In the post on complexity theory, I suggested that (viewed on the informational level, as networks of behavior) human relationships function as what are technically known as “complex adaptive systems.” One of the most important features of such systems is the fact that while they are often intricately “organized,” they are also difficult or impossible to either predict or control (at least through our habitual top-down mechanistic strategies). One of complexity theory’s fascinating ironies is that for all its analytical abstraction, its practical implications—(what it has to say about actually dealing with such systems) are centered around the vaguely-mystical concept of “intuition.”
To unpack that a bit, In complex systems, the “solutions” that tend to strike us as obvious and rational—direct intervention to restructure the system to make it behave better—tend to backfire, come with unintended consequences, or just not work as intended. The use of punishment in an attempt to disincentivize unwanted behavior is a good example: while it may temporarily decrease unwanted behavior from your partner, it is also likely to damage the relationship over time, and make the unwanted behavior harder to control in the future. On the other hand, while complex systems cannot generally be traditionally “controlled” (at least without “killing” them) they can often be effectively “steered” through large numbers of small “nudges” in a particular direction, based on immediate, local information and pattern-recognition informed by prior experience with the system—rather than on rational analysis of the entire system at once.
I previously provided I provided an example from this paper of teachers and West Point graduates attempting to manage the complex system of kindergarteners at recess, where it was observed that: (rather than foolishly attempting to “impose order” on the the children’s play as the officers did) “Experienced teachers allow a degree of freedom at the start of the session, then intervene to stabilize desirable patterns and destabilize undesirable ones; and, when they are very clever, they seed the space so that the patterns they want are more likely to emerge.” So rather than attempting to pre-design an “ideal recess” and imposing that template on the kindergarteners, the teachers achieved better results by followed their intuition about which interactions between the children are “good,” and which “bad,” and what sorts of “nudges” are likely to be effective in a given situation.
What is Intuition?
Also variously called “nonverbal knowing,” a “felt sense,” or a “gut feeling,” (closely associated with Kahneman’s “System 1”) intuition is a sense of what’s going on that’s arrived at without conscious thought or rational analysis, which nonetheless serves as an information source to the conscious mind. Because of its subconscious mechanisms, we often casually discuss intuition as the opposite of “reason” or “logic,” as if it arises from vaguely mystical sources. But, intuition becomes directly comprehensible as the result of “reinforcement learning” rather than the higher-order process of “insight learning” or the acquisition of explicit verbal knowledge.
In this sense, the process of “learning” intuition is more like learning a physical skill than learning a fact. When trying to teach someone to play catch skillfully: we typically rely on iterated feedback and successive approximation rather than a top-down conceptual explanation: “keep your eye on the ball,” “hold your elbow closer to your body,” “follow through.” Over numerous throws, this kind of incremental feedback, responding to each new attempt to approximate a “good throw” gradually helps build a sense of predicting the motion of a flying object and perform the sequence of perceptive predictions and physical actions involved in catching and throwing it efficiently and accurately.
Considering interpersonal intuition, if you and I are interacting, you will likely have some sense of what’s going on with me, whether or not I’m to be trusted, my current emotional state, how I seem to feel about you, etc. You will not generally need to think through or consciously analyze the situation in order to arrive at these conclusions. Especially if you know me well, you will also likely have a generally-accurate sense of what I’m likely to do in the immediate future, with your predictions decreasing in both certainty and accuracy as you extend the time horizon. You would, for instance, just know that I was likely to retrieve a mug, and pour coffee into it, if you had just witnessed me pick up the coffee pot, but you would likely be less accurate in predictions about what I’m likely to do tomorrow.
To some degree, we experience our intuition as fused with our perception. The same way as the world we experience is a simulation or guess about the nature of the world around us, my perception of my partner as intentional, as pursuing goals, wanting and feeling things—which I sense as automatically as seeing their physical body, or hearing their voice—is nonetheless based on a complex “predictive model” trained over a lifetime of experience with people as well as my lived experience with my partner in particular.
Intuition is intimately intertwined with emotion—both arise automatically, without conscious choice, but are nonetheless shaped by experience. In other words, my intuition about what’s going on in a complex situation is also at least partially synonymous with how I feel about it, as exemplified by the ominous cliche from movie dialogue: “I’ve got a bad feeling about this.” So, as a system of prediction “designed” (by evolution) to be effective in confronting complex phenomena in particular, shapeable by the evolution-adjacent process of reinforcement learning, intuition represents a critical technology in the effective management of the inherent complexity of human relationships.
