Clarifying the neural substrates of threat and safety reversal learning in humans

Revisiting deficits in threat and safety appraisal in obsessive-compulsive disorder

Current behavioural treatment of obsessive-compulsive disorder (OCD) is informed by fear conditioning and involves iteratively re-evaluating previously threatening stimuli as safe. However, there is limited research investigating the neurobiological response to conditioning and reversal of threatening stimuli in individuals with OCD. A clinical sample of individuals with OCD (N = 45) and matched healthy controls (N = 45) underwent functional magnetic resonance imaging. While in the scanner, participants completed a well-validated fear reversal task and a resting-state scan. We found no evidence for group differences in task-evoked brain activation or functional connectivity in OCD. Multivariate analyses encompassing all participants in the clinical and control groups suggested that subjective appraisal of threatening and safe stimuli were associated with a larger difference in brain activity than the contribution of OCD symptoms. In particular, we observed a brain-behaviour continuum whereby heightened affective appraisal was related to increased bilateral insula activation during the task (r = 0.39, pFWE  = .001). These findings suggest that changes in conditioned threat-related processes may not be a core neurobiological feature of OCD and encourage further research on the role of subjective experience in fear conditioning.

Brain activation during fear extinction recall in unmedicated patients with obsessive-compulsive disorder

Specific brain activation patterns during fear conditioning and the recall of previously extinguished fear responses have been associated with obsessive-compulsive disorder (OCD). However, further replication studies are necessary. We measured skin-conductance response and blood oxygenation level-dependent responses in unmedicated adult patients with OCD (n = 27) and healthy participants (n = 22) submitted to a two-day fear-conditioning experiment comprising fear conditioning, extinction (day 1) and extinction recall (day 2). During conditioning, groups differed regarding the skin conductance reactivity to the aversive stimulus (shock) and regarding the activation of the right opercular cortex, insular cortex, putamen, and lingual gyrus in response to conditioned stimuli. During extinction recall, patients with OCD had higher responses to stimuli and smaller differences between responses to conditioned and neutral stimuli. For the entire sample, the higher the response delta between conditioned and neutral stimuli, the greater the dACC activation for the same contrast during early extinction recall. While activation of the dACC predicted the average difference between responses to stimuli for the entire sample, groups did not differ regarding the activation of the dACC during extinction recall. Larger unmedicated samples might be necessary to replicate the previous findings reported in patients with OCD.

Measuring human trace fear conditioning

Trace fear conditioning is an important research paradigm to model aversive learning in biological or clinical scenarios, where predictors (conditioned stimuli, CS) and aversive outcomes (unconditioned stimuli, US) are separated in time. The optimal measurement of human trace fear conditioning, and in particular of memory retention after consolidation, is currently unclear. We conducted two identical experiments (N₁  = 28, N₂  = 28) with a 15-s trace interval and a recall test 1 week after acquisition, while recording several psychophysiological observables. In a calibration approach, we explored which learning and memory measures distinguished CS+ and CS- in the first experiment and confirmed the most sensitive measures in the second experiment. We found that in the recall test without reinforcement, only fear-potentiated startle but not skin conductance, pupil size, heart period, or respiration amplitude, differentiated CS+ and CS-. During acquisition without startle probes, skin conductance responses and pupil size responses but not heart period or respiration amplitude differentiated CS+ and CS-. As a side finding, there was no evidence for extinction of fear-potentiated startle over 30 trials without reinforcement. These results may be useful to inform future substantive research using human trace fear conditioning protocols.

Behavioral and neural responses during fear conditioning and extinction in a large transdiagnostic sample

•

Behavioral and neural responses during Pavlovian fear learning were examined in a large sample of healthy and individuals with anxiety and depression.

•

Latent profile models to threat were derived from behavioral and neural data.

•

Demographic, cognitive, and psychological variables did not robustly characterize latent profiles.

