Exposure Therapy Session: What Happens, Phase by Phase

What happens during an exposure session?

An observer walking into an exposure therapy session halfway through would see very little movement. Two armchairs set at an angle, a recording sheet on the low table, a stopwatch, and a conversation made of short sentences. The scene is plain on purpose. Anyone expecting drama comes away disappointed, because the work is not in the scene but in what the patient learns about the thing he fears.

The phases of an exposure session are four, and every exposure therapy session runs through them in that order: the framing, the approach to the stimulus, staying with it while the record is kept, and the closure. That order is not decorative. Each phase leaves something in writing that the next one needs, and skipping one turns the hour into a test of endurance. How the script is prepared before the patient arrives belongs to another point in the process and is taken as given here.

The stimulus changes with the presenting problem: a stalled lift, a loose dog, the gaze of an audience, or the height exposure scenarios delivered through a headset. The grammar of the hour, by contrast, barely moves. What varies from one week to the next is the level chosen inside a graded step-by-step progression, and the clinician recalibrates it from whatever was written down last time.

It is worth stating the aim of the hour early, because it orders everything else. An exposure therapy session is built so that, at some point, what the patient predicts does not happen. That mismatch is what leaves a trace — not how long the patient holds out, and not the minutes spent with the stimulus in front of him.

Phase 1. How an exposure therapy session opens

An exposure therapy session opens with a short review of the week: sleep, alcohol intake, changes in medication, events that may have shifted the baseline. This is not courtesy. A patient who has slept four hours arrives with different reactivity, and the level planned for today may turn out to be badly calibrated.

The framing of the exposure therapy session comes next. The clinician states in operational terms what is going to be done, how long the contact will last, and what part each of the two will play. Then comes the prediction: what the patient believes will happen, with what probability, and with what concrete consequence. That expectancy is written down in the patient's own words, before anything begins. Without a written prediction there is nothing to set the outcome against at the end.

The third step is consent to the task, which is not the informed consent for the treatment as a whole. It is an agreement for the day: this step, this length of time, this exit criterion. The patient can negotiate it and can decline it. Declining is clinical information, not a failing. Nor does the clinician promise relief, because what is on offer is information about the thing the patient fears.

The framing closes with a detail that looks minor: the signals. It is agreed how the patient will flag something without breaking contact — a hand gesture, or a number said out loud — and which scale that number belongs to, a matter with a chapter of its own. Where the clinician sits is agreed as well. Sitting directly opposite and holding the patient's gaze turns the professional into a safety signal, and then what the patient learns is that he can tolerate the situation while accompanied.

Phase 2. Approaching the stimulus, and staying with it

In an exposure therapy session the approach begins at the agreed level and without ceremony. A long preamble carries a cost: anticipatory anxiety ends up taking over the hour. Once contact has begun, the work consists in sustaining it with attention on the stimulus rather than on the patient's own bodily sensations.

The framework that accounts for this phase today is inhibitory learning, formulated by Craske and colleagues in 2014 and updated towards inhibitory retrieval in 2022. Extinction does not erase the original fear association: it raises a competing association alongside it, one the patient has to be able to retrieve when the moment calls for it. Hence the engine of the work is expectancy violation and not the clock.

Habituation is still there as an observable phenomenon, and arousal usually does come down where contact is sustained. What has changed is the weight given to it. A within-session fall in anxiety during exposure predicts medium-term outcome poorly, so it does not serve as the success criterion for the hour. Measurement with subjective rating scales deserves separate treatment.

For most of an exposure therapy session the adversary is not anxiety but covert avoidance: distracting oneself, drawing the clinician into conversation, ruminating, squeezing an object inside a pocket, glancing sideways. The clinician intervenes little, and almost always to return attention to the stimulus and to the prediction. Withdrawal of safety signals, variability of the trial, and a change of context are options the protocol may build into the hour itself.

Phase 3. What the clinician records during an exposure therapy session

While the patient stays in contact, the clinician writes. The sheet for an exposure therapy session carries at least six columns:

• start and end time of the contact;
• the hierarchy level that was administered;
• the patient's prediction verbatim, exactly as it was said;
• what actually happened, in one line;
• covert avoidance behaviours observed;
• interruptions, with the time and the reason for each one.

