I should have seen that one coming. Human desire is bound up with an animal reward-learning system we share with many other species. Our desire to approach a task is coded by the neurotransmitter dopamine and the level of dopamine is heavily influenced by the apparent likelihood of success.
Studies with primates show that dopamine peaks when the odds of a win are about 50:50. So seeing the shape of a Christmas present beneath the tree and feeling it with our hands, but not being able to be completely sure of what it is, is an important part of why we want it so badly. Because of our dopamine-based learning system, the wrapping paper that shields us from certainty makes us want it more than if we absolutely knew it was not, or was, the object of our desire.
Of course this type of animal reward learning is very different from the formal learning children experience in school. Here, children usually prefer a much higher probability of achieving their goal, possibly because success in school is seen as a reflection of intrinsic ability and how learners are viewed by others and by themselves. But this relic from our days of foraging remains a powerful influence on modern life, as evidenced by the burgeoning number of lotteries, scratch cards and online gaming sites. In all of these, our animal attraction to risk appears to outweigh our rational understanding of the house advantage.
Our research within the Neuroscience and Education Network at the University of Bristol shows that making a classroom more unpredictable does make it more attractive, as long as the additional uncertainty is not a reflection of the pupil's ability.
In a computer-based maths quiz, we gave every pupil the choice of receiving the question from Mr Certain, who would award a point for the correct answer, or Mr Uncertain. Mr Uncertain would toss a coin if they answered the question correctly, providing two points for heads and zero points for tails. The choice between these two virtual examiners was always made before the question was seen. By the end of the 30 question quiz, there was a clear and increasing preference (especially with boys) to be rewarded by Mr Uncertain, who showed a distinct lack of respect for the principle of reward consistency.
Making sure every pupil is rewarded consistently for effort and achievement is a well-established principle of good teaching, but our results suggest children's response to reward may be more complex than we thought. Now, NEnet at Bristol University is carrying out laboratory and classroom studies in primary and secondary schools to gain a better understanding of exactly how uncertainty is wrapped up with engaging pupils.
We are finding interesting ways in which uncertainty subverts the conventional wisdom about the way children learn. Those who usually succeed become more vulnerable; those who rarely win are given a new source of hope. It also allows learners to moderate their own feelings of success and failure, in ways more often seen in sport. Some part of a disappointing performance can be attributed to bad luck while winners may prefer to attribute their success entirely to skill.
What has surprised us more, however, is not just the extent to which uncertainty engaged our learners, but how they found this disruption of reward consistency so unproblematic. Any accusations of unfairness centred on the usual areas - those that might reflect on ability, such as whether someone was given an easier question or not. Elements of uncertainty that arose from pure chance were rarely cited as potentially unfair, even when they had a radical impact on scored performance.
In the meantime, I am now on my way back to the high street for the last-minute stressful Christmas shopping I wanted to avoid.
Paul Howard-Jones is a lecturer in education and programme director of the MSc in education, technology and society at the Graduate School of Education, University of Bristol.
Fiorillo, C.D., Tobler, P. N., Schultz, W. (2003). "Discrete Coding of Reward Probability and Uncertainty by Dopamine Neurons." Science, 299, 1898-1902.
Knutson, B. and Cooper, J.C. (2006) "The lure of the unknown," Neuron 51(3), 280-282.
Shizgal, P., Arvanitogiannis, A. (2003). "Gambling on dopamine." Science, 299, 1856-1858.