There is a character to metropolitan city living, a quality separating the city psychologically from the countryside, and from the smaller towns and cities around it. It is difficult to put your finger on it exactly, but it is there: it’s at least partly the presence of buildings without end, a city stretched out to the horizon, the place that people outside look to as the place of escape. The familiar strangers on the train platform; the crowds of tourists jostling about, the places that remain constant although the people change constantly.

Then there are the unseen constraints: old street names pronounced in new ways with strange accents, the discontinuous neighbourhoods, stranded small villages with blurry boundaries; the variegated sights, sounds, smells, the changing ground under foot. There is that peculiar character of a city awakening in the early morning; noise building here and there, people appearing, bikes and cars suddenly rushing past, the rumble of the trains and buses.

This quality, is so difficult to describe, is one so very different to the small towns and villages and other places you encounter, or perhaps have lived in, or perhaps even live in, but it’s there, exciting, frightening, innervating, ennervating, alive.

All roads lead to the city. All roads lead from the city. It is concrete, glass, steel, unyielding. Then a park, a green square, trees, bushes, seats, a different pace.

I love it.

I’m thinking here of Dublin, London, Paris, New York, Rome, Bangalore, Singapore, but not Florence, Venice, Galway, Cork, Manchester, Oxford.

There’s something different, almost intangible, about metropolitan city life, the cityscape itself, a separateness, a something feeling different in itself, and purely of itself.

A town, even a beautiful or historic one, feels bounded, hemmed in in some way: you can imagine its edges; you can see yourself walking out of it to the countryside around.

You can hold its whole image in your head.

A metropolis exceeds you: your grasp, your vision, your intuition.

A metropolis cannot be grasped, apprehended, visualised, all at once, for it has no single face: it contains villages, empires, ruins, commuters, tourists, office workers, ghosts, money, poverty, performance, exhaustion, architecture, bureaucracy, memory, anonymity, connections. A metropolitan city is not merely a place where people live, for the metropolis is a machine for producing encounters, ambition, escape, concealment, reinvention, renewal, change.

That may be the most intangible thing: a sense the city itself almost has an independent psychological existence, not just a backdrop for human life, but a defining, liminal, subtle presence with peculiar affordances all its own.

In such a city, the urbanity almost seems conscious: the confluences of people, the crowding and emptying railway platforms, the near collisions of speeding pedestrians at street corners, the squares where we sit and watch the world go by, the office blocks where we imagine we command the world as it goes by, the noisy parks at lunchtime, the cafés, the rivers and canals crossed by people who do not look up, for their phones have captured them:- all of this has a character, a qaulity, a way of being, independent of any one life passing through it.

The metropolitan city is an atmosphere, a hive mind encapsulating the individual mind, absorbing you, capturing you, rendering you small even as the stage you are on gets bigger.

It’s a place that watches no one, yet seems to accomodate everyone.

The distinction between the ‘beautiful city’ and the ‘metropolitan city with psychic mass’ feels central - perhaps the key point.

Florence and Venice are extraordinary, astounding, jawdropping, but they are too available to the eye: they are aesthetic objects, too easily reduceable to tourist porn in the way that New York or London cannot be. Rome and Paris are beautiful, but they are not reducible to their beauty alone: there is so much more going on in them.

Dublin now has this character too, albeit not imperial in scale, never having had the plundered treasures of empire to conjure itself into being with, but it has a peculiar something: a new busyness, crowdedness, literature, rain, Georgian order, institutional memory, pub life, sea edge, commuters, red brick buildings, old tenaments now repurposed, the ghosts, new buildings, new and old neighbourhoods, its raggedy unfinishedness.

Dublin has an all new tempo now - a speed, crowdedness, and urgency it lacked thirty years ago. It has become somewhere. Not just a city people left from, or returned to, or remembered sentimentally from abroad, but a place people come to. A place of arrival. A place of ambition, friction, anonymity, and pressure.

This is not simple progress, and not simple loss, for something else has been gained: scale, confidence, worldliness, the electricity of elsewhere arriving here. Something has been lost too: ease, cheapness, slack, the accidental rooms in which certain kinds of life can unfold. But psychologically the change is unmistakable. Dublin no longer feels like a city waiting to become itself. It feels like a city under pressure from having become itself.

