Neural circuitry changes associated with increasing self-efficacy in Posttraumatic Stress Disorder

Abstract

Cognitive models suggest that posttraumtic stress disorder (PTSD) is maintained, in part, as a result of an individual's maladaptive beliefs about one's ability to cope with current and future stress. These models are consistent with considerable findings showing a link between low levels of self-efficacy and PTSD. A growing body of work has demonstrated that perceptions of self-efficacy can be enhanced experimentally in healthy subjects and participants with PTSD, and increasing levels of self-efficacy improves performance on cognitive, affective, and problem-solving tasks. This study aimed to determine whether increasing perceptions of self-efficacy in participants with PTSD would be associated with changes in neural processing. Combat veterans (N = 34) with PTSD were randomized to either a high self-efficacy (HSE) induction, in which they were asked to recall memories associated with successful coping, or a control condition before undergoing resting state fMRI scanning. Two global network measures in four neural circuits were examined. Participants in the HSE condition showed greater right-lateralized path length and decreased right-lateralized connectivity in the emotional regulation and executive function circuit. In addition, area under receiver operating characteristics curve (AUC) analyses found that average connectivity (.71) and path length (.70) moderately predicted HSE group membership. These findings provide further support for the importance of enhancing perceived control in PTSD, and doing so may engage neural targets that could guide the development of novel interventions.

Introduction

In the wake of a traumatic event, a significant minority of individuals will develop Posttraumatic Stress Disorder (PTSD), a psychiatric disorder characterized by intrusive memories, avoidance of stimuli associated with the event, negative alterations in cognition and mood, and disturbances in arousal (APA, DSM-5, 2013). Estimates of lifetime PTSD in the United States range from 6.8% to 8.7%, (Kessler et al., 2005). Rates of PTSD among those serving in military operations often appear to be higher, as a recent meta–analysis reported that the estimated lifetime prevalence among those who served in Operation Enduring Freedom (OEF) and Operation Iraqi Freedom was 23% (Fulton et al., 2015).

Over the past two and a half decades, considerable progress has been made in characterizing the neural, genetic, physiological, molecular, endocrinological, and immunological alterations associated with PTSD (Heim and Nemeroff, 2009; Pitman et al., 2012; Shalev et al., 2017; Yehuda et al., 2015). Such work has led to numerous studies examining brain regions implicated in the pathogenesis of the disorder (Etkin, 2010; Fonzo et al., 2017; Patel et al., 2012; Rauch et al., 2006; Shalev et al., 2017). Recent work aimed at developing neurocircuitry models of PTSD through the integration of existing research has identified four circuits that may underlie the core symptoms clusters of PTSD (Shalev et al., 2017).

Specifically, Shalev et al. (2017) proposed that interconnected brain structures within these four circuits underlie fear learning, exaggerated threat detection, diminished emotional regulation and executive function, and altered contextual processing (see Table 1 for list of brain structures and neural circuits). Although PTSD has been conceptualized as the result of alterations in a distributed network of neural structures throughout the whole brain, Shalev, Liberzon and Marmar's (2017) neurocircuitry models offer a novel framework for examining the inter-relations among brain structures in order to further characterize the disorder and identify therapeutic targets. For example, fear learning has been associated with abnormalities in the amygdala, which are believed to underlie the acquisition of fear learning and failure to extinguish fear. Abnormalities in amygdala functioning in PTSD have been attributed to deficits in fear extinction, safety learning, and extinction recall (Jovanovic et al., 2012; Milad et al., 2009). Threat detection, characterized by attentional biases to threatening stimuli, hypervigilance, and exaggerated startle, has been linked with alterations in the amygdala, the dorsal anterior cingulate cortex (dACC), and the insula (Seeley et al., 2007). Alterations in emotional regulation and executive function are associated with decreased engagement of the dorsolateral (dlPFC), ventrolateral (vlPFC) and dorsomedial (dmPFC) prefrontal cortex in clinical and non-clinical studies (e.g. Buhle et al., 2014; Ochsner et al., 2012). Furthermore, studies with PTSD patients have found reduced connectivity within the fronto-parietal regions, which is believed to be associated with impaired executive functioning (Sripada et al., 2012). Contextual processing appears to be impaired in PTSD contributing to difficulty distinguishing between safe and threatening stimuli. Functional brain imaging research has shown the ability to distinguish between safety and threat cues is supported by signaling between the mPFC and the hippocampus (Lang et al., 2009), and studies have found volumetric and functional abnormalities of the hippocampus among individuals with PTSD (e.g. Pitman et al., 2012; Shin and Liberzon, 2010). Additionally, alterations in signaling in the mPFC have been shown to interfere with the processing of contextual information, and contribute to impaired safety-signal learning (Rougemont-Bücking et al., 2011).

