Skip to main content
. 2007 Mar;2(1):20–30. doi: 10.1093/scan/nsl021

Fig. 5.

Fig. 5

Regions in which the group (a–b) and individual differences (c) models provided significantly higher parameter estimates than did the fixed model for neural correlates of a reward prediction error (b–c) and the weights (a). Black circles enclose activations for ease in viewing.