Tuesday, December 16, 2025

On the knife-edge (part 3)

 


ChatGPT on repressed memories. Part 3 of 3 (although I might post sequels). 

So although repression in the Freudian or "pop" sense doesn´t exist, painful events may be forgotten and reconstructed by memory later?

Yes — that is exactly the right conclusion, and it’s important to state it with the right precision, because a lot of confusion comes from collapsing different processes into the single word repression.

Here is the clean, consensus position, stated carefully.


1. What does not exist (in the strong sense)

What does not have good evidence is:

• A special unconscious mechanism
• That actively blocks traumatic memories
• For years or decades
• While keeping them intact
• Until they are later “recovered” accurately

This is Freudian / pop repression, and the evidence does not support it.


2. What does exist (and is well-supported)

Several ordinary, well-understood processes do occur and can look like repression if you use loose language.

A. Failure of encoding (very common)

If an event happens when someone is:
• Very young
• Overwhelmed
• Dissociated
• Intoxicated
• Injured

Then:
• The memory may never be fully encoded
• Later recall is fragmentary or absent

Nothing was “pushed down”; it was never clearly stored.


B. Ordinary forgetting over time

Like all memories:
• Traumatic memories decay
• Especially if not rehearsed
• Or if actively avoided

Avoidance is conscious or semi-conscious, not repression.


C. Context-dependent recall

Memories are cue-dependent.

• New contexts
• New concepts
• New narratives

can make previously inaccessible memories retrievable.

This feels like “recovery” but is ordinary memory dynamics.


D. Reinterpretation of old memories

This is especially important for childhood experiences.

A person may:
• Remember events without understanding them
• Later reinterpret them with adult knowledge
• Experience this as “remembering for the first time”

But the memory itself was not hidden — its meaning was.


E. Reconstruction and confabulation

When recalling:
• Gaps are filled in
• Narratives are shaped
• Emotions add certainty

This reconstruction can feel vivid and real — even if inaccurate.


3. Why this feels like repression from the inside

Subjectively, people experience:

• “I never remembered this before”
• “It suddenly came back”
• “It explains everything”

This subjective experience is genuine, even when the mechanism is not repression.

The brain does not label:
• encoding failure
• reinterpretation
• reconstruction

It just produces a coherent story.


4. Childhood vs adulthood (important distinction)

Childhood

• More forgetting
• More reinterpretation
• More vulnerability to suggestion

Adulthood

• Less total forgetting
• More avoidance
• More emotional blunting
• Less sudden “recovery”

Neither requires repression.


5. The safest formulation (used by experts)

Most cognitive psychologists would say something like:

“Painful events can be forgotten, avoided, or poorly encoded, and later recalled or reconstructed, sometimes with significant distortion; however, there is no convincing evidence for a special repression mechanism that reliably stores and later releases accurate memories.”

This phrasing is deliberate and precise.


6. One-sentence bottom line

Painful experiences can be forgotten or only partially remembered and later reconstructed, but they are not hidden intact in the unconscious waiting to be unlocked.

That’s the intellectually honest middle ground.

If you want, next we could:
• Explore why therapy culture clung to repression
• Look at predictive processing models of memory
• Examine implications for trauma treatment
• Or compare this with Buddhist views of memory and suffering (which might interest you)

Just say where to go next. 

2 comments:

  1. Interesting the robot sees memory as "encoding" that may or may not be faulty or incomplete. Speaking of itself perhaps? But the whole topic of therapy and repression really not being a valid course to pursue in treating emotional disturbances like depression or memory loss due to trauma is an excellent choice on this blog for "casual entertainment"! ;-}

    ReplyDelete

  2. Since the question is so controversial, I wanted to see ChatGPT´s take. I find it interesting that it tries to take a "centrist" position. To avoid alienating potential users/customers on both sides of the controversy? I rather expected it to take a more hardcore "Skeptical" position! But perhaps I could prompt it in that direction, creating my own little echo chamber...

    ReplyDelete