The Familiar Illusion: Relationship, Mental Images, and the Limits of Certainty

Further Reading

A clinical reflection on Krishnamurti, Gibran, Rumi, and the internal models we sometimes mistake for the person

Jason Joseph, LPC, LMHC, CPC · Updated July 2026 · 23 min read

The Moment the Model Fails

In therapy, I often see people get caught on a particular painful moment: someone important to us does something that no longer fits the person we thought we knew.

The moment may seem ordinary at first: a partner’s tone changes, a friend doesn’t respond in the expected way, or someone we thought we understood does something we couldn’t have predicted. From the outside, the event may not look dramatic enough to explain the reaction. Internally, though, something has shifted. We return to these moments, sometimes obsessively, because the mind is responding to more than the event. It is responding to a collapse in certainty. The event hurts or confuses us, and it also threatens the model we had constructed of the person in our head.

A great deal of relationship distress is already described in familiar clinical language: attachment injury, abandonment fear, unmet needs, boundary violation, and betrayal. Those descriptions are useful, but they don’t fully capture the distress that occurs when someone’s behavior violates the internal image we’ve built of them.

I’m using image in a specific clinical sense. It’s narrower than fantasy or conscious idealization. I mean the working model of another person that forms through memory, repetition, injury, repair, and thousands of small interpretive acts that rarely announce themselves as interpretation. The image may contain accurate pattern recognition and hard-won information. It may protect us from ignoring what experience has already taught us.

The risk is that we forget it’s an image at all. Once that happens, the internal model can quietly begin to operate as if it were the person.

At that point, we no longer experience ourselves as making predictions. We experience the image as certainty. We don’t think, “Based on my history with this person, I expect them to behave in this specific way.” We think, “This is who they are.” Once that shift has occurred, a behavior that contradicts the model can feel more destabilizing than the behavior alone would explain. The mind is not responding only to an event. It is responding to a relationship map that no longer works.

That’s one reason certain relationship moments become strangely difficult to metabolize. The hurt may be real, but the looping quality of it often carries a quieter question underneath: if this person is who I thought they were, how could they have done that?

The mind then tries to make the relationship make sense again. It replays the conversation, searches for missed signs, reorganizes the past, and revises the meanings it had assigned to events. It may begin asking whether it misunderstood the person, the relationship, or its own trust in the first place. In therapy, this can be mistaken for simple rumination, and sometimes it is. In many cases, though, it’s also a repair attempt by a mind whose relationship map has failed.

The Image as Map and Substitute

Human beings don’t encounter one another as blank slates. We relate through memory, expectation, threat detection, and prediction. No relationship can function without some internal model of the other person. We need continuity. We need to remember who someone has been to us, what has hurt, what has repaired, and what history has taught us to notice.

Having an image of another person isn’t inherently a problem. What matters is the role it has begun to play: does it help us stay oriented to the relationship, or has it started to stand in for the relationship itself?

Some mental images or maps do help us when they remain flexible enough to be revised. They can allow us to remember the past without freezing the other person inside it. They can help us recognize that a partner gets quiet when overwhelmed, that a friend needs more time to respond, or that a relationship has shown us enough repeated harm that distance is necessary.

Other images begin to replace actual contact. They become too complete, too certain, or too resistant to revision. The living person starts to disappear behind the model, and we respond less to what’s happening now than to what the mental image insists must be happening.

The image can be positive or negative; either way, a person can become trapped inside our mental picture of them, whether we’re idealizing them or explaining them too completely. “I know exactly what you meant.” “You always do this.” “This is just who you are.” Statements like these may contain some truth, but they can also impose the image back onto the person rather than allowing us to notice where the image has failed to represent them fully.

At that point, the image has stopped functioning as a map and has started functioning as a substitute. A map helps us move through unfamiliar ground with more orientation. It becomes dangerous when we forget that the terrain remains larger, more complex, and more alive than the map can ever be. A map is not the territory, but it can become part of the territory we’re actually living inside.

