THE MACHINERY OF AVERSION
A Complete Guide to the No That Fires Before Thought
Why the Body Rejects Some Humans Without Asking the Self for Permission
What follows is not advice.
It is not a dating framework. Not a guide to becoming more attractive. Not a system for fixing what someone else feels in your presence. Not a workshop on overcoming your own revulsions.
It is mechanism.
The actual machinery that produces a no toward another human. The defense and screening assembly that fires before any conscious evaluation begins. The set of circuits that read another body, another face, another set of behaviors, and return a verdict the conscious self did not author and cannot easily override.
Most people live inside this system without ever seeing it. They feel the pull-back. The skin that does not want to be touched. The face that does not want to be looked at. The voice that grates. The presence they want to leave. They explain it to themselves with stories about taste, chemistry, type, vibe, energy. The stories are not the mechanism. The mechanism is older and far less interested in the conscious self’s preferences.
This document describes that mechanism.
What the reader does with the seeing is their business.
PART ONE: AVERSION IS A SCREEN, NOT A FEELING
The Standard Story Is Wrong
The folk account treats aversion as a single emotional reaction. Someone repels you. You feel it. You explain it.
This is backwards.
Aversion is an output. The output of a multi-system screening assembly that runs continuously, below conscious awareness, on every other human the body encounters. Several distinct circuits each take a vote. The votes are summed. If the sum crosses a threshold, the body produces a felt no.
The felt no arrives in consciousness as one experience. The mechanism that produced it is not one thing.
There is the disgust system, which evolved as an immune defense against pathogens. There is the major histocompatibility complex (MHC) similarity detector, which screens for genetic-cost matings. There is the asymmetric trait detector, which reads cues of poor genetic quality. There is the deception detector, which reads behavioral incongruence as unsafe. There is the identity-defense system, which flags information that threatens the desired self. There is the boredom register, which silences the dopamine signal when novelty stops arriving. There is the trust monitor, which collapses attraction the moment deception is discovered. Each runs on its own architecture. Each can produce a no on its own. Together they form the screen.
The conscious self is not informed of the votes. The conscious self is informed only of the result.
THE AVERSION SCREEN
┌──────────────────────────────────────────┐
│ PATHOGEN DISGUST vote: yes / no │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ MHC SIMILARITY vote: yes / no │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ ASYMMETRY / CUES vote: yes / no │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ DECEPTION / SAFETY vote: yes / no │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ IDENTITY DEFENSE vote: yes / no │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ PREDICTION SILENCE vote: yes / no │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ TRUST MONITOR vote: yes / no │
└──────────────────────────────────────────┘
│
▼
summed below threshold
of conscious awareness
│
▼
┌─────────────────────┐
│ FELT NO │
│ delivered to │
│ conscious self │
│ with no receipt │
└─────────────────────┘
The implication is uncomfortable. The no that fires when a particular human enters the room was not constructed by the self. It was constructed by an assembly the self does not see and cannot vote in.
This piece is the counterpart to THE MACHINERY OF DESIRE. Desire is the pull. Aversion is the push. They run on different circuits. They are not opposites of one degree. They are opposites in kind.
PART TWO: THE DISGUST SYSTEM
Three Domains
Paul Rozin and April Fallon (1987) proposed a structural account of disgust as an emotion built originally for one purpose and subsequently extended to others. The original purpose was rejection of contaminated food. The extension was social and moral.
Tybur, Lieberman, and Griskevicius (2009) proposed that human disgust operates across three functional domains, each with overlapping but distinguishable inputs.
Pathogen disgust. The original. Triggered by bodily fluids, decay, lesions, deformity, parasites, and signs of infectious disease. Functions as a behavioral immune system (Schaller and Park, 2011), screening for pathogen exposure before contact occurs.
Sexual disgust. Triggered by potential mating with low-quality, biologically inappropriate, or kin-coded partners. Functions as a screening layer that prevents reproductive cost.
Moral disgust. Triggered by violations of social norms. Functions as a group-cohesion enforcer.
The three domains share neural architecture. Wicker and colleagues (2003) demonstrated that the anterior insular cortex activates both when a person feels disgust themselves and when they observe disgust on another’s face. The insula is the disgust hub. It does not know whether the trigger is rotten food, an inappropriate partner, or a moral violation. It produces the same somatic signature across all three.
THE THREE DOMAINS OF DISGUST
┌──────────────────────────────────────────┐
│ PATHOGEN DISGUST │
│ │
│ Bodily fluids, decay, lesions, │
│ deformity, parasites, illness cues │
│ │
│ Function: behavioral immune system │
└──────────────────────────────────────────┘
│
▼
┌──────────────────────────────────────────┐
│ SEXUAL DISGUST │
│ │
│ Low-quality mate cues, kin coding, │
│ reproductive cost markers │
│ │
│ Function: mating screen │
└──────────────────────────────────────────┘
│
▼
┌──────────────────────────────────────────┐
│ MORAL DISGUST │
│ │
│ Norm violations, betrayal, taboo │
│ breach │
│ │
│ Function: group-cohesion enforcement │
└──────────────────────────────────────────┘
All three converge on the anterior insula.
Same somatic signature. Different triggers.
This is why physical revulsion has the felt-quality of moral wrongness. The body does not distinguish. Someone who triggers pathogen disgust will be experienced as morally suspect, even when there is nothing moral at stake. Someone who violates a moral norm will trigger physical nausea. The systems share output even when they do not share input.
