| Getting your Trinity Audio player ready... |

Why do loving partners so often end up locked in cycles of misunderstanding?
One person says something neutral; the other hears criticism. A late reply to a message is felt as rejection. A sigh becomes proof of disappointment. Before long, both partners are convinced they are responding to what the other is doing, while neither feels seen or understood.
Modern neuroscience and philosophy offer a powerful way of understanding these familiar moments of relational distress. According to philosopher and cognitive scientist Andy Clark, the human mind does not passively register the world as it is. Instead, it actively predicts what is happening and then checks those predictions against incoming evidence. This approach—known as predictive processing—helps explain why perception is usually smooth and efficient, but also why it so often goes wrong, especially in emotionally charged relationships.
Seen through this lens, many interpersonal conflicts are not caused by malice, selfishness, or poor communication skills alone. They arise because two predictive minds, each shaped by their own histories, expectations, and emotional states, are trying—and failing—to make sense of one another.
The Brain as a Prediction Machine
At the heart of predictive processing is a simple but radical idea: the brain is a prediction engine.
Rather than waiting for the world to deliver information and then reacting to it, the brain is constantly generating guesses about what it expects to perceive. These guesses—called predictions—are based on past experience, learned patterns, current context, bodily states, and emotional needs.
Incoming sensory information (sounds, sights, words, facial expressions) is then compared with these predictions. The difference between what was expected and what actually arrives is known as a prediction error.
• If prediction and input match, perception feels effortless and obvious.
• If they don’t match, the brain must reduce the error—either by updating its interpretation or by acting to change the situation.
Crucially, this process happens continuously and unconsciously. We don’t experience ourselves as guessing; we experience ourselves as seeing the world directly. But what we experience is always the brain’s best attempt to make sense of incomplete and ambiguous information.
Clark describes this process as hierarchical. Lower levels of the brain deal with raw sensory details—tone of voice, facial muscle movement, posture. Higher levels deal with meaning and intention—“She’s annoyed,” “He doesn’t care,” “This is a threat to the relationship.”
In intimate relationships, these higher-level interpretations are especially powerful, because what is at stake is not just understanding the world, but feeling safe, valued, and loved.
Why Relationships Are a Perfect Storm for Misperception
Human relationships—and romantic partnerships in particular—are uniquely vulnerable to predictive error.
First, the signals we rely on are often ambiguous. A brief text, a change in tone, a distracted look can mean many things. There is rarely enough information to determine meaning with certainty.
Second, relationships carry emotional weight. The brain assigns high importance—or precision, in predictive processing terms—to cues that seem relevant to attachment, rejection, or threat. This means prediction errors in relationships feel more urgent and more personal.
Third, partners develop strong priors—deep expectations about how the other person thinks, feels, and behaves. These priors are shaped not only by the relationship itself, but by earlier experiences of intimacy, conflict, neglect, or betrayal.
Once established, these expectations don’t simply sit in the background. They actively shape what partners perceive.
Overweighted Priors: Seeing What We Expect to See
One of the most important ways predictive processing can go wrong is through overweighted priors—when expectations dominate perception.
In relationships, this often sounds like:
• “He always dismisses me.”
• “She’s never really listening.”
• “If I matter, they would know what I need.”
Once such expectations become entrenched, the brain begins to interpret incoming information through them. Neutral or ambiguous cues are pulled into line with what is already expected.
A tired silence becomes indifference. A delayed reply becomes avoidance. A practical suggestion becomes criticism.
This is not dishonesty or deliberate distortion. It is perception doing what it evolved to do: reducing uncertainty by relying on what it already “knows.”
In couples with a history of conflict, these priors can become so strong that even clear counter-evidence fails to register. Apologies feel insincere. Acts of care are discounted. The brain effectively says, “This doesn’t fit my model, so I won’t take it seriously.”
Underweighted Sensory Evidence: When New Information Can’t Get In
The flip side of strong priors is underweighting sensory evidence—giving too little importance to what is actually happening now.
In distressed relationships, partners often stop updating their internal models of each other. They respond to who they think the other person is, rather than who they are becoming or trying to be.
This is why reassurance often fails in couples caught in negative cycles. One partner offers genuine care, but the other cannot register it as meaningful. The prediction—“I’m not important”—is treated as more reliable than the evidence in front of them.
Predictive processing helps explain why change in relationships is so hard. New behaviour must compete with old models that have been reinforced over years, sometimes decades. The brain, designed to favour stability, resists updating unless prediction errors are strong, repeated, and emotionally safe enough to be processed.