A pattern can be real without being the whole truth.

Krishnamurti, Gibran, and Rumi: Three Languages for Relationship and Image

Krishnamurti, Gibran, and Rumi approach this relational terrain from different directions. I’m not treating them as interchangeable writers or as representatives of one shared tradition. The connection here is narrower: each offers a different way of thinking about the space between the living person and the image, longing, or interpretation we bring to them. Krishnamurti gives the clearest language for the mental image that can stand as a barrier between people. Gibran describes an imagined beloved: the person love can partly understand but never finish knowing, along with the space required if love is not to become possession. Rumi raises the inward question of perception: what in the perceiver shapes what’s being perceived?

Placed alongside related clinical frameworks, including attachment theory, Imago, schema therapy, and ACT, these writers bring the central question into sharper focus. When does the mental image help us stay oriented to another person, and when does it begin to replace direct contact with the living person?

Krishnamurti’s language is the most direct. In a 1950 public talk in Bombay, India, he argues that communion is possible only when there is no barrier, and names “an idea, a prejudice” as one form such a barrier can take. Nearly three decades later, in a 1979 public talk in Ojai, California, he brings that concern into the language of image-making. Relationship, he says, is often based on “the images that you have built about each other,” and “this image is the divisive factor between the two.” Clinically, these passages point toward the same practical concern: the accumulated picture of another person can begin to stand between us and direct contact with the living person.

That’s a stark claim. It becomes clinically useful when treated as an inquiry rather than a slogan. Sometimes the idea of the person becomes more available to us than the person. The relationship may still involve two living people, but it also starts to involve two accumulated images trying to protect themselves, confirm themselves, or avoid being surprised.

I find Krishnamurti most useful here because his language sharpens a practical distinction: memory can help us stay oriented to another person, and memory can also become the thing that stands in the way of contact. If a person has repeatedly lied, manipulated, abused, or crossed boundaries, the image formed from those experiences may contain necessary information. It may protect us from minimizing danger, repeating an unsafe pattern, or returning to a relationship that has already caused serious harm. That’s not the kind of certainty I’m arguing against.

The concern is what happens when the evidence is more ambiguous, the relationship is still unfolding, or the image begins turning inference into fact. Then the image can harden into a complete account of the person and begin standing in for contact with them.

Gibran approaches the problem differently. His language is romantic, ornate, and sometimes sweeping; but beneath that style is a clinically useful distinction: another person may become deeply meaningful to us without becoming fully knowable, finished, or possessed. In his poetic novella The Broken Wings, he offers a striking image of the beloved as someone partly knowable and never finished. Selma becomes “a book whose pages I could understand” but “which I could never finish reading.” That one line preserves the possibility of real understanding, while rejecting the fantasy that another person can ever become finished, fully known, or finally contained.

We do know one another, meaningfully if imperfectly. We know gestures, values, injuries, promises, and limits. Some of that knowledge is precious; some of it is protective. Intimacy, though, doesn’t mean the other person has been fully read, or can be fully comprehended at all.

Gibran’s better-known language in The Prophet makes the ethical point more directly: love requires space, and even accurate understanding can become possessive when it hardens into certainty. The familiar wedding line about “spaces in your togetherness” can sound sentimental when it’s pulled out of context and read aloud under flower arches. Taken seriously, the line is demanding. Certainty may make another person feel easier to manage, while actually making genuine contact with them more difficult. The person we love is never truly honored by our minds making them small enough to fit inside our certainty.

Rumi adds a different caution, rooted less in possession than in perception. In the opening passage of the Masnavi, a six-book poem central to Sufi mystical literature, the reed flute says, “Each interprets my notes in harmony with his own feelings.” Its original context is a religious and poetic meditation on longing for the divine, not a psychological theory of relationship. Even so, the line names something psychologically recognizable. We don’t receive another person’s words, silences, gestures, or changes in tone from nowhere. We hear through longing, fear, memory, and expectation. Fatigue, culture, neurotype, and old injury can also shape what gets heard. Alongside “What did this person say?” there’s another question: “What in me is listening?”