Curtis, Aunger, and Rabie (2004) and Curtis, de Barra, and Aunger (2011) measured disgust responses across cultures and confirmed that the strongest cross-cultural triggers are precisely those that signaled pathogen risk in ancestral environments. The system is not learned culture by culture. The triggers are.
The insula produces a specific bodily signature. Nausea. Throat tightening. The face contortion. The pull-back. Avoidance. These outputs run regardless of whether the conscious self believes they are appropriate.
The Function
Disgust is a screening operation that runs on the body, not the mind.
Schaller (2006) and the behavioral immune system literature (Schaller and Park, 2011) describe the function. Pathogens kill. The immune response is metabolically expensive and probabilistic in outcome. A behavioral screen that biases the organism away from likely pathogen sources, before contact, is structurally cheaper than fighting an infection after exposure.
This screen produces false positives. It must, for the same reason the fear system jumps at sticks. The cost of avoiding a healthy person who happens to display a pathogen cue is small. The cost of contacting an infectious one can be lethal. The system is tuned for asymmetric error.
The implication for inter-human aversion. Someone who displays cues the disgust system reads as pathogen-relevant will be avoided. Visible skin lesions. Asymmetric facial features beyond a perceptual threshold. Body odor that signals illness. Coughing. Visible bodily fluids. The conscious self may know the person is not infectious. The screen does not consult the conscious self. The pull-back happens. The conscious self interprets the pull-back as personal taste.
This is also why the disgust system is the most easily weaponized aversion circuit. Cultural amplifiers, propaganda, dehumanization rhetoric. All operate by aiming the disgust system at human targets. Once the system is aimed, the moral and pathogen registers fuse. The targeted group is experienced as both morally polluting and physically contaminating. The aversion is felt as righteousness.
PART THREE: MHC SIMILARITY AND THE OLFACTORY VETO
The Below-Conscious Genetic Screen
The major histocompatibility complex is a set of genes that codes for the molecules immune cells use to identify pathogens. MHC diversity in offspring confers resistance to a wider range of infections. Mating with someone whose MHC profile is too similar to one’s own produces offspring with narrower immune repertoires.
The body has a screen for this. The screen runs on smell.
Wedekind, Seebeck, Bettens, and Paepke (1995) ran the now-classic t-shirt study. Men wore t-shirts for two nights. Women smelled the shirts blind and rated them for attractiveness. The women preferred shirts from men whose MHC profiles were dissimilar to their own. The signal was olfactory. The processing was below conscious awareness. The women could not articulate why some shirts were appealing and others were not.
Roberts, Gosling, Carter, and Petrie (2008) extended the work. Couples whose MHC profiles were similar reported lower sexual satisfaction, more attraction to outside partners, and more relationship instability. The MHC layer was operating in established relationships, not just initial encounters.
Havlíček and Roberts (2009) reviewed the human MHC literature and confirmed the structural pattern. MHC dissimilarity correlates with attraction. MHC similarity correlates with aversion or indifference.
THE MHC OLFACTORY VETO
┌──────────────────────────────────────────┐
│ MHC PROFILE A │
│ (the perceiver's immune coding) │
└──────────────────────────────────────────┘
│
│ compared via
│ olfactory signal
▼
┌──────────────────────────────────────────┐
│ MHC PROFILE B │
│ (the target's immune coding) │
└──────────────────────────────────────────┘
│
┌─────────┴─────────┐
▼ ▼
┌─────────────────┐ ┌─────────────────────┐
│ DISSIMILAR │ │ SIMILAR │
│ │ │ │
│ Pleasant scent │ │ Neutral or │
│ Attraction │ │ unpleasant scent │
│ bias up │ │ Attraction │
│ │ │ bias down │
└─────────────────┘ └─────────────────────┘
The Westermarck effect (Westermarck, 1891) is the developmental layer. Children raised in close proximity during a critical window form sexual aversion to each other regardless of biological relatedness. Wolf (1995) documented this in Taiwanese minor marriages, where unrelated children raised together as future spouses showed elevated divorce rates and reduced fertility. Lieberman, Tooby, and Cosmides (2007) identified the kinship cues the system uses: co-residence duration in childhood and observation of maternal care toward the other child.
The kinship detection system feeds the MHC screen and the sexual disgust register. Together they form a sibling-avoidance assembly that runs whether the conscious self acknowledges siblinghood or not. Step-siblings raised from infancy show the effect. Biological siblings raised apart often do not.
The aversion is not knowledge of relatedness. It is the output of a developmental cue-tracker that the brain installed without asking the self.
This is one structural source of the felt-no toward a specific human that the self cannot rationalize. The smell is wrong. The face is wrong. The body says no. The conscious self cannot find a reason. The reason is in the immune coding and the developmental record. Neither is available to introspection.
PART FOUR: ASYMMETRIC TRAIT DETECTION
Cues of Genetic Quality
The face and body of another human carry information about developmental stability, current health, and genetic quality. The brain reads this information rapidly, using detectors tuned over evolutionary time.
Fluctuating asymmetry is the standard measure. Bilateral organisms develop with mirrored left and right sides. Departures from perfect symmetry signal developmental perturbation: pathogen load, parasite stress, or genetic instability during growth. Thornhill and Gangestad (1993, 1999) and Rhodes (2006) reviewed the literature. Symmetric faces are rated as more attractive across cultures. Asymmetric faces, beyond a perceptual threshold, trigger aversion.
The threshold matters. The brain does not require perfect symmetry. Most faces show measurable asymmetry. Below the perceptual threshold, the asymmetry registers as nothing. Above it, the face begins to read as wrong. The aversion onset is non-linear. Small departures produce no signal. Larger departures produce sharp signal.
