Cognitive decline could start early

Cognitive decline can start much earlier than previously thought, new research found. Published in the British Medical Journal, the study revealed the brain's ability to function might begin its deterioration at the age of just 45.

Investigators at University College London looked at comprehension, vocabulary and memory skills of people aged between 45 and 70 over a ten-year period.

They discovered a 3.6 per cent reduction in mental reasoning for members of both sexes aged 45 to 49, despite previous studies on the subject suggesting cognitive decline does not get underway until a person is in their sixties.

Professor Archana Singh-Manoux from the Centre for Research in Epidemiology and Population Health in France and leader of the research team at University College London, noted a link between such changes in the body and the onset of dementia.

The expert added: "Rates of dementia are going to soar and health behaviours like smoking and physical activity are linked to levels of cognitive function."

Last year, Professor Lorraine Tyler, Professor of Cognitive Neuroscience at the University of Cambridge, gave 2011's British Academy/British Psychological Society Lecture on a related subject at the Royal Society in London. Professot Tyler explained how resilient the brain can be and what we can do to stave off the effects of age. Read more about Professor Tyler's lecture on The Resilient Brain as reported in The Psychologist.

I was fascinated by the paper published in the BMJ. However, I was far from convinced (McCarthy, 2012). The crucial tasks were all performed against restrictive time limits. The only task that did not involve time limits was also the only task that did not show any decline as a function of age.
Anyone familiar with assessing cognitive function using psychometric tests (or even testing reaction times in older people) will realise that tests which are time-limited are subject to substantial age effects (e.g. Rabbit, 1979). A prime example is the WAIS-III Digit Symbol-Coding task: a score of 79 would be at the 50th percentile for someone between the ages of 16 and 34 - but by the age of 45 only 25% of the population would be doing as well and by 65 only 5%. Of course, cross-sectional studies such as the WAIS III standardisation have limitations and it is interesting to see that the present study replicates these established findings in a longitudinal sample. Change with age is not just a "Flynn Effect".
However, changing scores do not necessarily mean "decline" (as in "deterioration") has set in at 45. There are many reasons why timed tasks might be performed more slowly by older people including lower motivation or a more cautious response style (Forstmann et al, 2011). For psychologists and neuroscientists the most interesting tasks are those showing change with age irrespective of time limits - but unfortunately this study did not include such measures.

Forstmann B. U., Tittgemeyer M., Wagenmakers E.-J., Derrfuss J., Imperati D., Brown S. D. (2011). The speed-accuracy tradeoff in the elderly brain: a structural model-based approach. J. Neurosci. 31(47): 17242-1724
McCarthy, R.A. (2012) Intellectual functions may be slower but no worse with age BMJ 344 (7844) 25
Rabbitt P (1979) How old and young subjects monitor and control responses for accuracy and speed. British Journal of Psychology 70:305-311