Emotional expressions evoke a differential response in the fusiform face area

Abstract

It is widely assumed that the fusiform face area (FFA), a brain region specialized for face perception, is not involved in processing emotional expressions. This assumption is based on the proposition that the FFA is involved in face identification and only processes features that are invariant across changes due to head movements, speaking and expressing emotions. The present study tested this proposition by examining whether the response in the human FFA varies across emotional expressions with functional magnetic resonance imaging and brain decoding analysis techniques (n = 11). A one vs. all classification analysis showed that most emotional expressions that participants perceived could be reliably predicted from the neural pattern of activity in left and the right FFA, suggesting that the perception of different emotional expressions recruit partially non-overlapping neural mechanisms. In addition, emotional expressions could also be decoded from the pattern of activity in the early visual cortex (EVC), indicating that retinotopic cortex also shows a differential response to emotional expressions. These results cast doubt on the idea that the FFA is involved in expression invariant face processing, and instead indicate that emotional expressions evoke partially de-correlated signals throughout occipital and posterior temporal cortex.

Modelling the perceptual similarity of facial expressions from image statistics and neural responses

“…Although our findings are consistent with neural models that suggest that this region is important for the representation of relatively invariant facial characteristics associated with recognition of identity (Allison et al, 2000;Haxby et al, 2000), they contrast with more recent studies that have shown responses in the FFA can be linked to the perception of facial expression (Harry, Williams, Davis & Kim, 2013;Wegrzyn et al, 2015). One potentially crucial difference between our study and these previous studies is that they asked only whether patterns of response to different facial expressions were distinct.…”

Insight into Emotion and Body Part Representation in Relevant Brain Areas: An fMRI Study

“…EBA had the highest emotion and body part classification accuracies which were not influenced by voxel numbers. The relative classification accuracies in other brain areas were almost in line with a previous study [7,10]. The present study represents an important step on understanding the emotion and body type representation in relevant brain regions.…”

“…We ranked the ROIs from high to low according to the classification accuracies, resulting in a sequence of EBA, OFA, FG, STS, IFG, insula and AMG (shown in Figure 2D). The ANOVA analysis for accuracies with ROI and ROI size showed a main effect for ROI (F (10,190 …”

N170 sensitivity to facial expression: A meta-analysis

“…In this sense, the influential model by Haxby et al (2000) proposed that IT cortex is in charge of invariant aspects of faces, such as identity, but is not involved in the processing of changeable characteristics such as expression. However, in recent years modulations of activity in IT by emotional expressions have been reported (Fox et al, 2009;Ganel et al, 2005;Harry et al, 2013;Kawasaki et al, 2012;Tsuchiya et al, 2008;Xu and Biederman, 2010).…”