Skin texture is the second cue. Skin signals current pathogen load, hormonal state, and age. Visible lesions, asymmetric pigmentation, parasitic markers, and acute infection cues are read by the disgust system as pathogen risk. Jones and colleagues (2004) showed that skin condition independently predicts attractiveness ratings beyond face shape.
Body proportions are the third. Singh (1993) documented the cross-cultural waist-to-hip ratio preference. Singh and Singh (2011) extended the analysis. The ratios that signal reproductive viability across cultures map closely to the ratios rated attractive. Departures register as cues of poor reproductive fit.
Voice is the fourth. Hodges-Simeon, Gaulin, and Puts (2010) and Pisanski and colleagues (2014) showed that vocal frequency, breathiness, and resonance carry information about hormonal state, body size, and current health. Voices outside the perceiver’s preferred range trigger reduced attraction. Voices that signal acute illness or hormonal abnormality trigger aversion.
THE ASYMMETRIC TRAIT SCREEN
┌──────────────────────────────────────────┐
│ FACIAL SYMMETRY │
│ Below threshold: no signal │
│ Above threshold: aversion onset │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ SKIN TEXTURE │
│ Lesions, asymmetric pigmentation, │
│ acute infection markers │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ BODY PROPORTIONS │
│ WHR, shoulder ratio, posture │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ VOCAL CUES │
│ Frequency, resonance, illness markers │
└──────────────────────────────────────────┘
Each detector runs continuously.
Each can produce a no on its own.
These cues are not consciously inventoried. The brain processes faces in roughly 100 milliseconds (Willis and Todorov, 2006). The trait detection runs in parallel with the conscious gaze. By the time the perceiver has looked at a face, the verdict is in. The conscious self constructs a story about why afterward.
The story is not the mechanism.
PART FIVE: BEHAVIORAL AVERSION CUES
What the Brain Reads in Action
A still face is one input. A moving body and a speaking voice are another. The behavioral channel carries cues that the still-face channel cannot.
Deception markers are the first class. Paul Ekman’s (1992, 2003) work on microexpressions documented the brief, involuntary facial movements that leak when an expressed emotion conflicts with a felt one. The brain of the perceiver registers these leaks below conscious awareness. The result is a felt-incongruence: the words say one thing, something else is reading wrong, the self does not know what.
Porter and ten Brinke (2008) extended the work. They showed that liars produce inconsistencies between high-stakes emotional displays and the underlying microexpressions. Observers detect the inconsistency at rates above chance, often without being able to articulate what they detected.
The mechanism is congruence-checking. The brain runs continuous integration across face, voice, posture, and verbal content. When the channels align, no signal. When they diverge, a not-safe flag is raised. The perceiver experiences the target as off, slimy, not quite right. The aversion is the output of the integration failure.
Low-status displays are the second class. Posture, gaze pattern, voice projection, and gesture rate carry status information that the brain reads automatically (Ridgeway, 2014; Anderson and Kilduff, 2009). Submissive posture, averted gaze, vocal hesitation, and shrinking body language can trigger aversion in perceivers calibrated to dominance hierarchies. The aversion is partly disgust (low rank as cue of poor fit) and partly safety-screening (the inability to defend the perceiver in a coalition context).
Unpredictability is the third. The brain runs prediction over the target’s behavior. Predictable behavior, even predictably hostile behavior, allows planning. Unpredictable behavior triggers a sustained activation in the BNST-driven anxiety circuit described in THE MACHINERY OF FEAR. The perceiver cannot relax around the target. The body reads sustained activation as cost. The cost is paid in aversion.
Approach speed is the fourth. Bowlby’s attachment work (1969, 1973) and the subsequent literature documented that approach behavior calibrated outside the perceiver’s expected timing produces a defensive response. The person who approaches too fast, too close, too soon, triggers a withdrawal response that reads as aversion. The same person, approaching at the perceiver’s expected pace, may produce no aversion at all. The behavioral mismatch, not the person, is the trigger.
BEHAVIORAL AVERSION CUES
┌──────────────────────────────────────────┐
│ CHANNEL INCONGRUENCE │
│ Face, voice, posture, words diverge │
│ Output: not-safe flag │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ STATUS MARKERS │
│ Posture, gaze, vocal projection │
│ Output: rank-based filter │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ UNPREDICTABILITY │
│ Behavior outside prediction range │
│ Output: sustained BNST activation │
└──────────────────────────────────────────┘
┌──────────────────────────────────────────┐
│ APPROACH MISMATCH │
│ Pace and proximity outside expected │
│ Output: withdrawal response │
└──────────────────────────────────────────┘
The behavioral channel is where most adult aversion is generated. Faces vary little across encounters. Behavior varies continuously. The repeated exposure to a person across many behavioral instances allows the brain to update its assessment in real time. A face that initially registered as attractive can become aversive over weeks of behavioral data. A face that initially registered as plain can become attractive through behavioral integration.
The behavioral data outweighs the static-trait data over time.
PART SIX: THE WANTING DRAIN
When Prediction Fails Repeatedly
The wanting circuit, described in THE MACHINERY OF DESIRE, runs on dopamine. It assigns incentive salience to cues that predict reward. The pull toward another human is the output of this circuit. The intensity of the pull is a function of accumulated positive prediction error.
What happens when the predictions fail.
Schultz, Dayan, and Montague (1997) and Schultz (2016) described the dopamine teaching signal in detail. When an outcome is better than predicted, dopamine fires above baseline. When an outcome matches prediction, baseline is held. When an outcome is worse than predicted, dopamine drops below baseline. The teaching signal updates the cue-outcome association.
