Two different faces of threat. Comparing the neural systems for recognizing fear and anger in dynamic body expressions
Introduction
Watching fear and anger behaviors makes the observer feel threatened and prompts him to prepare an adapted response. It has long been understood that the behavioral manifestations of anger and fear shown in the face, the voice and the whole body help to prepare the body for adaptive action (Darwin, 1872, Frijda, 1986). They also serve as communicative signals by warning observers about potential threats in the environment (de Gelder, 2006). Yet, anger and fear signals are quite different as far as the adaptative behavior they elicit in the observer. In contrast with fear, anger is often displayed with the aim of altering the behavior of the agent to which it is addressed (Frijda, 1986) and therefore appears to be a more interactive signal in the sense that it requires the observer to adapt or regulate his own behavior in tune with the ongoing interaction.
With fear and anger both amounting to threat signals, an important question concerns the specificity of the observers' reaction to perceived anger and fear behaviors in others and this issue has not so far been addressed in the literature. Overall, neuroimaging studies in humans that investigated the perception of fearful facial expressions have reported amygdala and fusiform cortex responses (Morris et al., 1996, Phillips et al., 1997, Vuilleumier et al., 2001). Electrophysiological studies in the monkey's amygdala have also underscored its sensitivity to facial expressions, gaze or vocalizations signaling threat (Hoffman et al., 2007, Kuraoka and Nakamura, 2007). These observations are consistent with the view that the amygdala plays a central role in processing threat related signals and linking them to appropriate defensive and attentional responses (Amaral, 2003, LeDoux, 1995, Vuilleumier et al., 2004). To our knowledge, only few imaging studies directly compared brain evoked responses to fear and anger static facial expressions (Phillips et al., 1999, Whalen et al., 2001, Williams et al., 2005). The results showed that compared with neutral expressions, the perception of both fear and anger faces enhanced amygdala BOLD response, yet fearful expressions seem to evoke the greatest responses. In parallel, neuroimaging studies using fearful and angry facial expressions have often revealed activations in the inferior frontal gyrus and lateral orbitofrontal cortex (IFG BA45 and OFC BA47) (Blair et al., 1999, Fitzgerald et al., 2006, Kesler-West et al., 2001, Sprengelmeyer et al., 1998), consistent with their essential roles in processing emotional expressions (Hornak et al., 1996). Interestingly, Murphy et al. (2003) in their meta-analysis show the highest proportion of lateral OFC activations in studies targeting anger vs. other emotions. Yet as a majority of neuroimaging investigations have been using the same static material, it remains unclear how amygdala and other brain regions are engaged during sensory processing of other emotional signals such as dynamic body-related ones.
As noted above, anger-based vs. fear-based threat manifestations may trigger rather different adaptive behaviors. Therefore using whole body images rather than only facial expressions may better reveal the underlying neurofunctional similarities in emotion related action structures (de Gelder et al., 2004). Hadjikhani and de Gelder (2003) showed that the perception of body postures expressing fear elicited amygdala and fusiform responses in the same way that did facial expressions. Nevertheless, perceiving fearful body postures was also associated with activations in other affective centers such as the OFC and the insula as well as action-related areas such as the inferior frontal gyrus (IFG) and the premotor cortex (de Gelder et al., 2004). Grosbras et al. (2006) recently used realistic video-clips of hand actions expressing anger and found increased activations in the superior temporal sulcus (STS), the dorsal premotor cortex, the dorsomedial prefrontal cortex (dmPFC), the IFG, the insula and the supramarginal gyrus. Two other experiments investigated the impact of movement on the perception of actions signaling fear and anger (Grèzes et al., 2007, Pichon et al., 2008). The perception of static and dynamic angry and fearful actions was associated with increased responses in the STS, the amygdala and adjacent temporal pole, the inferior frontal cortices, the pre-SMA and the dmPFC. Moreover, the perception of dynamic actions expressing fear specifically engaged the STS extending to the temporoparietal junction (TPJ) and the premotor cortex (Grèzes et al., 2007), whereas the perception of dynamic actions expressing anger increased responses in the anterior temporal cortices, the ventromedial PFC (vmPFC), the hypothalamus and the premotor cortex. Together, these results showed that besides modulating sensory and emotional regions, the perception of actions expressing a threat is also coupled with increased responses in brain regions associated to motor preparation (Hoshi and Tanji, 2004) and defensive responses (Brown et al., 1969, Graziano and Cooke, 2006).