The state is here, the universities are here, the tourists are here, the tech firms are here, the lawyers, writers, migrants, returning emigrants, students, consultants, civil servants, shoppers, drinkers, and commuters all pressed into the same narrow streets.

The result is not grandeur exactly. Dublin is not grand in the manner of Paris, nor vast in the manner of London, nor mythically vertical like New York. Its difference lies in compression. It is a city whose old scale has been forced to contain a new global role. That produces the tempo: the crowded platforms, the costly rooms, the cranes, the hotel lobbies, the hurried cafés, the little Georgian streets often carrying too much traffic.

Dublin could feel smaller than itself: a capital city with a provincial tempo, a place of great buildings and narrow prospects, of literary memory and economic constraint, of wit, talk, exile, and return: a place of ‘silence exile and cunning’. Its history was long and large, but its present, and presence, was underpowered.

That has changed. Dublin has become, or is becoming again, a place that matters. Not simply larger, not simply richer, but more consequential.

I do not think those of us who live in Dublin have quite metabolised this change. We still talk about the city as if it were the Dublin of thirty years ago: intimate, familiar, a little shabby, a little underpowered, a place whose charm lay partly in its limits.

But the city has changed category: much more crowded, more expensive, more international, more urgent, more economically consequential. It has regained gravity, but much of our public language still treats it as a large town with capital-city functions.

Nor do I think our politicians have recognised this fully. They govern Dublin as though its pressures were temporary inconveniences rather than signs of a changed urban condition. All under straing and in demand: housing, transport, public spaces, policing, planning, tourism, culture, night life, universities, migration, work: these are are the symptoms of a city that has become somewhere again.

The political task is not merely to fix Dublin’s deficits, important as that is. It is to understand what Dublin now is. A city that matters needs institutions capable of dealing with it, managing it, envisioning it.

And yet we behave as if this were normal, as if Dublin were always bound to become this: crowded, connected, expensive, multilingual, airport-facing, economically consequential. But it need not have happened. Dublin could have remained what it was for much of living memory: dowdy, charming, down-at-heel, clever, intimate, rain-darkened, a city of talk, literature, underused buildings, familiar faces, modest expectations and emigration.

Dublin could have remained a capital that felt smaller than itself, but instead it has become, or is becoming again, a place that matters - not rhetorically, not in nostalgia, but in the hard senses of work, money, migration, travel routes, pressure and consequence. The strange thing is not only that Dublin has changed; the strange thing is how little we have allowed ourselves to be astonished by this change.

However, to borrow from Robbie Burns: to see ourselves as others see us, Dublin looks less like a charming, rain-streaked literary capital than like a high-performing, high-cost, globally connected services city that has not fully built out the housing, transport or civic capacity required by its own success.

We have a sometimes-provincial political class managing a post-provincial economy; Dublin has not become a lesser version of a larger city: Dublin has become the larger version of itself that is was never fated to become, but somehow did and this happened before it has become domestically understood. Dublin has changed category: it’s not that Dublin has become or tries to outcompete London or some other metropolis - it is self-confidently its own place now, and can’t be bothered with self-aggrandising or hubristic comparisons anymore.

To my non-Dublin readers: you should come visit, stay, live here.

• Our (future) cities, our (future) selves

• Urban Walking (Galway, Dublin, London, Paris, New York, Miami Beach, Lucknow)

• Central Park Walking

• Post-pandemic cities

• The myth of the magical city: Why I hated Paris, why I was wrong, and what New York taught me

In Praise of Walking

College Green, After the Bell

“O tell me all about Anna Livia!” (Joyce, Finnegan’s Wake)

Step out of Trinity through Front Gate into the rarefied air of College Green; behind you are lecture rooms, exam halls, labs, libraries, tourists photographing the Campanile, students cutting diagonally across Front Square (because students walk whatever way they want). Ahead is Dame Street - what should the grandest of vistas: instead a melange of buses, taxis, cyclists, delivery riders, office workers, visitors looking the wrong way before crossing, the old Bank of Ireland building curved and self-important where the old Irish Parliament once sat. Edmund Burke is standing on his plinth facing the onslaught, willing to take on all-comers, waiting to demolish a poor argument, a bad idea.