Hyperactive Error Detection: When Everything Feels Like a Problem
In some individuals and couples, the system swings the other way. Instead of ignoring prediction errors, the brain becomes overly sensitive to them.
This is often seen in anxiety-prone relationships, where partners are constantly scanning for signs of trouble.
• A change in routine feels ominous.
• A brief withdrawal triggers alarm.
• Minor misunderstandings escalate rapidly.
Here, prediction errors are not smoothed over but amplified. The nervous system behaves as if the relationship is fragile and under threat, even when no clear danger is present.
Predictive processing frames anxiety not simply as “worry,” but as a state in which the brain expects uncertainty and threat, and therefore treats small deviations as highly significant.
In couples, this can lead to relentless checking, reassurance-seeking, or conflict escalation—all attempts to reduce unbearable uncertainty.
Filling in the Gaps: The Brain as a Meaning-Maker
Because interpersonal information is often incomplete, the brain routinely fills in missing details.
This process is usually helpful. It allows us to understand tone, intention, and emotion without needing explicit explanation every time. But it also creates fertile ground for misinterpretation.
Consider how often partners infer meaning:
• “The way you said that means you’re angry.”
• “If you loved me, you’d want to talk.”
• “Your silence says everything.”
From a predictive processing perspective, these are not conclusions reached after careful reasoning. They are rapid perceptual inferences, experienced as obvious facts rather than interpretations.
Two partners can witness the same behaviour and experience entirely different realities—not because one is lying or irrational, but because each brain is completing the picture in its own way.
Active Inference: How We Make Our Predictions Come True
Andy Clark emphasises that prediction is not only about perception. It also guides action.
We don’t just update our models of the world; we act in ways that make the world fit our expectations. This process is known as active inference.
In relationships, this can become a self-fulfilling loop.
If I predict rejection, I may withdraw or become defensive. My partner experiences distance or hostility and responds in kind. The outcome—disconnection—then confirms my original prediction.
From the inside, this feels like reacting to reality. From the outside, it is a tragic choreography in which both partners are trying to protect themselves from prediction error and emotional pain.
Why “Just Communicate Better” Often Fails
Advice about relationships often emphasises communication: say what you mean, listen carefully, clarify misunderstandings. While valuable, this advice underestimates how deeply perception itself is shaped by prediction.
When partners are emotionally dysregulated, stressed, or carrying unresolved hurt, they are not simply exchanging information. They are perceiving through competing predictive models.
Words alone cannot easily override priors formed through years of experience. This is why couples can talk endlessly and still feel unheard. The difficulty is not lack of data, but how that data is weighted and interpreted.
Changing the System: What Predictive Processing Suggests for Couples
Predictive processing does not offer quick fixes, but it does point to new ways of understanding relational change.
First, it highlights the importance of emotional safety. The brain is more willing to update its models when it does not feel under threat. Harsh confrontation often strengthens priors rather than loosening them.
Second, it suggests that change requires repeated, embodied experiences, not just insight. Small, consistent actions that violate negative expectations—kindness, reliability, responsiveness—gradually generate prediction errors that the system can no longer ignore.
Third, it reframes conflict. Instead of asking “Who is right?”, couples can ask “What predictions are we each bringing into this moment?” This shift alone can soften defensiveness and open curiosity.
Two Realities, One Relationship
Perhaps the most humane implication of predictive processing is this: misperception is normal.
Human experience is not a mirror of the world but a best guess, shaped by history, emotion, and need. In intimate relationships, where vulnerability is high and information is incomplete, misperception is almost inevitable.
Understanding this does not excuse harm, but it does invite compassion. Partners are not simply reacting to each other’s actions, but to the meanings their brains are generating in real time.
Andy Clark’s framework reminds us that the mind is doing its best to minimise surprise in a noisy, uncertain world. Sometimes that effort brings us closer together. Sometimes it drives us apart.
The task of a relationship, then, is not to eliminate prediction error, but to create conditions in which our predictions about one another can remain flexible, revisable, and open to being pleasantly surprised.
Final Reflection
Predictive processing shows that interpersonal conflict is often less about what happens than about how it is perceived. In couples, distress frequently arises when outdated expectations overpower present-moment reality.
Seen this way, healing is not just about better behaviour or better arguments, but about gently reshaping the predictive landscapes through which partners encounter each other.
Misperception, in this view, is not a failure of love or intelligence. It is the price we pay for having minds that must anticipate before they can understand—and for loving in a world where certainty is always just out of reach.
Discover more from Richard Askew | Actualise Psychotherapeutic Counselling
Subscribe to get the latest posts sent to your email.

