Abstract
Behavioral and electrophysiological correlates of the influence of task demands on the processing of happy, sad, and fearful expressions were investigated in a within-subjects study that compared a perceptual distraction condition with task-irrelevant faces (e.g., covert emotion task) to an emotion task-relevant categorization condition (e.g., overt emotion task). A state-of-the-art non-parametric mass univariate analysis method was used to address the limitations of previous studies. Behaviorally, participants responded faster to overtly categorized happy faces and were slower and less accurate to categorize sad and fearful faces; there were no behavioral differences in the covert task. Event-related potential (ERP) responses to the emotional expressions included the N170 (140–180 ms), which was enhanced by emotion irrespective of task, with happy and sad expressions eliciting greater amplitudes than neutral expressions. EPN (200–400 ms) amplitude was modulated by task, with greater voltages in the overt condition, and by emotion, however, there was no interaction of emotion and task. ERP activity was modulated by emotion as a function of task only at a late processing stage, which included the LPP (500–800 ms), with fearful and sad faces showing greater amplitude enhancements than happy faces. This study reveals that affective content does not necessarily require attention in the early stages of face processing, supporting recent evidence that the core and extended parts of the face processing system act in parallel, rather than serially. The role of voluntary attention starts at an intermediate stage, and fully modulates the response to emotional content in the final stage of processing.
Highlights
Human facial expressions are the main nonverbal channel for socio-emotional communication
In this epoch, encompassing the N170, the factorial mass-univariate testing (FMUT) returned a significant cluster for the Emotion main effect (F cluster mass = 4514.12, p < 0.0001, partial η2 = 0.225), while no significant clusters were detected for the Condition main effect, nor the Condition × Emotion interaction
The high-density electrode array combined with the mass-univariate approach that is not constrained by the choice of a specific electrode or region of interest, allowed characterization of the early Event-related potential (ERP) modulation at the whole scalp level, including both a negative modulation with a temporo-occipital scalp distribution resembling the N170 topography as is more typically reported [11,22,23,24], and a positive modulation with an anterior frontocentral distribution, as reported in some ERP studies of emotional processing employing the average mastoid reference
Summary
Human facial expressions are the main nonverbal channel for socio-emotional communication They convey essential information regarding others’ emotional state and intentions that is critical for proper social interactions. Given their biological and social relevance, information signaled by emotional expressions should be processed rapidly in order to help regulate behavior, for example when approached by a stranger with hostile or friendly intentions. Given their importance for survival, threat-related information, such as fearful expressions, may even be processed automatically, regardless of the attentional demands imposed by the situational context [1], as a bottom-up mechanism of early surveillance to threat stimuli [2]. As anticipated in the previous paragraph, stimuli of an emotional nature benefit from prioritized access to cognitive and neural resources from very early processing stages [4,5,6], more capture attention [7] and are resistant to attentional filtering when irrelevant [8,9]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
