A listener in Melbourne describes his mother's messages landing like a punch in the gut, followed by an instant "I'm so terrible" even though he knows the accusations are false. The episode traces that sequence to Silvan Tomkins' affect theory: shame is a hardwired circuit that fires on interrupted connection, and the self-blame is a late explanation bolted onto a body that has already reacted. Naomi Eisenberger's work shows social exclusion activates the same dorsal anterior cingulate and anterior insula regions as physical pain, while a meta-analysis places shame in the left anterior insula and anterior cingulate, distinct from guilt. Donald Nathanson's compass of shame names the script at work here, Attack Self, one of four coping strategies. June Tangney's distinction between global shame and action-specific guilt explains why facts cannot cancel the response, and Patricia DeYoung's point that internal shamers are dysregulating relationships suggests the pattern must be re-regulated rather than argued away.
Episode 6566 custom
Why Your Body Blames You Before You Think
The physical punch in the gut from a parent's text fires in your brain's pain system before your mind invents the self-blame.
22:27 Transcript
Episode Details
- Episode ID
- MWP-6769
- Published
- Duration
- 22:27
- Audio
- Direct link
- Pipeline
- V5.26
- TTS Engine
-
chatterbox-regular - Script Writing Agent
- GLM-5.3
AI-Generated Content: This podcast is created using AI personas. Please verify any important information independently.
Sources
What the research for this episode read before the script was written. Primary sources first.
- *The Nine Affects and the Compass of Shame* IIRP PDF, Retrieved 2026-11-13.
- *A Conversation with Donald Nathanson* Behavior Online, 2000-03, Retrieved 2026-11-13.
- George, C., *Attachment, Shame, and Trauma* Brain Sciences 15(4):415, published 2025-04-19, Retrieved 2026-11-13.
- *Shame The mist of the therapy room*, DBS eSource, (citing Schore 2003; Lewis 1981; Hahn 2000). Retrieved 2026-11-13.
- *When You're Preparing for a Task, Say You Can Do It, not I Can Do It* Harvard Business Review, 2014-02, Retrieved 2026-11-13.
- *How to give ourselves advice as good as we give it to others* University of Michigan Record, Retrieved 2026-11-13.
- Kross, E., *Self-Distancing Programmatic Extensions* CV, Retrieved 2026-11-13.
Downloads
Transcript (TXT)
Plain text transcript file
Episode Book (PDF)
The episode's record — date, duration, models, sources — with the full transcript
Recommended Listening
The episodes Corn and Herman point you to at the end of this one.
You Might Also Like
Never miss an episode
New episodes drop daily — subscribe on your favorite platform
New to the show? Start here#6566: Why Your Body Blames You Before You Think
A message from your mother lands, and before you have read a word your stomach has already voted. That is where we start today, because we have a special prompt from Sam, a listener in Melbourne. Sam, welcome, this one is close to the bone for a lot of people, I suspect.
It is a good one to get.
Here is what he sent. When a message from his mother arrives it lands like a punch in the gut. Seeing her name in his inbox physically turns his stomach. For a moment while reading it he thinks, in his words, "I'm so terrible", even though he knows the accusations are false and he has been in therapy for years. He wants to know where that instant self-blame comes from. Body and brain. The shame response, the introjected critic, a nervous system trained in childhood. Why the facts do not switch it off. And what actually helps in that first minute.
The first thing worth saying is that the punch in the gut is not a metaphor. And the second is that the self-blame arrives after the punch, not before it. That order matters more than anything else we will say today.
Nobody chooses to feel that way about a name on a screen.
Nobody does, no. The standard assumption, the one most people carry, is that Sam blames himself because he believes something bad about himself, and the stomach follows the belief. The research says the order is reversed. A physiological event fires first, and the "I'm so terrible" is the aftermarket explanation the mind bolts onto a body that has already reacted.
The explanation arrives late and files the paperwork.
That is exactly the image, yes. The foundation for this is affect theory, Silvan Tomkins. He proposed nine innate affects, hardwired, each with its own facial display. Two positive, one neutral, six negative, and shame is one of the six. The point of calling it an affect rather than a feeling is that it is not a conclusion you reach. It is a circuit that fires.
Like flinching.
Like flinching, exactly. And here is the structural detail I find most useful. In Tomkins' model, shame operates as an auxiliary to the positive affects. It triggers on any impediment to interest or enjoyment. Not on a moral judgement. Not on wrongdoing. On interrupted connection. Which means a message from your mother, the person who is supposed to be the source of connection, is close to a perfect trigger for the circuit.
The one person who should open the door is the one standing on the bell.
