Parametric analysis.
The parametric design analysis, with all the magnitudes of possible gains and losses included as parameters for each trial, revealed three main results: First, when examining the influence of wins, the dorsal ACC was significantly more activated during mixed-risk trials than in high- or low-risk trials (Table 3, Figure 6). Second, losses significantly activated the insular cortex (BA 13) and parts of the lateral OFC only during mixed-risk trials but not during high- or low-risk trials (Table 3, Figure 7). Third, for ambiguous trials, significant activation was found in the amygdala during mixed-risk trials with high ambiguity but not in response to mixed-risk trials with low ambiguity for the parameter magnitude of gains (Table 3, Figure 8). For ambiguous trials we found no significant activation with respect to the loss parameter.