learning is being alive

God why does learning feel so good? Because it's the most instinctively natural thing for your brain to do. Throughout evolution we've developed the instinct to look for uncertainties in our surroundings so that we can reduce them and increase our likelihood of survival. A living creature learning new information about its environment makes it feel more safe. But what sets us apart from other living creatures is that we're self-aware, which also means we know that there are things we don't even know we don't know yet. Even after we've gathered every piece of information about our environment our nervous system will still not be satisfied. Our uncertainty is never fully resolved and we can keep asking more and more questions. We have to keep learning in order to continuously satisfy that same level of feeling alive and whole, for our body to rest easy in the knowledge that it's continuing to do what it was evolved to do. Magic brain!

manifestation is a mixed bag

𝘏𝘶𝘮𝘢𝘯 𝘏𝘶𝘮𝘢𝘯 (𝘊𝘰𝘴𝘵𝘢𝘯𝘻𝘢) has a hidden dolphin in it. It’s a nod to Magic Eye pictures, which some of you will know are a deep lore core of my obsession with illusion (please look them up if you don’t know them). You just simply never forget a moment when you’re suddenly looking at a 3D animal simply by staring at a picture for long enough. Your brain is always actively building what you see based on memory, expectation and pattern- so what happens when you consciously change your perception? Can you control and manifest your whole reality? The answer is yes, obviously ;) but here’s where we need to be careful. Because the modern world loves controlling its environment just a little too much, and the risk of manifestation, I believe, is burnout.

Manifestation is packaged as freedom because it relies on the premise that you are in full control. And I don’t want to naysay that, it’s beautiful and I will always support it, it’s how I live my life. But with control comes responsibility, and with freedom comes responsibility too. Yes, it’s beautiful, you can create your own illusions, and by all means keep co-creating your reality. But the emphasis needs to remain on co-creating; there needs to be room for acknowledgement of those parts we can’t control. You can be strong, independent, self-reliant, but if you keep drinking from your own well you’re going to eventually run out of water. You’re going to be burnt up by your own freedom. You need to let people around you help you, you need to let your surroundings create themselves too. This painting is dedicated to my friend Costanza, one of the most passionate and conscious-perception-shifting people I’ve ever known.

optics and linguistics

Seeing is like talking — Short

In linguistics, Relevance Theory proposes that communication is an inferential process that relies on a shared cognitive environment, not just a linear transfer of information. Both people bring their own context to the conversation, and if those contexts doesn't overlap anywhere (read: if there is no shared cognitive environment) then there is just noise, no real communication. If I blab words or gesture signs that you have no prior knowledge of, we won't get very far, and even if I talk in clear words about something you have no frame of reference for, I'm still not exactly communicating anything useful to you.

The same structure appears in perception: key words + shared cognitive environment ☞ communication! sensory stimulation + relevant predictions ☞ vision!

Light hitting your retina is at first just sensory stimulation, not exactly yet "vision". For stimulation to turn into vision your brain's needs to be able to check this against its prior models, its predictions, its readiness to make sense of what it's receiving.

To use the example given by DeLanda in one of my fav books Materialist Phenomenology; when you're in a room filled with fog, there can be plenty of light stimulating your eyes but it will still tell you nothing about the contents of the room. Let the fog clear, and the same light becomes very informative. The difference isn't in the light, it's the same light, but what changed is the relationship between the light and the room's content. The fog is no longer scattering this light uniformly, so now it's carrying structural information about the room, giving your brain something to work with.

The world is not a static backdrop that you extract information from my friend, it is doing something, contributing something, without which your experience of it simply wouldn't exist. Without the world gifting you its abundantly rich sources of information, you would not be able to perceive or experience all the things you see as beautiful or meaningful. Just like a person you're having a conversation with, the world deserves just as much acknowledgement of what it's giving you as all other things that are alive.

50%

Seeing is like talking — Medium

In linguistics, Relevance Theory (Sperber and Wilson, 1986) proposes that communication is not simply a linear transfer of information. Rather, communication is an inferential process that depends on a shared cognitive environment. Meaning emerges from the overlap between what the speaker intends and what the listener brings. Both participants contribute context, expectations and prior knowledge, and without that shared ground, no amount of correct grammar produces understanding.

This turns out to be a surprisingly useful lens for thinking about perception.

Vision has traditionally been modelled as a feedforward process: light enters the eye, travels through processing stages, and produces a perceptual experience. Clean, linear, one-directional. But this model struggles to account for how rich and context-dependent our perceptual experience actually is. A more accurate picture is that perception is an inferential process — your brain holds an active model of what is probably out there, compares it against incoming sensory data, and continuously updates. What you see is not the world as it is; it is your best current hypothesis about the world, shaped by everything you have experienced before.

What Relevance Theory adds to this picture is the relational dimension. In communication, meaning requires a shared cognitive environment — neither speaker nor listener alone generates it. In perception, the same logic holds. Consider DeLanda's fog example: a room full of thick fog is stimulating your visual receptors, but you cannot see anything. Let the fog clear, and the same light is suddenly informing you about the room's contents. Nothing changed in the light itself. What changed is the relationship between the light and your perceptual apparatus — the conditions under which meaning can pass between world and observer.

Perception is not extraction. It is exchange. The environment is not a passive container of information waiting to be retrieved — it is an active participant in the construction of your experience. You bring your prior models and expectations; the world brings structure and constraint; and what you perceive emerges from the negotiation between them. Which has an implication worth sitting with: if your experience of the world is always co-constructed, then understanding yourself requires understanding what you are in dialogue with.

---

Seeing is like talking — Long

Two models of communication

Relevance Theory, developed by Sperber and Wilson in 1986, proposes that human communication operates on two levels simultaneously. The first is a shared semantic code: words carry meanings that speakers encode and listeners decode. The second, and more interesting, is inference: the listener does not simply decode what is said but actively constructs the most relevant interpretation based on context, shared knowledge, and prior expectation. Communication, on this account, is not a transfer of information between two passive endpoints — it is a collaborative act that depends on a shared cognitive environment, a set of facts and assumptions mutually available to both parties. Strip out either participant, or strip out the shared ground between them, and meaning does not degrade; it disappears entirely.