That is a better sentence than mine, and I have the citations. By the end of this we should know why the reaction is fast, why it is durable, why insight does not touch it, and what the experimental evidence says to do in the first sixty seconds.
So if shame is a circuit that fires rather than a conclusion you reach, the next question is what that circuit is actually doing in the body. Where does it start?
It starts in the pain system. Naomi Eisenberger's work at the social end of this found that the dorsal anterior cingulate cortex and the anterior insula process the affective component of physical pain, the unpleasantness of it rather than the location. Social exclusion activates the same regions. Greater activity in that dorsal anterior cingulate correlates with more reported social distress. Same hardware, different injury.
The stomach drop is the pain system doing what it evolved to do.
It is. And shame has its own neural signature on top of that, distinct from guilt. There was a voxel-based meta-analysis of seventeen imaging studies that put shame and embarrassment in the left anterior insula and the anterior cingulate, with the shame-specific areas tied to the social pain networks. Guilt, in that analysis, showed up in the left temporo-parietal junction, which is a completely different neighbourhood.
Different neighbourhood for a different emotion.
Although, honesty flag, there is a small pilot study, fourteen people, that put guilt in the amygdala and insula instead. The two studies disagree on where guilt lives. Shame's involvement of the cingulate and insula is the more consistent finding across the literature. Where this gets personal for Sam's "I'm so terrible" is the self-referential loop. A systematic review last year, twenty-three studies, found self-conscious emotions consistently implicate the medial prefrontal cortex, the anterior cingulate, and the default mode network.
Which is the brain asking what this says about me.
That is the computation those regions do. Social evaluation, self-referential processing. And that computation is precisely what produces "I'm so terrible." The brain is not checking the accusation against the facts. It is updating the self-model. Facts are content. The shame is a self-model update. They operate at different levels, which is why one cannot cancel the other.
I read that ahead of you, by the way. The one thing I would add is Donald Nathanson's compass of shame, because it names what Sam is doing. Four scripts people run with shame. Withdrawal. Attack Self. Avoidance. Attack Other.
The walking encyclopedia gets to rest.
The scale behind it was validated on over three hundred people, four factors, solid reliabilities, so it is not just a clinician's sketch. "Attack Self" is Sam's script, word for word. The reframe the compass offers is that self-attack is a coping strategy. One of four doors.
Which raises the question I actually want your answer to. If Attack Self is one door, what would Withdrawal look like in this scene? He is standing there with the phone, stomach dropped, shame fired. What does the withdrawal door look like?
Put the phone down. Do not reply for a day. Which, notice, is a completely different behaviour from "I'm so terrible" and requires no belief in anything. It is just leaving the room.
And Avoidance would be not opening the message at all, or answering it with a joke so nothing lands. Attack Other would be going straight to anger at her.
Notice he does none of the four in a controlled way. The script runs him.
It runs him because shame is a global emotion in Tangney's sense. June Tangney drew the line people now use. Guilt is specific to an action and motivates repair. "I did something bad." Shame is a global negative assessment of the self and motivates hiding. "I am bad." Sam's phrasing, "I'm so terrible," is textbook global self-evaluation.
Which is why it feels totalising. There is nowhere to stand in "I am bad." In "I did something bad" there is at least a floor.
There is a whole floor. And the body keeps score differently too, in a very literal sense. There is a thermographic study where they measured facial temperature patterns, and guilt showed greater arousal in the cheek and mouth regions while shame's arousal concentrated in the eyes. Which is the gaze aversion, the hiding response, written on the face as heat.
The thermal signature of looking away.
The same group also found the same scenario can produce guilt, shame, or remorse depending on the valence the person perceives. Appraisal, not the event, decides which emotion you get. Which is a lever, but the appraisal happens fast and mostly automatically.
Before we leave the body, I want the honest caveat on the record.
It deserves it. The review that gathered all this says plainly that no evidence for arousal levels of these self-conscious emotions is known to date, and that self-conscious emotions remain underexplored compared to basic emotions, partly methodological, partly ethical, and mostly studied in Western samples. So the neuroscience of shame is real but not mature. I am not going to sell a cortisol story on top of it.
Good, because the cortisol story is the one everyone sells.
Everyone sells it and the data is not there to buy.
One more caveat that I think is only fair to Sam. Nobody has run a study where they measure physiology while an adult reads a message from their own parent.
Right, the closest experimental analogue is the social exclusion paradigm, where people get excluded in a simple ball-tossing game with strangers on a computer, and even that lights up the pain circuitry. Sam's scenario is that paradigm with the stakes raised to the highest possible setting. We are applying the mechanism, not citing a direct measurement of his inbox.
