Skip to main content
. 2018 Feb 26;43(7):1581–1588. doi: 10.1038/s41386-018-0032-x

Fig. 2.

Fig. 2

ROI results: Parameter estimates reflecting reward and punishment prediction errors extracted from the right (a) and left (b) striatum, habenula (c), right insula (d) and ventral tegmental area (VTA) (e) in healthy controls and MDD. Error bars represent standard error. *p < 0.05