Date of Award

May 2017

Degree Type


Degree Name

Doctor of Philosophy



First Advisor

Christine L. Larson

Committee Members

Adam S. Greenberg, Deborah E. Hannula, Fred J. Helmstetter, Krista M. Lisdahl


Amygdala, Bed Nucleus of the Stria Terminalis, Connectivity, Threat


An influential model of the extended amygdala defines fear as the immediate response to phasic threat and anxiety as the prolonged response to unpredictable or sustained threat (Davis, Walker, Miles & Grillon, 2010). This model proposes that in response to unpredictable threat, the centromedial amygdala (CeA) activates the bed nucleus of the stria terminalis (BNST), which coordinates the anxiety response, and, in turn, inhibits the CeA. Connectivity between the BNST and both the basolateral amygdala (BLA) and hippocampus may also play an important role in the coordination of the anxiety response (Davis et al., 2010; Herman et al., 2003; Zhu, Umegaki, Suzuki, Miura & Iguchi, 2001). However, there is a dearth of human research investigating whether state anxiety is accompanied by increased connectivity between the BNST and CeA, BLA and hippocampus. To test whether sustained threat elicits increases connectivity between the BNST and these areas, I monitored participants’ resting brain activity via high resolution 7 tesla fMRI during two five minute resting state scans, one while under threat of unpredictable shock and one while safe. I predicted that each of these areas would exhibit greater connectivity with the BNST during periods of threat vs. safety. To test whether BNST connectivity during periods of threat is altered in anxiety prone individuals, I collected self-reported behavioral inhibition. I predicted that greater behavioral inhibition would predict increased connectivity during periods of threat (vs. safety) between the BNST and the CeA, BLA and hippocampus. I also tested whether connectivity in these areas changes over time following the onset of threat by examining connectivity for three time windows, corresponding roughly to the first, second and fifth minute following threat onset. I predicted positive threat vs. safe BNST-CeA connectivity during the first time window, negative threat vs. safe BNST-CeA connectivity during the second time window, and no difference in BNST-CeA connectivity during the final time window. I found a marginally significant trend toward greater BNST-BLA connectivity during threat vs. safety. I found no evidence for increased BNST-hippocampus connectivity during threat, or that BNST connectivity with either the BLA or hippocampus is modulated by behavioral inhibition. Threat condition, behavioral inhibition, and time window interacted to affect BNST-CeA connectivity, although a lack of significant follow-up tests makes interpreting this interaction challenging. Further research is needed to characterize how individual differences alter the time course of BNST-CeA connectivity during conditions of threat and safety, and the conditions under which threat may elicit BNST connectivity with the hippocampus and BLA.