Intuition, Instinct, and Taste
The related psychological functions of taste (the unconsciously-derived sense of the “goodness” of a given stimulus, or more generally, “how you feel about it”) and instinct (defined nontechnically here as “the learned sense of what to do in a given kind of situation”) work similarly—they can be shaped by experience but not consciously chosen or controlled. In software terms, as compared to “rational thought” intuition might be said to work more like an “AI model” than a traditional “program,” (but that’s a topic for another time.)
Intuition, taste and instinct also tend to play a critical role in our functioning in intimate relationships in particular. Our intuition represents much of the meaningful experience of our partner, which guides much of our behavior in relationships. Taste, especially in kinds of people and kinds of interactions, likewise shapes the choice of who to pursue interaction with, and what kinds of interactions to pursue with them. Instinct, within which I’m including functions like what it occurs to me to say, or what I think to do in a given situation, then interweave with instinct and intuition to form this partially-innate, partially-learned, subconsciously-functioning “software-system” of perception and action functioning in interactive parallel with our conscious, rational, process.
Relational Intuition
In some sense, I’m just retroactively justifying the already-common knowledge that relationships (especially in the love/friendship category) are generally governed by intuition rather than rational analysis, if only to note that complexity theory might help us explain why, and demystify the underlying mechanics somewhat. Of course, the fact that we’re verbally talking about it right now implies that we would like to get our conscious rationality involved, especially where our intuition and instinct gets us into relational trouble.
One major way intuition can lead us astray in relationships is when one, or both partners are accurately predicting one another’s observable behavior while at the same time inaccurately assessing their private behavior—in other words knowing in advance what their partner is going to do or say, while being wrong about why or about what’s going on internally for that partner. For example, in couples where one partner exhibits an anxious attachment style and one an avoidant one, it is common for the avoidant partner to interpret their partner’s attempts at connection as “trying to start a fight,” or otherwise aggressively motivated, leading to defensive responses from the avoidant.
In such a situation, both partners may be familiar enough with the steps in the pattern that plays out between them to be able to quite accurately write out a “script” in advance if they were asked to, while fundamentally disagreeing about what is happening—what they are each doing according to their own self-perception, or about their individual definition of the situation. Part of the difficulty is that when the behavioral predictions are validated, they are likely to feel increasingly confident about the accuracy of both their behavioral predictions and their motivational interpretations, when their skewed perceptions of one another’s private behavior may be playing a significant role in guaranteeing that their negative predictions of the other are ultimately confirmed. These predictions typically become more automatic and “obvious” the longer you know—or especially live with—someone. So in a sense, our intuition about what’s going on becomes ever more fused with reality, at the same time that our partner-model becomes increasingly resistant to being updated with new evidence.
More generally, in a range of situations, a partner’s explicit verbal communication might not be the most reliable source of feedback about what is actually going on with them. Especially when a relationship is struggling, there’s a tendency for authentic self-disclosure to decrease, which makes “strategic” sense the more that partners come to experience one another more as “competition” or “opposition” rather than a potential collaborator.
Engineering Intuition: Psychotherapy and Supervision
In keeping with the whole “engineering” theme of this site, many of us might be inclined (like the aforementioned West Point graduates) to attempt to simply impose reason on relationships that have ostensibly become “irrational,” (such as when their participants seem to be experiencing more damage than benefit from their interactions). Rather than choosing between reason and intuition, though, integrating these two approaches may hold the most promise. In other words, when we’ve arrived at the rational conclusion that we want to “improve” our relationships in some way, we might be best served to attempt to leverage our conscious, rational process to improve intuition, rather than attempting to force our intuition into alignment with our cognitive beliefs.
A clinical supervisor of mine described the decision to start therapy as what one does when, while wandering around lost in a snowstorm, one comes upon one’s own footprints. Whatever your level of faith or skepticism in psychotherapy as a thing, this metaphor usefully describes the tendency for intuition (despite its many strengths) to lead one in circles, or otherwise become problematically self-reinforcing. Though as a therapist myself I’m obviously biased in favor of its value, I think that at (at it’s best) psychotherapy functions to disrupt this circularity, by providing both experiential counterevidence to maladaptive intuitions about self and of others, and also as a context in which this evidence can actually be “taken in.” In other words, a primary aim of what therapists say is typically aimed at challenging intuition, or specifically those intuitions implicated in a given client’s functional difficulties. If my experiences with other people have led to an intuitive sense of “not belonging” in most social contexts, my therapist might usefully function as a context in which the signals of belonging disrupt the otherwise self-validating effects of my behavioral defenses against the experience of not-belonging.
