Knowing when to trust others: an ERP study of decision making after receiving information from unknown people

Abstract

To address the neurocognitive mechanisms that underlie choices made after receiving information from an anonymous individual, reaction times (Experiment 1) and event-related brain potentials (Experiment 2) were recorded as participants played three variants of the coin toss game. In this game, participants guess the outcomes of unseen coin tosses after a person in another room (dubbed 'the reporter') observes the coin toss outcomes and then sends reports (which may or may not be truthful) to participants about whether the coins landed on heads or tails. Participants knew that the reporter's interests were aligned with their own (common interests), opposed to their own (conflicting interests) or opposed to their own, but that the reporter was penalized every time he or she sent a false report about the coin toss outcome (penalty for lying). In the common interests and penalty for lying conditions, participants followed the reporter's reports over 90% of the time, in contrast to <59% of the time in the conflicting interests condition. Reaction time results indicated that participants took similar amounts of time to respond in the common interests and penalty for lying conditions and that they were reliably faster than in the conflicting interests condition. Event-related potentials timelocked to the reporter's reports revealed a larger P2, P3 and late positive complex response in the common interests condition than in the other two, suggesting that participants' brains processed the reporter's reports differently in the common interests condition relative to the other two conditions. Results suggest that even when people behave as if they trust information, they consider communicative efforts of individuals whose interests are aligned with their own to be slightly more informative than those of individuals who are made trustworthy by an institution, such as a penalty for lying.

Fig. 1

Sample trial from Experiment 1.

Fig. 2

Percentage of participants’ predictions that matched the reporter's reports in the common interests, conflicting interests and penalty for lying conditions in Experiments 1 and 2.

Fig. 3

Grand average ERPs to the reporter's reports in the common interests (solid line), conflicting interests (dotted line) and penalty for lying (dashed line) conditions. Negative voltage is plotted up.

Fig. 4

Grand average ERPs recorded from the midline frontal (Fz) and parietal (Pz) electrode sites in the common interests, conflicting interests, and penalty for lying conditions. ERPs are timelocked to the reporter's reports in each experimental condition. Negative voltage is plotted up.