Studying “fast chess” to see how decision making varies through the day

We've all had the experience of trying to make a tricky decision through the fog of fatigue, but there's surprisingly little objective evidence about how time of day affects the way we decide. Perhaps late-day tiredness makes us more rash, as we lack the energy to be considered. Alternatively, maybe it's our mid-morning zest that could lead us to be impetuous. Of course, our own chronotype is also likely come into play – perhaps morning people – "larks" – make better decisions in the morning, whereas evening people – "owls" – make better decisions in the evening.

One place to look for answers is in the data trails left by our online behavior. For a new paper in Cognition, a team led by María Leone has analysed the moves made by dozens of internet "fast chess" players, some of whom have played tens of thousands of two or three-minute games, consisting of an even greater number of moves. The results suggest that regardless of chronotype, we're inclined to make progressively faster, less accurate decisions as the day wears on, with the effect plateauing in mid-afternoon. 

The researchers asked the participating players – all of whom have playing data recorded on the Free Internet Chess Server – to complete a measure of their chronotype. Thirty-three per cent were owls, 9 per cent were larks and the remainder somewhere in between. As expected, the records showed the players had tended to play more games at their preferred times of day.

Ignoring early-game moves which are often based on rote-learned strategies, or moves made when game time was almost up, the analysis showed that player moves tended to be of a more conservative style in the morning, meaning the moves were judged as more optimal, but they were also made more slowly – the researchers call this a "prevention" focus, in the sense of preventing an error (past research has shown that this is the style people typically adopt when playing a superior opponent). By contrast, afternoon and night moves were more "promotion" focused, in that they were made more quickly, but less optimally. Although this pattern was found across all chronotypes, the trend for morning moves to be slower than afternoon and evening moves was more striking for larks.

It's worth noting that Leone and her colleagues do not interpret this as evidence that decision making simply deteriorates through the day. In fast chess, you lose if you run out of playing time, so there are clear advantages to deciding quickly. Consistent with this interpretation, there was no evidence that players' overall performance (as revealed by post-match changes to their online skill rating) varied with time of day.

This study provides a great example of how researchers are increasingly exploiting "big" online data to explore psychological effects in real-life settings. It's not clear how far we can extrapolate from these results to decision making in other contexts, but certainly the implication is that we're likely to be more cautious and considered in the morning, and more quick and bold in the afternoon or evening – perhaps worth noting when it comes to scheduling a business negotiation or even the time of day to do your grocery shopping.

Further reading

—Time to decide: Diurnal variations on the speed and quality of human decisions

Image via Rom/Flickr