By – Dr Srabani Basu
Associate Professor, Department of Literature and Languages, SRM University AP, Amaravati.
The greatest prison is rarely built with walls. It is built with predictions.
Long before we fall in love with another person, our brain has already fallen in love with certainty. It would rather inhabit a familiar storm than an unfamiliar sky. And perhaps that is why some of the hardest doors to walk through are the ones that lead to freedom.
Love is often described as irrational. We fall in love with people who hurt us, defend those who repeatedly disappoint us, and sometimes remain emotionally invested long after a relationship has become psychologically destructive. Friends wonder why someone “simply doesn’t leave.” Therapists hear the same question repeatedly. Even individuals trapped in toxic relationships often ask themselves, “Why am I still here when I know this is hurting me?”
The answer may not lie solely in love, dependency, low self-esteem, or even trauma. While these undoubtedly influence our choices, contemporary neuroscience suggests that something even more fundamental is at work. The human brain is not primarily designed to maximise happiness. Its foremost priority is to minimise uncertainty. More than seeking pleasure, it seeks predictability.
This seemingly simple principle may explain why people often choose familiar pain over unfamiliar peace.
For decades, neuroscience viewed the brain as a remarkably sophisticated machine that reacted to information arriving through the senses. Today, however, one of the most influential theories in cognitive neuroscience proposes something radically different. The brain is fundamentally a prediction engine. Rather than waiting for reality to unfold, it constantly generates models of what it expects to happen next and then compares incoming information against these predictions. Every perception, every emotion, and every decision emerges from this continuous dialogue between expectation and experience.
Neuroscientist Karl Friston’s Free Energy Principle and the broader framework of predictive processing suggest that the brain works tirelessly to reduce prediction error—that uncomfortable gap between what it expects and what actually occurs. In simple terms, the brain is continuously asking a single question: What is most likely to happen next?
Notice what this question is not asking.It is not asking, What will make me happiest?Nor is it asking, What will help me flourish?Its concern is far more primitive: What can I reliably predict?This distinction changes the way we understand human relationships.
Let us consider two individuals. One grows up in an environment where affection is consistent, disagreements are resolved respectfully, and emotional safety is the norm. The other’s childhood is marked by unpredictability where love is mixed with criticism, warmth is interrupted by neglect and, approval alternates with rejection. Both brains become extraordinarily efficient prediction machines, but they are predicting very different worlds.
For the first individual, kindness feels normal because it matches existing neural expectations.
For the second, inconsistency feels strangely familiar.
Years later, when both encounter romantic partners, their brains are not merely evaluating behaviour. They are comparing present experiences against deeply established predictive models.
This explains one of the most puzzling phenomena in toxic relationships.People often mistake familiarity for compatibility.
The relationship may be objectively unhealthy, yet subjectively it “feels right” because it resonates with long-established neural predictions. Conversely, a genuinely secure relationship can feel unsettling—not because it is unsafe, but because it violates the brain’s expectations.
Psychologists have long observed that familiarity powerfully influences attraction. The mere exposure effect demonstrates that repeated exposure increases preference, while attachment theory explains how early caregiving shapes later relational expectations. Predictive neuroscience offers an elegant bridge between these findings. The brain does not simply remember previous relationships; it uses them to forecast future ones.In many ways, every relationship becomes a prediction.
This also sheds light on why leaving a toxic relationship can provoke profound psychological distress, even when the individual intellectually recognises that separation is the healthier choice.
From the outside, walking away appears to reduce suffering.From the brain’s perspective, however, it introduces something far more alarming than pain.It introduces uncertainty.The unknown partner, the unknown future, the unknown identity and the unknown emotional landscape.Ironically, predictable suffering often feels neurologically safer than unpredictable freedom.
This preference for predictability becomes even stronger through one of psychology’s oldest discoveries: intermittent reinforcement.Behavioural psychologist B. F. Skinner demonstrated that behaviours rewarded unpredictably become remarkably resistant to extinction. Slot machines operate on precisely this principle. A gambler does not know when the next reward will arrive, yet the uncertainty itself strengthens persistence.
Human relationships can unknowingly operate in much the same way.A partner who alternates between affection and withdrawal creates a powerful cycle of hope. Every unexpected act of kindness becomes psychologically magnified because it arrives after emotional deprivation. Dopaminergic systems involved in reward prediction become highly engaged, not merely because rewards occur, but because they occur unpredictably. Neuroscientist Wolfram Schultz’s work on dopamine illustrates that dopamine is closely tied to prediction error; the difference between expected and actual outcomes, rather than pleasure alone.
This is precisely the reasonbehind individuals often report feeling “addicted” to relationships that simultaneously exhaust them.They are not addicted to suffering.They become trapped within cycles of uncertain reward, the next apology, the next affectionate message, the next promise, and the next glimpse of the person they first fell in love with.Each unexpected positive experience updates the brain’s predictions just enough to preserve hope, but not enough to establish lasting security.Hope, paradoxically, can become the mechanism that prolongs suffering.
There is another fascinating psychological process at work.The brain dislikes incomplete stories.Social psychologist Arie Kruglanski describes the human need for cognitive closure – the desire to resolve ambiguity and achieve certainty. When relationships remain emotionally unfinished, our minds continue searching for explanations, replaying conversations, reconstructing memories, and imagining alternative endings.
This persistent mental rehearsal is not simply emotional weakness.It is the predictive brain attempting to reduce uncertainty.The relationship may have ended externally.Internally, however, prediction continues.
“What if they change?”
“What if I misunderstood?”
“What if this one conversation could fix everything?”
