Daniel sent us this one — he wrote in after our driving phobia episode and it turns out he's been living exactly what we described. Grew up driving in Ireland, loved it, totally at ease. Moved to Israel, and now every time he gets behind the wheel after a gap of months, it's white-knuckle terror for the first ten minutes. Then his brain builds that temporary safety model, and suddenly the second and third drives are fine. But the fear comes back every time there's a long break, and now he's avoiding driving altogether, which he can't really afford to do. He's got two specific questions. One: what does "often enough" actually look like for making the fear extinction stick? And two: if you're sitting in a car park before driving, what can you do to prime your brain during those critical first ten minutes so the stress response decays before you're on the road?
That ten-minute window he described is textbook. The amygdala is screaming threat, the prefrontal cortex gradually builds a safety signal, and somewhere around the ten-minute mark, the safety signal wins. But here's what makes his case so interesting — the fact that drives two and three were uneventful, that's the extinction taking hold within a session. And then months pass and it's gone. That's the "return of fear" phenomenon, and it's one of the most well-documented frustrations in clinical anxiety research.
The brain learns safety, and then just... That seems like a terrible design.
It does, but from an evolutionary standpoint it's actually pretty sensible. If you learn that a predator isn't in a particular watering hole, that's useful for about a week. If you go back six months later, the prudent move is to be vigilant again. The brain treats a threat memory as the default and a safety memory as the update — and updates expire unless they're refreshed. What Daniel experienced is exactly what a 2015 meta-analysis by Vervliet and colleagues documented across dozens of studies. After a single extinction session, the return of fear rate is somewhere between forty and sixty percent. That's not a failure of treatment. That's just how the brain does extinction.
Forty to sixty percent relapse after one session. So Daniel's one good ten-minute drive — followed by months of nothing — was never going to hold.
And this is where the post-pandemic driving pattern makes this so widespread. People who used to commute daily now drive once every few weeks. Their extinction learning never gets consolidated. Every drive is essentially a first drive, neurologically speaking.
That hypervigilance in the first ten minutes — Daniel called it the most dangerous part of being a periodic driver. Is he right about that?
He's absolutely right. When the amygdala is in full alarm mode, it actually impairs hazard detection. The brain floods with false alarms, and you get tunnel vision and what researchers call auditory exclusion — you literally stop hearing ambient sounds properly. Sirens, horns, a motorcycle in your blind spot. The very things that keep you safe become harder to detect precisely when you need them most. It's not just uncomfortable. It's a genuine safety risk.
The fear that's supposed to protect you from danger is actually making you more dangerous.
That's the cruel irony of driving phobia. The hypervigilance spike doesn't sharpen your perception — it narrows it. You're scanning for threats so intensely that you miss the actual threats. Daniel's instinct to find a way to let that stress decay before he's in traffic is exactly the right instinct.
The first question he's asking — what does "often enough" look like — that's the practical anchor of this whole thing. Because "drive more often" is useless advice if nobody tells you what the actual frequency needs to be.
The answer isn't what most people would guess. The intuitive approach is "drive every day until the fear goes away," but that's actually less effective than spacing it out. Michelle Craske and her team ran a clinical trial in 2008 with a hundred participants, testing massed exposure — daily sessions — against spaced exposure, every two to three days. The spaced group had significantly better long-term extinction retention. Massed practice gave faster short-term results, but the relapse rates were higher. The brain needs the consolidation gaps between sessions to properly encode the safety learning.
That's counterintuitive. Most people would think more is better. Just hammer it until it sticks.
It feels like it should work that way. But extinction learning isn't like building a muscle. It's more like letting grout set between tiles. If you keep disturbing it, it never cures. The consolidation happens in the gaps.
Daniel's asking what "often enough" means, and the answer starts with "every two to three days, but not every day." That's already not what most people would expect.
That's just the initial phase. The Vervliet -analysis showed that after six to eight spaced sessions, the return-of-fear rate drops below twenty percent. That's the threshold where you can start spacing things out further — maybe once a week for maintenance. The key is never letting the gap stretch beyond the point where the fear returns to baseline, which for most people seems to be around five to seven days.
