Barlow's Model Applied to VRET: Why VR Activates Real Fear

Why We Need an Etiological Model to Understand VRET

VRET has become popular in recent years, backed by robust evidence of efficacy, but the relevant clinical question isn't just whether it works, but why it works. Without a clear etiological model, the clinician risks applying the technology as a black box whose indications, intensity, and stopping criteria are dictated more by the vendor's marketing than by clinical reasoning.

David Barlow's etiological model of anxiety disorders, developed over several decades and consolidated in his book Anxiety and Its Disorders, offers a coherent framework for understanding what kind of mechanism VRET is mobilizing. Its usefulness isn't only theoretical: the model informs practical decisions (when exposure is appropriate, with what hierarchy, how to read the patient's response, and when to close the exposure) that would otherwise be left to clinical intuition.

In this article we review the triple vulnerability model, its relationship to Mowrer's conditioned fear theory and later findings on extinction and relearning, and then bring it down to the concrete question: when a patient with dog phobia faces a virtual dog, why does their amygdala behave as if the dog were real?

Barlow's Tripartite Model: Three Vulnerabilities That Compound

Barlow proposes that no anxiety disorder is explained by a single variable. Its etiology is multifactorial and organized into three levels that act in cascade.

Generalized biological vulnerability captures the genetic and temperamental predisposition toward exaggerated alarm responses. It includes traits such as neuroticism, heightened amygdalar reactivity, and low activation thresholds in the autonomic nervous system. It doesn't determine any specific disorder, but it predisposes the entire family of anxiety disorders.

Generalized psychological vulnerability describes the set of beliefs and schemas acquired over the course of development that increase the likelihood of interpreting the world as unpredictable or uncontrollable. Low-contingency parenting patterns, caregivers modeling avoidance responses, or early experiences of losing control all contribute to this cognitive-emotional substrate.

Specific psychological vulnerability is what connects anxiety to a particular object or situation. This includes direct conditioning experiences (a childhood dog bite that precedes dog phobia), vicarious learning (watching a family member react with panic to heights), and informational transmission (sustained warnings about how dangerous a stimulus is). The interaction of the three vulnerabilities generates the specific symptomatic pattern of the disorder.

Conditioned Fear: Mowrer, Two Factors, and the Role of Avoidance

Barlow's model draws on the tradition of classical and operant conditioning. Mowrer's two-factor theory (1947) remains a useful reference: first, a neutral stimulus becomes associated with a fear response through classical conditioning (a neutral dog paired with an aversive bite). Second, avoidance of the stimulus is reinforced through operant conditioning because it relieves anticipatory anxiety.

The clinical problem is that avoidance, which functions as an immediate distress-reduction mechanism, prevents extinction of the conditioned fear. The patient is never exposed long enough for the autonomic response to come down and the learning to update. Therapeutic exposure, whether in vivo, in imagination, or in virtual reality, aims precisely to break this cycle: block avoidance, sustain exposure until the autonomic response drops, and allow the system to build new inhibitory learning that coexists with the original fear learning.

This is why cognitive desensitization alone ("it's not a big deal," "the dog won't bite you") rarely resolves a phobia. The original learning is somatic and autonomic, not purely declarative: the update has to happen at the same level.

Presence as the Bridge Between Virtual Stimulus and Real Response

If therapeutic exposure requires the conditioned-fear system to activate, why does a virtual scenario, which the patient knows isn't real, produce that effect? The answer lies in the concept of presence, central to the clinical VR literature.

Slater and Wilbur (1997) distinguished between immersion (the system's objective technical properties: resolution, latency, field of view, audiovisual fidelity) and presence (the user's subjective response, their sense of being inside the virtual environment). Immersion is a necessary but not sufficient condition: two patients with the same headset can experience very different levels of presence.

Diemer, Alpers, Peperkorn, Shiban, and Mühlberger (2015) synthesized the relationship between presence and clinical response in VR. They showed that the degree of evoked presence correlates with the emotional response (autonomic activation, subjective fear) and, through it, with therapeutic outcome. When presence is high, the limbic system responds to the virtual stimulus in a way qualitatively similar to how it would respond to the real stimulus. When presence is low, whether because the scenario is poorly designed, there's sensory mismatch, or the patient is cognitively distanced, activation falls short and the exposure loses therapeutic power.

This has an important practical consequence: VRET's effectiveness doesn't depend on the patient's cognitive conviction that the stimulus is real (the patient knows perfectly well they're wearing a headset), but on whether the perceptual-emotional system responds as if it were. The amygdala doesn't consult the prefrontal cortex before triggering the fear response; it reacts to perceptual patterns before conscious deliberation can step in.

