Neural correlates of emotion processing during observed self-face recognition in individuals with autism spectrum disorders
Introduction
In human society, the ability to interact with others in a socially appropriate manner is essential. In addition to basic emotions, such as happiness, fear, and anger, humans experience higher-order self-conscious emotions (e.g., coyness, shyness, pride, embarrassment, shame, and guilt). Unlike basic emotions, self-conscious emotions tend to arise through relationships with others and serve important interpersonal functions (Miller & Leary, 1992; Tangney, 1999; Lewis, 2000). For instance, when individuals are exposed to self-images via mirrors, photographs, or videos, they sometimes experience early forms of self-conscious emotions (e.g., coyness or shyness). This type of emotion emerges at around 18–24 months of age, and only after self-recognition appears (Lewis, Sullivan, Stanger, & Weiss, 1989). Furthermore, when individuals are exposed to visual self-images that deviate substantially from the individual’s ideals or standards, they can experience more complex negative self-conscious emotions (e.g., embarrassment) (Duval & Wicklund, 1972; Carver & Scheier, 1981; Carver & Scheier, 1998). Such negative emotions emerge at around 3–4 years of age, when the child has internalized rules or standards for self-evaluation. Therefore, this class of self-conscious emotions are called “self-evaluative emotions”, and they can be viewed as part of an alarm system that detects deviations of behaviors and attitudes relative to social standards. This system could play an important role in guiding appropriate social conduct.
Previously, we demonstrated that the anterior insula (AI) and the anterior cingulate cortex (ACC) are involved in the experience of embarrassment (Morita et al., 2008; Morita et al., 2012). These areas were more active when participants viewed self-face images, including those that would be expected to elicit feelings of embarrassment, than when they viewed images of others’ faces. These brain regions are co-activated when subjects experience a range of basic emotions, including disgust and fear, as well as social emotions including romantic love, injustice, and social exclusion (Blood & Zatorre, 2001; Eisenberger, Lieberman, & Williams, 2003; Wicker et al., 2003; Bartels & Zeki, 2004; Takahashi et al., 2008; Onoda et al., 2010; Moor et al., 2012). In addition, these regions are also co-activated in response to salient stimuli or events that do not necessarily elicit a specific emotional feeling (Craig, 2002). Therefore, the ACC and AI are thought to be components of a “salience network” that functions to identify the most relevant among several internal and extra-personal stimuli in order to guide appropriate behavior. In this framework, the AI serves as an integral hub in mediating dynamic interactions between other large-scale brain networks: the central executive and default mode networks (Seeley et al., 2007; Menon & Uddin, 2010).
Recently, we also obtained evidence that social situations in which participants are observed by others modulate activation patterns in the AI and ACC in distinct manners (Morita et al., 2014). In that study, we showed that individuals view self-face images while being observed by others, they experience a stronger feeling of embarrassment than when viewing the same images in the absence of an observer. Individual increases in the subjective feeling of embarrassment are positively correlated with individual increases in self-related activity (self vs. others) in the right AI, but not in the caudal part of the ACC. According to the Craig’s model of integration across the insula cortex, diverse information, including homeostatic, environmental, hedonic, motivational, social, and cognitive activity, is integrated in a posterior-to-anterior direction to produce subjective experiences representing the sentient self at a particular moment in time (Craig, 2009). Considering this view together, it is suggested that the right AI plays a crucial role in creating the subjective experience of embarrassment.
