When Should You Trust Regression?

Short answer: you can’t verify the historical accuracy of a recovered memory from inside the session. No practitioner can. And anyone who tells you otherwise is selling something. What you can do is run the regression in a way that lowers the odds of contaminating it, read the signs that a memory arrived organically rather than on cue, and be honest with your client about what hypnosis can and can’t do. Trust isn’t a switch you flip at the end. It’s a series of choices you make before you ever say “go back to a time when…”

Hypnosis can increase the ability to visualize more vividly. And for many people, “vivid” is a submodality that means “real.” If a person remembers, or imagines, a scene vividly, it can seem more real to them. So can we trust thoughts that come up as a result of regression techniques? Or could they simply be imaginings that seem real?

The honest answer is: not so clear. Let me walk you through how I think about it.

Why Hypnosis Makes Memories Feel Real, Not Necessarily Accurate

There’s a popular belief, and a lot of clients walk in with it, that hypnosis works like a mental video recorder. Hypnotize someone, rewind the tape, and out comes the buried footage exactly as it happened. It’s a great story. It’s also not how memory works.

The research on hypnotic hypermnesia, the idea that hypnosis unlocks accurate memories that normal recall can’t reach, is weak. What hypnosis reliably increases is confidence and the sheer volume of recall. It does not reliably increase accuracy, and it can increase error right alongside everything else (Leo, Bruno & Proietti, 2025). So your client comes out more certain about more details, some of which may be inventions. That combination, high confidence plus mixed accuracy, is exactly the thing that gets people into trouble.

THAT’S THE POINT. Vividness and certainty are not evidence. They feel like evidence. They aren’t.

Putting The Conscious Mind In Neutral

Perhaps you’ve done an arm levitation, either as a subject or a practitioner. You suggest an arm is getting lighter, and you’re not sure which finger will start to float first. Or maybe the wrist comes up. You’ve heard the eye-closure phrasing: “Don’t make it happen, don’t stop it from happening. Just concentrate on it and allow it to happen.”

We have to put the conscious mind in neutral for a bit and allow other parts of the mind to move things forward. That’s true for phenomena like arm levitation, and it’s true for regression.

Involuntary And Voluntary

If we can’t get the conscious mind to sit out, it will influence things. And that’s generally not what we’re going for when we hypnotize folks. We want to bring up the parts of the mind in charge of involuntary behavior, and let the parts in charge of volition take a breather. Why? Because we want to access involuntary material directly. Routing everything through the conscious mind is inefficient. And worse, the conscious mind taking the steering wheel, even a little, can derail the whole thing.

So can biases, prejudices, and existing beliefs.

How Regressions Get Contaminated

Regression wants the mind open and expectant, but not biased. Preconceived ideas interfere. They lead the process, and we want the process to be organic and neutral.

People are biased toward whatever theories they’ve already been exposed to. If a client believes their trouble could have been caused by alien abduction, they may “remember” an alien abduction, or they could have simply seen a movie about it. If they’ve read that anorexia is often a symptom of sexual abuse, it’s easier for them to imagine a scenario in which that happened. What already exists in the mind can shape what comes up. Even a stray idea can muddy the water.

And that’s before we add the therapist’s influence and the influence of the context, both of which can contaminate a regression fast. This isn’t a fringe worry. A recent scoping review found that therapists who believe memories of trauma are commonly repressed are more likely to use suggestive techniques, and those techniques are precisely the ones that generate false memories (Otgaar et al., 2025). Your beliefs leak into your language. Your language leads the client.

Since everyone has thoughts about what may have caused a problem, you can’t completely guard against preconceived ideas contaminating a regression. But in many cases you can do a lot to minimize the contamination. Most of that work happens before the trance.

Past-Life Regression And Age Regression Are Not The Same Risk

People lump these together. Don’t.