What remains unclear though is to what extent these responses are characteristic of perceiving a threat or whether some aspects thereof are specific to either fear or anger cues. To investigate this question, we used functional magnetic resonance imaging (fMRI) to record participants' brain haemodynamic activity while they were categorizing videos showing either fear, anger or a neutral action. We tested whether the amygdala is preferentially activated by fear signals. We also aimed at identifying the common and distinct regions associated with the recognition of fear and anger signals. From this, we drew three predictions: first, that the recognition of actions signaling threat increases the amygdala's response; second, that it also enhances the BOLD response in posterior temporal (STS, TPJ, fusiform) as well as inferior frontal (BA45 and BA47) regions; third, that the anterior temporal cortices and OFC are preferentially engaged during the recognition of anger signals.
Section snippets
Participants
16 right-handed volunteers (8 females; mean age = 25.6 years, standard deviation (SD) = 8; and 8 males; mean age = 23.5 years, SD = 2.6) with no neurological or psychiatric history participated in the imaging study. All provided written informed consent according to institutional guidelines of the local research ethics committee and were paid for their participation.
Stimuli
71 full-light 3 second videos (23 fear, 24 anger and 24 neutral) were used for the present experiment. Videos were chosen from a wider set
Behavioral results
Examination of the participants' average recognition rate revealed good recognition of the three expressions (mean 88.5%, SD = 4.7). Fear, anger and neutral movies (Fig. 1a) were recognized respectively, 81% (SD = 10.3), 86% (SD = 7.2), and 98% (SD = 2). A repeated measures ANOVA revealed a significant difference between emotions (F(2,30) = 25.74 P < 0.001, Greenhouse–Geisser sphericity correction) and post-hoc t-tests (corrected for multiple comparisons) showed that the latter result was driven by a
Discussion
The present study was designed to identify the neurofunctional basis of threat perception when observers are faced with fear and anger behaviors. This is the first imaging study that directly compares brain activity elicited by the recognition of dynamic actions signaling fear and anger. Our results clearly indicate that the recognition of fear and anger actions elicits similar activity in amygdala, posterior temporal cortices, dorsomedial and inferior frontal cortices. However, correlation
Conclusion
We show that viewing fear and anger behaviors elicit comparable activity increases in the amygdala and temporal cortices as well as in the ventrolateral and the dorsomedial prefrontal cortex. We submit that the activity in these areas may reflect the evaluation of the emotional significance of sensory events associated with an automatic regulative process exerted upon the emotional response elicited in the observer by actions signaling threat. Moreover, we observe specific activity when
Acknowledgments
We are grateful to Charlotte Sinke for subject recruiting and assistance in scanning participants, Lydia Pouga for help in collecting behavioral data, Sven Gijsen and France Maloumian for skilful technical assistance. This work was supported by the Human Frontier Science Program [HFSP-RGP0054/2004-C] and the European Union Research Funding FP6 NEST program [FP6-2005-NEST-Path Imp 043403].
References (100)
- et al.
Social perception from visual cues: role of the STS region
Trends Cogn. Sci.
(2000) - et al.
Topographic organization of projections from the amygdala to the visual cortex in the macaque monkey
Neuroscience
(2003) The roles of orbital frontal cortex in the modulation of antisocial behavior
Brain Cogn.
(2004)- et al.
Response of neurons in the macaque amygdala to complex social stimuli
Behav. Brain Res.
(1990) - et al.
A PET investigation of the attribution of intentions with a nonverbal task
NeuroImage
(2000) - et al.
Movement and mind: a functional imaging study of perception and interpretation of complex intentional movement patterns
NeuroImage
(2000) - et al.
Neural mechanisms subserving the perception of human actions
Trends Cogn. Sci.
(1999) - et al.
A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data
NeuroImage
(2005) - et al.
Age-differential patterns of brain activation during perception of angry faces
Neurosci. Lett.
(2005) - et al.
Beyond threat: amygdala reactivity across multiple expressions of facial affect
NeuroImage
(2006)