Azara Blog: Quick reactions allegedly correlated with long life

Blog home page | Blog archive

Google   Bookmark and Share

Date published: 2009/03/10

The BBC says:

People with quick reactions are likely to live longer than those less quick off the mark, a study suggests.

The joint Edinburgh University and Medical Research Council team looked at the response rates of more than 7,400 people, the journal Intelligence said.

Researchers found those with the slowest reactions were 2.6 times more likely to die prematurely.

They said quick reactions may be a sign of intelligence, which in turn was linked to healthier lifestyles.
Ash Ranpura, a neuroscientist at University College London, said: "There is evidence of a moderate correlation between how long it takes you to process information and intelligence. So these are very interesting findings."

Ranpura is wrong. These are not "very interesting findings". These are trivial findings. There ought to be some kind of correlation between just about any alleged measure of intelligence/health and just about any other alleged measure of intelligence/health. And it proves nothing, because correlations prove nothing. More pointless "health" related "research".

Update: Ash Ranpura (of the ICN at UCL) emails to clarify and correct:

Thanks for picking up the recent BBC story on cardiovascular health and reaction time. As you know doubt have guessed, the little quote from me at the end is printed out of context and doesn't adequately represent my views on this paper. Like you, I thought the paper wasn't hugely exciting, and I was surprised that the BBC was covering it. What I actually said to the reporter was that the findings might be very interesting if they provided the basis of a useful bedside diagnostic tool.

This is why I think you're unjustified in your blanket condemnation of correlational research. Health care workers look for factors which predict disease, because those factors help to select treatment plans. If the CRT measure used in the study is substantially predictive of cardiovascular health, then it is a cheap and easy way to identify patients who might need extra medical attention. It doesn't matter that CRT isn't causally related to cardiac health -- all that matters is that it helps clinicians diagnose patients.

You're wrong to say that any measure of intelligence or health will be strongly correlated with any other. My research demonstrates that vocabulary scores aren't predictive of mathematics scores in children, and that neither scores are predicted by postnatal birth weight or gestational age at birth (both correlate with general health). Moreover, vocabulary scores are predictive of the development of certain sets of brain areas (in the temporal lobe) while mathematics scores are predictive of parietal cortical development. So the scores index different kinds of cognitive functions and different aspects of biology, and all of those aspects can operate somewhat independently.

In short, the CRT test isn't amazing, but it's not completely useless either.

All material not included from other sources is copyright For further information or questions email: info [at] cambridge2000 [dot] com (replace "[at]" with "@" and "[dot]" with ".").