At least in theory, therapists also rely on learned intuition as a primary tool for affecting this change, with the process through which that intuition is “trained” providing a potentially useful model for “reshaping” intuition generally. While graduate training in clinical psychology typically involves quite a bit of traditional academic learning—research and theory on the mechanisms of human psychology, health, and pathology—the bulk of the training in the craft of therapy occurs in “supervision” and “consultation” which could be understood as directly focusing on building students’ therapeutic intuition, rather than “teaching knowledge” as we typically think of it.
In supervision, one or more students meet regularly with a more experienced clinician to discuss the student’s work with therapy clients, and review videos of the student conducting therapy. Though these discussions can take on a variety of forms, one central component is the supervisor simply offering what they would have said to the students’ client at each given moment in the session, providing a point of comparison to what the student actually did say to the client. Implicit in what my supervisors suggested saying to my clients was each supervisor’s understanding of each client’s psychology—what the supervisor was hearing (intuitively) from the client, and their “instinct” about what would be most helpful to say in a given moment, in service of the client’s individual goals.
Though my competence with smalltalk might lead you to believe otherwise, I inevitably arrived at the beginning of my graduate training with an existing intuition about how to understand people and talk to them, which inevitably formed the foundation for my interactions with clients. In addition to that pre-existing intuition, taste, and instinct, though, over the course of working with each supervisor, I developed an increasingly sophisticated sense of what they would think and say, which would then just “come to mind” —mostly effortlessly—while working independently with clients.
This process is a more intentional and explicit version of what happens between people automatically and effortlessly over the course of “getting to know each other.” A relational partner inevitably provides a similar feedback with their response (or absence of response) to any action you might take, or thing you might say. If you and I just met, we each would be entering our new relationship with a pre-existing intuition and instinct about people generally, which we would each gradually refine based on experience with one another, gradually developing a more refined, finer-grain sense of one another as particular people. Which then leaves you with the question of how to interpret my responses, and how to use those interpretations.
For therapists, this question of how to interpret and utilize client feedback is a central concern, complicated by the fact that a decent proportion of clients are ultimately struggling with their inability to give honest feedback, such that ostensibly positive verbal feedback (e.g. “I’m getting a lot out of therapy”) might be a further expression of their difficulty as readily as it could signal genuine improvement. In other words, if a therapist were to employ some sort of “satisfaction survey” or “symptom checklist” at the end of each session, it would be hard to know whether they were in fact measuring their clients satisfaction, depression, or simply how much they want their therapist to like them. Similarly, when your partner tells you they are “totally happy with the relationship,” it can be hard to tell if it’s because you are genuinely meeting all their needs as a partner, or they just don’t currently want to fight with you.
The ambiguous task of choosing which stimuli to pay attention to and which to ignore—or else how to weight different potential feedback sources— underlies the development of effective interpersonal intuition, especially one’s intuition about a particular person. As a therapist, I have come to put more weight on “functional” changes—when a client is suddenly saying or doing things that they wouldn’t have previously—than on explicit expressions of satisfaction or gratitude for my clinical performance. A related therapeutic strategy emphasizes attending to implicit or “metaphorical” content rather than explicit praise or criticism, assuming that the apparently unintentional connections a mind makes is a more accurate indicator of that mind’s subconscious functioning than conscious, intentional communication. To some degree this all circles back to the unsatisfyingly circular conclusion that one must rely on intuition as the best source for feedback about intuition—my gut sense of whether my partner’s feedback is to be believed.
In relationships, this often means paying attention to the convergence or divergence between words and actions. When your partner says they want more connection but consistently create distance, or claim to be fine while displaying clear signs of distress, your intuition about what’s actually happening may be more reliable than taking verbal communication at face value. At the same time, we must be cautious not to use this as justification for dismissing our partner’s explicit communication—the challenge is finding the right balance between listening to what they say and attending to your sense about the story behind their actions.
