تفاوت های فردی در رفتار اجتماعی واکنش آمیگدال به حالات چهره ترس در سندرم ویلیامز را پیش بینی می کند

Abstract Williams syndrome (WS) is a genetic condition often paired with abnormal social functioning and behavior. In particular, those with WS are characterized as being relatively hypersocial, overly emotional/empathic, and socially uninhibited or fearless. In addition, WS is associated with abnormal amygdala structure and function. Very little is known however about the relationship between specific social behaviors and altered amygdala function in WS. This study was designed to compare three models that relate abnormal social behavior with amygdala function in WS (indiscriminate sociability, emotional and empathic sociability and social fearlessness). We used a social behavior assessment procedure (Salk Institute Sociability Questionnaire), functional magnetic resonance imaging and an implicit emotion face processing task to test these models. Our findings provide support for a model of abnormal social fearlessness by showing that in WS, abnormal amygdala response to fear is paired with an increased tendency to approach strangers. Specifically, individuals with WS that exhibited less amygdala response to fearful facial expressions (compared to neutral) also exhibited an increased tendency to approach strangers. These findings contribute to our understanding of social and emotional functioning in neurodevelopmental conditions and provide evidence that in WS, amygdala response to fear modulates social behavior.

Introduction Williams syndrome (WS) is a neurodevelopmental condition caused by a hemizygous microdeletion on chromosome 7q11.23. WS is often paired with a distinctive, abnormal social and emotional phenotype. In particular, those with WS are often described as being relatively hypersocial (Bellugi, Adolphs, Cassady, & Chiles, 1999), overly emotional/empathic (Klein-Tasman & Mervis, 2003) and socially uninhibited or fearless (Gosch and Pankau, 1994 and Meyer-Lindenberg et al., 2006). In terms of the brain, evidence suggests that alterations of the amygdala may in part contribute to the observed abnormal social and emotional phenotype in WS (Haas et al., 2009, Martens et al., 2009 and Meyer-Lindenberg et al., 2005). Recently, brain-imaging studies have begun to explore the neural correlates of individual differences of social behavior in WS. For example, Martens and colleagues (2009) demonstrated that in WS, individual differences in social behaviors, such as with approachability biases, are associated with alterations in amygdala volume. Although studies have indicated that functional abnormalities of the amygdala occur in WS (relative to healthy controls) and that individual differences in social behaviors are associated with amygdala structure in WS ( Martens et al., 2009), very little is known regarding the relationship between individual differences in social behavior and amygdala function in WS. This study was designed to investigate the relationship between social behavior and amygdala function in WS. We used functional Magnetic Resonance Imaging (fMRI) and an implicit emotion face processing task to test three models that relate abnormalities of social behavior with amygdala function in WS ( Fig. 1). One model posits that individuals with WS tend to display abnormal indiscriminate sociability ( Doyle et al., 2004 and Einfeld et al., 1997). This model describes those with WS as being abnormally social and driven towards social interaction independent of emotional valence or arousal and is supported by studies showing that relative to mental and age-matched controls, individuals with WS tend to be rated as generally more “overly-friendly” ( Mervis & Klein-Tasman, 2000), people-oriented and gregarious ( Klein-Tasman & Mervis, 2003). In addition, as compared to controls, those with WS rate facial expressions as more approachable ( Bellugi et al., 1999) and tend to exhibit greater gaze duration towards socially relevant scenes ( Riby & Hancock, 2008) and faces ( Riby & Hancock, 2009). Some studies however, have also found normal approachability towards both negative and positive expressions in WS ( Porter, Coltheart, & Langdon, 2007). Taken together, several studies provide evidence that WS is associated with abnormalities in overt social behavior and attention towards socially relevant information such as facial expressions. We predicted that if functional abnormalities of the amygdala are associated with indiscriminate sociability in WS, then we would observe a