|Home | About | Journals | Submit | Contact Us | Français|
Aberrant subcortical-prefrontal connectivity may contribute to insula hyper-reactivity to threat in generalized social anxiety disorder (gSAD). A novel PsychoPhysiological Interaction (PPI) analysis was used to examine functional ‘coupling’ between the insula and prefrontal cortex in gSAD patients and healthy controls (HCs). During fMRI, 29 gSAD and 26 HC volunteers performed an Emotional Face Matching Task, involving the processing of fear, angry, and happy expressions. As expected, compared with HCs, gSAD patients exhibited greater bilateral anterior insula (aINS) reactivity for fear vs. happy faces; this group difference was less robust for angry vs. happy faces. PPI of insula connectivity when processing fearful faces revealed the gSAD group had less right aINS-dorsal anterior cingulate coupling compared to HCs. Findings indicate that aINS hyper-reactivity for fear faces in gSAD, compared to controls, involves reduced connectivity with a prefrontal region implicated in cognitive control and emotion regulation.
Generalized social anxiety disorder (gSAD) is a prevalent, debilitating disorder marked by social-evaluative fears (Kessler et al., 2005). Symptoms include negative beliefs (Heimberg, 1994, 2002) and autonomic changes in anticipation of, or during, feared situations (Cornwell et al., 2006; Davidson, 2000; McTeague et al., 2009), reflecting a heightened threat processing system. It has been generally established that the amygdala, a crucial region in fear circuitry (LeDoux, 2000), plays a role in gSAD and other types of anxiety (Freitas-Ferrari et al., 2010; Phan and Klumpp, 2010). Less often highlighted is the insula, shown to be hyper-reactive to salient stimuli in gSAD (Amir et al., 2005; Etkin and Wager, 2007; Evans et al., 2008; Shah et al., 2009; Straube et al., 2004, 2005) and anxiety-prone individuals (Simmons et al., 2006; Stein et al., 2007; Wright et al., 2003).
Broadly, the insula is proposed to link internal and external information to produce an awareness of mind-body experiences that aid in maintaining a context relevant homeostatic state (Craig, 2009; Jones et al., 2010). Notably, interoceptive awareness involves the anterior insula (aINS) (Craig, 2009), which is modulated via reciprocal connections with the prefrontal cortex (PFC) (Augustine, 1996; Craig, 2002, 2009; Mesulam and Mufson, 1982; Nelson et al., 2010; Seeley et al., 2007). Therefore, regulation of anxious responsiveness (Critchley, 2009) may involve intact aINS-PFC interactions, which if aberrant, has implications for anxiety disorders (Paulus and Stein, 2006, 2010). For example, patients with gSAD may experience fluctuations in bodily state (e.g., heart rate variability) when meeting someone new and in doing so engage in self-referential somatic-cognitive interpretations. Negative interpretations of internal state driven by exaggerated insula responsiveness and/or inefficient prefrontal regulation could lead to over-arching threat schemas that govern over-estimations of threat and associated anxiety (Beck and Clark, 1997). In social anxiety, evidence of frontal hypo-activation to threat when control functions are recruited and hyper-reactivity concerning threat perception (Freitas-Ferrari, et al., 2010) indicate the sensitivity of a task to engage more dorsal than ventral regions or vice versa effects direction of the observed frontal abnormality. Findings of increased dorsal and ventral ACC and medial PFC activity during threat processing suggest expression, appraisal and/or emotion regulation functions, sub-served by dorsal-caudal and ventral-rostral areas, respectively (Etkin et al., 2011) enhance threat processing in social anxiety.
We used functional connectivity based on task-related aINS activation to threat in patients with gSAD as a step towards elucidating the role of aINS-prefrontal interactions in the pathophysiology of gSAD. Based on evidence of deficient subcortical-PFC relationships in social anxiety (Danti et al., 2010; Phan et al., 2009), we hypothesized that patients would have enhanced insula reactivity and less insula-(pre)frontal connectivity when processing threat faces compared to healthy controls.
Twenty-nine gSAD patients and 26 HC participated. The Structured Clinical Interview for DSM-IV (First et al., 1995) was used for diagnosis and the Liebowitz Social Anxiety Scale (Liebowitz, 1987), as a complementary measure of social anxiety pervasiveness. See supplemental material for subject characteristics. All participants provided written informed consent, as approved by the local Institutional Review Board. See Table 1 for group characteristics.
During fMRI participants performed a modified Emotional Face Matching Task (EFMT) involving a trio of faces from a validated set (Gur et al., 2002), in which was one of the two faces (bottom) that expressed the same emotion as the target face (top) was selected. Matching face blocks were interspersed with “baseline” blocks, in which participants matched geometric shapes (circles, rectangles, or triangles) similar to instructions above. See supplemental material for more details on the EFMT.