This is a familiar enough account of language. What is less obvious is that the same structure appears in perception.

The inferential structure of vision

Vision has traditionally been modelled as a feedforward process: light enters the eye, is processed through successive stages of the visual system, and eventually produces a perceptual experience. Each stage builds on the previous one; the flow is linear and the direction is one-way. This model has the appeal of simplicity, and it captures something real about the mechanics of early visual processing. But it fails to account for how context-dependent, expectation-laden and actively constructed perceptual experience actually is.

A more accurate account is that vision operates inferentially. Before light even reaches the retina, attention-allocation is already shaping which parts of the visual field will be processed and how — "all perceptual experience is necessarily the end product of a categorisation process." After that initial phase, the visual system does not passively receive information so much as test hypotheses against it: your brain holds an active generative model of what is probably out there, compares its predictions against incoming sensory data, and updates where prediction and input diverge. What you perceive is not the world as it is, but your brain's best current inference about the world — continuously revised, shaped at every stage by prior knowledge and expectation.

Cognitive penetrability — the degree to which higher-level beliefs and expectations influence lower-level perceptual processes — is genuinely contested in the literature. The Müller-Lyer illusion is the classic counter-case: knowing that the two lines are identical in length does not stop you seeing them as different. Something in early visual processing resists top-down correction. But this is precisely what makes the inferential account interesting rather than trivial: perception is not simply belief in disguise. It is a distinct process with its own logic, one that interacts with cognition in complex and sometimes asymmetric ways. The inferential structure is real; it is just not unlimited.

What Relevance Theory adds: the relational dimension

Where predictive processing accounts of perception tend to focus on what happens inside the individual brain, Relevance Theory draws attention to something additional: the relational structure of meaning-making. In communication, meaning is not located in the speaker, or in the listener, or even in the words — it emerges from the shared cognitive environment that both parties inhabit together. Neither side alone generates it.

The same logic applies to perception, and DeLanda's fog example makes this vivid. Imagine a room filled with thick, evenly distributed fog. The light in the room is physically interacting with your visual receptors — it is, in the technical sense, stimulating them. But you cannot see anything. Now let the fog dissipate. The light has not changed. The room has not changed. And yet something fundamental has shifted: the light is no longer merely stimulating you; it is now informing you about the room's contents. The difference is not in the light itself. It is in the relationship between the light and your perceptual apparatus — the conditions under which information can pass between world and observer.

This is the structural parallel with Relevance Theory. Just as meaning in communication requires a shared cognitive environment — a mutual ground in which speaker's intention and listener's inference can meet — meaning in perception requires a relational context in which the world's structure and the observer's models can meet. Merleau-Ponty observed this parallel decades before Relevance Theory was formalised: "we perceive in conformity with the light, as we think in conformity with other people in verbal communication." The environment is not a passive container of pre-existing information waiting to be extracted. It is an active participant in the construction of perceptual experience.

The observer is always half the picture

This has consequences that extend beyond the philosophy of perception. If what you perceive is always the product of a negotiation between your internal models and the structure of the world — if the observer is constitutively part of the process — then there is no view from nowhere. Your perceptual experience is not a neutral readout of reality; it is a portrait of the relationship between you and your environment, shaped by your history, your expectations, your current state, and what the world is offering back.

The self that emerges from this picture is not a fixed inner entity that perceives an outer world. It is something more like a zone of ongoing exchange — always partially constituted by what it is in contact with. Hohwy's account of the "PP-self" makes this explicit: the sense of self is itself an inference, a subset of the inferred causes of sensory input that happen to be traced back to your own body and actions. You are, in this sense, something your brain constructs in the same inferential process by which it constructs the chair you are sitting on.

None of this dissolves the self or makes experience arbitrary. It does, however, suggest that self-knowledge cannot be purely introspective. If your experience of the world is always co-constructed with it, then understanding yourself requires paying attention to what you are in dialogue with — what environments, expectations, and prior models you are bringing to the exchange, and what the world is doing in response. Perception is not something that happens to you. It is something you do, together with everything around you.

---

References

Original essay

Can Relevance Theory (RT) inform our understanding of perception?

Relevance Theory (RT) is a theory of communication and cognition developed by Sperber and Wilson in 1986. It proposes that human beings process information by maximising improvement of their mental representation of the environment at the smallest processing cost. The theory posits that – rather than communication being simply a transfer of coded information – upon hearing an utterance, the listener in a conversation selects the interpretation with the most relevance based on highest contextual effect and minimal processing effort (Sperber and Wilson, 1987). The theory emphasises the importance of relevance in communication, where relevance is determined by the contextual effects and the processing effort required. RT suggests that the hearer selects the interpretation with the highest relevance based on the contextual effects and processing effort. (ibid.)

This essay will consider how RT as a framework for understanding various pragmatic phenomena, language processing and communication, can be applied to an entirely different form of contextual processing; namely, visual perception.

