Writing an Imaginal Exposure Script is harder than most people expect
Most clinicians rush through this step because it feels administrative. It isn't. The script becomes the core stimulus you expose the patient to repeatedly. If it's vague, poorly paced, or emotionally flat, the whole exposure session fails regardless of how well you handle the SUDS tracking. I've watched competent therapists waste three sessions on a single scene because they wrote a script that read more like a summary than a sensory experience.How to Build an Imaginal Exposure Therapy Script That Actually Works
Start with the worst memory or feared scenario you've identified in case formulation. Write it in first person, past tense, present continuous feel. Not "he walks into the room" but "I walk into the room and my hands are already shaking." The shift from third to first person is not stylistic decoration. It's the difference between describing an event and re-experiencing it. I had a patient with complex PTSD where the trigger wasn't a single event but a recurring pattern of humiliating interactions at work. The first script I drafted was a composite of three incidents. It was logically coherent and completely ineffective. Subjective Units of Distress stayed at a 2 out of 10 throughout the entire 45-minute session. The problem was that the script was too generalized. Nothing landed. The workaround was brutal but straightforward. I pulled the most specific humiliating moment from that pattern and wrote a new script around it alone. Every detail mattered. The fluorescent light buzzing overhead. The particular tone of the manager's voice. The exact words that were said. When we ran that single-scene script, her SUDS peaked at 85 on the second run. Same patient, same therapist, completely different outcome because the script had surgical precision instead of broad coverage.
Here's what the structure actually looks like in practice. You open with grounding context. Where are you. What time of day. What's happening immediately before the feared event begins. Then you move into the event itself with full sensory detail. Sounds, textures, body sensations, internal dialogue. End with the aftermath. The uncomfortable part is including the emotional and physical response you had or would have at the time. Not just what happened to you but what you felt inside your body. The script should be long enough to sustain 20 to 40 minutes of repeated listening. Most people write 600 to 1,200 words. Shorter scripts finish too quickly and the habituation curve never gets a chance to work. Longer scripts tend to lose emotional intensity toward the end because the narrator's voice fatigues. I usually aim for around 900 words as a sweet spot. A few technical points that matter more than people realize. Record the script yourself if possible. A recorded voice attached to the memory creates stronger conditioned responding than text alone, but a generic audiobook voice feels disconnected. Use a neutral tone. Don't overact. The voice should sound like you're describing something unpleasant, not performing a drama. Add pauses marked in the text for natural breathing. Readers rush through emotional content. The pauses force the pace down to something survivable.
The biggest mistake I see is writing scripts that avoid the worst part. Therapists get uncomfortable around the trauma peak and accidentally edit it out. They describe the approach to the feared situation but skip the confrontation, the, the humiliation, the impact. This is avoidance dressed up as clinical sensitivity. The exposure has to hit the core fear. If the patient is avoiding the memory, the script needs to remove that escape route. Another counter-intuitive point. Repetition matters more than variation. Run the same script multiple times in a single session until SUDS drops by at least 50 percent from the first run to the last run. Most patients report that the third or fourth run feels noticeably less intense even though the content hasn't changed. That's habituation. Switching to a slightly different version of the script mid-session resets the learning. Stick with the same words each time.
Get the Full Details

Pitfalls and Where This Method Breaks Down
Imaginal exposure through scripted recording doesn't work for everyone. Patients with severe dissociation may go blank during the exercise and gain nothing from it. The script becomes background noise. In those cases, grounding techniques need to be established first, or you should consider in vivo exposure or EMDR as alternatives depending on the diagnosis. There's also the risk of script dependency. Some patients become so attached to hearing their exact recording that they can't tolerate live imaginal exposure with the therapist. They need the recording to feel safe. This isn't inherently wrong but it narrows your clinical flexibility later. I usually introduce live narration alongside the recording within the first few sessions to prevent this bottleneck. Writing the script is also time-consuming. A thorough, sensory-rich script takes me about 45 minutes to an hour on average. Not every patient needs that level of detail. Simple phobias or specific trauma memories can work with shorter 300-word scripts. Complex PTSD requires the longer version. Judge the complexity of the material and match the script depth accordingly.
If you're looking for a template structure to start from, the components are fairly standard. Opening context paragraph. Build-up paragraph. Peak fear paragraph. Aftermath paragraph. Internal response paragraph. That's five paragraphs and roughly 700 to 900 words total. Expand any section that feels emotionally thin. Cut sections that read like summaries rather than experiences. The download link question comes up often. There's no single official version of an Imaginal Exposure Therapy Script because each one is patient-specific by definition. You'll find downloadable templates online that give you the five-paragraph structure. Those are starting points, not finished products. The actual script has to be written for the individual's memory and fear hierarchy.