Thread by @IntuitMachine

As an outsider in neuroscience, I always have been uncomfortable with their literature in that nothing is ever conclusive. It's like stamp collecting contradictory evidence. It's up to the reader to make up his mind of which set has no contradictions.

Also, it appears that almost all generalizations are wrong. The left-right brain hypothesis is wrong. The triune brain is wrong. System 1/2 is wrong. Cortical columns and uniformity of the neocortex are just imagined. The list is endless!

So as an outsider of neuroscience, you can't quote with confidence any generalization about the brain that comes from the field. At best we treat it as metaphors. All models in neuroscience appear to be wrong. What's worse is neuroscientists can't tell what's useful!

So we are stuck with metaphors. The brain is homeostatic. The brain learns through experience. The brain has a 'global workspace'. There are many rhythms in the brain. The brain works like 'a scientist'.

It's like there's a massive disconnect as to what has been discovered in neuroscience and what we can conclusively say about higher-level cognitive capabilities of the brain. It's like neuroscience isn't about cognition despite everyone pretending that it is.

The strangest thing is that many in the cognitive sciences will cite a conclusion from neuroscience to bolster their argument. Why do they do this? Is this just an appeal for our bias in reductionist ideas?

A pervasive example of this is the explanation of the harm created by social media algorithms. It's explained as exploiting the 'dopamine drip' that is intrinsic to brain function. You can't get more elementary than highlighting a function of a neurotransmitter.

So it's like I can argue that empathy is intrinsically fundamental to human cognition because... well there is a neurotransmitter (i.e. serotonin) that's function is to enhance empathy. A lot of people will buy into this reductionist explanation hook line and sinker!

Well that's the thing about intuitive explanations that are easy to understand. They are easy to understand because they appeal to cognitive biases. Humans are biased to seek out simple and reductionist explanations.

To appear to explain something complex in an intuitive way is to appeal to existing biases. So that's why explanations that cite the left-right brain divide, the triune brain, etc. Stuff that is essentially reductionist, has so much appeal.

This is the saddest thing because very few will accept your explanation if you begin from a concept that is counter-intuitive. For example, I have difficulty convincing people that all thinking is System 1 (i.e. Intuitive). Most people don't believe that this can be true!

GOFAI was so popular an agenda that it appeared intuitively true that human cognition derives from logical and symbolic processing. Centuries of Descartes' duality had conditioned the population to accept this.

But human general intelligence is difficult to formulate because it's built on top of many unintuitive ideas. Something that is intuitive is an old idea that we are immersed in.

The hive mind internet is now an example of an old idea that we've been immersed in for decades. The same can be said about user interfaces in computers and virtual worlds in games. I doubt you can explain this in an intuitive way to people who lived a century ago.

The human mind understands the world because it experiences the world. Language that conveys a world that we have yet to experience is understandable because we embed metaphors with a world that we are familiar with. Absent metaphors and we can't understand explanations.

So to understand the brain, we rely on metaphors and the metaphors that neuroscience has invented may just be good enough to be useful.

One characteristic of general intelligence is that we can hold in our heads multiple contradicting world views. General intelligence is the capability of selecting the useful one depending on context. We explain the world using metaphors that are useful in context.

@threadreaderapp unroll