In the context of inter-human relationships, the cue is the partner. The predicted outcome is the felt-quality of being in their presence. Validation. Pleasure. Sexual response. Safety. Stimulation. The wanting circuit predicts each of these from the partner’s cues.
Repeated negative prediction error reshapes the cue value.
THE WANTING DRAIN
Time 0: Prediction = high Outcome = high
Dopamine: positive Wanting: amplified
Time 1: Prediction = high Outcome = medium
Dopamine: small dip Wanting: held
Time 2: Prediction = high Outcome = low
Dopamine: large dip Wanting: drained
Time 3: Prediction = medium Outcome = low
Dopamine: small dip Wanting: collapsing
Time 4: Prediction = low Outcome = low
Dopamine: zero Wanting: zero
Time 5: Prediction = low Outcome = high
Dopamine: small bump Wanting: barely
registers
This is the slow death of attraction.
The partner has not changed. The predictions have updated. The cues that once produced anticipatory dopamine now produce baseline. The presence that once was pulled-toward is now neutral. The next phase, often, is pulled-away. Berridge and Robinson (1998) and the subsequent work distinguished wanting and liking. Wanting is the dopamine-driven pull. Liking is the opioid-driven hedonic response. Wanting can collapse while liking is preserved (the partner is still pleasant when present, but no anticipation builds). Wanting can also collapse below zero, into active avoidance, when the prediction error has been negative often enough that the cue itself becomes a predictor of cost.
The wanting drain is one of the most common pathways into aversion in long-term relationships. The partner’s cues now predict disappointment. The body registers the prediction. The pull becomes a push. The conscious self may not see what shifted. The shift is in the prediction-error log.
PART SEVEN: IDENTITY-INCOMPATIBLE INFORMATION
When the Other Person Threatens the Desired Self
The brain maintains a model of the desired self. The model is a target the self is trying to inhabit. Information that supports the model is integrated. Information that threatens the model triggers defensive operations.
Steele’s self-affirmation theory (1988) and the subsequent literature describe the defensive layer. When self-relevant information is threatening, the system can deploy several responses. Counter-arguing. Dismissal. Avoidance of the source. Active aversion to the source.
Another human can be a self-threat in several ways.
They have what the desired self wants and cannot easily get. The successful peer who triggers envy and downstream contempt. Festinger’s (1954) social comparison work and the subsequent envy literature (Smith and Kim, 2007) describe the architecture. Envy operates as a status-loss signal. The signal is unpleasant. The defense is to devalue the source. The devaluation registers as aversion.
They embody what the desired self is trying to avoid being. The peer who shows what the perceiver fears becoming. The aversion runs through the same circuit but with reversed valence. The perceiver pushes the target away because contact threatens identity contamination.
They challenge a belief the desired self is built on. The peer whose existence implies the perceiver’s life is wrong. Their values, choices, or framework. The aversion is felt as moral disgust because the moral and identity systems share output.
They see what the perceiver is hiding. The peer who knows. Their gaze itself becomes intolerable because it carries the threat of identity exposure.
IDENTITY-DEFENSE AVERSION
┌──────────────────────────────────────────┐
│ DESIRED SELF MODEL │
│ (the target the self is inhabiting) │
└──────────────────────────────────────────┘
│
│ threatened by
▼
┌──────────────────────────────────────────┐
│ TARGET HUMAN │
│ carries information that violates │
│ the self model │
└──────────────────────────────────────────┘
│
┌─────────┴─────────┐
▼ ▼
┌─────────────────┐ ┌─────────────────────┐
│ COUNTER-ARGUE │ │ DEVALUE / AVERT │
│ │ │ │
│ cognitive │ │ felt as disgust │
│ defense │ │ or contempt │
└─────────────────┘ └─────────────────────┘
The identity-defense system is the structural reason ideologically opposed humans often experience visceral aversion to each other. The belief is not the trigger. The belief is a marker of an identity model that the other party’s existence threatens. The aversion registers in the body as moral disgust, with full insular activation, even when no concrete cost has been incurred.
This is also the structural reason former intimate partners often become objects of strong aversion after a breakup that involved identity rupture. The ex carries information about a self the perceiver no longer wants to be. Contact reactivates that self. The aversion is the defense.
PART EIGHT: BOREDOM AS LOW-GRADE AVERSION
When the Dopamine Signal Goes Silent
Boredom is not the absence of feeling. It is the active register of insufficient prediction error.
Eastwood, Frischen, Fenske, and Smilek (2012) reviewed the boredom literature and converged on a definition. Boredom is the state in which the perceiver wants engagement, cannot find it, and registers the absence as aversive. Danckert and Merrifield (2018) added the predictive-processing layer. The brain runs predictions. When predictions are met perfectly across many cycles with no surprise, the dopamine teaching signal silences. The system reads silence as a failure of input.
In the context of another human. A partner who reliably produces the predicted response, with no novelty, no surprise, no friction-resolution event, eventually silences the dopamine response. The presence still produces baseline. Nothing more. The conscious self begins to experience the partner as boring.
Boring is a softer felt-quality than disgust or contempt. It still belongs to the aversion family. The body’s response to chronic boredom is the same response it produces to chronic low-grade negative input. The perceiver wants out of the situation. The wanting circuit, deprived of teaching signal, gradually withdraws investment.