Using a 3T GE Signa System (General Electric; Milwaukee, Wisconsin, USA), whole-brain functional images (i.e., blood oxygenated level-dependent [BOLD]) were acquired from 30 axial, 5-mm-thick slices using a standard T2*-sensitive gradient echo reverse spiral acquisition sequence (repetition time, 2000 ms; echo time, 25 ms; 64 × 64 matrix; 24 cm field of view; flip angle, 77). A high-resolution, T1-weighted volumetric anatomical scan (3-dimensional magnetization-prepared rapid gradient echo) was also acquired for anatomical localization. High quality and scan stability with minimum motion corrections was set at < 2 mm displacement in any one direction. See supplemental material for details of preprocessing and first-level analysis methods.
To test a priori hypothesis, a region of interest (ROI) approach localized to anatomically-based left and right insula masks (Tzourio-Mazoyer et al., 2002; Walter et al., 2003) with the anterior portion demarcated as y-axis=0 and forward was used to examine group differences (see Figure 1). Significance was set at p<0.05, False Discovery Rate (FDR) corrected for multiple comparisons using a small volume correction (Nichols et al., 2005; Worsley et al., 1996). Each participant parameter estimates of activation were extracted from the ROI (β weights, arbitrary units [a.u.]) and averaged across all voxels to examine the direction and variance in gSAD and HC separately. For completeness, a whole brain analysis was conducted; significance was set at p < 0.005, uncorrected for multiple comparisons. Regarding functional connectivity, the time series from significant voxels were extracted within this same left and right aINS ROI using conventional steps of the PsychoPhysiological Interaction (PPI) analysis as implemented in SPM5 to measure aINS-prefrontal connectivity; see supplemental material for details.
For accuracy and reaction times for accurate trials, there was a significant main effect of Emotion but no main effect of Group or Emotion × Group interaction. See supplemental materials for these results.
Voxel-wise between-groups t-test revealed gSAD patients exhibited greater activation to fear (>happy) faces than HC in the left aINS (peak MNI coordinates: [−40, 20, −4], k=294, Z-score=3.90; FDRsvc p<0.003) and right aINS [(38, 24, −4), k=37, Z-score=2.83; FDRsvc p<0.005) (Fig. 1A). Extracted β-weights of activation (BOLD response [a.u.]) from the aINS ROI revealed an exaggerated aINS response to fear (>happy) faces in gSAD, whereas HC showed no differences in aINS to fear (>happy) (Fig. 1B). See Table 2 for whole-brain results beyond insula. Regarding angry (>happy), group differences for left [(−40, 20, −4), k=87, Z-score=2.97] and right [(38, 26, −2), k=80, Z-score=2.19] insula only emerged at liberal significance threshold, puncorrected <0.05; therefore, subsequent PPI analysis was restricted to fear perception.
The only area showing a significant group difference in connectivity to the right aINS was the dorsal ACC (dACC) ([4, 32, 20], k=1339, Z-score=4.40) (Fig. 2A). Group differences in left aINS connectivity emerged in the dACC ([4, 38, 22], k=24, Z-score=2.89), albeit below our predetermined threshold for significance. Extracted β-weights of connectivity (parameter estimates, a.u.) showed HC had robust positive coupling whereas gSAD participants demonstrated an inverse coupling pattern (Fig. 2B).
As expected, gSAD versus HCs exhibited greater bilateral aINS reactivity for fear (>happy) faces. Similar results occurred for angry (>happy) though only at a more liberal significance thresholding procedure (Lieberman and Cunningham, 2009).
Fear-related connectivity analyses revealed gSAD had less right aINS-dACC connectivity than HCs. Right insula is a proposed interface between bodily arousal and its representation as a subjective feeling (Critchley et al., 2004). Significant connectivity results for right aINS support the notion that anxiety disorders are a consequence of altered interoceptive state (Paulus & Stein, 2006, 2010). The strong link between aINS and dACC, which help give rise to an awareness of mind-body interactions (Craig, 2009; Critchley, et al., 2004; Dosenbach et al., 2007), confirm that aINS and ACC interact when processing threat. Based on deficient subcortical-PFC relationships in social anxiety (Danti, et al., 2010; Phan, et al., 2009), we hypothesize the aINS exaggerated reactivity to threat in gSAD is due to deficiencies in cognitive control over threat, threat appraisal, and/or regulation of threat signals, functions of dACC (Bush et al., 2000; Etkin, et al., 2011; Ochsner and Gross, 2005). In support, healthy controls revealed a non-significant aINS response to fear faces yet robust aINS-dACC coupling.
Study limitations include PPI analysis, which does not permit interpretation of direction or causality of aINS-ACC associations and use of aINS as the only seed, which precludes identifying other brain networks relevant to gSAD pathophysiology outside of aINS.
In conclusion, gSAD is associated with an exaggerated aINS response, and altered aINS-dACC connectivity to salient threat suggests deficient subcortical-PFC coupling. Findings elucidate the role of insula reactivity and insula-ACC connectivity in the pathophysiology of gSAD.
This work was supported by a grant from the National Institutes of Health, National Institute of Mental Health Patient-Oriented Career Development Award K23MH076198 (KLP) and by a grant from the National Center for Research Resources (NCRR) UL1RR024 986 (HK). The content is solely the responsibility of the authors and does not necessarily represent the official views of NCRR or the National Institutes of Health.
All authors report no biomedical financial interests or potential conflicts of interest.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.