Daniel's months-long gaps are basically resetting the counter to zero every time. He's doing the hard work of extinction on each drive, and then letting it decay completely before the next one.
And that's exhausting and demoralizing, because each time he gets in the car, it feels like he's made no progress. But he has made progress — the fact that the second and third drives were uneventful shows the extinction mechanism is working fine. It's the spacing that's the problem, not his brain.
Which is actually encouraging. The machinery works. It just needs the right schedule.
That's the good news. The bad news is that for someone who doesn't need to drive daily, creating that schedule requires deliberate effort. You're not going to naturally drive every two to three days if you work from home or live near a bus line. You have to manufacture reasons to get in the car.
That's where the second question comes in. If you're going to manufacture those drives, what do you do in the car park before you turn the key? Because those first ten minutes of hypervigilance are the barrier. That's the wall Daniel keeps hitting.
There's solid research on this. A 2018 study by Schmidt and colleagues looked specifically at pre-drive diaphragmatic breathing for driving phobia. Two minutes of box breathing — four seconds in, four hold, four out, four hold — in the parked car reduced self-reported anxiety by an average of 2.3 points on a ten-point scale within three minutes. That's a meaningful drop. It's not going to eliminate the fear, but it brings it down from "I can't function" to "I can manage this.
3 points in three minutes. That's a lot of return on a very small investment.
The mechanism is straightforward. Diaphragmatic breathing activates the parasympathetic nervous system. It's literally signaling to your body that you're not in immediate danger. The amygdala can't maintain full alert when your breathing is slow and controlled — the physiological signals contradict the threat assessment.
You're hacking the body to convince the brain it's safe, rather than trying to reason your way out of panic.
Which is the right order of operations. You can't do cognitive work when your amygdala is in charge. You have to bring the physiological arousal down first, then engage the prefrontal cortex. Which is why the second technique Daniel should try is reality testing. Once the breathing has taken the edge off, you ask yourself: what exactly am I afraid of right now? And you rate the actual probability on a scale of one to ten.
Because the fear is usually vague and catastrophic. "Something terrible will happen." Naming it forces specificity.
"I'm going to cause a collision" — okay, what's the actual probability of that on this quiet residential street at eleven in the morning? Probably a two. "Someone's going to honk at me and I'll panic" — what's the actual consequence of being honked at? You feel embarrassed for thirty seconds and then it passes. The prefrontal cortex is good at probability assessment, but it only works if you give it a specific question.
The third thing — Daniel mentioned just sitting in the driver's seat, adjusting mirrors, touching the wheel. Is there actual research behind that, or is it just common sense?
There's a basis for it in what's called sensory grounding. Narrating your environment — "I am in a parked car, the engine is off, I am safe, my hands are on the steering wheel, the parking brake is on" — that activates the prefrontal cortex and pulls processing away from the amygdala. It's similar to techniques used for panic disorder. You're anchoring attention in factual observations rather than catastrophic predictions.
The pre-drive protocol is basically three steps. Breathe for two to three minutes until the physiological arousal drops. Then reality-test the specific fear. Then ground yourself in the actual sensory environment of the parked car. And only then turn the key.
The order matters. Don't try to reality-test while your heart is pounding. Don't try to ground yourself while your thoughts are racing. Each step sets up the next.
Daniel also mentioned something interesting — that he reckons the hypervigilance panic response is the most dangerous part of being a periodic driver. And you confirmed that with the tunnel vision and auditory exclusion piece. But there's another layer here, which is that periodic drivers often develop what researchers call safety behaviors — subtle things they do to feel safer that actually maintain the fear.
This is the Salkovskis work from 1999, and it's crucial. Safety behaviors are things like gripping the steering wheel with white knuckles, driving well under the speed limit, checking mirrors obsessively, avoiding left turns. They feel protective, but what they actually do is prevent the brain from learning that the environment is safe on its own. The brain attributes safety to the behavior — "I survived because I was gripping the wheel" — rather than to the driving context itself.