In the classic experimental work on this, Nicholas Spanos showed that the content of hypnotic past-life “memories” tracks the subject’s expectations and prior beliefs about reincarnation, not their hypnotizability and not any evidence of a prior existence (Spanos et al., 1991). Tell subjects that past lives are usually a different sex or a different culture, and their “past lives” oblige. The subject isn’t lying or play-acting; the experience feels completely real to them. But a real experience and a real memory are two different things. What they describe is being constructed on the spot out of their own expectations, not retrieved from anywhere.

Why does this matter to you as a practitioner? Because it’s a clean, well-studied illustration of the mechanism you’re up against in every regression: expectation shapes content. Past-life work makes it obvious. Age regression to a real childhood hides the same mechanism behind a plausible timeline. Same current running beneath both. One just has better camouflage.

So, Should Recovered Memories Be Believed?

I think that’s the wrong question.

I think recovered memories should be taken seriously. And clients should be informed that hypnosis can cloud the issue: anything surfacing in trance could be a real memory, or an imagining that seems real. Hypnosis is not truth serum. It’s not a magical “accurate memory” pill.

This isn’t just my opinion as a practitioner. The American Psychological Association’s long-standing guidance on memories of childhood abuse puts it plainly: without other corroborating evidence, there is no way to distinguish a true memory from a false one based on the memory alone (APA). Not by vividness. Not by emotional intensity. Not by how sure the client is. The thing that moves the needle on trust is corroboration from outside the session, and it’s not something you produce in your office.

The Canadian Psychological Association’s guidance for practitioners lands in the same place from the ethics side: you shouldn’t walk in with a verdict already formed: not “this abuse happened” and not “this is all confabulation.” Approaching with a preconceived conclusion, in either direction, is where competence breaks down (CPA).

Screening And Informed Consent Before You Regress Anyone

Here’s the part a lot of “regression technique” content skips entirely, and it’s the part that keeps you and your client safe.

Screen first. Some clients are poor candidates for exploratory regression:

• Anyone who walks in wanting to recover a specific memory. If a client’s goal is “prove that X happened to me,” I usually pass, or I’m extremely careful. The demand itself is a contamination risk.
• Clients with a pending or possible legal matter (more on that below).
• Clients who arrive already certain of the cause and want you to confirm it. You’re being recruited to co-sign a conclusion.

Get real informed consent. Not a signature on a form. It’s an actual conversation. The client should understand, in plain language, that hypnosis can increase both accurate and inaccurate recall, that feeling certain is not the same as being right, and that anything that comes up may need outside corroboration before it’s treated as fact. If you can’t have that conversation comfortably, you’re not ready to do the technique.

A note on scope: this article is training and informational content for practitioners. It is not a substitute for supervised clinical training, and it doesn’t override the licensure, scope-of-practice, and consent requirements in your jurisdiction. Working with trauma memories in particular sits close to, and sometimes inside, the practice of psychotherapy. Know where that line is for you.

Legal And Forensic Exposure

If there’s a legal matter pending, or a potential one, you shouldn’t regress a person without considering the legal consequences to them and to yourself.

In many U.S. jurisdictions, testimony that has been “refreshed” through hypnosis is restricted or excluded outright, precisely because of the false-memory and suggestibility problems above. The rules vary by state and by court, so I’m not going to hand you a blanket statement. The point is that hypnotically-influenced recollection can taint a witness’s testimony and, in some cases, make it inadmissible. There’s also malpractice exposure: therapists have been successfully sued over recovered-memory work that a court found to be suggestive.

You don’t need to be a lawyer. You need to know enough to stop, refer out, or get proper legal and clinical guidance the moment a case has a legal dimension. When in doubt, don’t regress.

How I Like To Use Regression: The Affect Bridge

With all of that said, regression is a genuinely useful tool when you use it for change work rather than as an evidence-gathering expedition. My go-to is the affect bridge, which uses an emotion as a breadcrumb trail back to the earliest, most charged memories carrying that feeling.

