Warmer night-time temperatures are linked to poorer associative memory performance
Behler, A.; Thienel, R.; Bayliss, N.; Simpson, F.; McAloney, K.; Adsett, J.; Martin, N. G.; Breakspear, M.; Lupton, M. K.
Show abstract
Ambient temperature is emerging as an environmental factor that may influence cognitive performance in ageing populations. This is particularly relevant in Australia, where people live across diverse climatic regions spanning alpine to tropical conditions. We examined daily temperatures and cognitive performance in 1,873 midlife and older adults (1,297 women, mean age 61.0 years) who completed the Creyos online battery (formerly Cambridge Brain Sciences). Twelve tasks assessed memory, visuospatial processing, language, attention, and executive function. Task scores were linked to postcode-level contemporaneous weather data. The scores were analysed in relation to maximum and minimum air and wet-bulb temperatures and postcode- and month-relative temperature percentiles. Regression models adjusted for age, sex, education, socioeconomic status, and climate zone, with season included for air and wet-bulb measures. Higher minimum, but not maximum, temperature was associated with poorer performance on Paired Associates, a task assessing associative memory. This pattern was observed for air temperature, wet-bulb temperature, and temperature percentile, suggesting poorer memory performance after warmer nights, both in absolute terms and relative to local seasonal norms. Temperature was not significantly associated with performance on any other task, including measures of short-term/working memory, visuospatial processing, language, attention, or executive function. These findings suggest a task-specific association between higher overnight temperature and poorer associative memory performance, rather than a general reduction in cognition. Further studies incorporating personal exposure and sleep measures are needed to clarify whether night-time thermal conditions affect cognitive health in midlife and older populations.
Matching journals
The top 8 journals account for 50% of the predicted probability mass.