That question doesn’t require us to distrust perception entirely. Sometimes what we perceive is definitively real. The condition of perception still matters. A fearful nervous system may detect a real signal and amplify it, while longing can build an entire future from a kindness, or old hurt can turn a delay into a verdict.

Together, these writers illuminate an old human problem. Krishnamurti gives us the image as a barrier between people. Gibran describes the beloved as never fully finished in our understanding and insists on the space love requires. Rumi turns attention inward toward whatever in the perceiver may be shaping perception. The person in front of us may be obscured by ignorance. More painfully, they may also be obscured by certainty built from partial knowing.

Clinical Neighbors: Images, Schemas, and Internal Models

Clinically, this isn’t unfamiliar territory, though different traditions would name different parts of it. Attachment theory gives us internal working models of self and other. Object relations theory studies internalized representations of significant others. Schema therapy examines enduring patterns that shape how a person interprets experience. Imago Relationship Therapy places the unconscious image of familiar love near the center of its couples work. ACT gives us cognitive fusion, where a thought is taken as reality itself.

These aren’t identical theories, and they shouldn’t be treated as if they’re saying the same thing. They do, however, share a practical concern: human beings rarely relate to the present moment in a purely direct way. We relate through expectation, history, fear, desire, and prior learning.

Trouble develops when the model of the person becomes more relationally powerful than full, ongoing contact with them

That question cuts across models. It appears in attachment injury, cognitive fusion, transference, schema activation, and limerence. It also appears in neurodivergent social pattern recognition and in the ordinary conflict of people who love each other and still manage to misunderstand each other spectacularly. No one clinical language owns it completely.

The distinction I keep returning to is between image as orientation and image as substitution. An orienting image helps us remember and respond. A substituting image begins to take over the encounter, telling us what must be happening before the encounter has fully unfolded. It may offer emotional protection while also leaving us alone inside our own certainty.

ACT offers another way to assess the image: is it helping the person move toward the life and relationships they value, or is it narrowing perception and reducing flexibility? In practice, that question is often more useful than deciding too quickly whether the image is simply accurate or inaccurate.

The Image I Carried

I first encountered Krishnamurti, Gibran, and Rumi as a lonely, as yet undiagnosed neurodivergent teenager in Lubbock, Texas.

There was a small bookstore there called The Odyssey. For me, it was less a store than a kind of oasis: strange books, philosophy, poetry, and a sense of belonging that felt almost entirely absent in the surrounding city. I was socially awkward, intellectually hungry, often mystified by other people, and trying to understand why interactions and relationships seemed to matter so much yet hurt so much, and make so little sense.

Another reader could have encountered these writers and never connected them at all. They come from different centuries, cultures, literary forms, and religious worlds. My pattern-seeking mind did something else: it drew lines between them. Whether that synthesis was accurate, conflated, or simply the work of a lonely teenager trying to make sense of human interactions, it hit me hard.

Over time, the synthesis became a saying I thought I’d remembered from Krishnamurti. Something like: “The closer you truly become to someone, the more they become like a fine mist, unknowable.”

I carried that quote in my mind for decades, and the embarrassing part is that I appear to have made it up. Or, more generously, my mind made an image out of several sources and then mistook the image for a quote. Krishnamurti’s language of images in relationship, Gibran’s beloved who could never be finished reading, and Rumi’s language of longing and interpretation all fused into something I remembered as if it had been written somewhere.

That mistake now seems like a lived example of the very problem I’m describing. Even my memory of the idea had become an image: built from real material, emotionally powerful, meaningful enough to endure, yet still not identical with the sources themselves.

