Original articleTask instructions modulate neural responses to fearful facial expressions
Introduction
Lesion and functional neuroimaging studies have demonstrated the roles of different brain regions in the perception of certain basic emotions. Many of these studies have employed as stimuli facial expressions from the series of Ekman and Friesen (1976), in which subjects view different identities displaying facial expressions of fear, disgust, anger, sadness, happiness, and surprise, in addition to a neutral expression. Using these stimuli, it has been demonstrated in human subjects that the amygdala is of particular importance for the perception of fearful facial expressions Adolphs et al 1994, Breiter et al 1996, Calder et al 1996, Morris et al 1996, Phillips et al 1997, and may modulate the visual cortical response to these stimuli (Morris et al 1998). Other neural regions also demonstrated to be involved in the response to fearful expressions are the hippocampus and amygdalohippocampal junction Critchley et al 2000, Phillips et al 1998, and ventral and dorsolateral prefrontal cortices Hariri et al 2000, Sprengelmeyer et al 1998.
A range of tasks has been employed in previous studies investigating neural responses to fearful facial expressions. In the majority of functional neuroimaging studies either no task Breiter et al 1996, Buchel et al 1998 or an indirect or implicit task of emotion judgment has been performed (e.g., a judgment of gender rather than the emotion displayed by the face, to allow subjects to attend to facial expressions without becoming aware of the nature of the study) Morris et al 1996, Phillips et al 1997, Phillips et al 1998; however, if the task involves generation of emotional meaning or interpretation, or the representation of the subjective emotional response, previous studies point to activation of bilateral medial/middle prefrontal cortex (Brodmann area [BA] 9) and dorsal anterior cingulate gyrus (BA 24/32) Lane et al 1997, Partiot et al 1995, Teasdale et al 1999.
In more recent studies, neural responses to a range of different positive and negative, but not specifically fearful, facial expressions have been compared during performance of explicit facial expression labeling and other more indirect or implicit tasks of emotion processing Nakamura et al 1999, Narumoto et al 2000. These studies have demonstrated in response to positive and negative facial expressions, increased blood flow (Nakamura et al 1999), and activation (Narumoto et al 2000) of right dorsolateral and ventral frontal cortices during performance of explicit, emotion labeling tasks compared with more implicit tasks. In another study, neural responses to facial expressions of fear and anger were examined during performance of different tasks (Hariri et al 2000). Here, subjects matched a target facial expression of anger or fear either with one of two other facial expressions, or with one of two emotion labels. Performance of the matching task was associated with significantly greater activation in bilateral amygdalae than performance of the labeling task, while performance of the labeling task was associated with significantly greater activation of the right dorsolateral prefrontal cortex (BA 44 and 45) than performance of the matching task; however, the designs of these three studies did not permit direct examination of the effect of type of task upon neural responses to a specific series of facial expressions because of the employment of different types of facial expression or different combinations of facial expression and word labels during performance of the implicit and explicit tasks.
The effect of task type during presentation of a single series of positive and negative facial, but not fearful, facial expressions has been examined in two previous studies Critchley et al 2000, Gorno-Tempini et al 2001. In the first, subjects viewed alternating blocks of mixed happy and angry compared with neutral facial expressions. Here, performance of an implicit (gender-decision) emotion-processing task was associated with greater activation in the amygdalohippocampal junction than during performance of an explicit, emotion-labeling task. The design of this study did not permit a distinction between neural responses to happy and angry facial expressions, however, because of the contrast between two different facial expressions (happy and angry) versus one facial expression (neutral) in both experiments. In the second study, subjects viewed neutral, happy or disgusted facial expressions, and performed either a gender-decision task or an explicit, emotion-recognition task. Performance of the explicit task was associated with significantly greater activation within the right middle and dorsolateral prefrontal cortices.
To our knowledge, there has been no examination of the effect of performance of different types of task upon neural responses to presentation of an identical series of fearful facial expressions. We report here a study in which the sequence of stimuli was identical in each of three experimental conditions and only the task performed in each of the conditions was varied. We examined neural responses to a single series of fearful and neutral expressions during three different conditions: 1) passive viewing without task performance; 2) performance of a gender-decision task, which did not involve explicit judgment of facial emotion; 3) performance of an emotionality judgment task — an explicit emotion-processing task. Findings from previous studies allowed us to predict that, compared with passive viewing and performance of the gender-decision task, performance of the explicit task would be associated with a reduction of the amygdalar and hippocampal responses to fearful faces, and that this reduction would be associated with greater right dorsolateral and/or ventral frontal cortical activation (although we were unable to predict the specific location of frontal activation).
Section snippets
Stimuli and procedure
Nine right-handed healthy male subjects without previous history of psychiatric or physical illness, head injury, psychotropic medication use, or substance abuse (mean age, 29, range 21–42 years; mean number of years of education, 18, range 16–22) participated in three 5-min experimental conditions. The study was approved by the Ethical Committee (Research) of the Institute of Psychiatry. Written, informed consent was obtained from each subject after the details of the procedure had been
Facial expression recognition accuracy
Mean accuracy of recognition of the fearful facial expressions was 84.7% (range: 37.5%–100%).
Passive viewing
Major regions activated more by fearful than neutral facial expressions included the left amygdala, left hippocampus, the left putamen, bilateral cerebellum, and left lingual gyrus (Table 1 and Figure 1).
Gender-decision task
Major regions activated more by fearful than neutral faces included visual processing regions (the right fusiform and superior temporal gyrus), the left posterior cingulate gyrus, the right
Discussion
Our findings indicate that different types of tasks affect the neural response to presentation of alternating blocks of fearful and neutral facial expressions. Specifically, we observed significantly increased activation within the left hippocampus and left inferior occipital gyrus during passive viewing compared with performance of the gender-decision task or the explicit emotionality judgment task. The increased hippocampal activity during passive viewing of fearful facial stimuli may reflect
Acknowledgements
This study was supported by the Wellcome Trust. Project Grant No. 053651. We thank Dr. Krish Singh for use of his program to display the neuroimaging data upon a spoiled gradient echo (SPGR) template in Figure 1.
References (38)
- et al.
Generic brain activation mapping in functional magnetic resonance imagingA non parametric approach
Magn Reson Imaging
(1997) - et al.
Response and habituation of the human amygdala during visual processing of facial expressions
Neuron
(1996) - et al.
Brain systems mediating aversive conditioningAn event-related fMRI study
Neuron
(1998) - et al.
Explicit and incidental facial expression processingAn fMRI study
Neuroimage
(2001) - et al.
Functional MRI of the human amygdala?
Neuroimage
(2001) - et al.
Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala
Nature
(1994) - et al.
Episodic memory, amnesia and the hippocampal-anterior thalamic axis
Behav Brain Sci
(1999) - et al.
Dissociation of working memory from decision making within the human prefrontal cortex
J Neurosci
(1998) - et al.
Dissociable neural responses to facial expressions of sadness and anger
Brain
(1999) - et al.
Methods for the diagnosis and treatment of stimulus correlated motion in generic brain activation studies using fMRI
Hum Brain Mapp
(1999)