Accuracy and consensus in judgments of trustworthiness from faces: Behavioral and neural correlates.

Abstract

Perceivers' inferences about individuals based on their faces often show high interrater consensus and can even accurately predict behavior in some domains. Here we investigated the consensus and accuracy of judgments of trustworthiness. In Study 1, we showed that the type of photo judged makes a significant difference for whether an individual is judged as trustworthy. In Study 2, we found that inferences of trustworthiness made from the faces of corporate criminals did not differ from inferences made from the faces of noncriminal executives. In Study 3, we found that judgments of trustworthiness did not differ between the faces of military criminals and the faces of military heroes. In Study 4, we tempted undergraduates to cheat on a test. Although we found that judgments of intelligence from the students' faces were related to students' scores on the test and that judgments of students' extraversion were correlated with self-reported extraversion, there was no relationship between judgments of trustworthiness from the students' faces and students' cheating behavior. Finally, in Study 5, we examined the neural correlates of the accuracy of judgments of trustworthiness from faces. Replicating previous research, we found that perceptions of trustworthiness from the faces in Study 4 corresponded to participants' amygdala response. However, we found no relationship between the amygdala response and the targets' actual cheating behavior. These data suggest that judgments of trustworthiness may not be accurate but, rather, reflect subjective impressions for which people show high agreement.

The Look that Binds: Partner-Directed Altruistic Motivation and Biased Perception in Married Couples

“…For example, it may be the case that although actor motivational orientation plays the most important part, certain attributes of the partner, such as the frequency with which s/he engages in accommodative behaviors, may still contribute, albeit to a much lesser degree, to biased partner perception processes. Likewise, consonant with previous findings that the evaluative context can impact trustworthiness evaluations (see Study 1 in Rule, Krendl, Ivcevic, & Ambady, 2013), there was a trend for participants who reported greater partner-directed altruistic motivation to be more likely to choose the untrustworthy looking morph as being more similar to the unfamiliar faces. Subsequent analyses revealed that this contrast effect emerged as a joint function of perceiver marital satisfaction and partner-directed altruistic motivation, which may explain why, in our main analyses, focused on the unique effect of altruistic motivation, this effect failed to reach significance (p = .06) .…”

The Neurobiological Basis and Potential Modification of Emotional Intelligence through Affective / Behavioral Training

“…Direct contrasts between each dynamic trustworthiness condition versus the neutral condition revealed left amygdala activation to Increasing Trustworthiness and right amygdala activation to Decreasing Trustworthiness. This is consistent with prior research showing that the process of evaluating facial trustworthiness activates the amygdala (Engell, Haxby, & Todorov, 2007;Rule, Krendl, Ivcevic, & Ambady, 2013;Winston et al, 2002), and with a large literature suggesting that the amygdala is involved in detecting facial cues associated with threat and fear (Killgore & Yurgelun-Todd, 2004Whalen et al, 1998), as well as other emotional expressions (Fitzgerald, Angstadt, Jelsone, Nathan, & Phan, 2006). Furthermore, such responsiveness of the amygdala to trustworthiness information appears to be more strongly correlated with facial features that are commonly agreed upon by consensus raters as a signal of untrustworthiness than to idiosyncratic judgments unique to the individual perceiver (Engell et al, 2007;Rule et al, 2013).…”

Trying to trust: Brain activity during interpersonal social attitude change

“…We did not observe greater amygdala activity during the control-of-distrust condition, where previous trust studies have shown increased amygdala activation in response to faces evaluated as both trustworthy and untrustworthy (Rule et al, 2013;Winston et al, 2002). It is currently unknown how task instructions to control the attitude of distrust may affect the way that the amygdala functions, and this remains an open question for future research.…”

“…The design of the task was based on prior behavioral research on the trustworthiness evaluation of faces (Rule et al, 2013;Todorov, 2008). Participants were presented with a series of 36 neutral faces and were instructed to evaluate how trustworthy they believed each face to be, on a 7-point Likert scale, with 1 being untrustworthy and 7 being trustworthy.…”

“…Many researchers have investigated the tendency to trust others by using tasks that involve the trustworthiness evaluation of faces or through economic games that infer trust via cooperation (Adolphs, 2002;Dimoka, 2010Dimoka, , 2011Krueger et al, 2007;Riedl, Hubert, & Kenning, 2010;Riedl, Mohr, Kenning, Davis, & Heekeren, 2014;Rilling et al, 2002). These studies have highlighted the role of limbic regions, such as the amygdala, when faces are judged to be either very trustworthy or very untrustworthy (Rule, Krendl, Ivcevic, & Ambady, 2013;Winston, Strange, O'Doherty, & Dolan, 2002). When using paradigms such as the prisoner's dilemma or the trust game, studies have shown a mixed pattern of results.…”