"What we call the principle of relevance is the thesis that every act of ostensive communication communicates the presumption of its own optimal relevance. We argue that the principle of relevance explains how the production of an ostensive stimulus can make the communicator's informative intention mutually manifest, thus leading to the fulfilment of the communicative intention." (Sperber and Wilson, 1987, p.704)​

Effective communication relies on a common ground of shared knowledge, assumptions and expectations; in order for the intended thought to be communicated with maximum relevance, the speakers need to be engaged in a mutually cognitive environment. (Sperber and Wilson, 1987). A cognitive environment is a set of facts that is manifest to an individual; the person’s ‘available mental information’. This means that, for cognitive impact to occur (ie. for the intended information to be understood by the listener), there is a prerequisite for both agents to share access to a certain context and facts. A cognitive environment can be compared “at a conceptual level, to notions of visual or acoustic environment at a perceptual level” (ibid, p.699). Where verbal communication relies on a cognitive space in which information is maximally relevant, visual perception relies on visual information processing to happen in conformity with light. A parallel of this kind was observed decennia before the introduction of RT by Maurice Merleau-Ponty;

“We perceive in conformity with the light, as we think in conformity with other people in verbal communication. And just as communication pre-supposes […] a certain linguistic setting through which a meaning resides in the words, so perception presupposes in us an apparatus capable of responding to the promptings of light in accordance with their sense (that is, in accordance both with their direction and their significance, which amount to one thing), of concentrating diffuse visibility and completing what is merely foreshadowed in the spectacle.” (Merleau-Ponty, 1945)

In verbal interactions, then, there are two ways in which communication occurs. The first consists of a shared code to be offered by the speaker and decoded by the listener – a straightforward interpretation of how communication occurs – and a second way which uses inference; ostensive-inferential communication. In the latter, the communicative process relies on a shared cognitive environment in which the communicator makes manifest his intention to make something manifest (“ostension”), providing the listener with evidence from which his informative intention can be inferred. (Sperber and Wilson, 1987, p.700).

Ostensive-inferential communication can be a particularly useful framework to apply to visual perception because of its emphasis on the role of the listener and the inferential process. We can summarise the process of verbal communication as described above in the following format:

mutual cognitive environment ☞ relevant ostentive-inferential communication ☞ manifestation of thought

If we were to make a similar schema adapted to perception, it would look something like the below:

pre-perceptual attention-allocation ☞ early-vision ☞ post-perceptual evaluation & selection ☞ inference ☞ manifestation of thought (adapted from Pylyshyn, 2006).

The initial stage of visual information processing happens before the light stimulus reaches the retina. Where and how the individual will naturally direct their attention when exposed to a visual stimulus is already heavily influenced; “all perceptual experience is necessarily the end product of a categorization process” (Bruner, 1957, p. 124). After this initial phase, the rest of the process of perception continues to rely more and more on the crucial feature of inference; after attention-allocation has been determined, the actual visual information is received, which then gets evaluated and filtered through recognising patterns (accessing pre-existing background information present in the individual’s cognitive environment), eventually leading to an inferred conclusion. (Pylyshyn, 1999, p. 343-344).

Perception is therefore reliant on a prior basis of assumptions and expectations that will influence the way the information is processed at every stage of the process; the individual’s cognitive environment will inform their inferential process of visual sensemaking.

The role cognitive penetrability, predictive processing and the Self

Continuing the thread of perception being a collaborative act between observer and environment, I want to introduce the lens of cognitive penetrability (CP) and predictive processing (PP).

CP refers to the idea that higher-level cognitive processes, such as beliefs, expectations, or knowledge, can influence or "penetrate" lower-level perceptual processes. In other words, when something is cognitively penetrable, the information from our beliefs and expectations affect how we perceive and interpret input. As discussed, what sets RT apart from other theories of communication is that it strongly emphasises the role of the inferential process; ie. it assumes the mind is being cognitively penetrated. Whether or not visual perception is similarly cognitively penetrable is a hotly debated topic and outside the scope of this essay, but certain case-studies point out key phenomena that may indicate crucial differences in the two information processing pathways.

It is clear that vision is cognitively penetrable to a certain extent; as already briefly touched on before, prior knowledge and expectations influence perception in the early stage of attention-allocation, post-perceptual evaluation and selection. Conversely, the effect of hints, clues or other knowledge-suggestions on the ability to distinguish certain ambiguities or to achieve a stable percept (Pylyshyn, 2006; Felin et al., 2017) makes a case for the cognitive impenetrability of vision, setting it apart from verbal communication. Take for instance the case of the famous optical illusion of the Müller-Lyer illusion; even when given the knowledge that the lines are the same length, the viewer will still perceive them as being different lengths. Consequently, applying this lens to both fields opens a line of promising questions in their unique potentials for cognition and understanding.

PP is a theory within philosophy of mind and cognition that, since its conception, has reached a wider field of machine learning and computational neurosciences. PP relies on the minimisation of errors in a predictive process. It has an application to fields like predictive coding and other computational contexts, but for the purposes of this essay I will consider PP purely as a philosophical framework. PP proposes that a system, such as the brain, holds an intrinsic model of the origins of its sensory input. This model enables the system to anticipate the sensory input it is likely to encounter. These predictions are then compared to the actual sensory input, and any disparities are used to update or refine the model. In other words, the theory suggests that perception and cognition involve a continuous process of generating and updating predictions based on sensory input. (Hohwy, 2020, p.2, 3).

Visual perception has traditionally been perceived as a feedforward model, roughly analogous to the “code” element of communication (I have touched previously on coded communication vs inferential communication; RT suggests that verbal communication relies both on a shared semantic code and the role of inference as described above). The core idea of feedforward model is that sensory input is processed in a sequence, with each stage building on the previous one. Feedforward models posit a linear structure separate from high-level cognition; the idea is that information flows from initial reception of sensory stimulus, through various stages and a final perceptual experience. In recent years, however, a radical new framework has begun to take form that considers perception as an entirely inferential process; a collaborative experience in which predictions about the outside world shape the way we process information (Walsh, 2020)​. From this perspective, perception is the process of identifying the perceptual hypothesis that best predicts sensory input and hence minimises prediction error (ibid.). It would seem that in the case of vision, there is a dynamic interplay between feedforward and feedback pathways, with higher-level cognition influencing and modulating the processing of sensory input.

