Neural computations of threat in the aftermath of combat trauma

Abstract

By combining computational, morphological, and functional analyses, this study relates latent markers of associative threat learning to overt post-traumatic stress disorder (PTSD) symptoms in combat veterans. Using reversal learning, we found that symptomatic veterans showed greater physiological adjustment to cues that did not predict what they had expected, indicating greater sensitivity to prediction errors for negative outcomes. This exaggerated weighting of prediction errors shapes the dynamic learning rate (associability) and value of threat predictive cues. The degree to which the striatum tracked the associability partially mediated the positive correlation between prediction-error weights and PTSD symptoms, suggesting that both increased prediction-error weights and decreased striatal tracking of associability independently contribute to PTSD symptoms. Furthermore, decreased neural tracking of value in the amygdala, in addition to smaller amygdala volume, independently corresponded to higher PTSD symptom severity. These results provide evidence for distinct neurocomputational contributions to PTSD symptoms.

From Value to Saliency: Neural Computations of Subjective Value under Uncertainty in PTSD

“…Previous studies of PTSD often focused on the neural processing of fear and trauma, and identified both functional and structural abnormalities in amygdala, hippocampus, and vmPFC (Hayes, Hayes, and Mikedis 2012;Admon, Milad, and Hendler 2013;Wolf and Herringa 2015;Rabellino et al 2016;Homan et al 2019). Other studies have looked into more general cognitive processes and found blunted neural activation to monetary rewards (Sailer et al 2008;Elman et al 2009).…”

“…An individual's attitude towards uncertainty and the ability to handle uncertainty may therefore affect one's ability to cope with traumatic events. The notion of uncertainty has been incorporated in many fear-learning studies attempting to unravel the behavioral and neural mechanisms of PTSD (Brown et al 2018;Homan et al 2019). Participants in these studies encounter probabilistic deliveries of adverse outcomes (e.g.…”

Serotonin depletion impairs both Pavlovian and instrumental reversal learning in healthy humans

“…Likewise, reacting emotionally to threats is critical for survival yet must be modified as danger shifts from one source to another. Indeed, impairments in behavioural and emotional flexibility are pervasive in psychiatric disorders (Apergis-Schoute et al 2017;APA 2013;Homan et al 2019;Remijnse et al 2006;Robbins et al 2019;Waltz et al 2007) and serotonergic dysfunction is widely implicated across diagnostic categories (Cunningham et al 2020;Dayan and Huys 2009;Dersken et al 2020;Phillips and Robbins 2020;Quednow et al 2020;Zangrossi et al 2020). Behavioural inflexibility is a hallmark of obsessivecompulsive disorder (OCD) and substance use disorders, for instance, where learned behaviours persist inappropriately despite adverse consequences (APA 2013).…”

“…Behavioural inflexibility is a hallmark of obsessivecompulsive disorder (OCD) and substance use disorders, for instance, where learned behaviours persist inappropriately despite adverse consequences (APA 2013). Aberrations in threat and safety learning are characteristic of post-traumatic stress disorder (PTSD; Homan et al, 2019;Milad et al, 2009) and other anxiety disorders (Kim et al, 2011;Marin et al, 2017), and are also a feature of OCD (Milad et al, 2013;Apergis-Schoute et al, 2017) and schizophrenia (Holt et al, 2012). Drugs thought to boost serotonin transmission -selective serotonin reuptake inhibitors (SSRIs) -are first line treatments for OCD (Fineberg et al 2013) and PTSD (Baldwin et al 2014).…”

Reward Prediction Error Signaling during Reinforcement Learning in Social Anxiety Disorder is altered by Social Observation

“…The updating of predictions is a mechanism that is hypothesized to be altered in anxiety disorders (see reviews by Grupe & Nitschke, 2013, Cavanagh & Shackman, 2015, Paulus & Stein, 2006, Paulus & Yu, 2012). In particular, altered updating may be characterized by (a) altered prediction errors (White et al, 2017, Homan et al, 2019) and (b) altered learning rates for disorder-relevant stimuli (Browning et al, 2015, Koban et al 2018, Homan et al, 2019) as formalized by reinforcement learning (RL) models. Specifically in SAD, we would expect excessive prediction errors to performance feedback as well as lower learning rates if the updating of predictions takes place in socially evaluative contexts.…”