Are older people slower thinkers? New study shows that mental speed remains high until 60

Aging is strongly associated with cognitive decline and its fundamental property - decrease of mental speed. Mental speed is required for timely and adequate responses in complex environments. Older people being slower thinkers is a notion that found strong empirical support. Studies over the past few decades haveconsistently reported a negative relation between mental speed and age. This suggested linear trend had notable consequences and general perception of elderly people, as well as in general study of human cognition. The majority of these findings rely on a measure called mean response times (RT) in simple cognitive tasks. Mean RTs serve as a criteria of basic speed of information processing.

However, averaging all available RTs has two important drawbacks, which makes the prevailing use of mean RTs debatable. Firstly, mean RTs lack information on distribution being just a single point. Secondly, mean RTs represent not a pure mental speed, but rather accumulate the outcomes of different cognitive processes. To address these problems, the researchers employ complex mathematical models, such as the diffusion model (DM) that allows estimating a pure mental speed through the drift rate parameter. The drift rate denotes the average rate of information perceived or absorbed per time unit. 

The parameters derived from DM have been validated both experimentally and neurophysiologically. Interestingly, the researchers often reported that mental speed does not differ between young and old. However, decision caution (ability to take risk) and non-decision times (the time required for understanding and motor processes) were often increased in older age. Though the DM method has been growing in popularity, most of the studies compared only small groups of young adults with groups of older adults. Small sample sizes and only two age groups make the existing DM models less reliable and limit their generalizability.

To overcome the limitations of previous studies, von Krause et al. created a deep learning framework based on a massive dataset of more than one million participants, containing data on RTs and accuracy rates. Regarding age, their dataset covered range from childhood till late adulthood (ages 10 to 80). The authors observed a clear nonlinear association between drift rate (as a criteria of mental speed) and age, which differs from that implied by mean RTs and is more informative than previous DM studies. Additional parameters, defining mental speed, were analyzed by model, including decision caution and non-decision time.

The study replicated previous results showing the age-related increase in mean RTs. Mean RTs decreased during teenage years, reaching minima around 20, and showed nearly linear increase after, which further accelerated after 60. Decision caution declined from age 10 until approximately 18 (showing this period as a time of least cautious decisions), and increased linearly until the age of 65 with further acceleration after 80. Non-decision time was minimal around ages 14 to 16, with people outside this range needing more time for non-decision process. The most remarkable finding concerned the drift rate. From age 10 to 30 drift rate increased, showing increased mental speed, and, most importantly, it remained stable across all middle adulthood (age 30 to 60). After 60, the drift rate started decreasing contributing to slower mental speed. These results shed light on cognitive changes connected with age and prove that for large parts of the human lifespan and work careers mental speed remains mostly stable. 

Source: link 

‍