17 Comments

I'm surprised you think the function of consciousness is to tell stories. Maybe you have a post on this, or a link to book. Or perhaps I don't get what you intend to mean.

My view is the mind is a model/prediction machine. Simple as that. You intuit the future you desire, and your brain has your body act to make that desire happen. Catch a ball, eat, whatever.

The next level is theory of mind modeling for social consciousness. But again, the brain intuits what you want, and then you act.

Stories are of course important to how people understand the world. But it's not clear to me they are the "main function."

Anyway. the prediction from my view is that there is a general difficulty to understanding motives/how people would act/react. The point of your post. If it's mostly a kind of blindness to our selfish desires, then baseline is we'd understand a lot of motvies and predict them well, EXCEPT when implies selfish motives. I think your original post better supports my view of a general condition, than your view of a general capability that has targeted blind spots.

Expand full comment

The main function of consciousness is to create a simple story of what I've been doing and why that I can remember and tell to others. That suggests that self-understanding is quite important.

Expand full comment

My position is complexity is more important than self-deception as cause of lack of self-understanding. I'm pretty confident you'd argue the opposite.

Perhaps where we'd align is that evolution could have selected for better self understanding. But there's little Darwinian survival benefit to this. So we don't have much of it.

Whether that lack of effective selection pressure to have better self understanding is due more to difficulty, versus due more to advantage for not knowing our own selfish motives, is not a disagreement in principal. Could be resolved by clever enough empirical evidence if we could come up with a way to measure.

Expand full comment

People with Williams syndrome come across as very childlike regardless of their chronological age. They can be incapable of lying or distrusting others. It's a bit like they never entered the arms race everyone else does via aging. It has been argued that when humans domesticated wolves into dogs, we essentially bred them in the direction of Williams syndrome.

Expand full comment

Lots of people call themselves economists. neoclassical economics, macroeconomics, various types of heterodox economics and the neoclassical synthesis are a set of incompatible intellectual approaches that can each be used to offer explanations or advice on related subject matter. Game theory is another tool that can be used to discuss that subject matter. So are evolutionary game theory, evolutionary psychology, computer science, and economic history. The people using each of those tools can call themselves economists or can call themselves something else, and as your history demonstrates, people skilled with mathematical tools from physics can extend that skill into social sciences more broadly and then specialize in economics.

I will try to write an essay on the division between intellectual divisions.

Expand full comment

We are almost always hiding some things from others AND hiding some truths about ourselves from ourselves.

The 1000 brains idea (haven't read book) seems to allow multi-variate analysis, in 1000 dimensions. For any individual, most incentives have positives (P# of brains) & negatives (N#), and this results in some probability estimate of overall net P or N.

Either including or further complicated by the desire to be some way, or appear to be that way.

Expand full comment

Yes, an arms race is a reasonable interpretation of our relative ease in reading kids. Not sure it is fair to put game theory outside of economics; economists have been central to developing and using it.

Expand full comment

Complex, but also there's a stochastic element, so fundamentally unpredictable in any precise way.

Expand full comment

Would you say that it’s difficult to predict the motives and emotions of children? My sense is that it’s not. If adults are difficult, one natural interpretation is an arms race between prediction and concealment. Such an arms race, however, suggests a much more adversarial human condition than economics tends to depict, one oriented less around coordinating to reduce scarcity and more around pursuit of information asymmetries. The humanities and the military sciences (including game theory) seem to depict this to a greater degree than the social sciences, partly because such a world-view suggests that the work of actually creating social sciences is unlikely to take place, and the apparent existence of social science is more likely to be a front in an information conflict. The nonexistence of social science would explain the replication crisis.

On this view, philosophy and linguistics might be the most contested battlegrounds of all, leading to linguistics decisively refuting the viability of big data methods but in some sense only offering inferior alternatives, and to philosophy essentially failing to integrate its admittedly most eminent figure, Immanuel Kant, into its discourse, either by building upon or by refuting him.