The “PP-self” refers to a new conception of the sense of self in the context of PP. It is seen as a subset of the inferred causes of sensory input as a consequence of our own actions, our selective attention-allocation and free learning in a dynamic environment. The PP-self arises from inferences of the hidden causes of sensory input, even when those causes are intimately connected with ourselves through our bodies (Hohwy, 2o2o). One can see how this operates in the same framework as the inferential process in verbal communication as outlined in RT; “_inference is the process by which an assumption is accepted as true or probably true on the strength of__​_ the truth or probable truth of other assumptions." (Sperber and Wilson, 1987, p.701)​

Applying RT to the problem of perception introduces a promising shift in how we have understood perception to date, leading to PP and other fields that challenge our understanding of sensemaking and cognition. Where RT posits that verbal communication happens by means of a cognitive agent selecting an utterance for maximum relevance, PP suggests that perceptual models are selected for to minimise prediction errors. In RT, an ostensive-inferential communication is passed on in a mutual cognitive environment; in perception the PP-self and active inference are used as facilitators of sensory input. In both cases, the emphasis ends up being on the individual collaborating with their environment.

To finish, consider a visual case that is analogous to the dynamic ways in which information can become manifest during verbal communication;

“But even if we stick to the case of a motionless observer, a section through the optic array is a much richer object than the unstructured, highly impoverished raw sense data that dominated empiricist thinking for centuries. The concept of sense data led to the idea that the external world contributed nothing to perception but stimulation, an idea which was expressed in some extreme cases by saying that the input to the process of perception was merely an irritation of the sensory surfaces. But a simple example can dispel this myth. Let’s imagine an evenly lit room completely filled with a thick fog. The diffuse light in the room is clearly acting as a stimulus, since it is energetically interacting with receptors in our eyes, arousing them and changing their state. On the other hand, the light in the room is not causing us to be aware of its contents since we are incapable of seeing any object. If we allow the fog to dissipate, the light will continue to be a stimulus but it is now, in addition, something else: it is not just stimulating us but also informing us about the contents of the room.” (DeLanda, 2022, p.21)​

This example beautifully illustrates the use of applying relevance theory, with its key focus on inference, to the problem of perception. It allows us to radically shift our perception of perception, away from being simply a visual code that enters our retina, and towards a process in which vision happens in collaboration with the light and the environment, in much the same way that RT points out that communication relies on collaboration of two agents inside a shared mutual space that plays just as much a role as the individuals themselves.

(1970 words)

References

1. Sperber, D. & Wilson, D. (1987). Precis of Relevance: Communication and Cognition. Behavioral and Brain Sciences, 10, 697–754.
2. Bruner, J. S. (1957). On perceptual readiness. Psychological Review, 64(2), 123–152.
3. Hohwy, J. (2020). New directions in predictive processing. Mind and Language, 35(2), 209–223.
4. Raftopoulos, A. & Zeimbekis, J. (2015). The Cognitive Penetrability of Perception. Oxford University Press; Pylyshyn, Z. (2006). Seeing and Visualising: It's Not What You Think. MIT Press.
5. DeLanda, M. (2022). Materialist Phenomenology. Bloomsbury Publishing, p.21.
6. Merleau-Ponty, M. (1962). Phenomenology of Perception. Routledge.
7. Hohwy, J. (2020). New directions in predictive processing. Mind and Language, 35(2), 209–223.

you are a complex system

You are a complex system — Short

A complex system is an entity not defined by its parts but by the interactions between those parts. Complex systems live at the threshold between order and chaos, constantly self-organising through feedback and feedforward loops: larger patterns shape the behaviour of smaller parts, which in turn generate new larger patterns. I highly encourage you to look more into complex systems. They're really cool and they can give you some real brain pain.

The weather or the brain are classic examples of complex systems, but perception, action and self-awareness are too: what looks like "you" is always the emergent result of ongoing interactions within your body, between your body and the world, between your past experience and the present moment.

If the self is a complex system, then your mental wellbeing can't be a static state to reach or a problem to finally solve. Uncertainty and change are what drive every living adapting system from the inside, they're not signs that something is wrong with you. The most useful thing you can do for yourself is learn to move with the complexity rather than against it.

You are a complex system — Medium

A complex system is an entity not defined by its parts but by the interactions between those parts. Complex systems live at the threshold between order and chaos, constantly self-organising through feedback and feedforward loops: at every scale, components interact and give rise to properties that couldn't have been predicted from any single element in isolation. Those emergent properties then in turn loop back to shape the components that produced them.

This is how the visual system works: instead of passively recording the world it generates a constant cascade of predictions and corrections, to construct a best guess at reality from the inside out. What you see is your brain's most probable interpretation of it that gets updated in real time, never a direct one-to-one mapping of the world.

• perception is an emergent phenomenon; a vivid experience rising from billions of tiny adjustments, none of which individually contain the experience itself. What you see arises from a complex dynamical system, in a way that cannot be predicted from any of its parts individually.

• action in an emergent phenomenon; any decision you make is actually a convergence of intentions, context, history, emotion, your environment, your physical state. Your behaviour arises from a complex dynamical system, in a way that cannot be predicted from any of its parts individually.

The point of thinking in complex systems is to realise that being uncertain, changing and contradictory doesn't mean something is wrong with you - you don't need to fight that or try to be more stable. Uncertainty is the very feature of living systems doing what living systems do.

Your sense of self (an emergent phenomenon in itself) is an ongoing process of growing a coherent story about who you are in response to what you perceive and how you respond from moment to moment. When that story is too rigid (for whatever reason, whether that's in your control or not) you get stuck; when it dissolves entirely, you lose your footing. The more you can strive to strike a balance, to be flexible and intentional with how you update our self-story in response to a new experience, the more you can create greater alignment between your inner and outer worlds. That sense of coherence will increase your felt sense of meaning and wellbeing.

You are a complex system — Long

Humans have an impulse for reduction: if your code fails, find the bug and fix it; if your car breaks, find the broken part and replace it. This is extraordinarily useful for but simple and complicated systems, but there is a huge class of phenomena that reductionism doesn't work with: these are complex systems.

A complex system is unpredictable, because its output will depend on how its components interact with each other rather than on the components themselves. Independent elements interact with each other and give rise to new elements that could not have been anticipated from looking at each element individually. The whole influences the components as much as the components influence the whole.

> “aggregate patterns form from individual behavior, and individual behavior in turn responds to these aggregate patterns: there is a recursive loop".

