Face-induced expectancies influence neural mechanisms of performance monitoring

Abstract

In many daily situations, the consequences of our actions are predicted by cues that are often social in nature. For instance, seeing the face of an evaluator (e.g., a supervisor at work) may activate certain evaluative expectancies, depending on the history of prior encounters with that particular person. We investigated how such face-induced expectancies influence neurocognitive functions of performance monitoring. We recorded an electroencephalogram while participants completed a time-estimation task, during which they received performance feedback from a strict and a lenient evaluator. During each trial, participants first saw the evaluator's face before performing the task and, finally, receiving feedback. Therefore, faces could be used as predictive cues for the upcoming evaluation. We analyzed electrocortical signatures of performance monitoring at the stages of cue processing, task performance, and feedback reception. Our results indicate that, at the cue stage, seeing the strict evaluator's face results in an anticipatory preparation of fronto-medial monitoring mechanisms, as reflected by a sustained negative-going amplitude shift (i.e., the contingent negative variation). At the performance stage, face-induced expectancies of a strict evaluation rule led to increases of early performance monitoring signals (i.e., frontal-midline theta power). At the final stage of feedback reception, violations of outcome expectancies differentially affected the feedback-related negativity and frontalmidline theta power, pointing to a functional dissociation between these signatures. Altogether, our results indicate that evaluative expectancies induced by face-cues lead to adjustments of internal performance monitoring mechanisms at various stages of task processing.

Electrophysiological Correlates of Cue-Related Processing in a Gambling Task: Early Outcome Evaluation or Outcome Expectation?

“…This result was consistent with the results from several previous studies that have reported CNV arising after the onset of cues (Goldstein et al, 2006; Silvetti et al, 2014; Novak and Foti, 2015; Osinsky et al, 2016). Furthermore, the slow negative wave was larger for cues that predicted a loss with 50% probability than for cues that predicted either a loss or a gain with 100% probability.…”

“…Novak and Foti (2015) found that cues that predicted potential losses elicited a larger negativity than cues that predicted potential gains. However, this finding that there were no significant differences for FRN between cues that predicted losses (the 0-cue condition) and cues that predicted gain (the 2-cue condition) was consistent with two other previous studies (Nieuwenhuis et al, 2007; Osinsky et al, 2016). It should be noted that the cues that predicted outcomes were with a certain probability in some of those studies that have found cues predicting losses elicited a larger negativity than cues predicting gains (e.g., 80% in study by Holroyd et al, 2011).…”

“…Our results were also in line with those of Foti and Hajcak (2012), who found that another expectation-related ERP component, namely the stimulus preceding negativity (SPN), was larger for uncertain cues than for certain cues. These results suggested that the cue-linked CNV effect found in our study may reflect uncertainty, and preparatory activation of the performance monitoring functions during reward expectation (also see Osinsky et al, 2016). …”

“…However, studies which have supported the idea that the brain codes reward expectation have usually found a P300 or CNV after 300 ms. These differing results may suggest that when a cue is presented, the brain does an initial evaluation of forthcoming outcomes first, and then uses cue information to form a reward expectation (also see, Osinsky et al, 2016). For example, when cues are presented, individuals may first make an initial appraisal about whether the desired outcomes are probable (early outcome evaluation), and then use the information provided by the cues to shape their expectation regarding any upcoming reward (reward expectation).…”

“…We expected that the cues that predicted outcomes would elicit not only an FRN-like negativity, reflecting the initial appraisal of whether the desired outcomes was probable (e.g., Holroyd et al, 2011), but also an ERP component (P300 or CNV), reflecting an expectation concerning the upcoming outcome (e.g., Broyd et al, 2012; Pfabigan et al, 2014; Silvetti et al, 2014; Novak and Foti, 2015; Osinsky et al, 2016). …”