On the other hand, we also found that being observed increased functional connectivity between the caudal ACC and medial prefrontal cortex (MPFC) when viewing self-face images. The MPFC is consistently activated by self-reflective processing, in which participants are required to think about their own mental or inner states (e.g., emotions or personality traits) (Northoff et al., 2006; Murray, Schaer, & Debbané, 2012). In addition, the MPFC is also activated by tasks that require participants to infer others’ mental states or take a third-person perspective (i.e., mentalizing) (D’Argembeau et al., 2007; Gallagher & Frith, 2003;;Frith and Frith, 2006). Several recent studies suggested that the MPFC was involved in inference of the more complex mental states of others. For example, the MPFC is recruited when thinking about how another person would appraise us (Ochsner et al., 2005;Amodio & Frith, 2006; D’Argembeau et al., 2007; Frith & Frith, 2008; Izuma, Saito, & Sadato, 2008, Izuma, Saito, & Sadato, 2010a; Sugiura, Sassa, & Jeong, 2012). All these evidence considered, in our previous study, we proposed that being observed during self-face recognition would increase accessing information about the self that is reflected in the eyes or minds of others. This would lead to an increase in functional connectivity between the caudal ACC and MPFC. That is, unlike the functional role of the AI, the caudal ACC seems to serve as a hub, integrating information about the reflective self that is used for self-evaluative processing.
Autism spectrum disorder (ASD) is characterized by persistent deficits in social communication and social interaction across multiple contexts, in conjunction with restricted, repetitive patterns of behavior, interests, or activities (Diagnostic and Statistical Manual of Mental Disorders, DSM-5, American Psychiatric Association, 2013). A great deal of research has demonstrated that individuals with ASD exhibit deficits in face perception (eye gaze or facial expression) at both the behavioral and neural levels (Dalton et al., 2005; Dawson, Webb, & McPartland, 2005; Pierce, Müller, Ambrose, Allen, & Courchesne, 2001). In contrast to the abundance of evidence regarding atypical face-processing, however, few studies have suggested that self-face processing is atypical in ASD. Indeed, the ability of autistic children to discriminate between their own faces and the faces of others appears to be intact (Akagi, 2003; Dawson & Mckissick, 1984; Reddy, Williams, Costantini, & Lan, 2010).
However, autistic children do differ from control children in their responses to self-images reflected in a mirror. Typically, developing children aged around 2 years exhibit self-conscious emotions or self-conscious behaviors (e.g., a coy smile) in response to a self-reflection. This early form of self-conscious is thought to occur when one is the object of others’ attention (Lewis et al., 1989). By contrast, autistic children exhibited relatively neutral emotions in this context (Akagi, 2003; Dawson & Mckissick, 1984; Reddy et al., 2010). Consistent with these findings, we have obtained evidence that adults with ASD respond in an emotionally atypical manner to self-face images. In that study, we found that cognitive evaluation (self-evaluation) of self-face images and emotional responses (i.e., self-conscious emotions) were less coupled in adults with ASD, in parallel with reduced activity of the right insula (Morita et al., 2012). Given that self-conscious emotions arise through social relationships between self and others, atypical self-conscious emotions in ASD patients may be related to their reduced responsiveness to being observed. Indeed, another group has reported that autistic adults exhibit weaker reactions to being observed during social behaviors. When asked to make real charitable donations in the presence or absence of observers, neurotypical controls donated significantly more in the presence of the observers than when alone, whereas the amount of donation by adults with high-functioning autism was not influenced by the presence of observers (Izuma, Matsumoto, Camerer, & Adolphs, 2011). Therefore, we hypothesized that being observed would have little impact on the emotional response associated with self-face recognition in individuals with ASD.
To test this hypothesis, we employed a modified functional magnetic resonance imaging (fMRI) paradigm in which participants rated the extent of embarrassment elicited by self-face images and those of others in the presence or absence of observers (Morita et al., 2014). Our previous study employed a dual MRI system with an interaction system that allowed two participants to be observed mutually and equally by each other in real time. By contrast, this study employed a single MRI system with one participant inside and an observer outside, similar to the system used in another previous study (Izuma, Saito, & Sadato, 2010b). For half of the sessions, participants were told that they would be able to see the face of their observer via a live video link, indicating that the observer was watching the face images that the participant was viewing. On the other hand, for the other half of the sessions, participants were told that they would see an empty chair, indicating that the observer was not watching the face images.