The safety behavior becomes a crutch, and the fear never extinguishes because the brain never gets the data point that says "nothing bad happened even though I was driving normally.
And the research shows that deliberately dropping those safety behaviors during exposure sessions — what Salkovskis called safety behavior fading — accelerates extinction significantly. You have to let the brain experience safety without the crutch.
Which means that by drive three or four of Daniel's retraining schedule, he should be identifying and deliberately releasing at least one safety behavior per drive. Loosen the grip. Take a left turn. Drive the actual speed limit.
That's going to feel worse before it feels better. The anxiety will spike when you drop the safety behavior. But that spike is the extinction learning happening. The brain is getting the data it needs: "I drove normally and nothing bad happened.
We've got a pretty clear picture now. The spacing needs to be every two to three days for the first two to three weeks, aiming for six to eight sessions to push the return-of-fear risk below twenty percent. The pre-drive protocol is breathe, reality-test, ground. And by session three or four, start dropping safety behaviors.
I want to emphasize one thing about the spacing. The goal isn't distance — it's frequency. A five-minute loop around the block counts. A ten-minute drive to the supermarket counts. Daniel doesn't need to plan epic road trips. He just needs to get in the car and drive for a few minutes, frequently enough that the safety learning doesn't decay.
That's actually a huge psychological relief. The task isn't "become a confident driver." The task is "do a five-minute loop every couple of days." That's manageable.
It's measurable. Put it on a calendar. Check it off. Each check mark is a data point for your prefrontal cortex: "I did this thing, and I survived, and it was fine." The calendar becomes a record of evidence against the fear.
The one thing we haven't addressed is what Daniel mentioned about the car park — that many people in Israel don't have one. If you're parked on a busy street, sitting in the car doing box breathing for three minutes while traffic whizzes past might not feel like a safe priming environment.
That's a real constraint. And it's worth exploring whether a five-minute walk around the block before getting in the car could serve the same physiological function. The breathing and reality testing can be done anywhere. The sensory grounding might need to adapt.
That's probably a whole other question. For now, I think we've given Daniel a pretty concrete answer to both of his. Often enough means every two to three days, six to eight sessions, then once a week for maintenance. And the pre-drive protocol is breathe, reality-test, ground — in that order, before you turn the key.
The underlying message is that his brain is working exactly as designed. The extinction mechanism is there. The return of fear isn't a personal failure — it's a well-documented neurological phenomenon with a known fix. He just needs the right schedule.
Which is somehow both reassuring and annoying. Reassuring that it's not broken. Annoying that the fix requires deliberate, scheduled effort rather than just wanting the fear to go away.
The brain doesn't care what you want. It cares what you do, and how often you do it. Fear extinction isn't a one-time fix, it's a learning process. The brain builds a temporary safety model during each drive, but without enough repetition, that model decays right back to the default threat model.
It's less like learning to ride a bike and more like learning a language. You can't do one Duolingo session and expect it to stick for six months.
That's a perfect comparison. The safety learning is context-dependent and it degrades without reinforcement. Every time Daniel takes a months-long break, the threat model reasserts itself because from the brain's perspective, the absence of driving means the safety update might no longer be valid.
Which is why his second and third drives were fine in the moment, but the whole thing reset after the gap. The learning happened, it just wasn't consolidated.
And consolidation requires both repetition and spacing. So Daniel's two questions — what does "often enough" look like, and what can you do in the car park to prime the brain — are really two sides of the same problem. One is about building the extinction so it lasts. The other is about lowering the barrier to entry so you can actually do the building.
The barrier to entry is no small thing. If every drive starts with ten minutes of panic, you're going to avoid driving. That's not weakness, it's basic behavioral economics. The cost is too high.
Which is why the pre-drive priming isn't a nice-to-have, it's essential. If you can knock that initial anxiety spike down from an eight to a four before you're in traffic, the whole experience changes. You're not white-knuckling through the extinction learning — you're actually present for it.
Let's make this concrete. Here are three things Daniel — or anyone in his situation — can do starting today.