positive relationship between individual differences in social approach related behaviors and amygdala response to all facial expressions (fearful, happy and neutral combined compared to scrambled images) ( Fig. 1 and Fig. 2). Model based hypotheses relating social behavior with amygdala function in WS. We ... Fig. 1. Model based hypotheses relating social behavior with amygdala function in WS. We predicted that if abnormal indiscriminate sociability is associated with amygdala function in WS (A), we would observe a relationship between social approach scores and amygdala response to facial expressions (fearful, happy and neutral combined) compared to scrambled images. We predicted that if abnormal emotional and empathic sociability is associated with amygdala function in WS (B), we would observe a relationship between emotional/empathic sociability scores and amygdala response to emotional facial expressions (fear and happy, combined) compared to neutral. We predicted that if abnormal social fearlessness is associated with amygdala function in WS (C), we would observe a relationship between social approach towards strangers and amygdala response to fearful facial expressions compared to neutral. Figure options Examples of stimuli used in the experimental paradigm. Participants were ... Fig. 2. Examples of stimuli used in the experimental paradigm. Participants were presented with photographs of faces conveying fearful, happy, neutral expressions and scrambled images. Participants were instructed to judge if each stimulus was either male, female or scrambled as quickly and as accurately as possible. Behavioral responses were collected within the 2000 ms following the onset of each stimulus. Figure options A second model posits that individuals with WS tend to display abnormal emotional and empathic sociability. This model describes those with WS as being highly emotionally responsive and particularly tuned to the affective states of others and is supported by studies showing that relative to controls, those with WS are rated higher in empathy ( Klein-Tasman & Mervis, 2003), temperamental intensity ( Tomc, Williamson, & Pauli, 1990) and tend to be relatively over-affectionate ( Davies, Udwin, & Howlin, 1998). In addition, as compared to controls, those with WS are more emotionally responsive during social interaction ( Fidler, Hepburn, Most, Philofsky, & Rogers, 2007), rate happy facial expressions as more approachable ( Frigerio et al., 2006) and utilize more emotionally expressive language ( Jones et al., 2000 and Losh et al., 2000). Other studies however, have demonstrated reduced expressive language in WS ( Laws & Bishop, 2004). Together, several studies provide evidence that WS is associated with abnormalities in emotional and empathic processing. We predicted that if functional abnormalities of the amygdala are associated with the tendency to display emotional and empathic sociability in WS, we would observe a positive relationship between individual differences in emotional and empathic approach related behaviors and amygdala response to emotional (both fearful and happy) facial expressions (compared to neutral faces) ( Fig. 1 and Fig. 2). Lastly, a third model posits that individuals with WS are socially uninhibited with strangers and that they display abnormal social fearlessness. This model describes those with WS as being relatively unresponsive to social fear paired with an inappropriate tendency to approach strangers and is supported by studies showing that relative to controls, those with WS exhibit an inability to detect and respect social danger signals ( Meyer-Lindenberg et al., 2006) and are less reserved towards strangers ( Gosch & Pankau, 1994). In addition, as compared to controls, those with WS exhibit lower amygdala response to fearful facial expressions ( Haas et al., 2009 and Meyer-Lindenberg et al., 2005) and are less able to perceive negatively valenced emotional facial and vocal expressions ( Plesa-Skwerer, Faja, Schofield, Verbalis, & Tager-Flusberg, 2006). Taken together, these studies provide evidence that WS is associated with an abnormal social fear response and an increased tendency to approach strangers. We predicted that if functional abnormalities of the amygdala are associated with the tendency to display social fearlessness in WS, we would observe a relationship between individual differences in social approach related behaviors towards strangers and amygdala response to fearful facial expressions (compared to neutral) ( Fig. 1 and Fig. 2).