THE BOREDOM REGISTER
Steady prediction match:
████████ ████████ ████████ ████████
no surprise, no signal, dopamine falls
Below baseline activation:
▏▏▏▏ ▏▏▏▏ ▏▏▏▏ ▏▏▏▏
body reads as cost, registers as aversion
Result:
Low-grade pull-away from the source of
insufficient signal, regardless of whether
the source is threatening or merely silent.
The integration with the wanting drain is direct. Wanting drain happens when predictions are violated negatively. Boredom happens when predictions are met too consistently. Both produce dopamine silence. Both terminate in aversion.
This is also the structural reason high-novelty, high-friction partners often produce sustained attraction even when they generate measurable distress. The dopamine signal stays alive. The wanting circuit stays engaged. The same partner, fully predictable and frictionless, would produce boredom and then aversion. The system requires prediction error to remain interested. The conscious self does not want to admit this. The mechanism does not consult the conscious self.
PART NINE: FEAR-WITHOUT-RESPECT
When Threat Becomes Cost
Fear and attraction can coexist. The high-status threatening figure who produces both anxiety and pull is a documented pattern (Buss, 1989; Sadalla, Kenrick, and Vershure, 1987). The pull is calibrated to the perceived benefits of access. Resources. Protection. Status by association. Genetic quality signaled through dominance.
The pull holds as long as the brain reads the threat as benefit-coupled. The high-status figure is dangerous, but the danger is paired with what the wanting circuit reads as gain.
The pull collapses when the brain reads the threat as benefit-decoupled. The figure is dangerous, the danger is not coupled with gain available to the perceiver, the threat is pure cost. At this point, the integration shifts. Fear becomes contempt-laced aversion. The perceiver no longer wants access. The perceiver wants distance.
THE FEAR-RESPECT INTEGRATION
State A: Threat present
Benefit available to perceiver
Output: attraction with anxiety
Wanting: high
State B: Threat present
Benefit unavailable / withdrawn
Output: aversion
Wanting: drained
The same threat produces opposite pulls
depending on benefit coupling.
This is one of the structural reasons abuse-pattern relationships terminate the way they do. As long as the abusive partner is read as high-status, resource-providing, protective, the threat-attraction pair holds. When the resources stop or the protection becomes performative, the integration flips. The threat becomes pure cost. Aversion replaces attraction in a single update.
The integration with THE MACHINERY OF FEAR is direct. Fear is not the same as aversion. Fear is the threat-detection register. Aversion is the screening output. They can dissociate (fear without aversion, in the high-status threatening case) and they can fuse (fear coupled with aversion, in the pure-cost threatening case). The perceiver’s conscious experience tracks the fusion, not the underlying circuits.
PART TEN: CONTEMPT AND TRUST COLLAPSE
The Two Most Destructive Aversion Forms
Most aversion is mild. Mild pull-back. Lukewarm response. Indifference. The disinterest gradient.
Two forms of aversion are operationally different and structurally far more destructive.
Contempt is the first. Ekman (1992, 2003) identified contempt as a distinct emotion, with a specific facial signature: unilateral lip corner raise, head tilt back. The felt-quality includes superiority. The target is read as beneath the perceiver. The valuation is not zero, as in indifference. It is negative, with a downward direction.
John Gottman’s (1994, 1999) decades of marital research found contempt to be the single strongest predictor of relational rupture. Couples who showed contempt for each other in lab interactions divorced at far higher rates than couples who showed any other negative emotion, including overt anger. Anger is engagement. Contempt is downward dismissal. Anger is recoverable. Contempt is structurally corrosive.
The contempt machinery produces specific outputs: condescension, eye-rolling, sarcasm, dismissive humor, refusal to engage seriously. Each output is read by the target as a downward valuation. Each output further reinforces the contemptuous frame in the perceiver. The loop self-reinforces.
CONTEMPT vs DISINTEREST vs DISGUST
DISINTEREST
Valuation: zero
Direction: none
Recoverable: yes (a single positive
prediction error can reverse)
DISGUST
Valuation: negative
Direction: lateral / pull-back
Recoverable: difficult (somatic
signature self-perpetuates)
CONTEMPT
Valuation: negative
Direction: downward
Recoverable: rare (requires reversal of
status frame, which the
mechanism resists updating)
Trust collapse is the second. Trust, described in THE MACHINERY OF TRUST, is the integrated assessment of a target’s reliability across cycles of interaction. The integration takes time to build. The integration can be ruptured in a single discovered-deception event.
When deception is discovered, the brain does not simply update the trust value downward by a unit. The brain reclassifies the target. Every prior interaction is re-evaluated under the new frame. The smile that was warm is now read as performance. The vulnerability that was bonding is now read as manipulation. The trust collapse is global. The aversion is total.
Janoff-Bulman (1992) described this in trauma terms. The shattered assumption. The world model in which the target was a particular kind of person collapses, and a new world model is rapidly constructed in which the target is a different kind of person. The new model often runs aversive for the rest of the relationship and beyond.
The combination of contempt and trust collapse, when both occur, produces the most stable aversion the system generates. Reversal becomes structurally improbable. The target is positioned both downward and outside the safety perimeter. The pull is fully replaced by push.
PART ELEVEN: WHY AVERSION IS STICKY
The Asymmetric Weighting of the No
Kahneman and Tversky (1979) and the subsequent prospect theory work showed that losses are weighted at roughly 2.25 times the magnitude of equivalent gains. The pain of losing 100 dollars is felt more intensely than the pleasure of gaining 100. This asymmetry runs through every domain that involves prediction and outcome.
In aversion, the asymmetry has direct operational consequences. One negative event with a target outweighs many positive events. The brain treats the negative event as more diagnostic of the target’s underlying nature. The prior shifts more on a single negative datum than on many positives.