Alongside what is said, the clinician notes what is not: posture, response latency, tone of voice, whether the gaze comes off the stimulus. That column is what later allows genuine time spent with the stimulus to be told apart from time spent with attention elsewhere. It is not written from memory once the hour is over; it is written while it happens.

The record is not defensive bureaucracy. The next exposure therapy session is decided with that sheet in front of the clinician rather than from recollection, and the same traceability supports the documentation of the session in the clinical record. NICE CG113 places exposure as a component of a cognitive behavioural treatment plan, never as a technique standing on its own. The record is what keeps the hour inside that plan.

Is it normal to feel anxiety during an exposure?

Yes, and that is the short answer expected by anyone who asks this before a first exposure therapy session. Anxiety during exposure is to be expected, and it is also the material the work is done with. An hour with no arousal at all usually indicates that the step chosen sits below what the case calls for.

The clinical criterion, though, does not look at intensity: it looks at function. The operative question is whether the patient is still processing the stimulus or has stopped taking information in. Inside the window of tolerance there is high anxiety and there is learning. Outside it there is arousal and no learning, and the hour stops paying its way.

A practical consequence follows from that. A rise in anxiety during exposure is not in itself grounds for stopping, and the clinician does not read every rise as an order to withdraw. What he does is sustain the contact and return attention to the stimulus with two or three short questions. What he does not do is reassure in mid-contact: a “nothing is going to happen” offered at that point works as a safety signal and takes the hour's learning with it.

The signal worth recognising is dissociation: a fixed stare, monosyllabic answers, a sense of unreality, blunting, stretches of the session the patient cannot recall afterwards. Under dissociation the stimulus is not processed, so nothing consolidates [CITATION TO VERIFY]. This is where tolerance of exposure and therapeutic dropout are decided. Signs of decompensation and sustained deterioration are a separate conversation, with criteria of their own.

When should an exposure be stopped?

Stopping an exposure is never a matter of improvisation. The exit criterion of an exposure therapy session is fixed before the start; it is not negotiated with a patient who is already activated. In practice there are four cases: the agreed criterion is met, the patient withdraws consent to the task, sustained dissociation or signs of decompensation appear, or a physical incident occurs.

Stopping by criterion and escaping are not the same thing, however alike they look from outside. An agreed stop is orderly, it is named out loud, and it is written down. Flight is abrupt, it is decided by the arousal of the moment, and it tends to confirm for the patient the very contingency he had been fearing [CITATION TO VERIFY]. Telling the two apart in the moment is half the craft.

Once the exposure has been stopped, the minutes that remain are not empty minutes. The usual course is to drop a level and repeat a step already consolidated, so the hour does not end at the point of maximum flight, and to put the exact time of the interruption in writing. After that the closure is conducted as in any other exposure therapy session, with the prediction reread and the task agreed.

Stopping an exposure well has a shape to it. The clinician marks the end in words, brings attention back to the room, removes the equipment slowly, and rereads the prediction before closing the sheet. A stopped session is not a failed session: it informs the calibration of the challenge. DSM-5-TR describes specific phobia in terms of disproportionate fear and of the avoidance that maintains it, and that second part explains why the manner of stopping weighs so heavily.

Phase 4. Session closure, the final reframe and the between-session task

The closure of an exposure therapy session is a phase with content, not a farewell. It opens by rereading the prediction noted at the beginning and setting it against what happened, and the comparison is formulated by the patient. The clinician resists summarising it himself: the revision of probability only counts where it comes from whoever held the expectancy.

There is one final reframe worth avoiding, the “you see? nothing happened”. That is reassurance, and reassurance shuts the door the hour had just opened. Session closure states a datum revised by the patient, labels the affect precisely, and agrees a retrieval cue: a word, an image, a gesture that carries the learning out into the street.

The between-session task moves the learning into other contexts, which is where the competing association has to be retrievable. It is agreed in writing, bounded in time and place, and recorded exactly like the consulting-room hour. It is a self-administered task inside a plan agreed with the clinician in charge, never an exercise taken on by the patient alone.

The last stretch of an exposure therapy session is administrative and clinical at the same time: a few minutes in the room, water, the note in the record, the next appointment. Anyone weighing up how this procedure fits their own practice can book a demonstration to see how a hierarchy is configured inside the system, or look at how independent psychologists run exposure work in a single-clinician practice.

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.