Participants were paid ∼80

euros for their participation. In each of 5 scanning sessions of ∼8 min each, subjects viewed 120 successive, full-contrast BMS-777607 molecular weight Gabor patches that were oriented at between −90° and 90° relative to the vertical meridian. Each stimulus was visible for 1500 ms, during which period subjects were required to make a categorization judgment by pressing the right or left button on the response pad. Auditory feedback consisted of an ascending (400/800 Hz) or descending (800/400 Hz) tone of 200 ms, and followed stimulus onset by a variable interval in the range of 3–7 s. On 25% of trials, correct or incorrect feedback engendered a small monetary gain or loss, which was totaled up and supplemented subjects’

compensation (range 20–30 euros). An interstimulus interval of 1 s intervened between feedback and the subsequent stimulus. Stimuli were drawn randomly from category A (60 trials) or B (60 trials) with no constraints, and response-category assignments were counterbalanced across subjects. Category means and variances were unstable and independent, and jumped unpredictably every 10 or 20 trials (4 episodes of 10 trials, 4 episodes of 20 trials, randomly DAPT clinical trial intermixed) to a new mean drawn from a uniform random distribution with a variance of either 5° or 20°. Values representing the probability of choosing category A over B under the Bayesian model were estimated using a hierarchical Bayesian learner that calculates best-guess estimates of the generative mean and variance of each category in a Markovian fashion. For each category, a generative model of the observations is assumed as follows (see Supplemental Experimental Procedures and Behrens et al. [2007] for a more extensive description of a related model). At each trial i, after Rolziracetam the true category has been revealed, the probability of observing the orientation i (given any possible mean and variance) may be written: equation(Equation 4) p(Yi|μi,σi)∼N(μi,σi)p(Yi|μi,σi)∼N(μi,σi)Hence,

each new data point contains information about the underlying mean and variance. However, the mean and variance are constant over runs of trials before jumps, or change points occur. Hence, the prior distribution, conditional on the previous trial, may be written as follows: equation(Equation 5) p(μi|μi−1,Ji)={δ(μi−μi−1)U(0,180)Ji=0Ji=1This equation states that the underlying category mean at trial i will be the same as that at trial i-1 if there has not been a jump (J = 0), or could take on any value if there has been a jump (J = 1). A similar equation may be written to describe the dynamics of σ, which varied in a log space. equation(Equation 6) p(σi|σi−1,Ji)={δ(σi−σi−1)U(2,40)Ji=0Ji=1Jumps J occur at random with probability v, termed the volatility.