Our work has been important in suggesting that GAD and GSP are indeed associated with different pathophysiologies. In recent work, we presented patients with GSP, patients with GAD, and no pathology individuals with angry, fearful, and neutral facial expression stimuli. Our data clearly suggested that the neural circuitry dysfunction in GSP and GAD differ. Specifically, we found that patients with GSP showed significantly increased activation to fearful relative to neutral expressions in several regions, including the amygdala. In contrast, patients with GAD showed significantly reduced activation to fearful relative to neutral faces compared to healthy individuals and patients with GSP but this was coupled with anomalously and significantly increased responses in a lateral region of prefrontal cortex. Importantly, a secondary analysis conducted with the subgroup of patients with comorbid GAD/GSP indicated that they presented with the pathologies associated with the GAD, but not the GSP. That is, these preliminary results strongly suggest the importance of a systematic examination and comparison of GSP, GAD, and comorbid GAD/GSP for a fuller disorder-specific understanding and development of new treatment targets. This paper was published in The American Journal of Psychiatry in 2008. In other recent work, we examined the neural responses to receipt of praise or criticism in GSP and GAD within a newly design novel statement paradigm. Participants were presented with positive, negative, and neutral statements (e.g., You are beautiful/ ugly/ human) that could be either about highly relevant and about themselves or less relevant about somebody else (e.g., He is beautiful). There were again two major results. First, there again was a clear double dissociation between the individuals with GSP and GAD with suggestions of a divergent subcortical response to social stimuli in the two populations. Second, the results indicated an important role for self-relevance and medial prefrontal cortex (MPFC), mediating self-relevance, in GSP. Thus, on this task patients with GSP showed significantly increased neural responses in MPFC and amygdala to negative comments about the self (criticism). However, the groups did not differ significantly in their neural responses to any of the other statement categories: neutral or positive comments about the self, or negative, neutral, or positive comments about others. This paper was published in the Archives of General Psychiatry in 2008. In subsequent work, we considered whether the increased neural responses in GSP to negative comments about the self reflects (a) increased agreement with those comments, and/or (b) increased concern or mentalizing about others negative perceptions. Participants were presented again with positive, negative, and neutral statements (e.g., You are beautiful/ ugly/ human) but this time they were always self relevant, and instead we manipulated whether the comments came from somebody else (e.g., You are beautiful), or was internally generated (e.g., I am beautiful). The results further underlined the critical role of MPFC in GSP pathology, and again demonstrated a meaningful effect of psychological context on neural-circuitry hyperactivity in GSP. Thus, on this modified comment task, patients with GSP showed significantly increased activation in MPFC to comments that came from somebody else (i.e., You comments), however, they did not show increased neural responses to comments that originated within the self (i.e., I comments). This paper is currently being prepared for publication. The importance of MPFC and contextual information in GSP was further demonstrated in other recent work, where we examined the neural response to social situations that could either involve an intentional transgression (e.g., Joanna does not like her food and spits it out), or an unintentional (embarrassing) transgression (e.g., Joanna chokes on her food and coughs it up). People with no social anxiety typically show increased MPFC responses to the intentional social transgressions, presumably because those acts typically involve more severe consequences (i.e., using the Joanna example, spitting out the food will anger the host, but coughing up the food will probably not have consequences). However, in our study patients with GSP showed the opposite pattern with increased MPFC responses to unintentional transgressions, presumably because (1) they care more about causing embarrassment, than anger, and (2) the intentional stories are not relevant to them because they are extremely unlikely to intentionally commit a social transgression. This highly context dependent response in GSP helps constrain existing models of the disorder and may thus guide future therapeutic formulations in the treatment of the disorder. This paper is currently being prepared for publication. In other recent work, we examined stimulus-reinforcement based decision making using a task that has previously been demonstrated to recruit the amygdala and vmPFC in healthy individuals. The data on this task also indicated a dissociation between GAD and GSP. Thus, only patients with GAD showed impairment on this task (relative to both patients with GSP and healthy individuals who did not significantly differ). Moreover, the patients with comorbid GAD/GSP also showed significant impairment on the task (relative to both patients with GSP and healthy individuals). However, the patients with comorbid GAD/GSP and those with GAD alone did not show significant differences in impairment. Thus, this work has underlined the importance of determining the separable pathophysiologies associated with the individual conditions. This paper was published in Psychological Medicine in 2009.