Results 3.1. Behavioral measures Mean reaction times (981.46 ms) and accuracy rates (72.16%) were calculated for all participants. Reaction time and accuracy rates for all conditions were entered into an ANOVA. There were no statistically significant differences observed between experimental conditions (fearful, happy, neutral and scrambled) in reaction time (F = .47, p = .70) or accuracy (F = .08, p = .97). Furthermore, there were no statistically significant differences observed between emotional (fearful and happy) versus neutral face conditions in reaction time (F = .01, p = .93) or accuracy (F = .06, p = .81). Reaction times during each experimental condition (fearful, happy, neutral or scrambled) were regressed against each of the sociability measures (indiscriminate sociability, emotional and empathic sociability and social fearlessness) independently. No significant relationships between either reaction time or accuracy and the sociability measures (SISQ) were found. 3.2. Indiscriminate sociability and amygdala response to facial expressions Social approach scores were regressed against amygdala response to facial (fearful, happy and neutral combined) expressions compared to scrambled images (Fig. 1A). There were no significant positive or negative relationships between social approach scores and either left (group averaged ROI: R2 = .33, p = .40; standardized ROI: R2 = .27, p = .30) or right (group averaged ROI: R2 = .11, p = .70; standardized ROI: R2 = .04, p = .61) amygdala response to facial expressions compared to scrambled images. No significant relationships were observed when age, sex and handedness were entered as covariates. Furthermore, no significant relationships were observed between social approach scores and either hippocampus and parahippocampal gyrus response to facial expressions. 3.3. Emotional/empathic sociability and amygdala response to emotional facial expressions Emotional/empathic sociability scores were regressed against amygdala response to emotional (both fearful and happy, combined) facial expressions compared to neutral facial expressions (Fig. 1B). There were no significant positive or negative relationships between emotional/empathic sociability scores and either left (group averaged ROI: R2 = .52, p = .13; standardized ROI: R2 = .51, p = .07) or right (group averaged ROI: R2 = .30, p = .44; standardized ROI: R2 = .30, p = .25) amygdala response to emotional facial expressions compared to neutral facial expressions. No significant relationships were observed when age, sex and handedness were entered as covariates. Furthermore, no significant relationships were observed between emotional/empathic sociability scores and either hippocampus and parahippocampal gyrus response to emotional facial expressions. 3.4. Social approach towards strangers and amygdala response to fearful facial expressions Social approach towards strangers scores were regressed against amygdala response to fearful facial expressions compared to neutral facial expressions (Fig. 1C). There was a significant relationship between social approach towards strangers scores and left amygdala response to fearful compared to neutral facial expressions (Fig. 3) (group averaged ROI: MNI coordinates: −20, −6, −6; 47 voxels; t = 6.41; p < .05 corrected; standardized ROI: MNI coordinates: −24, −6, −14; 13 voxels; t = 4.70; p < .05 corrected). In particular, the relationship was such that higher scores on social approach towards strangers (greater tendency to approach strangers) were associated with less amygdala response to fearful facial expressions (compared to neutral) (peak voxel: r = −.89, p < .001). Individual differences in social approach towards strangers scores associated ... Fig. 3. Individual differences in social approach towards strangers scores associated with left amygdala response to fearful, compared to neutral, facial expressions in WS. Social approach towards strangers scores are plotted on the x-axis. Contrast estimates within the left amygdala are plotted on the y-axis. Data were extracted from the peak voxel located at MNI coordinates: −20, −6, −6 (t = 6.41; p < .05 corrected). Cluster found to display a significant correlation is overlaid upon on a coronal slice of a representative WS brain normalized into standard stereotactic space. Figure options The relationship between social approach towards strangers and left amygdala response to fearful compared to neutral facial expressions remained significant after controlling for age, sex and handedness (group averaged ROI: t = 8.16, p < .05 corrected; standardized ROI: t = 4.43, p < .05 corrected). No significant relationships were observed between social approach towards strangers scores and either hippocampus and parahippocampal gyrus response to fearful facial expressions (compared to neutral). In order to examine the functional specificity of the relationship between social approach towards strangers and amygdala activation, social approach towards strangers scores were regressed against amygdala response to facial expressions compared to scrambled images and to emotional facial expressions compared to neutral facial expressions. There were no significant positive or negative relationships between social approach towards strangers scores and either left or right amygdala response to facial expressions compared to scrambled images or to emotional (both fearful and happy, combined) facial expressions compared to neutral facial expressions. Controlling for age, sex and handedness did not change this finding.