Rozin and Royzman (2001) named this negativity bias and reviewed the cross-domain literature. Across perception, attention, memory, and judgment, negative information dominates positive information of equivalent magnitude. The asymmetry is structural, not learned. It is one of the most reliable findings in the cognitive literature.
For inter-human aversion, the implication is that aversion, once established, is hard to reverse. The conditioned aversion response resists extinction in the same way conditioned fear does. The integration with the extinction discussion in THE MACHINERY OF FEAR is direct. Extinction does not erase the original learning. It layers an inhibitory memory on top. The original aversion is preserved. Spontaneous recovery, renewal, and reinstatement all apply.
THE STICKINESS OF AVERSION
Positive events with target: ████████████
Negative events with target: ████
Weighted by negativity bias:
Positive: ████████████
Negative (x2.25): █████████
Net felt-valuation:
Slightly positive on paper.
Felt as ambivalent in practice.
One additional negative event:
Positive events: ████████████
Negative events: █████ (now 5)
Weighted:
Positive: ████████████
Negative (x2.25): ███████████
Net felt-valuation:
Slightly negative.
The pull collapses on a single update.
This asymmetry is also why repair after rupture is structurally expensive. The brain requires many positive updates to compensate for a single negative one. Most relationships do not survive the math. Either the perceiver does not invest enough positive interactions to overcome the weighted negative, or the target does not produce enough positive prediction error to feed the update, or new negative events arrive faster than positive ones can compensate.
The aversion stays.
PART TWELVE: CULTURAL AMPLIFIERS AND THE GRAY ZONE
The System Aimed Externally
The aversion machinery did not evolve to be aimed at humans across the boundary of the perceiver’s tribe. It evolved to screen for pathogens, mate-quality, and behavioral threat within and at the edge of the local group. Cultural systems recruit the machinery for other purposes.
Out-group markers are the simplest amplifier. Tajfel and Turner’s (1979) social identity work and the subsequent literature documented that minimal markers (a randomly assigned color, an arbitrary group label) can activate in-group preference and out-group bias. Combined with the disgust system, out-group markers can be loaded with pathogen-relevance, producing dehumanization (Haslam, 2006). The out-group is experienced as morally polluting and physically contaminating. The aversion is felt as righteousness.
Taboo violations are the second amplifier. Tetlock, Kristel, Elson, Lerner, and Green (2000) studied taboo trade-offs and showed that humans react with strong moral disgust to violations of values they have classified as sacred. The target who violates a sacred value is experienced as polluted. The aversion is total.
Stigma is the third. Goffman’s (1963) work and the subsequent literature describe stigma as a mark that recategorizes the bearer. The categorization runs through the disgust and identity-defense systems. The bearer is experienced as fundamentally different in kind, and the difference is loaded with aversive valence.
These amplifiers do not generate aversion from nothing. They aim a system that already exists. The system was built to screen pathogens, mates, and behavioral threats. The cultural overlay tells it where to look. The output is the same somatic response. The trigger is now group-membership, taboo, or stigma rather than lesion, asymmetry, or deception.
THE CULTURAL AMPLIFIER LAYER
┌──────────────────────────────────────────┐
│ EVOLVED AVERSION MACHINERY │
│ pathogen, mate, behavior, identity │
└──────────────────────────────────────────┘
│
│ aimed by
▼
┌──────────────────────────────────────────┐
│ CULTURAL OVERLAY │
│ out-group markers │
│ taboo categories │
│ stigma │
└──────────────────────────────────────────┘
│
▼
┌─────────────────────┐
│ AVERSION DIRECTED │
│ AT TARGET │
│ felt as righteous │
│ experienced as │
│ natural / obvious │
└─────────────────────┘
The Gray Zone
Most of what the conscious self labels as aversion is not active aversion. It is absence-of-attraction.
The two are not the same mechanism. Active aversion is a defense response with full somatic activation. The body is engaged. The disgust system, the identity-defense system, or the trust monitor has fired. The push is real.
Absence-of-attraction is the null state. The wanting circuit has not pulled. The pull machinery has not engaged. There is no push. There is just nothing. The target produces no signal in either direction.
The conscious self often conflates the two. The phrase “I’m not into them” can mean either. The operational consequence is different. Active aversion, when it fires, is sticky and resistant to update. Absence-of-attraction can flip to attraction in the presence of a single positive prediction error event. The cue that shifts a person from null to interesting is small. The cue that shifts a person from active aversion back to interesting is large or impossible.
The integration with THE MACHINERY OF NOTHING is direct. The null state is not a small version of the negative state. It is its own register. Aversion has a positive presence; nothing has no presence. Aversion is felt; nothing is the absence of felt. The body keeps the records separately.
This distinction has direct operational consequences. The person trying to read another’s response is often reading the wrong register. They interpret null as rejection and respond with defense. The target was not rejecting. The target had simply not produced any pull machinery in the first place. The defensive response then produces actual aversion in the target through behavioral mismatch (Part Five). The null state has been converted into an active no through the perceiver’s own defensive misread.
Final Synthesis
Aversion is the output of a screening assembly. The assembly runs continuously below conscious awareness. The conscious self is informed of the result, not the votes.
The disgust system screens for pathogen, mate-quality, and moral violation, with all three converging on the anterior insula. The MHC similarity detector and the Westermarck kinship system run an olfactory and developmental veto on partners coded as genetic-cost. The asymmetric trait detector reads facial symmetry, skin texture, body proportions, and voice for cues of poor genetic quality. The behavioral channel reads channel incongruence, status markers, unpredictability, and approach mismatch as not-safe signals.