This is also what happens in relationships. Real material gets gathered: a tone, a wound, a moment of closeness, a rupture. The mind condenses scattered experience into a model. Then the model begins to feel like the source.

Later, through clinical training, supervision, personal therapy, and years of work with clients, I found related language in several clinical frameworks. ACT, attachment theory, Imago, and schema therapy all gave me more precise ways to think about these patterns. But the misremembered image I had carried for years remained powerful. It gave me a way of understanding a particular kind of relational suffering: the pain that comes when the person exceeds the model, or when the model fails so badly that we no longer know what to trust.

My personal history, and the somewhat random way I encountered these writers, does not prove they belong together. But it does explain how they became connected in my mind, and why that connection has stayed clinically alive for me all these years later. Highly analytical minds, whether neurodivergent or not, can be remarkably good at building relational maps. Those maps may help us survive. They may also, at times, keep us from being fully present with the person we were trying so hard to understand.

Neurodivergent Pattern Recognition and Double Empathy

This image-making process is human; it’s not exclusively neurodivergent. For many ADHD, autistic, and AuDHD adults, though, it can become especially vivid. Years of decoding unclear social rules, masking, translating across neurotypes, or remaining vigilant for misunderstanding can intensify the process.

For people who have spent years trying to decode unclear rules, indirect communication, sensory overload, and inconsistent feedback, pattern recognition can become a form of translation. That translation work can become especially intense when the stakes include the subtle threat of being judged as too much, too literal, too intense, or not emotional in the expected way. The mind watches tone, response time, silence, and deviations from expected pattern. Underneath, it’s often trying to answer one question: Am I safe here, or am I about to be misunderstood or blindsided?

That pattern recognition may be intelligent and adaptive, yet still become difficult to disengage from or revise. A relational model can begin as protection and then become a burden, especially when the system that helped a person survive ambiguity starts treating ambiguity itself as danger.

The double empathy frame is essential here. The older claim that autistic people lack empathy or lack the ability to understand other minds has done real harm, partly because it turns cross-neurotype misunderstanding into a one-sided deficit located inside autistic people. The double empathy problem gives us a better frame: misunderstanding across neurotypes is often mutual, shaped by differences in communication style, sensory experience, social expectation, and interpretive context.

Neurotypical people also form distorted images of autistic people. A flat tone may be treated as indifference, directness as hostility, processing time as evasion, sensory overload as selfishness, and a need for solitude as rejection. Difficulty performing expected facial or vocal signals may be mistaken for lack of care, rather than understood as a difference in expression, regulation, or communication. These are images, too.

No neurotype has a monopoly on forming images or seeing clearly. All of us interpret, predict, and construct, and our images need to remain responsive to the living person in front of us.

For neurodivergent clients, therapy often has to hold two truths at once. Their pattern recognition may be meaningful, protective, and painfully earned. Their nervous system may also treat a pattern as a verdict before the full situation has been understood. Therapy can help perception stay flexible enough to remain in contact with reality without asking the person to become less perceptive.

This includes self-trust. Many neurodivergent adults have been told repeatedly that their perceptions are too sensitive, too literal, too dramatic, or too confusing to others. If therapy simply says, “Maybe you’re projecting,” it may repeat the injury. A more careful approach asks what the person noticed, what it reminds them of, what pattern it seems to fit, and what would help them respond without abandoning either self-trust or careful evaluation.

The aim is more accurate contact, including contact with one’s own perception.

What the Image May Be Trying to Solve

A useful clinical question is: What’s this image trying to solve?

Sometimes the image helps us stay oriented. We remember what someone tends to need, how they tend to respond, what has hurt before, and what history has already taught us to notice. Without some kind of working model, relationship would become nearly impossible. We would have no continuity, no learning, and no memory of the other person’s preferences, limits, or ways of moving through the world.