•

Neuroimaging data did not evidence functional role of amygdala in fear learning.

•

Human fear learning recruited a distributed network of regions involved in interoceptive, cognitive, motivational, and psychomotor processes.

Background

Dysregulation of fear learning has been associated with psychiatric disorders that have altered positive and negative valence domain function. While amygdala-insula-prefrontal circuitry is considered important for fear learning, there have been inconsistencies in neural findings in healthy and clinical human samples. This study aimed to delineate the neural substrates and behavioral responses during fear learning in a large, transdiagnostic sample with predominantly depressive and/or anxious dysfunction.

Methods

Two-hundred and eighty-two individuals (52 healthy participants; 230 participants with depression and/or anxiety-related problems) from the Tulsa 1000 study, an ongoing, naturalistic longitudinal study based on a dimensional psychopathological framework, completed a Pavlovian fear learning task during functional magnetic resonance imaging. Linear mixed-effects analyses examined condition-by-time effects on brain activation (CS+, CS- across familiarization, conditioning, and extinction trials). A data-driven latent profile analysis (LPA) examined distinct patterns of behavioral and neural responses to threat across fear conditioning and extinction, while logistic regression analyses evaluated cognitive-affective predictors of latent profiles.

Results

Whole-brain analyses revealed a condition-by-time interaction in the anterior insula, postcentral gyrus, superior temporal gyrus, middle frontal gyrus, and cerebellum but not amygdala. The LPA identified distinct latent profiles across subjective and neural levels of measurement. Anterior insula profiles were characterized by marginal differences in age and state anxiety.

Conclusions

Our findings demonstrate that human fear learning recruits a distributed network of regions involved in interoceptive, cognitive, motivational, and psychomotor processes. Data-driven analyses identified distinct profiles of subjective and neural responses during fear learning that transcended clinical diagnoses, but no robust relationships to demographic or cognitive-affective variable were identified.

Parsing neural circuits of fear learning and extinction across basic and clinical neuroscience: Towards better translation

Over the past decades, studies of fear learning and extinction have advanced our understanding of the neurobiology of threat and safety learning. Animal studies can provide mechanistic/causal insights into human brain regions and their functional connectivity involved in fear learning and extinction. Findings in humans, conversely, may further enrich our understanding of neural circuits in animals by providing macroscopic insights at the level of brain-wide networks. Nevertheless, there is still much room for improvement in translation between basic and clinical research on fear learning and extinction. Through the lens of neural circuits, in this article, we aim to review the current knowledge of fear learning and extinction in both animals and humans, and to propose strategies to fill in the current knowledge gap for the purpose of enhancing clinical benefits.

Microbiota links to neural dynamics supporting threat processing

There is growing recognition that the composition of the gut microbiota influences behaviour, including responses to threat. The cognitive‐interoceptive appraisal of threat‐related stimuli relies on dynamic neural computations between the anterior insular (AIC) and the dorsal anterior cingulate (dACC) cortices. If, to what extent, and how microbial consortia influence the activity of this cortical threat processing circuitry is unclear. We addressed this question by combining a threat processing task, neuroimaging, 16S rRNA profiling and computational modelling in healthy participants. Results showed interactions between high‐level ecological indices with threat‐related AIC‐dACC neural dynamics. At finer taxonomic resolutions, the abundance of Ruminococcus was differentially linked to connectivity between, and activity within the AIC and dACC during threat updating. Functional inference analysis provides a strong rationale to motivate future investigations of microbiota‐derived metabolites in the observed relationship with threat‐related brain processes.