Among factors that might help to better understand the engagement of neural circuits in the pathogenesis of PTSD is human self-regulation in response to stress and trauma. It has been well documented that in the wake a traumatic event, perceptions of the self may be altered (e.g. Benight and Bandura, 2004; Ehlers and Clarke, 2000), and numerous studies have found that maladaptive perceptions of one's self are strongly associated with PTSD (Bryant and Guthrie, 2007; Dekel et al., 2013). Negative changes in self-appraisals may be a critical mechanism underlying PTSD as it relates to one's ability to engage in self-regulatory processes, which facilitate adaptions to stress and trauma. According to Social Cognitive Theory (SCT), the capacity to adapt to stress depends, in part, on one's ability to successfully employ personal (adaptive cognitive appraisals), environmental (social support), and behavioral (proactive coping) strategies that facilitate goal attainment and a sense of agency.

Self-appraisals about one's ability to respond to stress may then lead to development of self-efficacy perceptions around coping (Benight and Bandura, 2004). There is considerable evidence suggesting that self-efficacy influences how individuals respond to potentially traumatic events (for a review see Benight and Bandura, 2004). Self-efficacy is described as one's self-perceived ability to influence command over personal thoughts, behaviors, and emotions, as well as the external environment (Bandura, 1997). Lower levels of self-efficacy have been associated with PTSD among a wide range of trauma-exposed populations (e.g. Benight and Bandura, 2004; Bosmans and Van der Velden, 2015; Cieslak et al., 2008; Ehlers et al., 1998a, Ehlers et al., 1998b; Piotrkowski and Brannen, 2002; Saigh et al., 1995; Solomon et al., 1991). In addition, experimental research has found that increasing perceptions of self-efficacy leads to more adaptive cognitive and affective processing when exposed to stress (Geer et al., 1970; Glass et al., 1973; Litt et al., 1993; Sanderson et al., 1989). Importantly, self-efficacy may also be an important mechanism underlying recovery, as low levels of self-efficacy have been associated with poor treatment outcome (e.g. Ehlers et al., 1998a, Ehlers et al., 1998b).

Of particular relevance to the treatment of PTSD, an emerging body of research is showing perceptions of self-efficacy in coping can be manipulated, and doing so leads to improved performance on cognitive and affective processes that may aid in recovery from PTSD. For example, in two studies with healthy controls, Brown and colleagues (Brown et al., 2012a, 2012b) used a false feedback technique in which participants were told that, in response to a stressful event, they were either in the top 1% or bottom 30% of copers. Although the two groups did not differ on levels of self-efficacy before the induction, those in the high self-efficacy condition self-reported significantly higher levels of self-efficacy after the experimenter feedback. Following this induction, participants generated autobiographical memories and imagined personal future events with greater episodic specificity (2012a), performed better on a social problem-solving task (2012a), and reported fewer visual intrusions and less distress following a trauma film-paradigm (2012b). In addition, Zlomuzica et al. (2015) employed a similar false-feedback technique in healthy individuals before a fear conditioning task, and found that self-efficacy induction facilitated extinction learning as measured by self-report and electrodermal activity.

In an effort to increase self-efficacy in clinical populations without utilizing false-feedback, Brown et al. (2016) recently demonstrated that self-efficacy can be increased through the recall of autobiographical memories of successful coping. In a sample of OEF/OIF combat veterans with and without PTSD, participants were randomized to either a high self-efficacy or control condition. Those in the high self-efficacy condition were asked to recall three autobiographical memories associated with times they demonstrated successful coping, whereas in the control condition they were asked to recall any three important memories. Participants in the high self-efficacy condition subsequently reported higher levels of self-efficacy. In addition, those in the high self-efficacy condition imagined future events characterized by greater agency and demonstrated better performance on a military-related social-problem solving task (Brown et al., 2016). Employing a similar method for increasing self-efficacy through autobiographical memory, Morina et al. (2017) showed that torture survivors asked to recall self-efficacy related memories exhibited greater distress tolerance on a mirror-tracing task. Such findings suggest that increasing self-efficacy may help to engage cognitive processes adaptive for recovery.

