Neural substrates of facial emotion processing using fMRI

Abstract

We identified human brain regions involved in the perception of sad, frightened, happy, angry, and neutral facial expressions using functional magnetic resonance imaging (fMRI). Twenty-one healthy right-handed adult volunteers (11 men, 10 women; aged 18–45; mean age 21.6 years) participated in four separate runs, one for each of the four emotions. Participants viewed blocks of emotionally expressive faces alternating with blocks of neutral faces and scrambled images. In comparison with scrambled images, neutral faces activated the fusiform gyri, the right lateral occipital gyrus, the right superior temporal sulcus, the inferior frontal gyri, and the amygdala/entorhinal cortex. In comparisons of emotional and neutral faces, we found that (1) emotional faces elicit increased activation in a subset of cortical regions involved in neutral face processing and in areas not activated by neutral faces; (2) differences in activation as a function of emotion category were most evident in the frontal lobes; (3) men showed a differential neural response depending upon the emotion expressed but women did not.

Introduction

The ability to decode emotional information on the face is critical to human communication. Clinical and experimental research in humans and in non-human primates suggests right hemisphere specialization [10], [11], [12], [22], [26], [35], [47], [53], [54]. Electrical stimulation, lesion, and microelectrode recording studies have identified several intrahemispheric brain regions involved in facial affect processing. These regions include the right temporal lobe [27], [55], [63], the right or left basal ganglia and anterior temporal lobe [16], the right mesial occipital and right inferior parietal regions [2], the right somatosensory cortex [1], and the amygdala [3], [4], [17], [72].

Neuroimaging studies of facial emotion perception suggest that there may be two parallel pathways, one for conscious processing of stimuli and one for unconscious processing of simple and associative stimulus features [45], [52]. Several imaging studies have implicated the amygdala in the processing of both seen and unseen fearful faces in humans [6], [14], [49], [50], [51], [58], [59], [70]. Morris et al. [52] identified a subcortical pathway from the midbrain and thalamus to the right amygdala that is involved in the processing of unseen visual events (e.g. fear-conditioned faces). Their data suggest that fusiform and orbitofrontal regions may be components of a pathway subserving conscious identification. In a more recent functional magnetic resonance imaging (fMRI) study, Critchley et al. [20] found that explicit processing of facial expressions of emotion activated temporal cortex in comparison to implicit processing which evoked significantly greater activity in the amygdala.

While investigators are beginning to map out the pathways engaged in conscious and unconscious emotional processing, little is known about how these pathways compare across different emotions, especially the pathways involved in conscious processing. Many studies such as those by Blair et al. [8], Morris et al. [49], Phillips et al. [58], and Sprengelmeyer et al. [67] required subjects to perform a gender discrimination task with a button-press response when emotional faces were presented. Thus attention is directed away from the emotions, and the regions activated are likely to differ from those engaged by fully conscious emotional face processing [20].

Only a few neuroimaging studies have examined the response to more than one facial emotion without using a backward masking paradigm or a task unrelated to emotional face perception. Using fMRI, Phillips et al. [57] found that happy face processing was associated with increased activity in the left anterior cingulate, bilateral medial frontal region, bilateral posterior cingulate, left supramarginal gyrus, and right putamen. They did not find any areas to be significantly more active in comparisons between sad and neutral faces. Also using fMRI, Breiter et al. [14] found bilateral activation of the amygdala and fusiform gyrus associated with viewing fearful versus neutral faces. They also found activation of the left amygdala and bilateral fusiform in comparisons of happy and neutral faces. In a ¹³³Xe inhalation study, Gur et al. [33] found that happy and sad emotional face discrimination tasks activated the right parietal lobe more than did an age discrimination task. Furthermore, the happy discrimination task activated the left frontal region relative to the sad discrimination task. In sum, the literature identifying brain regions engaged in the conscious processing of different facial emotions is limited and inconclusive.

Our fMRI study was designed to investigate explicit processing of happy, sad, angry, frightened, and neutral faces by normal volunteers. Subjects were instructed to ‘concentrate on each person’s expression’ when viewing a series of emotionally expressive faces. We did not require a behavioral response such as a button press because we did not wish to force subjects to engage in an artificial decision making task [60]. For each subject, four separate runs corresponding to the four different emotions were presented. We predicted that activations observed during the emotion conditions, over and above those seen during the neutral condition, would occur in the left and right fusiform gyri and in the left and right amygdalae and that these effects would be greater in the right hemisphere than in the left. We also sought to compare and contrast the brain regions involved in the perception of the four different emotions, although the existing literature did not permit us to make specific predictions in this regard.