(below the line - a Dublin psychogeographical tour, and lots of music; and D4 guy; how to speak Dublish; and the culchies among us; and an outsider’s take)

Read more

But how do we train them? Paradoxically, by behaviourist means. I’ll discuss this more below.

For prior context, see the last post:

Reinforcement learning from human feedback is one obvious example: desired outputs are differentially stabilised by evaluative signals. We reward some behaviours, discourage others, and shape performance through consequence. In practice, we treat these immense internal systems as black boxes whose behaviour must be adjusted from the outside. We do not yet understand them well enough to engineer their inner workings directly, so we fall back on an updated form of behavioural shaping.

Skinner has returned through the back door because when faced with systems too complex to understand completely, consequence-based control remains useful for changing their behaviour. We don’t need theory of mind to comprehend an LLM; we just need to engage in shaping its behaviour by manipulating the consequences of the text sequences it generates.

The irony: critics of radical behaviourism thought it had been superseded by richer accounts of internal mechanism - yet, in the age of deep learning, we often find ourselves managing immensely intricate internal architectures in thoroughly behaviourist ways.

tacit knowledge

What do we do about tacit, implicit, non-declarative, implicit knowledge?

Scaling word models, which is to say feeding machines vast quantities of declarative content, produces extraordinary fluency of text - reams and reams of it. But it has not, by itself, delivered general intelligence in any full, implementable, sense, because a great deal of human intelligence is non-declarative. It is procedural, embodied, subcortical, and difficult or impossible to articulate explicitly. It is the peculiar weighted feel of an instrument in surgeon’s hand while probing deep tissues, the subtle timing of a comedian, the balance of a cyclist at variable speeds on a bumpy road negotiating oncoming traffic, the practical judgement of a seasoned craftworker, the composer who knows that judicious silence and dropped beats are essential to music too.

We know more - much more - than we can tell (and sometimes we hallucinate - telling more than we can know - we don’t have access to our own cognitive processes - we can only report on their outputs).

Subscribe now

There’s load more BTL:

• breaking the black box open;

• how would Skinner’s schedules of reinforcement be integrated into modern RL?;

• fixed ratio schedules and deterministic rewards;

• variable ratio schedules and stochastic reward;

• fixed interval schedules and temporal credit assignment;

• variable interval schedules and partially observable environments;

• variable interval schedules and partially observable environments;

• reward schedules as environment design;

• why this matters for modern AI;

• the deeper point;

• the history we ought to tell

Read more

Read more

I suspect B.F. Skinner’s nam…

Read more

Read more

Here are recent pieces - a mixture of free to read and subscriber on…

Read more

Read more

Notes: this is probably the last of these longer discursive pieces I’m going to do here for a while - I am going to change the focus in favour of shorter pieces revolvin…

Read more

Read more

your brain’s “idle” mode is a powerful social and time-travel engine

• The idea: When you’re ‘doing not…

Read more

Read more

Read more

Read more

At the risk of self-praise, I’m very happy with the Alien/Xenomorph and P…

Read more

BL, UF: they have precisely no idea how humans brains process information; …

Read more

Read more

a coherent portrait of an intelligent and techologically advanced species hunting for ritualistic…

Read more

The theory: if we somehow match teaching to a learner’s preferred modality (typically labelled ‘visual, auditory, reading/writing, or kinaesthetic’, sometimes shortened to VARK), learning will improve (somehow).

This is a big claim. Imagine being confronted with a task to learn: a written script (verbal learning), or the piano (visuomotor learning), or navigating a complex work situation with competing demands (social learning), or learning a song (auditory); or the new complex ticket machine at your train station (visual, auditory, kinaesthetic); or learning to drive a car, a bike or a computer game (visual, auditory, reading, kinaesthetic).

Transform the task somehow to your preferred learning style, and voilà, you’ll learn more quickly, efficiently, and retain the material for longer, and presumably be able to use it more effectively.

On the contrary, if you ignore your preferred learning style, then you’ll learn more slowly and less effectively.

Seductive idea, isn't it? (First question: how do you know your preferred learning style? Do you really know it? On reflection, do you have a preferred style for learning how to swing a racket, a golf club, or to learn a new complex coding procedure? Course you don’t.)

