences could be helpful in preventing cognitive decline of the elderly.

Psychophysiological factors as predictors of second language writing achievement in a computer-based test.

Sleep quality, personality, and cognitive load potentially increase second language writing (SLW) anxiety and subsequently affect SLW achievement. This study investigates the predictions of sleep quality, personality (social inhibition/ negative affectivity), and cognitive load (content/ computer) toward SLW anxiety and achievement in a computer-based test. Participants included 172 voluntary undergraduates majoring in English as foreign language. SLW anxiety in a computer-based test, sleep disturbance, personality and cognitive load was assessed with the SLW Anxiety Inventory, Pittsburg Sleep Quality Index, Type-D Personality, and cognitive load questionnaires. A structural equation modeling approach was applied to examine the interdependence among the observed variables. An adequate-fit SLW anxiety model was built (X2 = 6.37, df = 6, p = 0.383, NFI = 0.97, CFI = 1.00, RMSEA = 0.02; R-squared multiple correlations: SLW anxiety in a computer-based test = 0.19, computer-based SLW achievement = 0.07). The structural model showed that sleep disturbance (+0.17), social inhibition personality (+0.31), and computer-induced cognitive load (+0.16) were significant predictors of SLW anxiety in a computer-based test. Subsequently, SLW anxiety in a computer-based test (-0.16) and computer-induced cognitive load (-0.16) were significant negative predictors of computer-based SLW achievement.