Why Talking About the Past Can Keep You Stuck
The neuroscience of memory, and what it actually takes to change it
Most people who have spent time in therapy, coaching, or any serious personal development work have had a version of this experience.
You go deep. You find the thing. You understand where it came from, what it means, how it shaped you. The insight is real. Something shifts.
And then, months later, the pattern returns. Not entirely. Not unchanged. But recognisably the same thing, back again, as though the work never reached it.
The standard explanation is that you need more work. More sessions. More depth. More time.
What if the issue is not the amount of work? What if it is what the work is actually doing to the memory?
Memory is not a recording
Most of us carry an implicit model of memory as something like a recording. An experience happens, the brain captures it, and it is stored somewhere inside us, fixed and permanent. What we can do, on this model, is learn to respond differently to the recording. Build new habits around it. Develop better ways of coping with it. Over time, perhaps, it fades.
This model is intuitive. It is also incomplete.
What neuroscience has established over the last twenty-five years is that memory is not a fixed record. It is a living trace. And under specific conditions, that trace can be genuinely rewritten.
Not managed. Not built around. Rewritten.
The mechanism is called reconsolidation. And understanding it changes the question from “how do I cope with my past?” to “what does it actually take to change it?”
The Discovery that changed the field
In 2000, a researcher named Karim Nader published a study that disrupted the prevailing model of how memory works.
He and his colleagues demonstrated that when a consolidated memory, one already encoded and stored in long-term form, was retrieved, it did not simply play back. It briefly entered an unstable state. The molecular structure holding that memory in place temporarily came apart. The memory became, for a window of time, open to modification.
Then it restabilised.
Retrieval transiently destabilises consolidated memory, creating a memory updating window that maximises the impact of new learning. During that window, the memory could be weakened, strengthened, or reconsolidated with new information.
This was not a refinement of the existing model. It was a structural revision of what memory is.
Memory is not written once and filed. It is rewritten every time it is retrieved. The question is what it gets rewritten with.
The Window and what opens it
Here is the part that most discussions of reconsolidation leave out. And it is the most practically significant detail in the entire field.
The window does not open simply because a memory is recalled. Retrieval alone is not enough. A second condition is required.
A mismatch between what was predicted by a consolidated memory and what actually happens during reactivation is postulated as a key factor to bring memory again to this labile state. This mismatch is called prediction error.
The memory only becomes open to change when something about recalling it violates what the system expected. A surprise. A disconfirmation. Something that signals to the brain: the existing record needs updating.
Without that violation, retrieval does not open the reconsolidation window. It simply replays the memory. And replay, without destabilisation, does not produce change at the level of the trace. The inhibition of repeated recall of negative narratives is actually part of what allows modification to occur. Rehearsing the story coherently, making meaning of it, building a narrative around it, does not destabilise the trace. Without destabilisation, the original encoding remains intact beneath whatever has been added to it.
This is a precise description of what a significant proportion of therapeutic and coaching work does. It revisits the memory. It builds understanding around it. It adds new content alongside the original encoding.
What it does not do is destabilise the original trace. Which means the original encoding remains intact beneath whatever has been added to it.
The Competing Traces Problem
When a memory is retrieved but not destabilised, and new information is introduced alongside it, the brain stores two things. The original trace. And the new learning.
These are not the same thing as an updated memory. They are competing memories.
If exposure therapy operates through the formation of an inhibitory memory trace, clinicians should aim to optimise this inhibitory learning. Treatments ought to focus on violating the patient’s expectations about negative outcomes that might occur upon exposure to the feared stimulus, as large prediction errors are thought to generate strong inhibitory memories.
The original trace remains. The new pattern sits alongside it as an alternative. Under low pressure, in safe conditions, the new learning may well dominate. But under sufficient stress, in conditions that resemble those in which the original trace was encoded, the system defaults to what it knows most deeply.
This is why the pattern returns. Not because the work was not real. Not because the insight was false. But because the original encoding was never touched. Two versions exist in the system, and the older, stronger, more deeply consolidated one reasserts itself when conditions demand it.
Conceiving of change through the lens of inhibitory learning and competing memory traces could serve to enhance treatment outcomes.
What This Means for the work
Genuine reconsolidation, actual updating of the original trace rather than building new learning alongside it, requires a specific sequence.
First, the memory needs to be retrieved in a way that brings its felt structure into the present. Not just the story about it. The actual encoding. The emotional and physiological reality of it, present enough that the brain is genuinely running the original trace rather than a cognitive summary of it.
Second, something about that retrieval needs to violate what the trace predicted. Not a better narrative. Not reassurance. Something that genuinely contradicts the structure of what the memory has been holding as true. Minor environmental changes define whether retrieval induces memory reconsolidation or the initiation of a new memory trace even before fear extinction can be observed. The difference between opening the window and adding another layer can be subtle. But the consequences are not.
Third, what is introduced during the window needs to be integrated before the trace restabilises. The research suggests timing is part of the mechanism, though the precise parameters in human clinical contexts remain an active area of investigation.
This sequence is precise. And it is not what most approaches to change are designed to produce. Most approaches are designed to add new content alongside existing memory. That produces real results at the level of state. It does not update the structure of what is already there.
The Distinction That Changes Everything
There is a difference between working with the content of a memory and working with its structure.
Content is what the memory is about. The event, the people involved, the story you have built around it, the meaning you have made of it. Most therapeutic and developmental work operates here. It is not without value. But it does not reach the level at which the trace is encoded.
Structure is how the memory is held. The implicit conclusions it contains about who you are, what is safe, what is expected, what is possible. These are not always conscious. They do not always have language. They often predate the ability to articulate them. And they are the level at which the reconsolidation window, when it opens, allows genuine change.
Confronting the structure of a belief rather than its surface content is the lever. Not because it is more emotionally intense, but because it is more likely to generate the prediction error that the system requires before it will allow the trace to be rewritten.
Talking about the past does not change it. It adds to it.
Reaching the past, opening the window, and introducing something that genuinely violates what the trace has been holding, that is a different kind of work entirely.
Sarah Merron works with people ready to move beyond surface-level change. Her work operates at the structural level of belief, identity, and physiology rather than at the level of content and story.
Find Out More About Training & Private Work