Learning styles theory is not just a claim about some preference you have (like a preference for salty over sweet foods); it is a serious claim about the mechanisms of learning and memory, as well as a set of predictions about the learning and memory systems of the brains of learners.

However, pursuing learning styles as a theory of learning and memory has been one of the deadest of dead ends: a pointless, nonsensical, take on learning and memory, a view of learning discussed solely within a group of aficionados who have been having silly and pointless conversations with themselves for decades, even though the idea has failed every empirical test, and makes no contact with actual studies of learning and memory, at any level, from synapses to brain systems to collective memory.

And it is a daft idea if you step back and consider the anatomy and evolved cognitive functions of the brain in any detail at all:

In experimental terms, learning styles proponents assert an aptitude–treatment interaction (ATI): a person × treatment effect in which people assigned to their preferred ‘style’ will learn more from instruction delivered in that style, and will learn less when paired to a mismatched style.

That decisive crossover pattern has not materialised - for it does not exist. Instead, we have an intuitively-appealing story floating free of neurocognitive theory, resting on little-to-no or weak measurement, and which resists being killed off when critical tests show it is wrong (because the committed want to believe in it!).

My book ‘Talking Heads’ (US Kindle Ed available here) has an extensive discussion of shared reality, memory, and collective remembering. No learning styles here.

The prevalence of belief in learning styles among educators remains strikingly and stupidly high, possibly explaining the persistence of the learning style neuromyth despite null findings.

The effective alternatives (retrieval practice, spacing and interleaving)are rooted in well-replicated cognitive mechanisms that generalise across learners.

Finally, contemporary neuroscience shows memory depends on distributed, interacting cortical–temporal–diencephalic circuits, and not isolated ‘visual’ or ‘auditory’ modules, undermining the biological plausibility of fixed sensory ‘styles.’

If you enjoyed this post, and want more essays on how neuroscience and psychology can help us build better lives, subscribe below. You’ll get new ideas from BrainPizza.

the seduction of personalisation

Learning styles continue as a conjecture about learning because they take some very weak claims, which are then taken to a stronger set of mistaken conclusions.

It is true that disciplines traffic in different representational media, from maps and diagrams in geography, to equations and graphs in physics, to timelines and narratives in history. And multi-modal materials for instruction may help learners.

From these premises, a stronger claim is made: that the durable part of learning improves when instruction is matched to a person’s stable modality preference.

The rhetoric is that of personalisation, but the reality is a label applied to people rather than an analysis of the tasks they face.

A credible theory of learning makes contact with mechanisms and boundary conditions. It specifies what is changing in the learner (encoding, storage, retrieval); it says when a predicted effect should appear and when it should not.

Learning styles accounts offer none of this. ‘Preference’ is treated as a cause; ‘matching’ is treated as a cure, leaving us to imagine a ‘visual learner’ who ought to learn better from pictures and an ‘auditory learner’ learning geometry by talk rather than diagrams (as if they don’t have a visual system!).

This is not a theory; this is ‘not even wrong’.

a test the idea cannot pass

The only fair test are experiments that:

(i) classify learners with a stable, valid measure;

(ii) randomly assign them to matched or mismatched treatments;

and (iii) evaluate outcomes that matter (delayed retention and transfer), while estimating the crossover interaction directly.

(BTL: what modern cognitive science actually says; neuroscience says not to styles; harms and opportunity costs; why the myth persists; from labels to learning architecture; policy and practice: what to stop, what to start; the bottom line; annotated bibliography, with brief notes on methods and relevance)

Read more

Continuing this series on experimentally-tested, effective, empirically-demonstrated, and theoretically-founded methods of enhancing learning and memory.

When you go to the gym, you expect gains entail (a little) pain, or effort, exertion, and discomfort.

Learning demands the same!

(PS: coming soon, why learning styles do not exist, so you can save yourself the trouble of thinking about them ever again… and further posts on action-oriented learning, applying optimal strategies from learning science to sport, music and other domains where expert knowledge is vital.)

what is interleaving?

Interleaving means rotating across related topics, concepts, or problem types within the same study block.

Suppose you're doing some difficult geometry problems: instead of finishing all the sphere problems and only then moving to cubes, you purposely alternate: sphere; cube; pyramid; back to sphere; and so on.