The image may also be trying to protect us. It can form around hurt, betrayal, neglect, rejection, or fear. It may say, “Don’t trust this too quickly,” or “This has happened before,” or “You need to pay attention.” Sometimes that protective image carries real wisdom. It may be the part of the mind that remembers what the hopeful self would rather ignore. The risk is that protection can become so organized around threat that ambiguity itself starts to look like evidence.

The image may be carrying longing. Another person can begin to hold the promise of being understood, chosen, rescued, or finally allowed to stop feeling alone. From the inside, it may feel less like fantasy than recognition. But when the image becomes too loaded with hope, the living person may not be able to bear the weight of what the image needs them to be.

The image can also become a forecast. The mind begins to treat the person as a prediction: they will leave, criticize, disappoint, dismiss, or fail to show up. Forecasting images can be built from real patterns. When they harden, the future becomes strangely unavailable. The other person is no longer free to surprise us, repair with us, grow with us, or even fail in a new way.

In actual relationships, these functions usually blur together. Orientation, protection, longing, fear, and prediction rarely arrive one at a time. The relevant question is whether these functions remain responsive to what’s actually happening, or whether they begin deciding in advance what the other person’s words and behavior must mean.

A frozen image can’t take in new information, register repair, or tolerate complexity. It may be positive or negative, romantic or suspicious, admiring or contemptuous. Either way, it stops behaving like a map and starts behaving like a prison.

When the Image Becomes Fixation: Limerence

Limerence deserves brief mention because it shows the mental image becoming more powerful than contact with the actual person.

Dorothy Tennov introduced the term “limerence” to describe an involuntary state of romantic preoccupation, intrusive longing, and intense desire for reciprocation. The word is now widely used online, where it can be helpful but is sometimes applied so broadly that it becomes another catch-all explanation. Its usefulness here is specific: in limerence, the imagined relationship can become so charged with longing and fear that the person in the mind feels more emotionally powerful than the person in reality.

A delayed message can become evidence in a private trial about worth and hope. A small kindness may keep an imagined future alive. A cool response may feel annihilating. The mind doesn’t merely think about the person; it begins to organize itself around them.

For ADHD, autistic, and AuDHD people, several factors can intensify the loop: hyperfocus, rejection sensitivity, sensory memory, or intermittent reward. That does not make neurodivergent people uniquely irrational in love. It means that, for some people, attention, longing, and nervous-system activation can bind to a person with unusual force.

The boundary issue matters. Intense feeling doesn’t confer access to another person. The actual person remains separate and autonomous. They’re allowed to say no, disappoint, withdraw, choose someone else, or simply not participate in the story. Gibran’s concern with non-possession becomes clinically relevant here. Love or longing can become a kind of intellectual and emotional possession when the other person is held inside an imagined future or private drama they may not have consented to inhabit.

Therapy can ask what emotional work the limerent image is doing. It may hold loneliness, proof of desirability, or the wish to be finally chosen. Once those questions can be asked directly, the image may still be painful and compelling, but it’s less able to present itself as destiny.

Keeping the Image Responsive to Reality

Therapy does not need to destroy the image. In most cases, doing so would be both impossible and undesirable. The goal is to help the image become visible, flexible, and accountable to the reality of the relationship.

When the distress seems tied to a collapse in the internal model, one clinical question often matters most: What image of this person was violated here?

From there, several related questions become useful. What did you believe this person would never do? What role had they come to occupy in your nervous system? What did their behavior seem to prove? A therapist might also ask what older story was activated, what information complicates the pattern, and what would change if the image were held as a hypothesis rather than a verdict.

In the talks collected as The Mirror of Relationship, Krishnamurti also returns to the problem of pattern. True relationship, in his language, requires alertness, adjustment, and a certain pliability of mind. Habit and fixed pattern interfere with that movement. This lands close to the clinical problem of the frozen image. We’ve formed a model of the other person, and the model may have become too rigid to move with the living relationship.