Triple Vulnerability Plus Presence: Connecting to the Clinical Case

Seen through Barlow's model, a VRET session unfolds the following elements. Biological vulnerability determines the raw intensity of the patient's autonomic response to the stimulus (a patient with high amygdalar reactivity will reach elevated SUDS scores even at lower hierarchy levels). Generalized psychological vulnerability modulates how the patient interprets their own distress and their sense of control during exposure (a patient with low tolerance for uncertainty may ask to end the session early). Specific vulnerability defines which particular stimulus activates the response and, therefore, what exposure hierarchy to build.

Presence is what builds the bridge between these three levels and actual clinical activation. If the virtual scenario evokes sufficient presence, the patient enters the therapeutic activation range. If not, the clinician observes persistently low SUDS scores and a "flat" session that fails to generate habituation. This reading is useful because it turns presence, which can seem like a merely aesthetic variable, into a measurable clinical variable that needs monitoring.

Tobias Schubert's team developed the IPQ (Igroup Presence Questionnaire) as a brief instrument for measuring subjective presence after a VR session. When patients persistently report low presence and clinical progress stalls, it's worth reviewing technical variables (headset latency, spatial audio quality, anthropometric fit) and clinical variables (framing, exposure instructions, the patient's window of tolerance).

Implications for the Clinical Workflow

Barlow's framework plus presence generates several practical decisions. First, indication: VRET makes sense when there is a clearly identified specific vulnerability and a virtual scenario capable of evoking that vulnerability with sufficient presence. For diffuse or hard-to-represent specific vulnerabilities (certain complex trauma, obsessions with moral content), the tool fits less well.

A related decision is the initial psychoeducational framing. Explaining to the patient, in accessible language, why their system responds to the virtual stimulus as if it were real reduces anticipatory dissonance ("I know it's not real, so why does it affect me?") and eases entry into the hierarchy. The classic explanation that draws on classical conditioning and the autonomic response usually works well, especially when paired with the metaphor of an "old alarm system" that reacts before rational deliberation can intervene.

Second, the hierarchy: building the hierarchy isn't just about ranking stimuli by intensity, but calibrating the point at which presence is sufficient to activate the fear response without overwhelming the window of tolerance. In practice, a well-designed hierarchy starts with moderate-presence, low-activation stimuli and progresses toward high presence and significant activation.

Third, reading the session: the psychologist watches the SUDS curve but also asks, in a structured way, about the sense of presence. A session with low SUDS scores can be due to legitimate habituation or to insufficient presence; these are very different situations that call for opposite adjustments.

Fourth, ending the session: the ability to exit the virtual scenario quickly is a relevant therapeutic asset because it frames exposure as a voluntary, reversible experience, which reinforces the patient's sense of control and, according to García-Palacios et al., favors adherence compared with in-vivo exposure when the latter involves a heavier logistical burden.

Limitations of the Model and Areas Under Discussion

Barlow's model is an integrative proposal, but it isn't the only one, nor is it beyond debate. Purely cognitive models (Clark, Wells) place the emphasis on dysfunctional cognitive processes as the central causal variable; more recent transdiagnostic models redirect attention toward emotional regulation processes common to several disorders. None of them excludes VRET; each simply frames it slightly differently.

As for presence, the findings from Diemer et al. and Schubert's line of research are robust but correlational. The strict causal question ("does increasing presence increase clinical response?") is better studied in the lab than in practice. For the clinician, the practical heuristic remains reasonable: maximize technical presence and attend to subjective presence as a process variable.

Finally, VRET doesn't resolve emotional regulation deficits or the interpersonal components of the disorder on its own. It's a powerful exposure tool that fits within a broader clinical approach, not a substitute for the therapeutic process.

One additional limitation deserves comment: operationalizing "specific psychological vulnerability" in adult patients with complex clinical histories isn't always straightforward. In practice, the clinician builds a functional formulation of the presentation that integrates documentable conditioned learning, vicarious experiences, and informational transmission, but much of the original learning may be inaccessible to the patient's conscious awareness or autobiographical narrative. Barlow's model doesn't require reconstructing the history for the intervention to work, which is a clinical advantage: corrective relearning happens through the new exposure experience, not through retrospective elucidation of the original learning.

Conclusion: The Headset Doesn't Trick the Brain — It Activates It

Barlow's framework offers a coherent answer to why VRET works. It doesn't work because the patient "believes" the dog is real; it works because the perceptual-emotional system responds to visual, auditory, and proprioceptive patterns before conscious deliberation intervenes, and because the confluence of the three vulnerabilities leaves that system especially primed to activate in response to the relevant virtual stimulus.

To see how this translates into concrete clinical scenarios, you can check out the dog phobia exposure scenario and the companion piece on presence and immersion in clinical VR. If you'd like to see the full clinical workflow, you can book a guided demo.

This article is for informational purposes for psychology professionals. It is not clinical advice for any individual case and does not replace the judgment of the licensed psychologist in charge. VRET is professional clinical-support software, not a CE-marked medical device.