Neural mediators of subjective and autonomic responding during threat learning and regulation

Threat learning elicits robust changes across multiple affective domains, including changes in autonomic indices and subjective reports of fear and anxiety. It has been argued that the underlying causes of such changes may be dissociable at a neural level, but there is currently limited evidence to support this notion. To address this, we examined the neural mediators of trial-by-trial skin conductance responses (SCR), and subjective reports of anxious arousal and valence in participants (n = 27; 17 females) performing a threat reversal task during ultra-high field functional magnetic resonance imaging. This allowed us to identify brain mediators during initial threat learning and subsequent threat reversal. Significant neural mediators of anxious arousal during threat learning included the dorsal anterior cingulate, anterior insula cortex (AIC), and ventromedial prefrontal cortex (vmPFC), subcortical regions including the amygdala, ventral striatum, caudate and putamen, and brain-stem regions including the pons and midbrain. By comparison, autonomic changes (SCR) were mediated by a subset of regions embedded within this broader circuitry that included the caudate, putamen and thalamus, and two distinct clusters within the vmPFC. The neural mediators of subjective negative valence showed prominent effects in posterior cortical regions and, with the exception of the AIC, did not overlap with threat learning task effects. During threat reversal, positive mediators of both subjective anxious arousal and valence mapped to the default mode network; this included the vmPFC, posterior cingulate, temporoparietal junction, and angular gyrus. Decreased SCR during threat reversal was positively mediated by regions including the mid cingulate, AIC, two sub-regions of vmPFC, the thalamus, and the hippocampus. Our findings add novel evidence to support distinct underlying neural processes facilitating autonomic and subjective responding during threat learning and threat reversal. The results suggest that the brain systems engaged in threat learning mostly capture the subjective (anxious arousal) nature of the learning process, and that appropriate responding during threat reversal is facilitated by participants engaging self- and valence-based processes. Autonomic changes (SCR) appear to involve distinct facilitatory and regulatory contributions of vmPFC sub-regions.

Time-Dependent Recruitment of Prelimbic Prefrontal Circuits for Retrieval of Fear Memory

The long-lasting nature of fear memories is essential for survival, but the neural circuitry for retrieval of these associations changes with the passage of time. We previously reported a time-dependent shift from prefrontal-amygdalar circuits to prefrontal-thalamic circuits for the retrieval of auditory fear conditioning. However, little is known about the time-dependent changes in the originating site, the prefrontal cortex. Here we monitored the responses of prelimbic (PL) prefrontal neurons to conditioned tones at early (2 h) vs. late (4 days) timepoints following training. Using c-Fos, we find that PL neurons projecting to the amygdala are activated early after learning, but not later, whereas PL neurons projecting to the paraventricular thalamus (PVT) show the opposite pattern. Using unit recording, we find that PL neurons in layer V (the origin of projections to amygdala) showed cue-induced excitation at earlier but not later timepoints, whereas PL neurons in Layer VI (the origin of projections to PVT) showed cue-induced inhibition at later, but not earlier, timepoints, along with an increase in spontaneous firing rate. Thus, soon after conditioning, there are conditioned excitatory responses in PL layer V which influence the amygdala. With the passage of time, however, retrieval of fear memories shifts to inhibitory responses in PL layer VI which influence the midline thalamus.

A Decision Architecture for Safety Computations

Accurately estimating safety is critical to pursuing nondefensive survival behaviors. However, little attention has been paid to how the human brain computes safety. We conceptualize a model that consists of two components: (i) threat-oriented evaluations that focus on threat value, imminence, and predictability; and (ii) self-oriented evaluations that focus on the agent's experience, strategies, and ability to control the situation. Our model points to the dynamic interaction between these two components as a mechanism of safety estimation. Based on a growing body of human literature, we hypothesize that distinct regions of the ventromedial prefrontal cortex (vmPFC) respond to threat and safety to facilitate survival decisions. We suggest safety is not an inverse of danger, but reflects independent computations that mediate defensive circuits and behaviors.