Using this experimental setting, we initially confirmed that neurotypical individuals experienced increased embarrassment when viewing self-face images in the presence of an observer. Next, we tested whether the extent of subjective embarrassment and the neural substrates for embarrassment would be modulated by observation in individuals with ASD. For standard analyses and psychophysiological interaction (PPI) analyses, we defined regions of interest (ROIs) based on peak coordinates in emotion-related regions (caudal ACC and AI), where activation patterns during self-face processing were shown to be modulated by observation in an independent study conducted on neurotypical individuals (Morita et al., 2014).
Section snippets
Participants
Fourteen males with high-functioning ASD (mean age ± standard deviation [SD] = 24.5 ± 6.7 years) were recruited at the Department of Neuropsychiatry of the University of Fukui Hospital or the Department of Psychiatry and Neurobiology of the Kanazawa University Hospital in Japan (Table 1). The authors (Hirotaka K. and Toshio M.) diagnosed the participants based on the classifications described in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) (American Psychiatric Association, 2000
Behavioural data
The average public self-consciousness scale scores were 41.4 ± 7.0 in the control group and 40.4 ± 9.0 in the ASD group. The average private self-consciousness scale scores were 33.4 ± 5.5 in the control group and 36.0 ± 6.1 in the ASD group. There were no significant differences between groups in either of the subscales (public, t (30) = 0.34, p = 0.75; private, t (30) = −1.24, p = 0.22). Fig. 2 shows the range of embarrassment ratings measured during the fMRI session. A three-way ANOVA, face type (SELF,
Effect of being observed on emotional response to self-face images
First, we confirmed that the presence of an observer led to an increase in the subjective feeling of embarrassment provoked by self-face images in neurotypical individuals, which is consistent with our previous findings (Morita et al., 2014). In this study, in addition to the embarrassment ratings, the coupling between embarrassment and photogenicity ratings for self-face images was also modulated by the presence of an observer. In the absence of an observer, neurotypical individuals
Conclusions
In neurotypical individuals, being observed by others had an impact on emotional processing associated with self-face recognition. First, the presence of an observer enhanced subjective feeling of embarrassment, and also enhanced the coupling strength between emotional response (embarrassment) and cognitive evaluations (photogenicity) of the self-images. In addition, the extent of embarrassment was closely correlated with activity in the right AI. This suggests that the right AI is involved in
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
TM (first author) was involved in conducting the experiment, analyzing and interpreting the data, and drafting the article. HK and MI were involved in recruiting the participants, diagnosing the participants with ASD, and conducting the experiment. DNS and TF were involved in recruiting the participants and conducting the experiment. TM (sixth author) was involved in recruiting the participants and diagnosing the participants with ASD. KI was involved in conducting the experiment. HO and RK
Acknowledgments
We wish to express our sincere appreciation to the participants and their families, who generously and courageously participated in this research. We thank the staff of the National Institute for Physiological Sciences and the University of Fukui for their support. This work was funded in part by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (20119001, 23119725, 21220005, 25293248, 15H01846), and the Takeda Science Foundation. Part of this study was
References (89)
- et al.
The neural correlates of maternal and romantic love
NeuroImage
(2004) fMRI studies of temporal attention: allocating attention within, or towards time
Cognitive Brain Research
(2004)- et al.
The functional anatomy of inspection time: an event-related fMRI study
NeuroImage
(2004) - et al.
Here I am: the cortical correlates of visual self-recognition
Brain Research
(2007) - et al.
Functional brain correlates of social and nonsocial processes in autism spectrum disorders: an activation likelihood estimation meta-analysis
Biological Psychiatry
(2009) - et al.
Experiencing oneself vs another Person as being the cause of an action: the neural correlates of the experience of agency
Neuroimage
(2002) - et al.
Psychophysiological and modulatory interactions in neuroimaging
NeuroImage
(1997) - et al.
Analysis of fMRI time-series revisited
NeuroImage
(1995) - et al.
Classical and Bayesian inference in neuroimaging: theory
NeuroImage
(2002) - et al.
Stochastic designs in event-related fMRI
NeuroImage
(1999)