The brain continues generating simulations because unresolved prediction consumes cognitive resources.In this sense, closure is not merely emotional.It is computational.
The same predictive architecture also influences memory itself. Contrary to popular belief, memory does not function like a video recorder. Each recollection is reconstructed rather than replayed. During every act of remembering, memories become temporarily malleable before being stored again, a phenomenon known as memory reconsolidation. This explains why individuals often remember the extraordinary moments of tenderness while gradually minimising months or years of emotional neglect. The predictive brain preferentially preserves memories that sustain its existing relational model, making it easier to justify staying than to rewrite the entire narrative of one’s life.
The consequence is profound. Over time, the relationship ceases to exist merely between two people.It becomes embedded within one’s identity.It becomes embedded within one’s identity.At this point, leaving is no longer experienced merely as the loss of another person. It begins to feel like the loss of oneself.
Modern psychology has consistently shown that our identities are constructed through narratives. We are, in many ways, the stories we repeatedly tell ourselves. “I am the one who fixes people.” “I never give up.” “I can love someone enough to heal them.” “This relationship has taught me too much to abandon now.” These narratives gradually become cognitive schemas which are deeply held mental structures that organise how we perceive ourselves and others. Aaron Beck’s work on cognitive schemas and Jeffrey Young’s Schema Therapy illustrate how these enduring patterns influence perception, emotion, and behaviour.
Once a relationship becomes woven into the architecture of identity, every attempt to leave creates what psychologists call cognitive dissonance – the psychological discomfort that arises when behaviour conflicts with deeply held beliefs about oneself. The mind instinctively attempts to reduce this discomfort, often not by changing the belief, but by changing the interpretation of reality.
The insult becomes “he was just stressed.”
The manipulation becomes “she had a difficult childhood.”
The repeated betrayal becomes “everyone deserves another chance.”
These explanations are not necessarily deliberate acts of self-deception. They are the brain’s remarkably sophisticated attempts to preserve coherence. A coherent story—even a painful one—is neurologically easier to maintain than an identity in fragments.
This may also explain why well-meaning advice from family and friends so often fails. They see behaviour. The individual inside the relationship experiences an entire predictive world. To someone standing outside, the solution appears obvious. To the brain inside the system, leaving represents the collapse of a carefully constructed internal model that has often been built over decades.This distinction is crucial.People do not remain in toxic relationships because they are weak.They remain because their brains are extraordinarily efficient prediction machines attempting to preserve internal stability.
This perspective also offers an interesting bridge to Neuro-Linguistic Programming (NLP). Although NLP should not be regarded as established neuroscience, one of its foundational premises resonates with contemporary cognitive psychology: people respond not directly to reality but to their internal representations of reality. In NLP, this is expressed through the well-known presupposition, “The map is not the territory,” borrowed from Alfred Korzybski’s General Semantics.
Whether one approaches this idea through predictive processing or through NLP’s concept of internal maps, the implication is remarkably similar. Human beings rarely respond to events as they objectively are. We respond to what our brains have learned to expect those events to mean.
Consider two individuals receiving the same delayed text message from their partners.One thinks, “They must be busy.”The other immediately concludes, “They are losing interest.”The external event is identical.The internal prediction is not.What differs is not reality but the predictive model through which reality is interpreted.
NLP also highlights the role of anchors, language patterns, beliefs, and internal representations in shaping subjective experience. While these concepts arise from a different intellectual tradition than neuroscience, they provide useful practical tools for increasing awareness of the mental filters through which people experience relationships. Awareness, in itself, does not erase old patterns. But it creates the possibility of interrupting them.
The encouraging news is that predictive models are not fixed. One of the defining characteristics of the human brain is neuroplasticity; its ability to reorganise itself in response to repeated experience. Every genuinely safe relationship, every healthy boundary, every act of self-respect, and every experience that contradicts an old expectation becomes what neuroscientists call a prediction error. If experienced consistently enough, these prediction errors gradually persuade the brain to revise its model of the world.
This process is neither dramatic nor instantaneous.The brain does not abandon familiar predictions simply because we understand them intellectually.Insight is the beginning of change, not its completion.Change occurs when lived experience repeatedly tells the brain, “Your old prediction is no longer the most accurate one.”
This has profound implications for healing.Recovery from a toxic relationship is not merely about “moving on.” It is about teaching the brain to expect something different. Safety must become predictable. Respect must become familiar. Calm must cease to feel suspicious. Consistency must become more recognisable than chaos.
Perhaps this explains why some individuals initially feel uncomfortable in genuinely healthy relationships. They describe them as “boring,” “too quiet,” or “lacking chemistry.” What they are often experiencing is not the absence of love but the absence of unpredictability. The nervous system, conditioned to equate emotional intensity with intimacy, struggles to recognise stability as a form of connection.
In reality, mature love is rarely dramatic.It is reliable.It is emotionally coherent.It allows the brain to expend less energy anticipating danger and more energy engaging in growth, creativity, and genuine intimacy.Ultimately, the opposite of a toxic relationship is not simply a healthy partner.It is a healthy prediction.The greatest prison is not another person’s behaviour.It is the internal model that convinces us that such behaviour is normal, inevitable, or deserved.
The most courageous act, therefore, is not merely walking away from another human being. It is allowing the brain to experience enough new evidence that it finally dares to make a different prediction about life itself.
Perhaps freedom begins at the precise moment when the mind stops asking, “What am I used to?” and starts asking a far more transformative question:
“What if the unfamiliar is not dangerous, but simply healthier than anything I have known before?”
The human brain has spent millions of years learning to predict survival.Our challenge is to teach it something evolution never explicitly promised:That sometimes, the safest future is the one it has never predicted.