Perception is construction

If you've spent some time reading around my website, you'll know by now that the visual system is a beautiful case study of a complex system. You can skip this section if you get it already.

When you look at a map of the data that reaches the visual cortex from the eyes, it is immediately obvious that the brain is not receiving a photograph of the world. In fact, approximately 80% of the signals arriving at the visual cortex are not received from outside but generated by the brain itself. What we see is not a readout of external reality but a construction based on past experience; a constantly updated prediction of what is most probably out there. "These signals in the past indicated there was a chair, so I must be looking at a chair" - so you see a chair.

The same retinal image could therefore, in theory, have arisen from an enormous number of different sources. The visual system has to resolve this ambiguity by selecting the interpretation that best fits both the incoming data and its prior model of the world. It's a recursive, self-organising negotiation between multiple levels of the perceptual architecture simultaneously: lower-level feature detection feeding upward, higher-level predictions feeding back down, the whole system settling into a coherent interpretation moment by moment.

As mentioned in the introduction (in more complicated words): when simple relations exist between a large number of elements, the structure as a whole will have abstract features that recur regardless of the specific values of its components. In the visual system, this is exactly what happens: depth, motion, objects, and eventually meaning emerge from interactions between neurons that individually detect nothing more than edges and contrasts. Each level is qualitatively different from the one below it. This is emergence.

Perception is therefore adaptive and experience-dependent, not fixed. Your brain models the world through relationships between patterns and their probable source based on everything your have ever experienced, and continuously updates those models at the same time. Seeing is quite literally a creative act!

Action as emergence

Intentional human behaviour follows the same logic. A given action is not the direct output of a mental decision but the emergent result of an interplay between high-level intentions, bodily readiness, emotional context, environmental constraints, and real-time sensory feedback, each level dynamically influencing the others.

Once you think about your actions and choices in terms of context-sensitive dynamical adaptations, rather than linear cause-and-effect processes, you can really change the way you understand your own behaviour. What you think and do is shaped by a continuous interaction between top-down cognitive goals and bottom-up opportunities and resistances, and new behaviours emerge from this negotiation that could not have been predicted from either alone.

This isn't meant to diminish any of your power; you are still an agentic human being. It's actually empowering to realise your actions are emergent properties of a relational context instead of simple being outputs of pre-formed decisions, because that context is something you can actively participate in shaping.

The self as a process

The free energy principle, a theoretical framework developed by Karl Friston, brings perception, action, and self-awareness together. It proposes that all adaptive biological systems (read: all things that are alive) operate by minimising the difference between their internal model of the world and the sensory inputs they receive. To be alive is to be constantly engaged in prediction and error-correction: this applies to perception and behaviour, but also extends to identity, belief, and meaning.

Enactivism argues the same; the self here is a hierarchical generative model, spanning perceptual, cognitive, emotional, physical layers. The self is seen as a model that is constantly updating through its active engagement with the world. Through this lens, self-awareness is then an emergent product arising from this process. When our self-model becomes too rigid (too much order), we get stuck in patterns that no longer fit; it gets overwhelmed and we lose our sense of continuity. When it doesn't hold together at all (too much chaos), we have no ground to stand on. But we can be flexible and intentional with how we update our self-story in response to a new experience (the perfect edge between order and chaos) we experience what feels like coherence, meaning, and insight. (Naturally, we can't take for granted that everybody is privileged with the same necessary physical or environmental conditions to be able to do this, but this is speaking generally and theoretically).

Our sense of meaning arises from how well we are tracking what matters to us in a situation, in order to be selective with what to adopt (John Vervaeke's notion of "relevance realisation")- and this tracking is a dynamic skill that can be cultivated. Consider mindfulness through this lens: meditation is the development of meta-awareness, the capacity to observe and be deliberate the process of your own attention and meaning-making.

Living with complexity

So. The self you are trying to understand is not a fixed thing waiting to be discovered. Your self is an ongoing process that you're always already participating in, which means that uncertainty and contradiction are not signs that you're messed up; they're actually the natural condition of any living, adapting system. It's uncertain, but that's what makes it alive; you're laying down the path in walking.

Psychological flexibility -- the capacity to update your internal model deliberately and carefully -- is key to living a flourishing life. Wellbeing is a quality of movement that embraces the complexity of the self as being a process of becoming.

References

References

1. Arthur, W. B. (2013). Complexity economics: A different framework for economic thought. In Complexity Economics, p.5.
2. Elkins, J., & Fiorentini, E. (2020). Visual Worlds: Looking, Images, Visual Disciplines. Oxford University Press.
3. Hayek, F. A. (1967). Studies in Philosophy, Politics and Economics. University of Chicago Press.
4. Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and Brain Sciences, 36(3), 181–204; Pylyshyn, Z. (2006). Seeing and Visualising: It's Not What you Think. MIT Press.
5. Purves, D., Lotto, R. B., Williams, S. M., Nundy, S., & Yang, Z. (2001). Why we see things the way we do: Evidence for a wholly empirical strategy of vision. Philosophical Transactions of the Royal Society B, 356(1407), 285–297.
6. Juarrero, A. (2002). Dynamics in Action: Intentional Behavior as a Complex System. MIT Press, p.27.
7. Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138.
8. Thompson, E., & Varela, F. J. (2001). Radical embodiment: Neural dynamics and consciousness. Trends in Cognitive Sciences, 5(10), 418–425; DeLanda, M. (2022). Materialist Phenomenology. Bloomsbury Publishing.
9. Muth, C., & Carbon, C.-C. (2013). The Aesthetic Aha: On the pleasure of having insights into Gestalt. Acta Psychologica, 144(1), 25–30; Friston, K. J. et al. (2017). Active Inference, Curiosity and Insight. Neural Computation, 29(10), 2633–2683.
10. Vervaeke, J., Lillicrap, T. P., & Richards, B. A. (2012). Relevance Realization and the Emerging Framework in Cognitive Science. Journal of Logic and Computation, 22(1), 79–99.
11. Vervaeke, J., & Ferraro, L. (2013). Relevance, meaning and the cognitive science of wisdom. In The Scientific Study of Personal Wisdom; Lutz, A. et al. (2015). Investigating the phenomenological matrix of mindfulness-related practices from a neurocognitive perspective. American Psychologist, 70(7), 632–658.
12. Varela, F. J., Rosch, E., & Thompson, E. (1991). Embodied Mind: Cognitive Science and Human Experience. MIT Press, p.237.

your eyes are intelligent

Picture this

80% of the signals arriving at the visual cortex aren't coming from the eye at all, but are generated by the brain . What you experience as vision is a phenomenon resulting from your brain constantly hypothesising what's out there, rather than your eyes snapping a photo of the world.

