Anxiety as a disorder of uncertainty: implications for understanding maladaptive anxiety, anxious avoidance, and exposure therapy

Abstract

In cognitive-behavioral conceptualizations of anxiety, exaggerated threat expectancies underlie maladaptive anxiety. This view has led to successful treatments, notably exposure therapy, but is not consistent with the empirical literature on learning and choice alterations in anxiety. Empirically, anxiety is better described as a disorder of uncertainty learning. How disruptions in uncertainty lead to impairing avoidance and are treated with exposure-based methods, however, is unclear. Here, we integrate concepts from neurocomputational learning models with clinical literature on exposure therapy to propose a new framework for understanding maladaptive uncertainty functioning in anxiety. Specifically, we propose that anxiety disorders are fundamentally disorders of uncertainty learning and that successful treatments, particularly exposure therapy, work by remediating maladaptive avoidance from dysfunctional explore/exploit decisions in uncertain, potentially aversive situations. This framework reconciles several inconsistencies in the literature and provides a path forward to better understand and treat anxiety.

A meta reinforcement learning account of behavioral adaptation to volatility in recurrent neural networks

“…Maladaptive adaptation to volatility is a core feature of several psychiatric disorders, including anxiety (Aylward et al, 2019; Huang et al, 2017), depression (Pulcu & Browning, 2019), obsessive-compulsive disorder (Apergis-Schoute & Ip, 2020; Moreira et al, 2020), substance use disorder (Zuhlsdorff, 2022) and gambling disorder (Addicott et al, 2015; Jara-Rizzo et al, 2020; Wiehler et al, 2021). A common observation in affective disorders such as anxiety and depression is that learning rates are not adapted to the true volatility of the environment (Brown et al, 2023; Pulcu & Browning, 2019). Cognitive control required for adaptation could be generalized from previous experiences of volatility (Hohl et al 2023), resonating with our neural network modeling results.…”

Resetting the Hippocampal Buffer: A Neurocognitive Account of Psychedelic Therapy for Anxiety-Related Psychopathology

“…Interestingly, compounds that activate 5-HT2ARs (indirectly via SSRIs and directly via psychedelics) and block 5-HT2ARs (e.g., trazodone and atypical antipsychotics) both downregulate 5-HT2A receptors (Van Oekelen et al, 2003). Stress appears to upregulate 5-HT2ARs particularly in frontal regions that are thought to facilitate learning about threats (Murnane, 2019) and could minimize uncertainty (Brown et al, 2023), and 5-HT2AR knockout mice exhibit less anxiety-like behavior (Weisstaub et al, 2006). Consistent with these findings, patients with GAD, but not depression (Anand et al, 1994), exhibit a larger anxiety response than healthy participants in response to mchlorophenylpiperazine (Germine et al, 1992), a hallucinogenic 5-HT2AR agonist with additional activity at other 5-HT receptors.…”

“…Together, these findings suggest that anxiety-related psychopathology manifests hyperreactive SN function to perceived threat that in turn results in ECN deployment geared towards exercising top-down control to adaptively respond to environmental demands. This environmental engagement may serve to reduce uncertainty by sustaining vigilance to one's environment (Brown et al, 2023;McGovern et al, 2022). In contrast, the aforementioned environmental engagement process may serve to limit vmPFC/DMN-mediated experiential learning processes that could otherwise more adaptively regulate anxiety in the long-term (e.g., fear extinction, belief updating).…”

“…In contrast, generalized anxiety disorder (GAD) is characterized by a broad, pervasive, and difficult-tocontrol chronic apprehension, which is generally less specific in focus than social anxiety disorder or panic disorder. More broadly, pathological anxiety can also be considered a heightened intolerance to uncertainty that results in disproportionate learning to expect threat (Brown et al, 2023;McGovern et al, 2022) and increased distress and dysfunction (Watson et al, 2022). Here, we consider anxiety-related psychopathology as the overall category of interest and target for treatment, and our proposed model and related conclusions apply broadly to this category.…”

Associations between anxiety-related traits and fear acquisition and extinction - an item-based content and meta-analysis [Stage 1 Registered Report]

“…It has been speculated that the literature may suggest differential associations between anxiety-related traits and fear conditioning depending on the predictability of the experimental situation (see for a review). Recently, it has been proposed that anxiety-related traits may in fact be more sensitive to uncertainty rather than the amount of threat (Brown et al, 2023). In a strong, unambiguous experimental situation (e.g., explicit CS+ → US contingency instructions, 100% reinforcement rate) anxiety-related individual differences may manifest only to a limited extent, (Lissek et al, 2006)while weak, ambiguous experimental situations (e.g., no explicit CS+ → US contingency instructions, low reinforcement rate) may show stronger anxiety-related effects (Lissek et al, 2006;Torrents-Rodas et al, 2013).…”