Thursday, 19 November 2009

Want to predict a footie result? Don't even think about it

Imagine you've just paid an expert good money for their verdict and they say to you: "Can you hang on a couple of minutes whilst I don't think about this". You'd be forgiven for thinking they've gone silly. They may have. But another possibility is that you've chosen a shrewd expert who's totally up-to-speed with the latest decision-making research: Ap Dijksterhuis and his colleagues have just shown that people with expertise in football are better at predicting match outcomes when they spend time not consciously thinking about their predictions.

In an initial experiment, 352 Dutch undergrads were divided into football experts and non-experts, based on their self-ratings, and they were all asked to make predictions (home or away win, or draw) about four forthcoming football matches in the top Dutch league - the Eredivisie. The students were shown the four pairs of competing teams for twenty seconds, and then one third of them were asked to make immediate predictions; one third were asked to think consciously for two minutes before making their predictions; and a final third engaged in a distracting, numerical memory task for two minutes before making their predictions.

For the non-experts, it didn't make any difference to their success whether or not they were able to spend time considering their predictions - they were correct between forty and fifty per cent of the time regardless. By contrast, the experts' predictions were significantly more accurate when they were distracted for two minutes, compared with when they made an instant or a considered prediction (approx 60 vs. 50 per cent accuracy). In other words, the experts were most accurate when they spent time not consciously thinking about the problem at hand.

This may seem bizarre but it's entirely consistent with Dijksterhuis's Unconscious Thought Theory and with the folk wisdom that says it's a good idea to sleep on a problem. According to Dijksterhuis's theory, the subconscious is sometimes less prone to the biases that afflict the conscious mind, thus ensuring that an expert gives due weight to the most important factors.

This was borne out in a second experiment, much like the first, in which students predicted the outcomes of World Cup football matches. Again, distracted experts made the most accurate predictions. This time, however, the researchers also asked participants to estimate the teams' world rankings - apparently this is the most reliable predictor for the outcomes of World Cup matches. For experts who spent time consciously considering their match predictions, there was no correlation between their knowledge of team rankings and their prediction accuracy. By contrast, for the experts who spent time not thinking about their predictions, there was a correlation between their ranking knowledge and predictive accuracy. Not consciously thinking about the problem at hand seemed to ensure that experts paid due attention to the most important factor affecting match outcomes.

The researchers warned that subconscious thought is not always superior to conscious thought. But they concluded: "Our results mean that unconscious thought may well be helpful in more situations than some people currently think."

ResearchBlogging.orgDijksterhuis A, Bos MW, van der Leij A, & van Baaren RB (2009). Predicting Soccer Matches After Unconscious and Conscious Thought as a Function of Expertise. Psychological science : a journal of the American Psychological Society / APS PMID: 19818044

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

Bookmark and Share


spldbch said...

Wow -- that's interesting research. It seems to suggest that we really CAN "think too much" about things:-)

David Lockhart said...

These are interesting findings, however I interpret them slightly differently. Free association tasks have been shown to produce the most common response much more frequently when done after such a distraction period. This is thought to be because incidental aspects of cognitive set are flushed by the distraction task. The same may be happening here rather than "unconscious thinking". As an example of the sort of thing I mean, suppose a subject had seen an ad for a self-defense class immediately prior to the experiment, because of this priming they might put undue focus on which is the better defensive team in their intial assessment. The distraction task fills working memory with a bunch of irrelevant things, and afterwards this bias is no longer present. This is a bit like your computer running better after you reboot it.

jproach said...

So basically people are recalling two teams world rankings to compare them. I fail to see how this relates to intelligent thought. Why not ask a math or logic problem?

The "distracting, numerical memory test", could be acting as a brain exercise. Which allows them to recall long term memories easier.

Anonymous said...

I'd appreciate if someone could reconcile the apparent contradiction of these study results with those found below under "Unleash the crowd within", where the accuracy of making numerical predictions inproved if one consciously re-considered the question after making a first guess, and then made a second guess, and averaged the two.

Post a Comment

Note: only a member of this blog may post a comment.