Notice what it is not: it’s not “let’s go find out what terrible thing happened to you.” It’s tightly scoped to a specific feeling connected to a specific presenting problem. Here’s the shape of it:

1. Identify the triggering emotion. A client comes in to stop smoking (or nail-biting, or a phobia; the content doesn’t matter). We find the specific emotion that fires off the behavior. Say it’s a particular flavor of anxiety.
2. Amplify and hold the feeling. Have them get in touch with that exact feeling, cleanly, without a story attached yet.
3. Follow it back. Let the feeling lead. Suppose it traces back to age 16, a big test at school. They went out afterward, had their first cigarette, and felt a rush of freedom.
4. Do the change work. Now we use whatever technique fits: anchoring better feelings of freedom to a new behavior, or reprocessing a negative memory to drain its charge.

If you want the emerging response to come from the client rather than from you, ideomotor signaling is a cleaner elicitation channel than asking out loud. I’ve written about that separately, and it pairs well with this. [internal link: ideomotor responses post]

But here’s the important bit: we didn’t go on a witch hunt. We went after a specific emotion, in a limited context. That scoping is itself a contamination control. The narrower the search, the less room for you to accidentally write the story.

Signs A Regression Was Lower-Risk, Not “Proof It’s True”

I’m going to choose my words carefully here, because this is where practitioners get themselves in trouble.

There is no test, none, that tells you from inside the session that a recovered memory is historically accurate. So treat what follows not as a validity test, but as risk-reduction indicators: signs the memory arrived organically rather than being led, cued, or manufactured. They lower your worry. They don’t prove the memory.

Researchers describe recovered memories as sitting on a continuum, from least trustworthy when they surface inside a suggestive, expectation-loaded context, to more trustworthy when they arrive spontaneously, triggered by an ordinary cue rather than by someone hunting for them (Dodier et al., 2023). The signs below are what the “arrived organically” end of that continuum tends to look like in your office:

Note the trap in that right-hand column: “amnesia” or “repression” framing correlates with the highest confabulation risk. In experimental models of hypnotic confabulation, suggesting that a memory is blocked and then lifting the block produces the most temporal confusion and invented detail (Cox & Barnier, 2015). If your process leans on “it’s all locked away and we’re going to unlock it,” you’re building the conditions for a false memory, not avoiding them.

Frequently Asked Questions

Can hypnosis recover forgotten memories accurately?

It can help surface material a person hasn’t thought about in a long time. It does not reliably improve accuracy, and it tends to raise a person’s confidence in whatever surfaces, accurate or not. Treat recovered content as a lead to be corroborated, not as established fact.

How can I tell if a regressed memory is real?

From inside the session, you can’t confirm historical accuracy. There’s no reliable test. You can look at whether the memory arrived organically versus being cued, and you can seek corroboration outside the session. Vividness, emotional intensity, and the client’s certainty are not indicators of accuracy.

Is it safe to do regression on trauma or abuse memories?

That’s high-risk territory that sits close to clinical psychotherapy and, in many places, inside it. Screen carefully, get genuine informed consent, avoid suggestive “unlock the repressed memory” framing, and refer out when a case exceeds your training or has any legal dimension.

Is hypnotic regression admissible in court?

Often not. Many U.S. jurisdictions restrict or exclude hypnotically-refreshed testimony because of suggestibility and false-memory concerns, and the rules vary by state and court. If there’s any legal aspect to a client’s situation, stop and get proper legal guidance before proceeding.

What’s the difference between age regression and past-life regression?

Both are real trance experiences, but past-life content has no possible corroboration and has been shown experimentally to be driven by the subject’s expectations and beliefs. Age regression to a real childhood is subject to the same expectation-shaping mechanism; it’s just harder to spot because the timeline is plausible.

The Takeaway

Trusting a regression isn’t about a feeling of certainty at the end. It’s about how you set it up: screening out the wrong candidates, getting honest consent, keeping your own theories out of the room, scoping the work tightly, and reading the signs that a memory came up on its own. Do that, and regression becomes what it should be, a precise tool for change, not a witch hunt for buried “truth.”

The affect-bridge sequence above is an abbreviated version of a few of the techniques in my regression program. If you want the full methodology (the screening scripts, the elicitation language, and the reprocessing techniques), you can learn more about my regression program here.

Enjoy,
Keith