View all news

John Fennell and Roland Baddeley publish landmark article in Psychological Review

30 July 2012

Within Psychology and Economics, there has been a cottage industry identifying "biases" and irrationalities in human decision making. One much studied example of such "irrational" behaviour is that if you tell someone the probability of an event happening, they don't behave as if they believe this probability, but rather a distorted version.

Within Psychology and Economics, there has been a cottage industry identifying "biases" and irrationalities in human decision making. One much studied example of such "irrational" behaviour is that if you tell someone the probability of an event happening, they don't behave as if they believe this probability, but rather a distorted version. Specifically they underestimate large probabilities and over estimate small probabilities. In a paper by John Fennell and Roland Baddeley just published in Psychological Review, it is shown that this behaviour is in fact rational (and potentially optimal), as long as there is some uncertainty associated with probability statements (which in the real world situations there always is). In short, people treat a statement of 80% chance as equivalent to 8 out of 10, not 80,000 out of 100,000, and this uncertainty allows previous experience to bias our beliefs about probability statements.

Fennell, J., & Baddeley, R. (2012). Uncertainty Plus Prior Equals Rational Bias: An Intuitive Bayesian Probability Weighting Function. Psychological Review. doi: 10.1037/a0029346

 

Edit this page