Expand full comment

Emotions are, essentially, the distilled evolutionary wisdom of a million generations of hominids. In situation X it is advantageous if you have feeling Y. But if you _think_ about it, if you wonder why you are feeling Y, that spoils it. We're bad at understanding emotions because our brains and genes are conspiring to keep us from looking behind the curtain.

Expand full comment

I agree that there are real complexity obstacles to self-understanding, but that by itself doesn't tell you if that is the main obstacle to self-understanding. We could also be purposely hiding some things from others.

Expand full comment

I've been thinking a lot about how much self insight we have while reading Jeff Hawkins new book on 1000 brains. Where he argues the neocortex cortical columns are all separate models, and these ~150k models vote to create our perceived reality and decisions.

Mercier and Sperber argue in Enigma of Reason that we use reason mostly to guess others beliefs, and argue for our own beliefs. And further, when we reflect on why we do something, we have barely more insight into our guesses for others as for ourselves.

How this ties together is that if we have 1000 brains competing models, our decision process is profoundly opaque, as much due to complexity as anything else. What we have are intuitions on what to do, and we do them, but we don't have access to where these intuitions come from. Machine learning has a parallel, where the classifier doesn't know why it classified, since it's an emergent property at the system level, not a causal step by step logical argument. Beehives also have a similar property, where bees do their waggle dance to pick location for new hive, but the individual bees promptly forget their dance. The decision emerges only at the hive aggregate level. So it's an intuition coming from the system, which itself has no logical step by step understanding of where the intuition comes from. Since at a formal logical level, that doesn't even exist. We just have intuitions and do what we do.

Anyway, sorry for long comment. But the short version is our inability to predict emotive responses and reactions is (my own view FWIW) primarily because they are incredibly hard to predict. Our minds models vote to make decisions at the emergent level, with no access to all the submodels which voted different, or the relative rank of alternate actions. So our motives/emotions are hard to predict not *primarily* because they are deliberately hidden. But because it's just really hard to do. Simple as that. Though of course hiding our selfish motives also happens as you've written about, so we hide our selfish thoughts. But most broadly I'd argue it's due to complexity of emergent thought.

Expand full comment

This is almost a post about mirror-neurons without ever mentioning mirror-neurons, which is understandable given the contentious nature of the topic.

If you accept that there is some sort of neurological process that is responsible for our capabilities surrounding the points you laid out then doesn't it just make sense to assign a gaussian distribution to a population to reflect our abilities? Thus 'we' are not bad at it, but average.

Expand full comment

Yes of course some people are better than others. The point is to notice how bad we are on average.

Expand full comment

Yes, predicting large aggregate statistics ("thermodynamics") is easier than predicting the behavior of an individual ("gas molecules").

But there do seem to be people that have extraordinary skill at even individual emotion prediction and even redirection. This is generally thought of negatively, as "manipulation". It's not really a part of mainstream "science", but you can see it in ("used car") salespeople, or politicians (Obama), con artists (Trump). The Pick-Up Artist (PUA) community could be seen as applying scientific methods to the prediction and manipulation of individual emotional behavior in the dating arena.

Are those not examples of the kind of understanding that you were searching for? They tend not to appear in peer-reviewed scientific journals, but they otherwise seem to have the right character.

Expand full comment

I'd be a little hesitant to say that you or the people in your examples find these things hard to predict. I mean imagine how good an alien or a computer program would be. We laugh at ai attempts at this kind of thing because it makes mistakes that are so absurd to us that we don't even see them as options...but the fact we exclude them as options is evidence we are quite good at these predictions on some objective scale not that we are bad at them.

I think it's more accurate to say that we have trouble making predictions which distinguish between cases other people are also bad at distinguishing. It's not that companies don't know that people want cheaper computers with more memory, greater battery etc.. or that film makers know that people prefer griping films to 2 hours of paint drying. Rather it's that in those cases where we know it's a hard call so pay attention to it it really is a hard call. If we were better at this we would just move the line about what's super obvious and goes without saying. Maybe it's that there is an abrupt shift from obvious to pretty inscrutable but I'm unsure.

Expand full comment