another Le Monde column

Another column in Le Monde (Sciences) had most unjustly escaped my attention: it mentioned Thomas Bayes on the very front page and I missed it till my most recent breakfast! This article was written by a neuroscientist columnist reporting on current research led by Tali Sharot, UCL, on the prediction mechanisms (if not on her book). The argument is not only that the brain operates in a Bayesian fashion, actualising predictions based on current observations (as exposed at Bayes 250), but also that the updating is not “objective”! While this may sound as if the neuroscientists have entered the debate between objective and subjective Bayesians, the study actually reports a bias toward optimism, when comparing predictions with “objective statistics”. The article concludes on the psychological advantages of this optimism bias. Not so much about Bayesian statistics, then, even though having almost everyone (subconsciously) working with his/her optimistic prior sounds rather cool!