34. At Human Scale

There's a metaproperty - a property of properties of objects - that I've noticed that I've never read anyone remark on. That metaproperty is something like "being human-legible" or "being unsurprising for possibly ancestral-environment reasons" or just "being at human scale". It's a little hard to operationalize well, but one stab would be something like: we have a property of objects - density, say - that can have a number reasonably attached to it. The lower and upper extremes of the range of that property give an object or material surprising properties, because they don't act like anything found in nature, but only in human-built environments. If an object has this property at the extremes, it's not human-legible in this way; if it falls well within that range, by contrast, it's human-legible or at human scale.

I'll give a few clarifying examples. As above, density is one of the best examples of this metaproperty: not for nothing are tungsten spheres so captivating, and likewise aluminum's ability to float on water and aerogel's extreme lightness are cause for delight or disappointment. By contrast, water, wood, and ceramics hold no such reason for surprise. Flammability is another example, where the relevant number might be something like flash point or flame front speed: kerosene, natural gas, and flash paper are all examples of materials not at human scale for flammability. By contrast, wood is once again relatively unsurprising here, as are cloth, sand, and water - it's not always the case that both extremes of the property need be surprising, though an unsurprising extreme can itself give rise to further "not at human scale" surprises - just think of metallic sodium or chlorine trifluoride. Abstract properties like "market value per gram" can fail to be at human scale at this way, too: think of the psychological weight an SD card with a bitcoin wallet might have, as opposed to a lump of clay, or even something like paper money or a gold coin. For one last example, consider poisonousness or intoxication potential: drinks like beer, hard cider, and coffee would be legitimately fairly tricky to poison yourself with; you'd likely run into digestive constraints and psychological difficulties before the ethanol or caffeine proved overly harmful of itself. On the other hand, distilled spirits - especially Everclear - and powdered caffeine are hazardous and deserving of caution.

So why does this phenomenon arise? One possibility is something like a cognitive bias in modelling: in the same sort of way that we might model other people as being like us, or like others we're familiar with - wanting similar things, thinking in similar ways, knowing similar facts - we might also be modelling objects and materials as being similar to the objects of ordinary life, and getting surprised when they fail to share properties. Another is the dictates of evolutionary psychology and its attendant selection pressures, that old saw. For one last stab at the underlying mechanism, it might even be a matter of unconscious predictions about how objects will change, react, or affect us, rather than just expectations about their properties, and the ways that those expectations are violated.

Comments

Post a Comment

Popular posts from this blog

20. A Sketch of Helpfulness Theory With Equivocal Principals

You may rightly wonder what else I’m up to, if it’s not just blogging and the occasional adventure. The short answer is that I spend most of my time and mental energy taking part in a summer research fellowship in AI safety called PIBBSS, where my research direction has to do with questions like “how does an agent help a principal, even if the agent doesn’t know what the principal wants?” and “how much harder is it for the principal to teach the agent what they want, and how much worse are the outcomes, if the principal doesn’t know for sure what they want?”. The long(ish) answer is this post. I’ll avoid jargon, mathematical notation, and recommendations to read other long papers, but some amount of that will be unavoidable. Here’s the basic setup - called a “Markov decision process”, sometimes with adjectives like “decentralized” or “partially observable” - which forms the ground assumption for the entire subfield my research direction lies in, called “inverse reinforcement learning”:...
Read more

4. Seven-ish Words from My Thought-Language

Seven entries from a dictionary that will never exist, even though it should. The words in [brackets] show up with some frequency in my thoughts, and I struggle with English’s semantic poverty. Some of them show up in other languages, like Korean or Lojban. Others among them, people I’ve spoken to share in wishing were wider-used words. [untranslatable 1: word-clogged/interference-silent] (adj): Of a person or their state of mind. Descriptive of the phenomenon where a person has multiple different things to say about a specific topic, or multiple different responses or reactions to a remark an interlocutor has made, only one of which can be taken. To be tongue-tied not out of a lack of things to say, but out of an overabundance of mutually exclusive things to say, all of equal priority, especially when each thing to say would result in following importantly different conversational paths. Often used instead in its natural verb form, which means “to be silent as a result of [word-clog]”...
Read more

11. Why the First “High Dimension” is Six or Maybe Five

Image
(Epistemic status: morally correct, in the mathematician’s sense; here to give flavor and intuition without too much rigor.) People toss around the concept of “high-dimensional space” a lot without having a good idea of where that starts. Interpretability researchers staring at vectors in hundreds of dimensions are clearly contemplating a high-dimensional space - loss landscapes packed full of saddle points and spheres where any given pair of vectors is almost certainly nearly orthogonal - but do they know what it means to be a high-dimensional space? Surely we can do better for ourselves than Justice Stewart’s infamous maxim - “I know it when I see it”! What do we care about, when we want to tell how many orthogonal axes we need in a space to start calling it high-dimensional? I claim that we should care about looking at unit-radius disks - like, literal filled-in two-dimensional circles - and in particular, checking when, whether, and how pairs of those unit-radius disks overlap when...
Read more