The nervous system is a product of evolution, a long series of modifications produced by the success or failure of action choices, so brains must be active not passive to evolve, and the same is true for learning. For example in one study, two kittens who couldn’t yet perceive depth were strapped to a carousel, but only one could walk to turn it (Figure 6.20) (Held & Hein, 1963). Both kittens saw the same input, but only the one that actively walked learned to see depth, because only it could relate its body position to what it saw. Learning in general is based on experience not input, so it requires action choices just as evolution does.
Autonomy, the ability of a system to make its own decisions, is then a brain feature because both evolution and learning require it. The brain of every creature is in a basic feedback loop with its environment (Figure 6.21), so it must initiate this loop to evolve or learn. Conversely, brain parts that aren’t active, like the passive kitten in the carousel, neither learn nor evolve. The theory that the brain is a machine that responds only to stimuli then doesn’t work because it contradicts evolution but last century, psychology split into two camps on the issue of what controls the human feedback loop:
a. Behaviorism argued that people are machines driven by outside events so stimuli define the loop entirely, including how our brains respond.
b. Constructivism argued that the brain directs the loop by actively constructing reality, so it can produce more sentences than could be learned from stimulus-response associations (Chomsky, 2006), and most psychologists agreed.
They differed on what initiates the basic feedback loop, as behaviorism was input-driven but constructivism was brain-driven (Figure 6.21). A circular process can be initiated at any point, so the brain can be driven by external events, but brains must control the feedback loop to learn or evolve. Behaviorism reinvents Newton’s mechanistic universe, as did Dawkin’s selfish gene (5.7.2) and Crick’s neuron pack (6.1.1), but biological evolution and psychological learning both require an autonomy of free action that machines simply don’t have.
Yet if the brain is a decentralized hierarchy of autonomous units, how does evolution update it when better control systems become available? For example, when Microsoft upgraded its keyboard DOS operating system to clickable Windows, it just replaced it. The result was easier to use and more powerful, but suddenly all the knowledge that experts had taken decades to learn was largely obsolete. If nature worked like this, mammal brain systems would make their reptilian version obsolete, to lose not decades but hundreds of millions of years of evolutionary experience!
Evolution doesn’t discard past advances but builds on them, so an American mental health physician proposed Triune theory, that our brain is a reptile brain overlaid by a mammal brain overlaid by a human brain (MacLean, 1990) (Figure 6.22). The reptile brain was the hind-brain cerebellum, the mammal brain was the mid-brain limbic system, and the human brain was the neocortex. Our brain is then three brains in one because it evolved from a reptile brain that handled movement and instincts, to a mammal brain that added emotions, then human brain that allowed language and thought.
This model describes autism as an out-of-control reptile brain and chronic anxiety as an out-of-control mammal brain, so neither necessarily imply a lack of intellect. For example, Temple Grandin‘s autism not only didn’t stop her getting a PhD but also made her an expert on animal behavior because she could experience what they do. She explains how the triune brain explains animals as follows:
“To understand why animals seem so different from normal human beings, yet so familiar at the same time, you need to know that the human brain is really three different brains, each one built on top of the previous at three different times in evolutionary history. And here’s the really interesting part: each one of those brains has its own kind of intelligence, its own sense of time and space, its own memory, and its own subjectivity. It’s almost as if we have three different identities inside our heads, not just one.” (Johnson & Grandin, 2006).
Yet as the triune model gained popularity, biologists rejected it because evolution doesn’t create layers as geology deposits strata, so as one critic put it: Your brain isn’t an onion with a tiny reptile inside. Evolution didn’t build a reptile brain, then a mammal brain, then a human cortex because it isn’t a linear production line, nor does it add without precedent, as bat wings are modified forelimbs that existed before. Hence even reptiles care for their young and have a primitive cortex that can solve problems (Patton, 2008).
Triune theory also didn’t explain birds, who evolved from reptiles as mammals did. Birds are the feathered descendants of dinosaurs who survived their extinction, so if evolution built a human brain over a mammal brain, what are birds? They aren’t mammals yet are warm-blooded and remarkably smart, even compared to mammals (Jarvis & et al., 2005). For example, crows can bend a wire into a hook to get food their beak can’t reach (Weir et al., 2002) (Figure 6.23), which children can’t do until about eight and even then, only half succeed (Cutting et al., 2014). Birds then are more like feathered apes than reptiles and urban crows are especially smart:
“On a university campus in Japan, crows and humans line up patiently, waiting for the traffic to halt. When the lights change, the birds hop in front of the cars and place walnuts, which they picked from the adjoining trees, on the road. After the lights turn green again, the birds fly away and vehicles drive over the nuts, cracking them open. The birds wait patiently with human pedestrians for a red light before retrieving their prize. If the cars miss the nuts, the birds sometimes hop back and put them somewhere else on the road.” (Earthfire Institute).
Biologists conclude that birds evolved from reptiles as mammals did, but by a different path as their cortex is smooth not folded, then converged to achieve the same cognitive benefits (Lefebvre et al., 2004). Yet if brains don’t update by layering, how do evolutionary advances replace their predecessors? The answer it seems is that instead of replacing the old with the new, as we do with computers, it just shifted the feedback loop locus of control based on autonomy.