The wanting circuit drains over time when predictions fail repeatedly. The drained circuit produces neutral first, then aversive valence. Identity-incompatible information from another human triggers active defensive aversion through the self-affirmation system. Boredom, the silenced dopamine signal, registers as low-grade aversion when prediction error stops arriving.
Fear and attraction can coexist when threat is benefit-coupled. They dissociate into pure aversion when the threat becomes pure cost. Contempt operates as downward valuation and is the strongest predictor of relational rupture. Trust collapse, triggered by discovered deception, reclassifies the target globally and rewrites prior interactions under an aversive frame.
Aversion is sticky because of the negativity bias. One negative event outweighs many positives at roughly 2.25 to 1. The conditioned aversion response resists extinction the same way conditioned fear does. Spontaneous recovery, renewal, and reinstatement all apply.
Cultural amplifiers aim the machinery externally. Out-group markers, taboo violations, and stigma load the disgust and identity systems with targets that the evolved machinery would not have selected on its own. The aversion is felt as righteous and natural. It is neither.
Most of what the conscious self labels as aversion is absence-of-attraction, not active aversion. The two are different registers. The null state has no push. Active aversion has full push. The flip from null to interesting is cheap. The flip from active aversion back to interesting is expensive or impossible.
Aversion is not failure of love.
It is a specific defense and screening assembly.
It exists for reasons that long predate the conscious self.
The mechanism does not consult the conscious self before firing.
What the reader does with this is their business.
CITATIONS
Disgust System
Rozin, P. & Fallon, A.E. (1987). “A perspective on disgust.” Psychological Review, 94(1), 23-41.
Tybur, J.M., Lieberman, D. & Griskevicius, V. (2009). “Microbes, mating, and morality: individual differences in three functional domains of disgust.” Journal of Personality and Social Psychology, 97(1), 103-122.
Curtis, V., Aunger, R. & Rabie, T. (2004). “Evidence that disgust evolved to protect from risk of disease.” Proceedings of the Royal Society B, 271(Suppl 4), S131-S133.
Curtis, V., de Barra, M. & Aunger, R. (2011). “Disgust as an adaptive system for disease avoidance behaviour.” Philosophical Transactions of the Royal Society B, 366(1563), 389-401.
Wicker, B., Keysers, C., Plailly, J., Royet, J.P., Gallese, V. & Rizzolatti, G. (2003). “Both of us disgusted in my insula: the common neural basis of seeing and feeling disgust.” Neuron, 40(3), 655-664.
Schaller, M. (2006). “Parasites, behavioral defenses, and the social psychological mechanisms through which cultures are evoked.” Psychological Inquiry, 17(2), 96-101.
Schaller, M. & Park, J.H. (2011). “The behavioral immune system (and why it matters).” Current Directions in Psychological Science, 20(2), 99-103.
MHC and Kinship
Wedekind, C., Seebeck, T., Bettens, F. & Paepke, A.J. (1995). “MHC-dependent mate preferences in humans.” Proceedings of the Royal Society B, 260(1359), 245-249.
Roberts, S.C., Gosling, L.M., Carter, V. & Petrie, M. (2008). “MHC-correlated odour preferences in humans and the use of oral contraceptives.” Proceedings of the Royal Society B, 275(1652), 2715-2722.
Havlíček, J. & Roberts, S.C. (2009). “MHC-correlated mate choice in humans: a review.” Psychoneuroendocrinology, 34(4), 497-512.
Westermarck, E. (1891). The History of Human Marriage. Macmillan.
Wolf, A.P. (1995). Sexual Attraction and Childhood Association: A Chinese Brief for Edward Westermarck. Stanford University Press.
Lieberman, D., Tooby, J. & Cosmides, L. (2007). “The architecture of human kin detection.” Nature, 445(7129), 727-731.
Asymmetric Trait Detection
Thornhill, R. & Gangestad, S.W. (1993). “Human facial beauty: averageness, symmetry, and parasite resistance.” Human Nature, 4(3), 237-269.
Thornhill, R. & Gangestad, S.W. (1999). “Facial attractiveness.” Trends in Cognitive Sciences, 3(12), 452-460.
Rhodes, G. (2006). “The evolutionary psychology of facial beauty.” Annual Review of Psychology, 57, 199-226.
Jones, B.C., Little, A.C., Burt, D.M. & Perrett, D.I. (2004). “When facial attractiveness is only skin deep.” Perception, 33(5), 569-576.
Singh, D. (1993). “Adaptive significance of female physical attractiveness: role of waist-to-hip ratio.” Journal of Personality and Social Psychology, 65(2), 293-307.
Singh, D. & Singh, D. (2011). “Shape and significance of feminine beauty: an evolutionary perspective.” Sex Roles, 64(9-10), 723-731.
Hodges-Simeon, C.R., Gaulin, S.J.C. & Puts, D.A. (2010). “Different vocal parameters predict perceptions of dominance and attractiveness.” Human Nature, 21(4), 406-427.
Pisanski, K., Fraccaro, P.J., Tigue, C.C., O’Connor, J.J., Röder, S., Andrews, P.W., Fink, B., DeBruine, L.M., Jones, B.C. & Feinberg, D.R. (2014). “Vocal indicators of body size in men and women: a meta-analysis.” Animal Behaviour, 95, 89-99.
Willis, J. & Todorov, A. (2006). “First impressions: making up your mind after a 100-ms exposure to a face.” Psychological Science, 17(7), 592-598.