Predict this

The brain being a prediction organ is the core idea of predictive processing, developed across the work of Karl Friston, Andy Clark, and Anil Seth among others, and is not just relevant for vision.

The brain has an internal model of the world, a vast and layered web of expectations built up from all the experiences you've ever had. It uses this model to predict what sensory input (of any kind) it's about to receive. When the actual received input matches your internal model, nothing much happens - you just sense the experience and move on with your life. But when the input doesn't match your model, the brain registers a prediction error and updates the model accordingly (not that you have much conscious awareness of this happening).

Two people looking at the same thing don't necessarily see exactly the same image or process it the same way: since they'll both have their own unique generative models built from different lives, they'll construct different visual experiences as a result. The input data is the same, what you make of it is unique.

Active inference

In addition to the constant updates your brain makes to its internal model, it also seeks out the input that will give it the most useful bits of information. In other words, it actively looks for incongruous signs in your surroundings to make sure it has as little uncertainty as possible about what you'll encounter your environment - this is active inference. The way your eyes move across a scene is a process that might seem random, but is actually a very targeted "information-gathering strategy" guided by your current model of what's likely to be relevant or surprising right now. This goes from micro-movements you're not consciously aware of, to bigger actions you might more easily notice (e.g. when you lean in to look at something more closely, you're using your body to gather more info and check if you're predictions are accurate, to update them if not so). The more you can make your predictions align with the world, the less uncertainty or surprise there is, and so the more likely you subconsciously think you'll survive.

Long story shorter: perception is something you do. You're doing it continuously, mostly unconsciously, but with so much sophistication: your eyes are like little scientists forming hypotheses, designing experiments, collecting evidence and updating conclusions. Your eyes are intelligent!

Picture yourself

Now this is all good and well, but the exciting thing about predictive processing is that you don't just model the outside world; your model also includes yourself.

The same predictive machinery also produces perceptual hypotheses about your sense of your own body, your emotions, your identity. While your brain is inferring the hidden causes of its sensory input, those causes include the ones traceable back to your own body and your own actions. The sensory experience of being you in this location in this body with this story, is something your brain is actively generating and updating all the time, not a fixed story it consults.

This means your sense of self has the same properties as any other perceptual model. Meaning it's built from previous experience, it varies in its accuracy, it updates itself. It also means that like any other model it can get stuck. A self-model that can't update produces the kind of psychological stuckness we may be familiar with, of feeling stuck in old patterns, of struggling to adapt to a situation. But the more you become aware of the nature of your own perception, but more you might become able to be flexible with your self-model. Growing and evolving into a more fulfilling version of yourself happens more easily the more you can effectively "integrate your prediction errors" without catastrophising. (Naturally, we can't take for granted that everybody is privileged with the same necessary physical or environmental conditions to be able to do this, but this is speaking generally and theoretically).

The world is a 3D mirror

If you are constantly predicting your world based on your internal models, and those models include your sense of yourself, then your experience of the world is always partially a portrait of you. You are always in the picture. Seeing is a creative act. What you choose to look at and how you choose to look at it matters more than the purely receptive model of vision would ever suggest.

References

References

1. Elkins, J., & Fiorentini, E. (2020). Visual Worlds: Looking, Images, Visual Disciplines. Oxford University Press.
2. Clark, A. (2016). Surfing Uncertainty: Prediction, Action, and the Embodied Mind. Oxford University Press; Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138; Seth, A. (2021). Being You: A New Science of Consciousness. Dutton.
3. Friston, K. J. et al. (2017). Active Inference, Curiosity and Insight. Neural Computation, 29(10), 2633–2683.
4. Hohwy, J. (2020). New directions in predictive processing. Mind and Language, 35(2), 209–223.

quantum transcendence and art

It's in reading Helgoland, Carlo Rovelli's wonderful account of the origins and implications of quantum physics, that I finally managed to understand why I don't fully understand what they mean when they say certain particles have been observed to change their behaviour when observed. As I said, I still don't understand the physics so I won't attempt to explain that part, but what I did understand was the implied conclusion to draw from this: according to quantum mechanics, objects do not have intrinsic properties that are true independently of their relations. There is no such thing as a self-contained thing, there is nothing that exists outside relationships - the relationships of a thing are what the thing consists of, and nothing more. Maddeningly, this isn't just a claim about perception but a claim about the structure of reality as a whole.

This reframing sent me on a flashback to something I wrote years ago as a student, about the relationship between a painting and its observer. I argued in an essay that without a perceiving subject (ie. a human), a painting is just a piece of canvas with some paint on it - not a work of art. But Rovelli just informed me that the work was in fact already real before any human came into the picture. It was already relational, already participating in a world of interactions that were making it into what it is. What changes when a human goes and stands in front of it is not that it suddenly becomes real, but that a specific kind of exchange becomes available in which the human's perceptual system - with all its prior models and expectations and history - becomes one of the interacting parties. The meaning of the work doesn't belong to either side of that interaction; it's a property of the interaction itself.

That little human is not an observer in a fixed world occasionally granted access to some meaning or other. That little human is a participant in a world that is alive with relation, and her encounters within said world are part of what that very world is made up of.