First, make a calendar. The next three weeks, schedule six to eight drives, spaced every two to three days. A five-minute loop around the block counts. The goal is frequency, not distance. Mark each one on the calendar and check it off when it's done.
The checking-off part matters more than people think. Each check mark is evidence. The prefrontal cortex needs data, and a calendar full of "I drove and nothing bad happened" is exactly the kind of data that extinguishes fear.
Second, before every drive, spend two to three minutes in the parked car doing box breathing and reality testing. Do not start the engine until your heart rate drops below eighty beats per minute, or your subjective anxiety is below a four out of ten. The Schmidt data shows this can lower anxiety by over two points in three minutes. That's the difference between white-knuckling through the first ten minutes and actually being present for them.
Third — this is the one that'll feel worst before it feels better — identify one safety behavior you rely on and deliberately drop it on the third or fourth drive of the retraining phase. If you're gripping the wheel like it's a life raft, loosen your hands. If you're avoiding left turns, take one. The Salkovskis research shows this is what forces the brain to attribute safety to the environment rather than the crutch.
The sequence matters here. First you establish the rhythm with the breathing and the short drives. Then, once you've got a few sessions under your belt and the extinction is starting to consolidate, you push it further by dropping the safety behavior. Don't try all three on day one.
Which is the kind of thing I'd do, and then wonder why I felt worse.
And then you'd take a nap about it.
I'd take a nap regardless. But the point stands — this is a progression, not a checklist to blitz through. Build the schedule, learn the pre-drive protocol, then start stripping away the crutches.
One thing Daniel mentioned that we didn't quite get to — he acknowledged that having a car park to sit in is a luxury a lot of people in Israel don't have. If you're parked on a busy street, cars whipping past, someone honking because they want your spot, sitting there doing box breathing for three minutes isn't exactly a calming exercise.
That's a real constraint. The breathing and the reality testing are portable — you can do those walking. What you lose is the sensory grounding in the car itself, which means the transition from "I'm safe on the pavement" to "I'm in the driver's seat" still has a step change. The anxiety will spike when you sit down.
The walk helps but doesn't replace the in-car protocol.
It lowers the starting point. If your anxiety is at an eight when you approach the car, a five-minute walk with breathing might bring it to a five. Then you get in and it jumps to a seven, but a seven is better than an eight. And you can do a shortened version of the grounding once you're inside — thirty seconds of narrating, a few more breaths — before you turn the key. It's not ideal, but it's workable.
Which feels like a theme of this whole topic. None of this is ideal. The ideal is not having the phobia in the first place. But the workable version is right there.
The other question worth leaving people with — we're talking about driving phobia in 2026, and autonomous driving is creeping into the mainstream. In ten years, will we still be having this conversation? Or will driving phobia become a niche problem because fewer people drive manually at all?
I suspect the phobia just shifts. Instead of "I'm going to cause a crash," it becomes "I'm going to be in a crash I can't prevent because the AI is in control." That's a different flavor of the same fear — loss of agency.
We already see early versions of that with people who refuse to use adaptive cruise control or lane-keeping assist. The fear of ceding control to a machine might turn out to be harder to extinguish than the fear of making your own mistakes.
Something to revisit when Daniel's son Ezra is learning to drive — or learning to trust the car to drive itself.
Now: Hilbert's daily fun fact.
Hilbert: In the 1780s, a unique sign language emerged among deaf communities on the Cape Verdean island of Santiago, developed in near-total isolation from European sign languages. If you were to convert its documented vocabulary into modern American Sign Language equivalents, roughly seventy-three percent of its signs would register as entirely novel — meaning a modern ASL user would understand about one in four gestures and be completely lost on the rest.
A deaf person from eighteenth-century Cape Verde walks into a room with a modern ASL user, and three-quarters of what they sign is gibberish to each other.
That's actually a pretty good metaphor for crossing cultures in general.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop. If you've got a question like Daniel's — something you're genuinely wrestling with — email the show at show at my weird prompts dot com.
We'll be back next week.
