Skip to main content
. Author manuscript; available in PMC: 2015 Apr 1.
Published in final edited form as: Nat Neurosci. 2014 Aug 31;17(10):1319–1321. doi: 10.1038/nn.3798

Figure 2.

Figure 2

Experimental paradigm and behavioral results. (a) Experimental paradigm. In the Message condition, the participant in the role of the signaler is presented with two options, A and B, associated with different monetary consequences. For example, Option A corresponds to $15 to the participant and $5 to an anonymous signal recipient, i.e., ($15, $5), and Option B corresponds to ($5, $15). There are furthermore two actions available to the participant in the form of two statements describing the monetary consequences of the options to the recipient. Specifically, the participants must choose between sending a truthful message (Message 2) that sacrifices economic self-interest in favor of honesty, or a false message (Message 1) that satisfies self-interest at the expense of being honest. See Online Methods for details. (b) Amount given. In the Choice condition, all cohorts gave similar amounts to the recipient (Healthy Comparison: $7.44 ± .22; DLPFC: $6.65 ± .38; OFC: $6.79 ± .35; Kruskal-Wallis test, p > .10, two-tailed). In the Message condition with identical monetary consequences but with the addition of honesty concerns, healthy participants increased giving by $2.94 ± .44. In contrast, DLPFC cohort’s giving increased by less than half this amount ($1.05 ± .43), and significantly lower than those of the healthy comparison cohort (Wilcoxon rank sum test, p < .001, two-tailed). Finally, OFC participants were nearly identical to healthy participants ($3.01 ± .55; Wilcoxon rank sum test, p > .50, two-tailed), and significantly different from DLPFC participants (Wilcoxon rank sum test, p < .001, two-tailed). (c) Conflict and no conflict trials. On trials in the Message condition where honesty motives conflicted with those of self-interest (Top), DLPFC patients made a significantly lower proportion of honest choices (36.7% + 5.75%) compared to OFC and healthy comparison cohorts (OFCs: 75.7% + 5.44%; healthy participants: 83.3% + 3.00%; Fisher’s exact test, p < .01 for both, two-tailed). In contrast, on trials where conflict was absent (Bottom), no significant differences existed between cohorts (Fisher’s exact test, p > .20, two-tailed). All error bars indicate SEMs.