The ACT workability question helps here. Is the image helping the person stay oriented and clear, or is it pulling them into vigilance, longing, or control? Does it help the person move toward the kind of relationship and self-respect they value? Or does it keep them circling the same internal courtroom, trying to reach a verdict that will finally make the uncertainty stop? Cognitive defusion can create enough psychological distance for the client to examine the image more carefully. The thought “They don’t care about me” could become “My mind is telling me the story that they don’t care about me.” That shift can create just enough distance for someone to pause and examine the thought without first having to decide whether it’s true or false.

Other frameworks can help clarify what gives the image its emotional force. Schema work may reveal themes such as abandonment, mistrust, emotional deprivation, defectiveness, or unrelenting standards. Attachment work can help locate the threat: Will I be left? Will I be engulfed? Will I be punished for needing? Will I disappear if I adapt too much? Regulation and nervous-system work also matter, because no amount of cognitive interpretation helps if the body is still in threat. Sometimes the therapeutic work is simply to slow the movement from observation to interpretation. “They were quiet” can quickly become “They were angry,” and a delayed response can begin to mean “They don’t care.” Even disagreement may be interpreted as evidence that the situation is unsafe. The goal is to notice when perception has hardened into certainty before the situation has been fully understood.

For neurodivergent clients, pattern mapping needs to be especially careful, respectful, and neurodiversity-informed. The therapist’s task is not to argue the client out of their perception. It is to help the client slow the process enough to distinguish signal from alarm, pattern from total conclusion, and memory from present contact. Many clients need help recovering trust in their perception, while also learning how to test it with care.

It helps to distinguish curiosity from self-abandonment. A curious response leaves room for the question, “What else might be happening here?” without dismissing what we noticed. It considers what supports an interpretation and what complicates it. Self-abandonment begins when that openness turns into automatic self-dismissal: “I’m probably too sensitive.” Making room for genuine complexity is different from erasing ourselves in order to preserve the relationship, avoid conflict, or appear reasonable.

This approach isn’t meant to promote endless uncertainty. Some relationships really are unsafe, dishonest, or otherwise emotionally costly. Sometimes the image that needs updating is not “they are bad,” but “if I understand them enough, this can still become safe.” Discernment remains essential, and boundaries remain essential. Complexity should not be allowed to become another way to stay trapped.

When the relationship isn’t clearly unsafe, when the evidence is mixed, or when the emotional reaction seems larger than the event alone can explain, it may help to ask what mental image has been activated. This helps identify the hidden injury: not only what happened, but what happened to the model. The model may need revision, the nervous system may need reassurance, and the client may need to grieve the loss of the image they had been holding. Sometimes what collapses is not the relationship itself, but the fantasy of what the relationship might have become.

Knowing Meaningfully, Not Totally

We can know another person meaningfully without knowing them totally. Human beings wouldn’t survive without real knowledge of one another, and relationships would be impossible if every moment had to begin again from nothing. We know histories, injuries, promises, repeated failures, and patterns. Some of that knowing is loving. Some of it is protective. Some of it tells us when to stay, when to ask, when to wait, and when to leave.

Fixed knowing has already finished the other person. It explains before it listens, predicts before it checks, and protects itself from surprise. Living knowing does something different: it remembers without freezing. It doesn’t pretend the past didn’t happen, and it doesn’t require us to disregard patterns of harm or loosen necessary boundaries. It leaves enough room for the person to remain in motion, to grow and change beyond the image we already have of them. It can say, “This is what I know so far,” without needing to say, “This is all you are.”

That may be the more difficult kind of knowing. It is less satisfying than certainty and less intoxicating than fantasy. It asks us to discern without possessing, remember without freezing, and love without turning the other person into the solution to our own ache.

Sometimes what hurts in a relationship is not only what the other person did, but the collapse of the certainty we had built around who we thought they were.

The work is not to give up on knowing. It is to know without forgetting that the person is always larger than the image.