Neural Processing of Fearful and Happy Facial Expressions: Effects of Fixation to Facial Features and Task Demands
MetadataShow full item record
The current literature regarding the time course of facial expression processing is inconsistent and early emotion effects are debated. Facial expressions are well-known to be characterized by “diagnostic” facial features (e.g., the smiling mouth for happy expressions and wide open eyes for fearful expressions) and these “diagnostic” features have been suggested to modulate the neural response to facial expressions; however, a systematic investigation of the impact of facial features on the neural processing of facial emotions is lacking. Thus, in an attempt to elucidate the time course of facial expression processing, and these early emotion effects, the main objective of the current thesis was to investigate whether fixation to facial features influenced the neural response to facial expressions. Combining EEG and eye-tracking using a gaze-contingent procedure, three experiments tested whether fixation to the “diagnostic” facial features of a given emotion was driving these previously reported early emotion effects on well-known ERP components (P1, N170 and EPN) during a gender discrimination (Experiment -Exp.1), explicit emotion discrimination (Exp.2) and an oddball detection (Exp.3) task. Given that experimental procedures have also been highly inconsistent in the previous literature, the impact of task on the time course of facial expression processing was directly tested within-subjects in Exp.4. Differential effects for fearful and happy expressions were seen at posterior sites, earlier and mostly occipital for happy expressions and later and mostly lateral for fearful expressions with no differences seen between tasks (Exp.’s 1 to 4), and these emotion effects interacted with fixation to facial features (Exp.’s 1 to 3). Happy cues from the mouth were required for early processing of happy expressions (i.e., happy gist) likely driven by low-level differences and the later semantic processing of the emotional content of the face. Fearful cues from both the mouth and the eyes were important for semantic processing of the emotional content of the face. Importantly, no interaction between emotion and fixation location was seen on the N170 (index of processing of structure of the face) arguing for separate processing of structural and emotional aspects of the face. Differential effects of fixation location were seen for the P1 and N170, with a sensitivity to face position (low-level) on the P1, followed by an eye sensitivity seen on the N170 component, possibly reflecting the activity of an eye-detector in the processing of the face structure. Overall, this thesis has helped to elucidate the debated early emotion effects in the temporal domain and has extended our current understanding of the role of facial features and task demands during facial expression processing. Results also highlighted the need for controlling for fixation in ERP emotion research and the importance of quantifying neural activity around P1 and N170 peaks as emotion effects may be missed by simply measuring these commonly studied ERP markers.