Max Bennett: belief

Transcript context

…How do we actually know what the cognitive abilities were of early animals, and and why should we care? Great question. So I think there's 2 reasons why we should care about the evolution of our brains and intelligence. The first is to understand who we are. So the the scope of what it means to be a human is not constrained to what it means to be a Homo sapiens. So so much ink has been spilled on the last 70000 years of us being Homo sapiens. But if aliens were to come down and engage with us and analyze us as a species, most of the things they would observe about us don't come from our legacy as homo sapiens. They come from our legacy of being a primate and our legacy of being a mammal and our legacy of being a vertebrate and our legacy of being an animal in general. And so I think if we wanna understand what it means to be a human being, I I don't think we can skip the full 600000000 year story of how we came to be. And I think in there is so much rich history and insight about what it means to be us. So I think that's 1 really key reason. It's it's our it's our legacy. It's our history, how we came to be. The second perhaps more practical reason is I think understanding the evolution of the human brain and the evolution of human intelligence is a key tool in our toolbox to understanding how the brain works and how human intelligence works. It's by no means the only method. It might not even be the main method, but it's a very useful method to add to the toolbox. So the the problem with going into the human brain and trying to directly reverse engineer it is that evolution doesn't work in clean ways. It doesn't work the way human designer would. It doesn't work from first principles. It tinkers. And so when we go into the brain, we see all of this messiness. There's redundant systems. There's vestigial systems. Know, new things evolve that make old things redundant, but they're still there. Lots of processing is duplicated in different regions. And so 1 way to understand the brain is to continuously probe it as the human brain is, which is fine. But another method that's also useful and can impose constraints for us is to actually track the history of how it came to be. And that can provide insights as to when this brain modification occurred, such as when the neocortex evolved or when the basal ganglia evolved, what were the new abilities that this enabled? And how did it affect the prior brain regions that were already there? And so this can give us insight into how the brain works today. So I think in the toolbox that we have of ways to reverse engineer the brain, I think this is just an underappreciated 1 that is worthy of being included. And of course, understanding how the human brain works has so many different applications. Helps us with mental health. Helps us with understanding why people do what we do. It helps us with building AI systems. I think there's lots of insights to garner for the brain. So that's that's why to do it. with mental health. Helps us with understanding why people do what we do. It helps us with building AI systems. I think there's lots of insights to garner for the brain. So that's that's why to do it. Now how to reverse engineer what behavioral abilities existed in our ancestors is a really interesting question. Of course, we can't go back in time. So what we can do though is there are mechanisms to reverse engineer what their brains looked like, and there are mechanisms to reverse engineer what abilities they had. So what their brains look like, we can do just by looking at other animals in the animal kingdom. So for example, we can look at all of the existing primates and all of the existing nonprimite mammals, And we can see what are the common brain structures that exist between them. We can look at genetic analysis, meaning what things seem to derive from similar roots. And we can back into what do we what seems to be common and shared amongst them and thus what do we think was actually existing in the brains of the first mammals. We do the same thing with fish and reptiles to do that with early vertebrates and we can do that with invertebrates to try and infer what was existing in the first bilaterians. In other words, the first animal with brains. So we can compare different brains to try and back into what the brains looked like. For behavioral abilities, there are sort of 3 ways you do this, and you need this is like my, sort of approach to trying to infer behavioral abilities. I call them the in group condition, the out group condition, and the stem group condition. So in order to make the arguments that a behavioral ability emerged at a certain location in our evolutionary history, so for example, a behavioral ability like episodic memory evolving with the first mammals, you need to satisfy these 3 criteria. The in group conditions stipulates that most ancestors or sorry, most descendants of this species, in other words, most mammals, should show this ability. Doesn't mean all of them, abilities get lost all the time, but most of them should show this ability. And the neural mechanisms by which the ability emerges should come from homologous regions. What that means is a shared neural underpinning. So if, for example, mammals show episodic memory, but they come from neurological regions that independently evolved along different mammal lineages, and that suggests it wasn't present in the first mammals. But if they all emerge from regions that emerged with early mammals, that's good evidence that this ability also emerged mammals. The outgroup condition says most, doesn't have to be all, but at least many non mammal, so outgroup, non mammal vertebrates, so so the group right above should not show this ability. And if they do show the ability, it should emerge from nonhomologous regions. In other words, parts of their brain that evolved independently. So for example, birds definitely show episodic memory. But when we look into the brain regions from which episodic memory emerges, it's clearly non homologous. It's a part of the brain that mammals, that early vertebrates did not have.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.