Testing the Triple Network Hypothesis in a Largescale Biopsychological Sample: Neural Responses to Psychosocial Stress

To date, several attempts have been made to unravel the mechanisms underlying the central nervous systems response to acute stress. Endocrine, affective/behavioral, heart rate, and neural responses have been measured most frequently in animals and humans. Although findings derived from animal studies have contributed extensively to our knowledge on specific stress response patterns of various entities and their interactions (Herman et al., 2005, 2016; Hermans et al., 2014), these findings are not necessarily transferable to humans. In this context, the hypothalamic-pituitary-adrenal (HPA) axis is of special importance as this neuroendocrine system has proven to be involved in adaptive and maladaptive outcomes modulated by acute and chronic stress exposure (De Kloet et al., 2005). Hence, direct and indirect projections of limbic structures, such as the hippocampus, amygdala, and medial prefrontal cortex (mPFC) have been discussed to interact with the HPA axis. In animals, the hippocampus and mPFC are thought to have inhibiting effects on HPA axis responses, whereas the amygdala is believed to act excitatory on the HPA axis. In humans, on the contrary, these modes of action have not yet been clearly understood as several studies exist reporting opposite effects of the mentioned structures under stress exposure (Berretz et al., 2021; Herman et al., 2005; Hermans et al., 2014; Jankord & Herman, 2008; Noack et al., 2019). Thus, although studies on stress have provided a wealth of data delineating the effects of acute and chronic stress, much remains to be done to fully understand how the human brain processes stress and how pathology or resilience are developed in the face of adversity (Lupien et al., 2009). With the advent of accessible and affordable neuroimaging techniques, it was a logical next step in human stress research to scrutinize the interaction between responses of the brain and HPA axis regulation. This research requires paradigms that are suitable for scanner environments and that induce both, significant neural and HPA axis responses. After the successful establishment of the Montreal Imaging Stress Task (MIST) (Pruessner et al., 2008), recently the ScanSTRESS paradigm was developed (Lederbogen et al., 2011; Streit et al., 2014). ScanSTRESS is based on the most widely used psychosocial stress protocol – the Trier Social Stress Test (TSST) (Kirschbaum et al., 1993) – prompting the subject to perform mental arithmetic and mental rotation tasks under time pressure, while being monitored by an investigator panel giving negative feedback. Thus, ScanSTRESS predominantly aims at inducing social-evaluative threat, negative feedback, and uncontrollability to provoke stress on a psychological level (Dickerson & Kemeny, 2004; Kirschbaum et al., 1993). ScanSTRESS and the MIST were employed successfully in several studies and significant mean cortisol and heart rate increases as well as elevated self-reported stress responses were found. However, especially HPA axis responses were not always fully consistent as expected and neural activation patterns did show variability across studies (Berretz et al., 2021; Noack et al., 2019). Key pathways involved in stress processing that have been clearly identified in animal models are the salience network (SN) and the central executive network (CEN) (Hermans et al., 2014). In humans, especially structures within the SN, like the amygdala, insula, and dorsal anterior cingulate cortex (ACC), have proven their involvement when facing stressful events (Akdeniz et al., 2014; Henckens et al., 2012; Khalili-Mahani et al., 2010; Pruessner et al., 2008). However, several studies in humans failed to find distinct CEN activation in response to psychosocial stress (Akdeniz et al., 2014; Boehringer et al., 2015; Khalili-Mahani et al., 2010; Lord et al., 2012; Pruessner et al., 2008). This difference could partly be explained by the stress-eliciting components of the paradigms; negative feedback, as a typical component in human psychosocial stress paradigms (MIST and ScanSTRESS), may not result in a pronounced CEN activation (Van Oort et al., 2017). Instead, key regions of the default mode network (DMN) which have been related to self-evaluative processing (Mak et al., 2017) are addressed under psychosocial stress, namely the mPFC, posterior cingulate cortex (PCC), and angular gyrus (Quaedflieg et al., 2015; Vaisvaser et al., 2013, 2016; Van Oort et al., 2017; Veer et al., 2012). Remarkably, other regions involved in stress processing, like the parahippocampal gyrus and hippocampus (Jankord & Herman, 2008), are strongly related to the DMN. In summary, it remains unclear which neural structures of the human brain - singular or networks - are responsive to acute psychosocial stress. To date, it has not been investigated if the same structures are responsive in different samples. We therefore want to examine if reactions in the SN and DMN, but less so in the CEN, can be detected in different samples exposed to ScanSTRESS following the hypothesis of Van Oort et al. (2017). In this way, we want to test whether the postulated so-called triple stress network (formed by the SN, CEN, and DMN; Menon, 2011) can be delineated or reduced to two networks (SN and DMN) with regard to psychosocial stress processing. Our study therefore has two aims: First, we will review the existing literature on psychosocial stress induction in fMRI environments (ScanSTRESS and MIST studies) in terms of cortisol, affective, heart rate, and neural responses resulting in a systematic review. Second, we want to test the triple network hypothesis on the basis of a large-scale biopsychological sample exposed to the ScanSTRESS paradigm incorporated in a mega-analysis. Hence, based on data from approximately 500 female and male subjects originating from (partly) already published studies, we will investigate if psychosocial stress components lead to responses in the SN and DMN at the expense of the CEN. We want to explore this by looking at task-based activations and deactivations on the one hand and psychophysiological interactions (PPIs; O’Reilly et al., 2012) on the other. PPI addresses the task-dependent analysis of functional connectivity. Therefore, a PPI analysis consists of determining which voxels in the brain strengthen their relationship to a seed region of interest in a given context, e.g., during acute stress exposure. In the context of psychosocial stress processing, only one study to date has examined the extent to which the connectivity of the triple network changes during acute stress exposure, in a sample that included only adolescents (Corr et al., 2022). Hence, we aim to assess the connectivity of the triple network based on five regions of interest (ROIs) representing the respective network as a hub. The SN hub is the left and right frontoinsular cortex (FIC), the DMN hub is represented by the posterior cingulate cortex (PCC), and the CEN hubs by the left and right dorsolateral prefrontal cortex (dlPFC). Because of the continued inconsistency of findings and the fact that this will be the first study to examine a large-scale biopsychological sample subjected to such an elaborate design of stress induction in an fMRI environment, our hypotheses – which are presented below – are undirected.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.