However, studies have yet to examine whether increasing levels of self-efficacy may lead to neural changes associated with recovery. Specifically, this study aimed to test whether using autobiographical memory self-efficacy induction (Brown et al., 2016) would lead to changes within four neural circuits that have been implicated in PTSD (Shalev et al., 2017). Brain imaging research with healthy individuals and PTSD populations would suggest that experimental manipulations aimed to increase self-regulatory processes, such as perceptions of self-efficacy, would engage the prefrontal cortex (e.g. Curtis & D'Esposito, 2003; Heatherton, 2011; Goldberg, 2001; Miller and Cohen, 2001). Considerable work has found that ventromedial prefrontal cortex (vmPFC), lateral prefrontal cortex (lPFC), and the anterior cingulate cortex (ACC) play key roles in self-regulation in response to affective stimuli (for a review see Heatherton, 2011). Moreover, neuroimaging studies of emotion regulation have consistently found that the PFC is engaged in top-down regulatory processes when individuals are attempting to regulate responses to negative stimuli (e.g. Davidson et al., 2000; Ochsner et al., 2004; Ochsner and Gross, 2005). Therefore, although the four neural circuits proposed by Shalev et al. (2017) were examined in this study, it was predicted that changes in the emotional regulation and executive function (ER/EF) circuit were most likely to be associated with an increase in self-efficacy.

To investigate the potential impact of self-efficacy on the four neural circuits associated with PTSD, graph-theory network analysis was utilized to examine the topological properties of brain networks (e.g. Bassett and Bullmore, 2006). Graph theory has been translated and applied to model the brain as a complex network represented by sets of nodes and edges (Bullmore and Sporns, 2009; Wang et al., 2016). Studies of healthy individuals employing graph theory have shown that human brain networks exhibit many complex organizational features, including small-worldness (a brain network characterized by high efficiency of information transfer at a low cost) and modularity (the partition of a brain network into smaller functional communities of modules, Achard et al., 2006; He et al., 2007). Graph theory analyses have also been able to distinguish psychiatric from healthy individuals in numerous studies (Bassett et al., 2012; Leistedt et al., 2009; Lynall et al., 2010; Zhang et al., 2011). For example, compared to trauma-exposed individuals without PTSD, PTSD patients exhibit decreased path length, as well as increased clustering coefficient, global efficiency and local efficiency (Lei et al., 2015). Additionally, a recent study of adolescent females with PTSD found that greater network modularity and assortativity, as well as lesser efficiency were associated with better response to treatment (Cisler et al., 2016). In contrast, a recent study of children diagnosed with PTSD showed an increase in path length, and decreased local and global efficiency (Suo et al., 2015). The authors suggest that differences in findings may be related to the way in which stress affects the brain at different stages of human development, which warrants further research (Suo et al., 2015).

This study is the first to look at whether recalling autobiographical memories of self-efficacy may have the potential to alter neural circuits associated with PTSD pathology. OEF/OIF combat veterans with PTSD were randomized to either a high self-efficacy induction (HSE) in which they recalled autobiographical memories related to self-efficacy or a no self-efficacy induction (NSE) control condition. Functional MRI (fMRI) was then used to measure resting state functional connectivity (rsFC) in previously identified neural circuits functioning in fear learning, threat detection, emotional regulation and executive function, and contextual processing. Although meta-analyses provide a powerful method for identifying alterations in resting state activity in PTSD (e.g. Disner et al., 2018; Hayes et al., 2012; Koch et al., 2016; Patel et al., 2012; Ramage et al., 2013; Wang et al., 2016), due to the heterogeneity of findings, the regions of interest in this study were based on those neural circuits identified in the framework proposed by Shalev et al. (2017). This model specifically aims to link abnormalities in neural circuits with the core symptom clusters in PTSD affording it potential utility as a conceptual and empirical guide in the apriori identification of specific dysregulated circuits to be used in the assessment of interventions (Sheynin and Liberzon, 2017).

Two network measures of brain connectivity, average connectivity and path length, were analyzed for each circuit to assess for changes associated with high self-efficacy induction in participants. Given previous findings from graph theory analyses showing that PTSD was associated with decreased path length and greater connectivity, it was predicted that increasing self-efficacy among individuals with PTSD would be associated with reduced connectivity and greater path length. Furthermore, although this hypothesis was examined in all four neural circuits, given previous findings on the primary role of the prefrontal cortex in self-regulation, it was expected that these patterns would be most evident in the ER/EF circuit.