Blocking means practising one topic or skill at a time - something students tend to do - concentrating on one thing at a time. Interleaving mixes related topics within the same session—for example, instead of AAA–BBB–CCC you instead rotate between items as follows: ABC–ABC–ABC. Across many studies, interleaving tends to produce better long-term retention and method selection than blocked practice, especially when you must distinguish between similar problems, even though blocking often feels easier during study.

To push the gym analogy a bit: block training is like doing all your bench-press sets, then all your squats, then all your rowing: AAA, BBB, CCC.

Interleaving is circuit training: bench, squat, row; repeat: ABC, ABC, ABC. The circuit feels tougher, because you must reset grip, stance, and breathing each time; switching teaches you to choose the right movement under pressure and builds more durable, transferable form.

Blocked sessions feel smoother in the moment, but mixed sessions pay off when it’s time to perform.

Similarly, each topic switch forces you to decide what kind of thing you’re facing and which method fits. Repeatedly having to pay attention to differences (shape, sizes, angles, spatial relationships) trains discrimination: you get better at spotting ‘what it is’, before ‘learning how to do it.’

Because each topic is revisited after a gap, you also get built-in spacing; pulling the right rule back from memory strengthens it remember the discussion of retrieval practice; link below). The learning and practice session feels less smooth than blocked practice, but the learning is more durable.

The challenge for the learner is mostly cognitive and metacognitive. Cognitively, these switches create interference: your working memory must drop one procedure and load another; decision time increases; mistakes rise at first.

Leave a comment

You can’t rely on the last problem’s pattern; you must choose afresh, which is effortful. Metacognitively, interleaving feels like worse performance, because it’s slower, less fluent, and prone to more errors: your intuition says it isn’t working.

But your intuition is wrong.

Blocked practice - sticking to the one topic for long periods of time -by contrast, feels easy and rewarding in the moment because you are exercising the same rules again and again; it inflates confidence without building flexible knowledge.

implementing interleaving to enhance learning and memory

Subscribe now

(btl: why interleaving works; interleaving and retrieval practice reinforce each other; when to interleave (and when not to); how much to switch; designing interleaved sets; common pitfalls; a practical protocol for interleaving; protocol card: “interleave + retrieve”; readings)

Read more

A bee can see UV light, and we cannot. Our world of botanic experience and theirs is utterly and completely different (and to complicate matters, they have compound eyes, which are very, very different to ours).

A sheep has peculiar-looking (to us), horizontal, rectangular, pupils that are highly sensitive to movement at the periphery of their vision (valuable if you want to avoid hungry, roaming, wolves, dogs, or other predators), and we do not, so we cannot see the world in the way a sheep sees the world: grass and flock in front, possible predation at the periphery).

Cats have horizontal pupils, hugely sensitive to motion on the ground below and in front of them - handy if you’re a predator looking for scuttling mice.

But not us.

We humans have forward-facing mobile eyes with round pupils mounted in an exceptionally mobile head. We do not have side-mounted eyes like horses and rats - our frontal field of view overlaps very substantially. We also have 3d colour vision exceptional vision - unsurprising, as we have roughly 40 distinct visual areas in our complex and elaborate brains.

Last time:

And we humans alone can learn to associate sounds with strange little squiggles and use these squiggles to communicate with other humans on the far side of the world, leading us to see and know the world in new ways.

If you enjoyed this post, and want more essays on how neuroscience and psychology can help us build better lives, subscribe below. You’ll get new ideas from BrainPizza.

This little tour of some of the senses in the animal kingdom should be enough to convince you that the way we humans can see 👀 and know the world 🌎 is very different to what other creatures can see and know of the world.

What you can see, hear, and smell is not the world in its entirety, just your piece of it; your umwelt - the world that you live in.

Umwelt is a concept from biology, psychology, and philosophy refering to the subjective perceptual world of an organism: the ways the world exists for that organism, based on its sensory and cognitive apparatus.

Aliens

What is the umwelt of a xenomorph or a Predator? I’m going to discuss these in later posts📫📫 (partly because there are some great Alien and Predator shows around at the moment, and speculating on their cognitive neurobiology is fun).

(Lots more btl, inc: core notion (inc ‘the world of the tick’); the world is plural; why it matters)

Read more