equates the mind with the brain and puts cognition—every last bit of it—in the head. Clark calls this the brain-bound model. If this model is correct, it means that the body exists exclusively for sensory perception (i.e., information inputs), with motor movements (i.e., action outputs) playing no role in cognition itself. It also means that cognition depends entirely on neural activity. If that’s the case, we should, in theory, be able to pop the top off a human skull, scoop out the brain, dump it into a cognition tank hooked up to ocular and auditory processors and say, “Voila, there it is, a thinking, intelligent, conscious being.”18
When you read an article about the mind that features an image of the brain, you’re probably dealing with the brainbound model. It is the culturally dominant conception of the mind, one exemplified by the computer scientist and futurist Ray Kurzweil. Kurzweil envisioned a day when our brains could be uploaded to the cloud, and we would exist as conscious creatures made of pure information.19 While this view has many proponents, it’s worth noting that even neuroscience has been coming around to the idea that what the brain does is intimately connected to the messy reality of our biology. The neuroscientist Alan Jasanoff summed up the “fundamental lesson of neuroscience” this way: “The brain cannot be all there is.”20 The more science learns, the more questions arise about the brainbound model.
Because of this, Clark proposed an alternative model called the extended mind.21 In this model, neural processes don’t handle each and every cognitive task. Some will happen in the head, while others might happen in the world. The basis for this idea is that we evolved in a physical world, which also means we evolved cognitive tooling that relies on our brains and our bodies and anything in the world. Sometimes that means thinking happens entirely in the head with mental representations. Other times it means the cognitive act depends on information outside the head—external representations—and interactions with other worldly resources. When the mind is extended, to use Clark’s vivid wordage, “Cognition leaks out into the world.”22
We can draw a line between these two models (see Figure 2.3). On one side is brainbound, with mind and brain co-located in the skull and the body serving only as input and output. On the other side is extended, with cognition spread across brain, body, and anything in the world: whiteboards, smartphones, sticky notes, maps, notebooks, and even other people. The extended model doesn’t dismiss what the brain does. But neither is it biased toward electrical signals whizzing through squishy gray tissue.
FIGURE 2.3 Contrasting models of human cognition. Brainbound locates all of cognition in the brain. Extended spreads it across brain, body, and the world.
A key difference between these two models is how much they depend on neurons. Brainbound assumes neuronal hegemony: thinking is restricted to what neurons do. Extended argues for neural frugality: thinking can happen with neurons, but it doesn’t have to. Sometimes our thinking happens outside the head because it can be faster, or easier, or just plain better to do the work out there.23 This isn’t the brain being lazy. Instead, it’s more like the busy executive who effectively delegates certain tasks to the people around her rather than doing all the work herself.
Mental Representations and the Big Divide
The question before us then is twofold. First, how does this most exciting idea change our view of how the mind works? And second, what does it mean for how we create understanding from information?
We’ve long understood that Starbucks and Calvin Klein use marketing to influence what we think. But it is surprising, and intriguing when science finds that a cappuccino or a lab coat can influence how we think. The temperature of our morning coffee can make the grumpy bus driver seem friendlier? Wearing a long white jacket can help us focus on a task? It sounds a bit crazy.24 In one sense, it’s not that controversial: of course, the outside world influences what happens in our head. Imagine taking a calculus test, not in a classroom, but outside, on a glorious day, at the beach. Who wouldn’t find it hard to concentrate when you could be lounging or surfing or reading a great book. But embodiment is making a much deeper claim: our ability to think depends on the world outside our head. We can never fully escape how the way we interact with our environment shapes our cognitive powers.
Embodiment is far from settled science. Just how far to take this idea is a matter of much debate and even more research. It is helpful to think of embodiment, not as a singular theory, but as an umbrella term that includes many different challenges to the computational theory of mind. You will find Clark’s extended mind under this umbrella, along with ecological psychology, distributed cognition, enactivism, activity theory, and something called radical embodied cognition, to name just a few. Although they each have various intellectual roots, and the distinctions between them are often fuzzy, they all share a belief that no robust theory of mind can come from studying neurons alone.
There is one topic, however, where there is a notable disagreement. Do mental representations exist or not? This is the big dividing line. Some groups under the embodiment umbrella, most notably radical embodied cognition, posit that we don’t need mental representations. To those who stand under this part of the embodiment umbrella, symbols in the brain are as mythical as unicorns. Since even the most advanced brain scanning technology has failed to observe even a single mental representation, this idea has an obvious allure. Experiments with creatures ranging from robots and crickets to children and baseball players have provided intriguing evidence for this position.25 Even so, explaining the mind without mental representations is so difficult that Anthony Chemero opened his book Radical Embodied Cognitive Science with the following qualifier: “One of the things I try to make clear is that it is actually very difficult to reject internal representations, and that radical embodied cognitive science must be more radical than most of its proponents realize.”26
The debate over mental representations means we can, and should, expand our brainbound vs. extended framework. Our revised diagram, shown in Figure 2.4, puts brainbound at one end of a spectrum (cognition is all about symbols in the head), radical at the other (cognition involves no mental representations whatsoever), and extended somewhere in the middle. Exactly where in the middle is not easy to say. It depends on where one stands under the embodiment umbrella. That middle area covers a lot of different ideas. The later chapters in this book, for example, often draw from distributed cognition, but for our purposes, it’s sufficient to overlook these differences since this is the region that accepts mental representations as part of an embodied view of the mind.
FIGURE 2.4 A spectrum of perspectives on the nature of human cognition.
What Is the Truth About Embodiment?
What is the truth about embodiment? Don’t ask the scientists because nobody knows. Not really. This is how science works. Hypotheses are put forth, experiments are designed, evidence is collected, debates rage, and scientific knowledge slowly moves forward. For decades, the evidence has been building in favor of an embodied view of mind. But here’s the thing: the rock-bottom truth doesn’t matter. Not for our purposes, at least.
The embodiment debate stems from a simple yet eternal question: How does the mind really work? Yet where scientists are looking for truth, the rest of us seek something else. All we require is a perspective on cognition that allows us to make substantial headway on a wide range of messy, complex,
