How to Run Experiments on Your Own Nervous System

A practical protocol for recalibrating your intuition about public speaking.

Feb 21, 2026

A close up of a thermometer on a wooden wall — Photo by Eric Prouzet on Unsplash

In a previous essay, I argued that your pre-talk “intuition” — that sinking certainty that this is going to go badly — isn’t a mystical warning system. It’s a prediction, generated by your brain based on bodily signals filtered through learned associations.

The good news: predictions can be updated. The associations that make your gut catastrophise can be retrained through deliberate practice.

This essay is the how. Not theory, but protocol — a set of concrete experiments you can run on your own nervous system to build a more accurate internal compass for speaking.

The core principle: exposure as data collection

You’ve probably heard that exposure helps with anxiety. Face your fears, do the thing, eventually it gets easier. This is true as far as it goes, but “just do more presentations” is about as useful as “just be more confident”.

The research on exposure-based treatments for public speaking anxiety (including virtual reality and video self-modelling approaches) shows reliable reductions in fear and avoidance. But here’s what makes the difference between exposure that works and exposure that just feels like suffering: how you frame what you’re doing.

Acceptance-based exposure — the kind informed by ACT principles — tends to outperform pure habituation (”gut it out until you stop feeling scared”) on longer-term outcomes. The difference isn’t the exposure itself but the stance you take toward what you’re experiencing.

So instead of treating each speaking situation as an ordeal to survive, treat it as a data collection exercise. You’re not trying to prove you can handle it. You’re trying to find out what actually happens when you speak while anxious — and whether your gut’s predictions match reality.

The protocol: four interlocking practices

What follows isn’t a rigid program but a set of practices you can adapt to your own speaking situations. The goal is to generate the “error signals” your intuitive system needs to recalibrate.

1. The pre-talk snapshot

Before any speaking situation — even low-stakes ones like a meeting comment or a classroom question — take 30 seconds to capture your gut’s prediction.

Ask yourself:

If I trust my gut right now, what does it predict about how this will go?
On a 0-100 scale, how anxious do I feel?
What’s my gut telling me will happen? (Be specific: “I’ll blank and people will stare” rather than just “it’ll be bad”)
What bodily sensations am I noticing? (Racing heart, tight throat, churning stomach — just note them without interpretation)

Write this down, even in shorthand. The act of externalising makes the prediction concrete and testable.

2. The body scan calibration

This one builds interoceptive accuracy directly. Werner’s research found that people with more accurate body-sensing reported less speaking anxiety — not because they felt less, but because they interpreted what they felt more accurately.

The drill: Before low-stakes speaking situations, do a quick scan of five areas: heart, breath, throat, stomach, hands. Rate each 0-10 for activation intensity. Don’t try to change anything — just notice and rate.

Afterward, compare your ratings with how the situation actually went. You’re looking for patterns: Does my throat-tightness rating predict anything useful? Or does it spike regardless of outcome?

Over time, this trains you to read your body as a meter rather than an alarm. High activation stops meaning “disaster imminent” and starts meaning “system is online.”

Bonus drill: At rest and after light movement (a brisk walk, climbing stairs), estimate your heart rate for 30 seconds, then check against a device or pulse count. The goal isn’t a lower number — it’s a more accurate estimate. This builds the basic interoceptive skill that transfers to speaking situations.

3. The post-talk reality check

This is where the recalibration actually happens. After the speaking situation, capture:

Actual anxiety level (0-100) during the talk — often lower than anticipated
What actually happened? (Did you blank? For how long? Did anyone visibly react? Did you make your key points?)
Any surprises? (”I felt 8/10 anxious but still got through all three points”)
How does this compare to my pre-talk prediction?

Be honest and specific. The point isn’t to spin a positive story but to collect accurate data. Sometimes your gut will be right — some talks do go badly. But you’re looking for the pattern across many data points, and for most anxious speakers, that pattern reveals systematic over-prediction of catastrophe.

When there’s a big gap between prediction and reality, note it explicitly. This is the error signal your intuitive system needs.

4. The language shift

This one’s subtle but powerful. When you notice anxious sensations, pay attention to how you’re labelling them — internally or out loud.

“My heart is pounding, I’m panicking” is an interpretation, not an observation. The observation is just: “My heart is pounding.”

The practice: when you catch yourself in catastrophic labelling, try a neutral or functional relabel. Not fake positivity (”I’m excited!”) but accurate re-description:

“I’m activated” instead of “I’m panicking”
“My system is preparing” instead of “I’m falling apart”
“This is adrenaline” instead of “This is proof I can’t handle this”

You’re not lying to yourself. Adrenaline is what’s happening. Your system is preparing. The catastrophic labels are the interpretive layer your intuition has learned to add automatically — and that layer can be revised.

Building your hierarchy

For this protocol to work, you need speaking situations to practice on. If your only exposure is high-stakes presentations twice a year, you won’t generate enough data points for meaningful recalibration.

List speaking situations from lowest to highest stakes. Be creative about the low end:

Asking a question in a meeting
Making a comment in a seminar
Leaving a voice memo for a friend
Recording yourself talking through an idea (no audience)
Speaking up in a casual group conversation

Each of these is an opportunity to run the snapshot → scan → speak → reality-check loop. The reps matter more than the intensity.

The stance that makes it work

None of this works if you approach it as a performance to be judged. The protocol requires a scientist’s stance: curious, observational, genuinely interested in what the data shows.

This is where acceptance-based principles come in. Research by England and colleagues found that baseline mindfulness predicted more positive self-statements about speaking after exposure training. How you attend to your experience shapes what you get from practice.

So: notice the anxious thoughts and sensations, but treat them as passing events to observe rather than commands to obey. Practice willingness statements when you can — “I’m willing to feel this activation in order to say something that matters” — not as affirmations but as stance-setting.

You’re not trying to feel calm. You’re trying to feel accurately — to let your gut update its predictions based on what actually happens when you speak while activated.

What to expect

Recalibration isn’t instant. Your intuitive system has been trained by years of catastrophic associations; it won’t revise overnight.

But if you run this protocol consistently across a range of speaking situations, you should start to notice shifts within a few weeks: pre-talk predictions becoming less apocalyptic, post-talk gaps between expectation and reality narrowing, bodily sensations feeling more like familiar activation and less like emergency.

Your gut will still have opinions about speaking. The goal is for those opinions to be based on better data.

Sven Brodmerkel (PhD) | Off-Script At Work

Discussion about this post

Ready for more?