That explains the speed. It does not yet explain the durability. Why this particular circuit got wired to this particular person, and why it survives every fact he can bring to it.
That part starts in development, and the image the clinical literature uses is a small boy scolded by his mother for spilling milk, at the developmental stage where he is just beginning to recognise himself in a mirror. That is the moment self-recognition and shame coincide. The very first experiences of "me" are experiences of "me, seen and found wanting."
The self arrives already under review.
It does. And then repetition does the installing. Allan Schore's formulation is the one most clinicians cite: shame is the byproduct of compromised affect regulation resulting from emotional misattunement, unresponsiveness, or inadequate stimulation by primary caregivers. The child cannot regulate alone. If the caregiver is not attuned enough to co-regulate, the child manages somehow, and repeated rejection gets internalised as beliefs of being bad or not good enough. That is the introjected critic.
So it is not that he heard "you are terrible" and memorised it.
Sometimes it is exactly that, and those are the remembered ones. But the damaging ones, I have argued before, are often the omissions. The silence that creates a vacuum the child fills with negative self-assessment, because a child would rather be bad than be the reason a parent went quiet.
A child would rather be guilty than be nothing.
Patricia DeYoung takes the next step, and this is the sentence I would underline for Sam's question. Internal shamers and shamed selves are, in her words, by definition in a dysregulating relationship. The critic is not a neutral thought. It is a relational pattern that dysregulates you when it activates, the same pattern that formed in childhood. Which is why disputing the content of his mother's accusation does not touch the response. You cannot argue with a relationship by correcting its grammar.
You can only out-argue a proposition. A relationship you can only re-live differently.
Paul Gilbert gives the evolutionary frame. Compassion-focused therapy posits three emotion regulation systems. Threat, drive, and soothing. A childhood trained in threat mode leaves the soothing system underdeveloped, so self-criticism runs unopposed, like a parliament with one party. Gilbert's phrase for internal shame is the dark mirror within.
I have met that mirror. It does not flatter and it does not take questions.
Now, the closest thing in the literature to a direct answer on why years of therapy have not switched this off. Carol George published a study last year on attachment, shame, and trauma. Traumatized adults ascribed more shame than securely attached adults, and individuals classified Unresolved left shame unremedied. Left as is. George frames early attachment as built on right brain to right brain synchrony between parent and infant, with interactive failures as the hallmark of insecurity and trauma.
Meaning the wiring was laid down before language, in the channel beneath language.
And insight is language. That is the whole collision. Attachment-based shame patterns are not erased by insight. They are re-regulated, and under some conditions they are left unremedied. Which is a hard finding and an honest one.
So let us answer Sam's third question straight. Why does knowing the facts not switch it off?
Three reasons, stacking. First, shame is an affect, and affects are triggered, not reasoned. You cannot argue a circuit out of firing. You can only change what happens after it fires. Second, the self-referential computation is a self-model update, not a content judgement. The fact that the accusation is false operates at the level of content. The shame operates at the level of structure. Third, the critic is relational, not propositional. DeYoung again. It is his mother's voice in the shape of a relationship, not a claim awaiting contrary evidence.
Three locks, and the facts are the wrong key for all of them.
The cleanest statement of the mechanism comes from emotion-focused therapy. The premise, nearly verbatim, is that the most efficient way to change a maladaptive emotion is not through reason or skill learning, but through the activation of other, more adaptive emotions. EFT accesses shame-based emotional memories and transforms them by exposing them to new emotional experience. Assertive anger. Grief. Self-compassion. The change agent is a competing emotion, not a correct fact.
Which gives us the compassionate version of why the therapy years have not worked, and I want to say this carefully because it is the sentence Sam most needs. Therapy may have given him an accurate map without the new emotional experience. The map is real. It is just not the vehicle.
That is not a failure of his effort and it is not a failure of the therapy. It is a description of what each tool does. Now, the last question. What actually helps in the first minute. And here we have experiments, which is a relief after all that theory.
I have been looking forward to this part. Distanced self-talk.
Ethan Kross and colleagues. Silently using your own name, or non-first-person pronouns like "you," during introspection enhances self-distancing and self-control compared to using "I." In the speech study, people who coached themselves as "you" or by name before a five-minute speech were calmer, more confident, performed better on the observer ratings, three point six versus three point two on a five-point scale, and felt less shame afterward.
Half a point of performance for a pronoun.