I still stand by my old self, or at least her intuition for writing the essay she did back then. Because considering the above, art makes these relational and participatory foundations unusually visible. A painting, or a sculpture, or a piece of music, offers a structured visual (or auditory) field that is intentionally open enough to invite your perceptual (or auditory) system to start pattern-matching and meaning-making before you've had time to rationally decide what you think. The meaning that surfaces belongs not only to the artist, nor to you alone, nor to the work in question, but to the way all of these relate to each other.

What my own drawings and paintings and installations and toys keep returning to is the nature of that threshold between stimulation and meaning. Light hitting your eye is not yet vision; pigment on paper is not yet an image - something needs to cross over and give rise to something new. Call it quantum mechanics, call it transcendence - the movement from material to material that leads to an experience will never cease to amaze me. Rovelli writes that to him, physics seems to be "the place where the weave between the structure of reality and the structure of thought [is] closest" and yeah I get you man, that checks out.

This is the main reason why I always include an element of participation in my work; translucent or reflective surfaces, refracting images, foldable pieces, toys - they're all attempts to highlight that relational element. You as the observer have to move around them, change the way you look, fold open the paper, etc; it's supposed to slow down the moment of crossing over. When a hidden image surfaces as you shift your angle, or when the drawing on the perspex begins to interact with whatever is behind it, something is happening that neither the object nor the viewer produces alone.

Anyway - to conclude. If you've been looking around these pages for a while you might have seen the phrase "the rainbow is only real because you are"; I hereby revise that statement to "the rainbow is only real because you are but so is everything else because of everything else"

why tarot rocks

This is old. This was written when I was thinking a lot about perception. I still love that stuff, but I’ve also learned by now that magic is real.

why Tarot rocks — Short

Tarot rocks because it takes the content of your mind, mixes it with ancient archetypes, and then gives you a new lens on things. (Obviously, there is also something much spookier at work in a tarot reading, but that's a question for another time (maybe)).

The images on the cards offer themselves to your imagination and activate your own pattern-recognition before your nice familiar rational thinking can intervene. A tarot reading builds a story that distils your beliefs into coherent myths, which have the wonderful quality of being both universal and person-specific at the same time.

You're being offered a new frame through which to look at your own life -- which gives you wonderful agency. The agency to intervene and decide which reflexes and instincts you want to keep, and which to update, so you can feel fresh year-round.

why Tarot rocks — Medium

One of the reasons why tarot rocks is, I think, because it operates by a different logic than the one we usually apply to questions of "truth".

In SSOTBME, Ramsey Dukes proposes that the foundational difference between science and magic lies in their way of knowing:

• science operates through falsification (a pattern is real if it cannot be disproven (for now))
• magic operates through verification (a pattern is real if it can be found at all)

These are different tools for different questions - and tarot is using the second more than the first. It's trying to reveal meaningful patterns in your life, not trying to produce a universal law.

Naturally I want to look at things from a perceptual standpoint. From a perceptual standpoint, what's happening when you look at your tarot cards, the visual imagery bypasses that mediating layer of verbal language. In other words: the feeling comes before it has a name, the body knows before the mind has constructed an explanation. Tarot cards are deliberately polysemic: each image holds multiple possible interpretations open at once, which means your eyes are instantly settling on patterns. You are looking, and something inside you is looking back.

Then there's the step afterwards of course, because what distinguishes tarot from other meaningful images is the active construction of a story. A reader assembles the symbolic meanings of the cards into a story that contains the person in front of them, and a well-built story, as both Campbell and neuroscience would agree, "does something" to us that other forms of communication don't: The brain is a prediction machine, constantly updating its internal models of reality. When something external suddenly reorganises that model (aka, when the story feels like it really lands) there is a real restructuring of that internal model.

In short, tarot can give you a real sense of agency. Your situation hasn't changed, but you're offered an alternative story, so the lens through which you're looking at your situation can change.

(That being said - the archetypes on the cards are ancient, and when a reading lands with that particular bodily quality of recognition, there is probably something spookier at play than just your own neuroscience, but I don't feel particularly equipped to tackle that right now.)

why Tarot rocks — Long

Science and Magic

You might say tarot is not really magic, but simply a projection tool or a Rorschach test with prettier images. For now please let's assume this isn't the case for our current universe. When someone sits down with a deck and gets a strange feeling in their gut when they find that the cards have said something true, let's assume there is something more complex at play.

Try to considering the tarot from a place of ambiguity; not through the lens of a dismissive skeptic or a believing mystic, but a from place that considers both to might equally true. In Ramsey Dukes's SSOTBME, a powerful text on the mechanism of magic, he posits that the foundational difference between science and magic lies in their way of knowing:

• science operates through falsification (a pattern is real if it cannot be disproven (for now))
• magic operates through verification (a pattern is real if it can be found at all).

The logic of science and the logic of magic don't have to be conflicting ideas; they are different tools for different questions -- and tarot is using the second more than the first. It's trying to reveal meaningful patterns in your life, not trying to produce a universal law. Infinite patterns are real regardless of whether we can prove them. By sitting down with a deck of cards, a person is sitting down with magic; they connect to a different kind of reality that can lead to real and real powerful insights.

what you see is what you know

A tarot reading is often done by a psychic for a client, but doing a reading for yourself can be incredibly powerful in uniquely different ways, thanks to the way our brain is designed to handle new information.

Both the visual imagery on the cards and their symbolic interpretation offer a way to notice your own instincts and intuitive associations; they show you your own meaning-making processes that were already running in the background without you being aware. Seeing these processes from the outside allows you to intervene and change whatever you think needs changing.

Before going to read the description of the meaning of a card, it can be extremely helpful to just look at it first. The tiny artwork on the card communicates a meaning that bypasses the mediating layer of verbal language and logical reasoning.

We tend to default to language when trying to understand ourselves, but visual perception is older, faster, and in many ways much more direct. When you look at an image, your brain responds immediately before waiting for you to translate anything into words or thoughts; the feeling comes before the name, the body knows before the mind has had a chance to construct an explanation.