Behavioral Cues and Deception
Ekman, P. (1992). “An argument for basic emotions.” Cognition and Emotion, 6(3-4), 169-200.
Ekman, P. (2003). Emotions Revealed. Times Books.
Porter, S. & ten Brinke, L. (2008). “Reading between the lies: identifying concealed and falsified emotions in universal facial expressions.” Psychological Science, 19(5), 508-514.
Ridgeway, C.L. (2014). “Why status matters for inequality.” American Sociological Review, 79(1), 1-16.
Anderson, C. & Kilduff, G.J. (2009). “The pursuit of status in social groups.” Current Directions in Psychological Science, 18(5), 295-298.
Bowlby, J. (1969). Attachment and Loss, Vol. 1: Attachment. Basic Books.
Bowlby, J. (1973). Attachment and Loss, Vol. 2: Separation. Basic Books.
Wanting, Liking, Prediction Error
Berridge, K.C. & Robinson, T.E. (1998). “What is the role of dopamine in reward: hedonic impact, reward learning, or incentive salience?” Brain Research Reviews, 28(3), 309-369.
Schultz, W., Dayan, P. & Montague, P.R. (1997). “A neural substrate of prediction and reward.” Science, 275(5306), 1593-1599.
Schultz, W. (2016). “Dopamine reward prediction error coding.” Dialogues in Clinical Neuroscience, 18(1), 23-32.
Eastwood, J.D., Frischen, A., Fenske, M.J. & Smilek, D. (2012). “The unengaged mind: defining boredom in terms of attention.” Perspectives on Psychological Science, 7(5), 482-495.
Danckert, J. & Merrifield, C. (2018). “Boredom, sustained attention and the default mode network.” Experimental Brain Research, 236(9), 2507-2518.
Identity Defense and Social Comparison
Steele, C.M. (1988). “The psychology of self-affirmation: sustaining the integrity of the self.” Advances in Experimental Social Psychology, 21, 261-302.
Festinger, L. (1954). “A theory of social comparison processes.” Human Relations, 7(2), 117-140.
Smith, R.H. & Kim, S.H. (2007). “Comprehending envy.” Psychological Bulletin, 133(1), 46-64.
Mate Selection Under Threat
Buss, D.M. (1989). “Sex differences in human mate preferences: evolutionary hypotheses tested in 37 cultures.” Behavioral and Brain Sciences, 12(1), 1-49.
Sadalla, E.K., Kenrick, D.T. & Vershure, B. (1987). “Dominance and heterosexual attraction.” Journal of Personality and Social Psychology, 52(4), 730-738.
Contempt and Relational Rupture
Gottman, J.M. (1994). What Predicts Divorce: The Relationship Between Marital Processes and Marital Outcomes. Lawrence Erlbaum.
Gottman, J.M. (1999). The Marriage Clinic: A Scientifically-Based Marital Therapy. W.W. Norton.
Janoff-Bulman, R. (1992). Shattered Assumptions: Towards a New Psychology of Trauma. Free Press.
Negativity Bias and Loss Aversion
Kahneman, D. & Tversky, A. (1979). “Prospect theory: an analysis of decision under risk.” Econometrica, 47(2), 263-292.
Rozin, P. & Royzman, E.B. (2001). “Negativity bias, negativity dominance, and contagion.” Personality and Social Psychology Review, 5(4), 296-320.
Cultural Amplifiers
Tajfel, H. & Turner, J.C. (1979). “An integrative theory of intergroup conflict.” In W.G. Austin & S. Worchel (Eds.), The Social Psychology of Intergroup Relations (pp. 33-47). Brooks/Cole.
Haslam, N. (2006). “Dehumanization: an integrative review.” Personality and Social Psychology Review, 10(3), 252-264.
Tetlock, P.E., Kristel, O.V., Elson, S.B., Lerner, J.S. & Green, M.C. (2000). “The psychology of the unthinkable: taboo trade-offs, forbidden base rates, and heretical counterfactuals.” Journal of Personality and Social Psychology, 78(5), 853-870.
Goffman, E. (1963). Stigma: Notes on the Management of Spoiled Identity. Prentice-Hall.
Related Machineries
-
THE MACHINERY OF DESIRE. The pull-system this piece opposes. Wanting is the dopamine-driven approach signal. Aversion is the multi-system push. The wanting drain (Part Six) is the structural bridge: when prediction error chronically goes negative, the wanting circuit collapses, and what remains is often the active push.
-
THE MACHINERY OF FEAR. Fear and aversion overlap but are not identical. Fear is the threat-detection register. Aversion is the screening output. Fear-without-respect (Part Nine) is the integration: threat coupled with benefit produces attraction-with-anxiety; threat decoupled from benefit collapses to aversion. Aversion stickiness (Part Eleven) runs on the same extinction-resistance architecture as conditioned fear.
-
THE MACHINERY OF NOTHING. The gray zone (Part Twelve) lives here. Most of what is labeled aversion is the null state, not active push. The two registers are different and have different operational consequences. Mistaking null for negative is one of the most common reading errors in inter-human perception.
-
THE MACHINERY OF TRUST. Trust collapse (Part Ten) is one of the two most destructive aversion forms. Discovered deception reclassifies the target globally and rewrites all prior interactions under an aversive frame. The trust integration runs on cycles. The collapse can happen in a single event.
Document compiled from primary source research across evolutionary psychology, social neuroscience, behavioral immunology, prospect theory, and the literature on disgust, attraction, and relational rupture. Every structural claim traces to a named primary source.