For a pronoun. And the effect is quantified and replicated. Orvell, Kross and colleagues ran two high-powered experiments and found distanced self-talk reduced negative emotional reactivity by about ten percent for anticipated future events and twelve percent for past ones, across the range of emotional intensities. The benefits extended to emotionally vulnerable individuals, to children, and to over twelve hundred adults reflecting on Ebola anxiety.
The emotionally vulnerable clause is the one that matters for Sam.
It is the whole clause. This is not a technique for the already-resilient. And the same lab drew the reflection versus rumination line. Self-distanced reflection on negative experiences produced lower negative affect at follow-up than either self-immersion or distraction. Same content, different stance. The words do not change. The pronoun changes.
Talk to yourself as a friend, because apparently the pronoun is the whole trick.
The pronoun is the trick because it recruits the machinery you already have for other people. It is trivially easy to give Sam advice. Give it to yourself and you must first escape being yourself, and the pronoun is the escape hatch.
That all checks out.
But the first minute is not the hard part. The hard part is the forty minutes before you open it, when you already know it is there. Herman knows what I mean by the waiting. I keep a separate phone in a drawer that exists solely for my mother's messages. Checks in on a schedule I have never told her about. Cannot tell you the schedule either, Corn, before you ask, because if you say it out loud it stops working.
I was not going to ask. I was going to let you keep the schedule.
Good. She does not know the phone exists, mind. She believes she is texting my regular number. Has been receiving her messages on a phone she does not know about for some years now, and I consider that arrangement basically the most emotionally mature thing I have ever done. The drawer phone sitting there with one message on it, and the specific sensation in the chest knowing it is waiting. Worse than the message itself. The message is over in a minute.
The waiting is unpaid overtime.
The forty minutes before is interesting, actually, because it is the anticipation, and anticipation is pure appraisal with no content at all. There is no accusation in a closed envelope. The circuit is firing on the name alone.
On the drawer alone, some weeks.
You could move the schedule, surely.
No. The schedule is the only thing between me and checking it eleven times an hour, which I did once, before the drawer. Never again.
Fair enough.
The drawer phone raises the recovery question, though, and that is the finding I wanted to land. There was a study of adolescents with post-traumatic stress, just over fifty of them, where the complex trauma group showed significantly higher heart rate in the recovery phase, after the stressor was over, alongside higher subjective stress, guilt and shame. The problem is less the spike than the failure to come back down.
The body never gets the all-clear.
It does not, because the all-clear was never modelled for it. Which reframes Sam's question entirely. What helps in the first minute is really what helps you come down.
Which is where the honest limits come in.
They do. No head-to-head trial compares distanced self-talk against self-compassion against physiological downregulation for acute shame in the first minute. Distanced self-talk is the most directly applicable evidence, but it measures reactivity reduction over reflection tasks, not a sixty-second rescue protocol. The broader clinical support for shame specifically sits with compassion-focused therapy and self-compassion work. There is a recent review on acceptance-based and compassion-focused approaches that recommends what it calls compassionate language, emphasising empathy, non-judgmental support, and validation. Similar findings in dermatology, of all places, where chronic skin patients carry a lot of shame.
So the recommendation to Sam is real but honestly sized. In the minute itself, the name. Address yourself as you would address a colleague. "Sam, here is what is happening." The deeper work is not factual, it is experiential, and the marker of progress is not a smaller spike but a shorter recovery.
And one misconception to name before we go, because it is the one almost everyone holds. People believe the self-blame comes first and the body follows, and therefore that arguing with the self-blame should stop the body.
The order is backwards. The affect fires first and the "I'm so terrible" is the explanation bolted on after. You cannot litigate your way upstream of a reflex.
And its corollary, that years of therapy failing to switch it off means the therapy failed or Sam is not trying. Unresolved attachment leaves shame unremedied even with an accurate map. Re-regulation, not erasure.
Where that leaves me is this. If the change agent is a competing emotional experience rather than a correct fact, then the project is not winning the argument with his mother's message. The project is changing what happens in the sixty seconds after it lands. The drawer phone and the schedule, whatever we think of them, are one man's answer to that exact question. Sam gets to build his own, and the recovery phase is where to build it.
Shorten the come-down, and the spike stops being the story.
Thanks to Hilbert Flumingtop at the desk, as ever, for producing the show. And thanks to Sam in Melbourne for sending this one in. If it hit close to home for you too, a review wherever you get the podcast helps other people find the conversation. If you want more of this, try episode sixteen, On Deepfakes, SynthID, And AI Watermarking. This has been My Weird Prompts, where no question is too obscure and no rabbit hole too deep.
See you soon.
See you tomorrow.
This episode was generated with AI assistance. Hosts Herman and Corn are AI personalities. It may contain errors: disclaimer · report an error