The Person Who Cannot Stop Believing What Their Autonomic Nervous System Has Been Telling Them for Years

by Stan Tatkin, PsyD, MFT, PACT Developer

There is a particular kind of couple case that makes even seasoned clinicians quietly dread the session clock. You know the one. Progress is made, sometimes good progress. And then, just as you’re about to land the plane, one partner reaches back into the past, pulls out an old injury, and hurls it squarely into the present. Session over. Sandcastle kicked. Everyone leaves worse than they arrived.

In our Consultation Group: Advanced, this theme keeps coming up. Clinicians describe partners who seem unable to move forward, despite transparency, reassurance, evidence, insight, and good intentions on both sides. Suspicion lingers even when nothing new is happening. Plans for the future repeatedly collapse under the weight of old material. The past intrudes, uninvited, like a relative who refuses to take the hint.

Traditionally, we’ve blamed the usual suspects: the amygdala, trauma, attachment anxiety, or “poor emotional regulation.” All of that is partially true — and insufficient. Something more precise is going on.

When Memory Colonizes the Future

What these cases share is not simply anxiety or distrust, but a failure of temporal integration. The autonomic nervous system becomes organized around preventing a future surprise rather than evaluating present reality. When that happens, memory stops behaving like memory. Instead of informing the present, it begins colonizing the future.

I’m proposing a clinical term for this pattern: Accrued Threat Memory Interference (ATMI). ATMI describes a relational phenomenon in which unresolved or repeatedly activated threat memory interferes with present-moment perception, executive reality testing, and most importantly the couple’s ability to move into the future together.

This is not delusion. It’s not psychosis. It’s not bad faith. And in many cases, it’s not even trauma in the traditional sense. It’s what happens when a relationship accumulates threat without sufficient repair, predictability, or shared governance.

A (Hypothetical) Case You’ll Recognize

Imagine a couple: Let’s call them Alex and Jordan.

They’ve been together for years, generally stable, somewhat distant, and highly independent. During the pandemic, they’re suddenly shoved together. Afterward life returns to normal: separate routines, travel, long stretches apart. Somewhere in that transition, Jordan notices something ambiguous. A strange message. A schedule that doesn’t quite line up. Nothing provable, just off.

Jordan begins monitoring. Carefully. Thoroughly. Over time, the investigation becomes exhaustive: emails, texts, receipts, timelines. Alex cooperates fully. There is no smoking gun. No confession. No evidence that stands up.

Five years later, they’re in therapy. Alex is exhausted and near tears. Jordan is convinced there must be a “there” there. Despite acknowledging that Alex has no history of lying or cheating, Jordan cannot let it go. The future, including retirement plans, shared goals, even next year, feels impossible to discuss.

What’s striking is that Jordan is thoughtful, intelligent, and reality-based in every other domain. This isn’t paranoia. It’s something else.

What the Brain Is Doing Instead of Thinking

In cases like this, the problem is not a lack of executive function. It’s that executive function is preempted.

When relational uncertainty persists, the brain shifts from present-moment appraisal to sustained anticipation. Systems involved in ongoing threat monitoring, including the bed nucleus of the stria terminalis (BNST), bias the autonomic nervous system toward vigilance rather than resolution. The subjective experience isn’t panic; it’s a chronic hum of something’s not right.

Moment-to-moment cues, such as tone, timing, absence, are still processed rapidly (thank you, amygdala), but instead of being resolved in the present, they’re handed off to predictive systems asking, What might this mean later?

Memory joins the party, not as historian, but as fortune teller. Fragments of past suspicion or injury are retrieved without proper timestamps and used as evidence for what could happen again. Reviewing the past doesn’t calm things down; it strengthens the loop. The brain isn’t trying to find truth. It’s trying to prevent being surprised.

At this point, conspiracy thinking becomes understandable. A bad story is often preferable to no story at all.

When Both Partners Are Hobbled

Now imagine a second hypothetical couple: Renee and Sam.

Both partners carry their own accumulated threat memory. Each has been hurt before by each other and by prior relationships. Both are vigilant. Both are convinced the other is the risk. Every attempt at planning triggers someone’s memory, which then triggers the other’s defensiveness. They can’t design a future because neither autonomic nervous system will stand down long enough to allow it.

This is ATMI in stereo. And it’s brutal.

Here, therapy often devolves into mutual litigation. The past is endlessly reviewed, as if one more piece of evidence might finally settle things. It never does. The problem isn’t insufficient understanding; it’s that the system itself is organized around threat prediction.

Why Insight Isn’t Enough (And Sometimes Makes It Worse)

One of the hardest things for clinicians and partners to accept is that insight does not undo ATMI. Neither does reassurance. Neither does transparency alone. In fact, repeated review of evidence often strengthens threat memory by reinforcing the idea that danger must be continuously monitored.

This is why these cases feel “intransigent.” It’s not that the partners are resistant. It’s that the autonomic nervous system has learned a rule: Do not move forward until the past is fully resolved. Unfortunately, that condition can never be met.

The PACT Correction: Design Beats Explanation

From a PACT perspective, ATMI highlights a core principle: Relationships don’t become safe through explanation; they become safe through design.

When threat memory has accrued, safety must be made obvious, predictable, and enforceable in the present. This means:

• Clear principles of mutual governance
• Rapid, reliable repair
• Predictable behavior that reduces uncertainty
• Agreements that prevent future harm rather than rehash past injury

Secure functioning is not about convincing the brain it’s safe. It’s about organizing the relationship so the brain doesn’t have to guess.

Prognosis and Case Selection (The Part We Don’t Like to Talk About)

ATMI carries real prognostic weight. Couples early in their relationship — or those without extensive accrued threat — often reorganize quickly once secure-functioning principles are in place. Couples with decades of unresolved injury, slow repair, and chronic distance may not.

In these cases, the clinician’s task is not just to assess willingness, but capacity: the capacity to prioritize the future over the past, to accept that memory is unreliable, and to commit to governance even when it contradicts subjective certainty.

Sometimes the most ethical intervention is helping partners recognize when continued couple therapy is reinforcing false hope rather than producing safety.

Final Thought

ATMI challenges a comforting illusion: that if we could just uncover the full truth of the past, safety would follow. It doesn’t. Safety emerges when partners stop asking the autonomic nervous system to solve the relationship and instead build a system in which good things reliably happen, bad things reliably do not, and both partners are demonstrably in each other’s care.

When the future is clear and enforceable, the past finally loses its grip — and memory can return to where it belongs.