Taste, Attraction, and Creativity
In the discussion of functional behavior, I talked about the idea of different sorts of outcomes being differentially “reinforcing” to the behaviors which resulted in them. Humans, like all organisms capable of learning, are “hardwired” to be reinforced by a few basic survival-relevant things such as food, sex, or the relief from pain (called “primary reinforcers” in Behaviorist terms). But, we can also learn to be reinforced by other outcomes, (secondary reinforcers) theoretically through association with primary reinforcers. Money is a good example of this, which is generally reinforcing because it allows us to access primary reinforcers. In predictive modeling terms, reinforcers are understood as “prior preferences,” or “preferred states” reflecting outcomes that you want to be able to predict, motivating active behavior to realize those “desirable” predictions.
Practically, all this theoretical abstraction is usefully distilled into the idea of taste, which (at least for humans) is unique to each individual person, reflecting something like that person’s individual “reinforcement profile.” The metaphor of the sense associated with the tongue is apt, in that we differentially enjoy different shaped molecules and ratios of molecules, and the experiences of the associated neural impulses they excite from our tastebuds. Likewise, we often have strong preferences for the shapes of the bodies of other people in our visual field, and the shapes of their minds, and behavior, and the interactions that emerge between us. The concept of an acquired taste is also highly relevant to relational psychology, in that many of us hope to learn to like different shaped relationships or partners than the ones we’ve been in in the past.
Though taste in friends is potentially just as important relationally, the role of taste in sexual attraction and sexual/romantic relationships is particularly obvious and central. They are also a useful demonstration of the unconsciousness of intuition: while we might have guesses about why we fall in love with one person and not another, or why some situations or activities are more sexually arousing than others, the mechanisms behind these preferences are not available to direct observation, or subject to conscious control.
Speaking of sex, Rick Rubin, the legendary music producer turned creativity guru, writes extensively about the role of taste in the creative process, and about the process of developing and refining taste over the course of a creative career. He describes the phenomenon of a “tingle” experienced in recognition of something like “artistic truth,” which he proposes can be “acquired” or “developed” through exposure art, culture, and the natural world, suggesting that we “follow the tingle” in the process of our own creative expression. He also describes an iiterative, evolutionary process as one compares alternative directions for that creative expression: trying out various options (mutation) and choosing the one that provides the most tingle (selection). We might try something similar in our relationships: trying out various ways of being together, and doing more of the ones that feel good. This might also be a decent argument for the value of sexual experimentation, (but I’ll leave that up to you).
Returning to the general experience of differential attraction to other people, there’s a fair amount of consensus in psychological circles that our taste in partners and friends generally has something to do with the relationship between their personality and ours—how the shapes of our minds fit together, or what their various characteristics “mean” to us given our accumulated life experience. Here I’m talking just about the contributors to the emotional, phenomenological experience of attraction to another person, not the conscious, cognitive evaluation of how many boxes they check on our “dating checklist.”
There’s a school of thought that a big part of the sexual experience people are generally after is the sexual experience of self, of being allowed to experience oneself as sexy and desired or just as a valid sexual being. This is complicated by the variously repressive taboos around sexual expression in many cultures, and from there, it may be true that the phenomena of kinks, fetishes, and individual turn-ons sometimes function to provide psychological “permission” to “be oneself” sexually—effectively becoming a “legal defense” against subconscious internal “prohibitions” or taboos around sexuality generally. Relatedly, many people in some stage of psychosexual self-discovery find repeated themes (or more generally shapes) in their “erotic/arousal maps” and their broader psychological organization—such that sexual fantasies can sometimes serve as a guidebook to creative problem-solving elsewhere in a person’s life.
Relatedly, we may be drawn to people whose minds or identities “solve a problem” psychologically for us, or else mean something desirable about us, or for us. If I feel alienated from my own emotional experience, for example, I might tend to be drawn to partners whose uninhibited emotional expression “gives me access” to a sense of affective vitality, or alternatively to people whose similar (rather than contrasting) personalities implicitly confirm I’m seeing things the right way.
To boil that down a bit: (metaphorically, at least) taste is kind of just pattern recognition—our preferences for different shaped things. We might want to try to develop or refine that taste—i.e. learn to enjoy different shapes, which may be accomplished by trying new things; or more cognitively, by telling ourselves different stories about what different things mean. I enjoyed the taste of crickets a lot more once I learned they were pretty much the same sort of organism as shrimp and lobsters… just saying.