This is true of anything you see, and tarot has carefully constructed a specific visual vocabulary that uses this mechanism to wonderful effect. Each card is polysemic: the image offers multiple possible interpretations at the same time, suggesting various patterns that are open to verification. Pulling multiple cards then combines these myriad meanings with each other, and before you know it you have a real playground of myths, patterns and stories that your eyes are settling on before you've had a chance to realise what's going on.

The archetypal imagery of the major and minor arcana is designed to hold meaning open instead of pinning it down. Offer yourself the chance to not know immediately what each card means; look at the symbolic environment in front of you and ask what arises when you encounter it. Even if you know tarot well and have an intuitive understanding of each card already, this perceptual exercise is still worth doing.

What results is a very special kind of cognitive engagement: you are looking, and something inside you is looking back. The meaning is not in the card, but the card is a key to your own locks.

story time

Naturally though, looking is only the first step. What distinguishes tarot from other images you can find meaning in is the active construction of a story. A reader takes the symbolic meanings of each card and weaves them into a narrative that applies to the person sitting in front of them.

Myths, as Joseph Campbell understood them, are symbolic representations of universal patterns: their function is orientation, giving people a felt sense of where they sit within a broader human and cosmic story. The reason they work is because of affect: the more we can recognise ourselves in something, the stronger our felt connection to it, and the more fully we understand it. We are instinctively inclined to make sense of things in terms of ourselves. This is also just how narrative works: humans have always preferred receiving information through story, because storytelling aligns with the way we already structure experience internally. Stories are more engaging and more memorable not because we choose to find them so, but because we are quite literally built that way.

When a myth, an image, or a story produces a felt recognition rather than just an intellectual response: when something you perceive is in harmony with your inner nature, you are pitched back into yourself. In a tarot reading, this is the moment when the story (whether you're experiencing it through your eyes or building it alongside a reader) suddenly lands, and you feel like you're being seen.

see for yourself

What makes this kind of recognition possible is the distinctly human skill of meta-representation: your ability to form an image of your own mental state . We can mentally visualise not just the world around us but our own representations of it too. The mind can take itself as an object, a move that allows us to investigate and question our beliefs and mental models, and perhaps build them into a different picture. Meta-representation is a crucial bridge between understanding and insight.

Tarot beautifully and deliberately activates this meta-representational power: its images are familiar enough to capture your attention (you vaguely recognise something) yet undefined enough to prevent any one final interpretation, which keeps your mind engaged rather than placated. This is flow: when the complexity of what you're looking at is nearly equivalent to your ability to understand it, the result is effortless engagement rather than anxiety or boredom .

The moment an image or a story "lands" or "clicks", is a moment of predictive alignment in neuroscience. The brain is not just a receiver of information but a prediction machine, constantly generating and updating its internal models of what is out there based on what it already knows . When something external suddenly aligns with your own internal predictive models, it's a visceral experience of being seen.

Powerful works of art do not write our stories for us, but they reveal the ones we already live by . A good tarot reading does the same thing in an active guided way. The real reason why tarot rocks is because you're not just getting answers but agency; your situation hasn't changed, but the lens through which you're looking at it can. You're not being handed a given meaning, but you're given the change to co-construct a new one with your own material.

about my deck

Tarot readings are a key part of my personal and social life. The cards offered endlessly fascinating additions to my research into how visually complex art facilitates insight and self-awareness by activating the brain's perceptual problem-solving mechanisms. The tarot is an extraordinarily sophisticated application of exactly these principles. And it's beautiful!

I made a tarot deck that maps the four suits onto the four dimensions of the predictive perception framework: Wands onto active inference; Cups onto affect; Swords onto cognitive thought; Disks onto material embodiment. The major arcana move through the full arc of individuation from the Fool's first step into uncertainty to the Universe's moment of integration.

PS

(That all being said - the archetypes on the cards are ancient, and when a reading lands with that particular bodily quality of recognition, there is probably something spookier at play than just your own neuroscience, but I don't feel particularly equipped to tackle that right now.)

References

References

1. Dukes, R. (1979). SSOTBME: An Essay on Magic. The Mouse That Spins.
2. Semetsky, I. (2005). The Hieroglyphics of the Other: The Major Arcana of Tarot as a Semiotic System. Semiotica, 2005(157), 47–74.
3. Campbell, J. (1986). The Inner Reaches of Outer Space: Metaphor as Myth and as Religion. Alfred van der Marck Editions.
4. Mellmann, K. (2012) ‘Is Storytelling a Biological Adaptation? Preliminary Thoughts on How to Pose That Question’, in Telling Stories: Literature and Evolution. pp. 30–49.
5. Sperber, D. (1985). On Anthropological Knowledge: Three Essays. Cambridge University Press.
6. DeLanda, M. (2022). Materialist Phenomenology: A Philosophy of Perception. Bloomsbury Academic.
7. Clark, A. (2016). Surfing Uncertainty: Prediction, Action, and the Embodied Mind. Oxford University Press.
8. Friston, K. (2010). The free-energy principle: a unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138.
9. Jung, C. G. (1936). The Archetypes and the Collective Unconscious. Collected Works Vol. 9, Part I. Princeton University Press.

you must change your life

If you look at things as if they had the power of looking back at you, everything comes to life in response.

From Rilke’s famous poem, ending with: "... for here there is no place that does not see you. You must change your life"

Your own perception acts as a nucleus for a burst of life that you can apply to whatever you choose to really look at. The best part of this is that once you realise everything is looking back at you, you suddenly find yourself surrounded by things that are, by virtue of their own acts of perception, bringing you continuously to life. Let it encourage you, not to feel pressured to change your life, but to be allowed to change your life, in the reassuring knowledge that you are being looked after on your quest.

Drawn from one of my favourite podcast episodes titled “You Must Change Your Life”, from one of my favourite podcasts titled “Weird Studies” (first shown to me by one of my favourite people).