Unconditioned response to an aversive stimulus as predictor of response to conditioned fear and safety: A cross-species study

Safety signals predict the non-occurrence of an aversive event, thereby inhibiting fear responses. Previous research has shown that conditioned safety learning is impaired in patients suffering from post-traumatic stress disorder (PTSD). Using a translational approach, the present study aimed to investigate whether individual responses to an aversive unconditioned stimulus (US) in rats (basic science), non-traumatized (pre-clinical) or traumatized humans (clinical) predicts their response to a conditioned fear or safety stimulus. Using three different archival datasets, the unconditioned response (UCR) to the US during fear or safety conditioning was assessed in rats, non-traumatized humans, and trauma-exposed humans. The response to learned fear (CS+; context) and safety (CS-) was measured by the modulation of the startle response (rats, traumatized humans) or skin conductance response (non-traumatized humans). Our results showed that all groups with low UCR and those with high UCR from the rodent or non-traumatized human samples displayed lower fear response to the CS- than to the CS+ . Traumatized humans with high UCR showed similarly high responses to the CS+ and CS-. While all groups showed a positive association between the UCR and CS+ response, the UCR correlated positively with the CS- response in traumatized humans only. Our findings suggest that an elevated response to aversive stimuli predicts deficits in conditioned safety memory in those at risk for trauma-related disorders and confirms that impaired safety learning could be a valid biomarker for these diseases.

Threat and safety reversal learning in social anxiety disorder - an fMRI study

Social anxiety disorder (SAD) has been linked to maladaptive forms of fear regulation, including flexibly distinguishing between learned threat and safety signals. Few studies have examined this in young, unmedicated SAD patients, including its neural basis. We aimed to characterize the neural, subjective, and autonomic correlates of reversal learning in patients with SAD and compare them to matched patients with major depressive disorder and to healthy control participants. All participants completed a threat-safety reversal learning task during functional magnetic resonance imaging. Successful threat-safety updating was associated with significant activation of primary regions of interest (anterior cingulate, insula and ventromedial prefrontal cortex), however, no significant differences were observed between them, consistent with subjective reports of task-evoked anxiety and affect. Contrary to expectations, we did not observe threat and safety reversal learning to be significantly impaired in young people with SAD.

Revaluing the Role of vmPFC in the Acquisition of Pavlovian Threat Conditioning in Humans

The role of the ventromedial prefrontal cortex (vmPFC) in human pavlovian threat conditioning has been relegated largely to the extinction or reversal of previously acquired stimulus-outcome associations. However, recent neuroimaging evidence questions this view by also showing activity in the vmPFC during threat acquisition. Here we investigate the casual role of vmPFC in the acquisition of pavlovian threat conditioning by assessing skin conductance response (SCR) and declarative memory of stimulus-outcome contingencies during a differential pavlovian threat-conditioning paradigm in eight patients with a bilateral vmPFC lesion, 10 with a lesion outside PFC and 10 healthy participants (each group included both females and males). Results showed that patients with vmPFC lesion failed to produce a conditioned SCR during threat acquisition, despite no evidence of compromised SCR to unconditioned stimulus or compromised declarative memory for stimulus-outcome contingencies. These results suggest that the vmPFC plays a causal role in the acquisition of new learning and not just in the extinction or reversal of previously acquired learning, as previously thought. Given the role of the vmPFC in schema-related processing and latent structure learning, the vmPFC may be required to construct a detailed representation of the task, which is needed to produce a sustained conditioned physiological response in anticipation of the unconditioned stimulus during threat acquisition.SIGNIFICANCE STATEMENT Pavlovian threat conditioning is an adaptive mechanism through which organisms learn to avoid potential threats, thus increasing their chances of survival. Understanding what brain regions contribute to such a process is crucial to understand the mechanisms underlying adaptive as well as maladaptive learning, and has the potential to inform the treatment of anxiety disorders. Importantly, the role of the ventromedial prefrontal cortex (vmPFC) in the acquisition of pavlovian threat conditioning has been relegated largely to the inhibition of previously acquired learning. Here, we show that the vmPFC actually plays a causal role in the acquisition of pavlovian threat conditioning.