To (over)simplify a handful of these threads, the popular discussion of “love languages” conveniently brings together intuition, taste, and instinct in the idea that we all have familiar ways of both expressing and receiving love which may or may not align with our partner’s. If my taste aligns with “words of affirmation”—verbally saying or hearing loving things—while my partner is more aligned with physical touch, the intuitive “meaning” I assign to my partner’s absence of verbalized love, and to their frequent physical overtures is likely to feel—to me—like something besides love.
Learning to Like It
One of the most powerful applications of these ideas is the possibility of deliberately reshaping our taste—learning to be reinforced by different things than we currently are. This isn’t about forcing yourself to like what you don’t like, but rather about creating conditions where new preferences might naturally develop.
There are obvious benefits to expanding what we find reinforcing. On an individual level, one of the single most powerful determinants of success in most areas of life might be one’s ability to enjoy the process of getting better at something rather than only being good at it. If you can learn to enjoy the process of working through conflict rather than avoiding it, or find appreciation for the lively parts of them that are also inconvenient or uncomfortable for you, your relationship stands to gain a huge amount of flexibility and resilience, in addition to your individual reward of enjoying your relationship more.
This resonates with philosophical traditions like Stoicism and Buddhism, which emphasize finding satisfaction in process rather than specific outcomes. Creating conditions for new enjoyment often means deliberately exposing ourselves to potentially reinforcing experiences that we might otherwise avoid. This could mean trying new activities with your partner, having conversations outside your comfort zone, or experimenting with different approaches to conflict. With repeated exposure in a supportive context, previously neutral or even aversive experiences can become reinforcing.
Relational Creativity: Grow Don’t Build
Rubin’s advice to “follow the tingle” reminds me of ideas from software engineering that might prove surprisingly relevant to relationships. In “The Mythical Man-Month,” Fred Brooks famously argues that complex software systems are better “grown” than “built”—that attempts to design perfect systems from the top down inevitably fail, while incremental, evolutionary approaches succeed. The parallels to relationship development are striking.
When engineers approach complex systems with a blueprint mentality—attempting to specify every detail in advance and then implement according to plan—they typically find themselves fighting against the inherent complexity of the system. Similarly, approaching relationships with rigid expectations or trying to “design” the perfect partnership often leads to frustration and disappointment.
Instead, both software and relationships benefit from what might be called “evolutionary design”—starting with simple, functional interactions and gradually evolving them through iterative feedback cycles. You try something, see how it works, keep what functions well, modify what doesn’t, and repeat. This approach acknowledges the inherent complexity and unpredictability of both systems while providing a practical path forward, valuing “evolution over revolution,” recognizing that sustainable change in complex systems typically happens through many small adjustments rather than dramatic overhauls. The relationship that results may not resemble what either partner might have designed from scratch, but it will likely be better adapted to the unique characteristics of both participants.
The software engineering principle of “release early, release often” finds its relationship parallel in the willingness to engage authentically with your partner in small, consistent interactions rather than waiting for the “perfect moment” for a major relationship conversation. Each small interaction provides feedback that shapes the next interaction, gradually evolving the relationship system over time.
So what does all this mean for our practical approach to relationships? If we accept that relationships are complex systems best approached through intuition guided by experience, what might that look like in practice?
A lot of it might just be about adopting an experimental mindset—being willing to try things, notice what works, doing more of that, and less of what doesnt. A relationship can be a collaborative creative process of finding patterns that work for both of you, rather than one person imposing their vision on the other. By integrating rational understanding with intuitive experience, we may be able to leverage our human intellect to engineer conditions where the real “magic” of intuitive evolution can take place.
Sources
Michael Bader (2003) Arousal: the secret logic of sexual fantasies
Fredrick Brooks Jr. (1995) The Mythical Man-Month: essays on software engineering
Lisa Feldman-Barrett (2017) How Emotions are Made: the secret life of the brain
Eugene Gendlin (1982) Focusing
Jay Hayley (1963) Strategies of Psychotherapy
Jack Morin (1996)The Erotic Mind: unlocking the inner sources of passion and fulfillment
Emily Nagoski (2021) Come As You Are: the surprising new science that will transform your sex life
Rick Rubin (2023) The Creative Act: a way of being
Timothy Wilson (2004) Strangers to Ourselves: discovering the adaptive unconscious
