Models that integrate sensory evidence to a threshold can explain task accuracy, response times and/nconfidence, yet it is still unclear how confidence is encoded in the brain. Classic models assume that/nconfidence is encoded in some form of balance between the evidence integrated in favor and against/nthe selected option. However, recent experiments that measure the sensory evidence’s influence on/nchoice and confidence contradict these classic models. We propose that the decision is taken by many/nloosely coupled modules each of which represent a stochastic sample of the sensory evidence integral./nConfidence is then encoded in the dispersion between modules. We show that our proposal can account/nfor the well established relations between confidence, and stimuli discriminability and reaction times,/nas well as the fluctuations influence on choice and confidence.

This work was supported by CONICET-Argentina (to L.P. and M.S.), the Spanish Ministry of Science and Technology Grant BFM2002-02042 (to D.C. and J.L.R.), by National Science Foundation Grant DMS-0245242 (to C.F.), by Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR - 2014SGR856 to AI and GD), by MINECO (PSI2013-42091-P to AI and GD) and the James McDonnell Foundation 21st Century Science Initiative in Understanding Human Cognition - Scholar Award (to M.S.).