The Brain as a Bayesian Machine: From Computational Cognitive Neuroscience to Psychotherapy
3. The Invisible Obstacles to Exposure: Attention, Covert Safety Behaviors, and Reassurance Seeking
3.1. Approaching a situation is not the same as fully confronting it
In the previous chapter, we discussed how exposure can change predictions of danger through new experience. In clinical work, however, it is important to establish not only whether a person approached the feared situation, but also how they confronted it and what information the brain, as a cognitive system, ultimately received.
A person may give a public presentation, use the subway, or attend a meeting while continuously performing actions that, in their view, prevent a catastrophe—safety behaviors. Some of these actions are visible; others operate through attention, mental preparation, or obtaining reassurance from other people. Understanding exposure therefore sometimes requires examining what cannot be seen from the outside.
3.2. Selective attention: which data get through?
Imagine a person with social anxiety speaking in front of a group of twenty people. Nineteen respond neutrally or positively, while one briefly shows a dissatisfied facial expression. Afterwards, the person says: “I saw the way they were looking at me. It was obvious that I was speaking badly.” The problem here is not only how the person interpreted the meeting. From the outset, their attention disproportionately selected information consistent with danger. Neutral or positive responses may have remained in the background, while one potentially negative signal became the basis for evaluating the entire meeting.
In Bayesian terms, belief updating depends not only on the weight assigned to the data already received, but also on which data are selected in the first place. If a person mainly gathers signals from the environment that match their prediction of danger, the new experience may provide less contradictory information than an outside observer assumes. Here, we are dealing with a phenomenon commonly called selective perception.
The therapeutic aim is not simply to look for positive responses. That could create another bias. The aim is to make attention more flexible and representative: to notice neutral information, the unfolding situation, and one’s own immediate experience as well. One of the simplest things to do in such cases is to ask: “Apart from that one response you noticed, what else was happening?”
3.3. Covert safety behaviors: approaching a situation while still protecting yourself
A safety behavior does not always take the form of a water bottle, medication, or standing near an exit. It may be a mental action or such a habitual form of behavior that the person does not even notice its protective function. For example, during a public presentation, a person may mentally construct every sentence in advance, constantly monitor their voice, speak quickly to finish as soon as possible, or avoid pauses and spontaneous answers. None of these actions is inherently problematic, but what matters is why the person does it and what they predict will happen if they do not.
Preparing for a presentation can help someone communicate the material clearly. However, the same preparation can serve a different belief: “If I do not choose every word in advance, I will become confused and embarrass myself.” In the second case, a successful presentation may reinforce the perceived need for control. It is therefore useful to ask: “What were you doing to prevent the thing you feared?”, followed by: “If you had not done that, what do you think would have happened?” These questions can reveal a prediction that has not yet been tested.
The behavioral experiment can then be designed more specifically: allowing a brief pause, answering one question without a prepared sentence, or temporarily shifting attention from one’s own voice to the content of the conversation. In this case, what is being tested is the belief that the safety behavior is necessary.
3.4. Reassurance seeking: when uncertainty is immediately “closed off” after an experience
A person may complete an exposure exercise but immediately afterwards ask: “Was I speaking normally?”, “Did you notice that my hands were shaking?”, or “Are you sure I did not make a bad impression?” These questions may seem like a simple wish for feedback. Sometimes that is exactly what they are. However, feedback and reassurance seeking do not serve the same function. Feedback can provide new, actionable information. Repeated reassurance seeking often aims to reduce distress immediately and obtain complete certainty.
The difference can be recognized through the repetition of the question and the effect of the answer. If an answer provides brief relief, but doubt then returns and the person needs to ask again, the process can become a safety strategy. The person may then learn less from their own experience and instead reinforce the following belief: “To feel calm, I need someone else’s confirmation.” What is maintained is not only a belief about a particular danger, but also a more general assumption: “I must not remain uncertain.”
In therapy, it is possible to gradually delay repeated requests for reassurance, first independently summarize the actual outcome, and notice how distress changes without obtaining another confirmation. This does not mean giving up necessary information or support. What is being tested is the belief that a person cannot continue to act without immediate reassurance.
3.5. Should all safety behaviors be eliminated immediately?
It is important to avoid a rigid rule such as “Every safety behavior is harmful.” Research findings on their effects are mixed, and clinical decisions require consideration of the behavior’s function and the situation (Blakey & Abramowitz, 2016; Meulders et al., 2016).
Some behaviors may initially help a person engage in exposure. For example, the therapist’s presence may make it possible to take the first step. Later, however, it is important that learning does not remain restricted to that condition. Otherwise, the belief “I can only do this when my therapist is beside me” may persist. The question, therefore, is whether the support helps the person gradually act independently or becomes an obligatory condition for safety. Its reduction can be agreed upon in advance and organized in stages, while examining what the person learns from each experience.
It is also necessary to distinguish safety behaviors that maintain anxiety from precautions proportionate to actual danger. Exposure does not require disregarding medical advice, independently changing prescribed medication, or ignoring real danger. The target is the additional protection whose perceived necessity arises from an exaggerated prediction of danger.
Summary
During exposure, it is important to notice not only a person’s behavioral participation, but also the direction of their attention, their covert safety behaviors, and their need for reassurance afterwards.
These processes may determine what information the person receives, how they explain the outcome, and how independently they can act following the new experience. One useful clinical question is therefore: “Did this experience help you participate more freely in the situation, or did it reinforce the conditions without which you still do not dare to participate?”
References
Բեգոյան Ա.Ն. Տագնապամարիչ. տագնապային խանգարումների հաղթահարման ուղեցույց / Ա.Ն. Բեգոյան. – Եր.: Հիլֆմանն Պրեսս, 2020. – 368 էջ։
Blakey, S. M., & Abramowitz, J. S. (2016). The effects of safety behaviors during exposure therapy for anxiety: Critical analysis from an inhibitory learning perspective. Clinical Psychology Review, 49, 1–15.
Meulders, A., Van Daele, T., Volders, S., & Vlaeyen, J. W. S. (2016). The use of safety-seeking behavior in exposure-based treatments for fear and anxiety: Benefit or burden? A meta-analytic review. Clinical Psychology Review, 45, 144–156.