Botox, Heideggerian Robots, and Protestant Communion

An introduction to embodied cognition with thoughts on Christian prayer

Eva Gonzalès, Une loge aux Italiens, huile sur toile, vers 1874. It’s located at the Musée d’Orsay, and not to glaze myself, but I did admire it for nearly ten minutes.

The University of Edinburgh leads the world in the philosophy of “embodied cognition.” So, I got a hefty dose of it during my master’s degree. Simply put, embodied cognition refers to the thinking we do with our bodies. Sounds like an oxymoron? Let’s dive in.

(If this essay has typos, I apologize. I’m running on instant noodles, coffee, and two bowls of pipe tobacco—and it’s 1 AM.)

Botox and emotion reading

Take an especially cerebral task: Math. Counting “in your head” from one to one hundred is easy, but multiplying takes a little more brainpower. Say it takes 70 brain power points to multiply two large numbers in your head. (Brain power points is totally a real science word.) Now, imagine instead of purely counting in your head, you can mouth the numbers and use your fingers. This might shave off 15 brain power points. But if the task required 70 brain power points otherwise, then your physical activity contributed to the task of thinking. Therefore, it seems plausible to say your physical activity just is thinking.

This is particularly salient in processing emotions. One study analyzed participants before and after getting Botox, which freezes certain face muscles. “Results found that BTX treatment of laughter lines was associated with increased depression scores. Further, BTX treatment was associated with reduced emotion recognition ability and sexual function” (emphasis mine). In other words, recognizing emotions in other people is affected by one’s face being restricted in movement. To categorize someone else’s anger, we scrunch our own brow. Fascinating!

The computational theory of mind

This contrasts with a popular view in philosophy of the human mind called the computational theory of mind. This view considers the mind to work like a computer. The mind has “inputs,” processes these as “symbols,” and then spits out “outputs.” The inputs are sensory experiences, the outputs are our behavior. Our sense data gets packaged into a symbol, then the symbol is processed by a bunch of algorithm-like rules. Conscious experience is something like the software to the hardware of the brain; our five senses are like the keyboard being typed by the world.

Seems intuitive, right? The success of computers stems from this core principle: thinking involves something like mathematical operations, and grammar and language can be coded and universalized. The legendary computer scientist and mathematician, Alan Turing, theorized about “thinking machines” in his 1950 paper, “Computing machinery and intelligence.” This was published not in a math or scientific journal, but in the philosophy journal, Mind. This essay helped spawn the symbol-processing computer, or a “Turing computer.” Your iPhone, computer, Nintendo Switch, and calculator are all Turing machines.

The strong claim of computational theorist is something like this: in theory, a language/symbol processing machine could think just as well as, and similarly to, humans.

The upshots and predictions of embodied cognition

Problem. If the insights from embodied cognition are correct, then the same parts of our bodies that do the sensing or feeling also do the thinking. So the analogy to the computer seems flawed, insofar as it’s a strong, black-and-white claim with boundaries around the brain being the mind and our bodies serving as inputs/outputs.

Why does any of this matter, you ask? It’s a oft-repeated query to philosophers. I do it for the love of the game, but here are a couple of interesting predictions and theories put forward by the study of embodied cognition.

Boston Dynamics

Robotics vs. “good old fashion AI” (Hubert Dreyfus)

Embodied cognition implies that true intelligence, at least, human-like intelligence, requires embodiment. This leads to the further conclusion that AI will never succeed until it incorporates sensation and agency. The philosopher Hubert Dreyfus argues that the human mind is a dynamic relationship between our bodies and the world—a two-way street. So, a computer that purely processes the world in symbols will fail to have a behavioral feedback loop that helps it learn from experience. The word “dynamic” is a common phrase in embodied cognition. You might recognize it from Boston Dynamics, the robotics company making robots who can do back flips. Dreyfus was a proponent of research programs into “Heideggerian robots.”

On this view, LLM AIs, like Chat GPT, will reach a fundamental limit. Here’s the prediction: LLMs will hit a ceiling of intelligence without sensation, agency, and feelings. Without these elements, they’ll never really “get” context, or be anything like true human minds.

Humans as cyborgs; the extended mind (Andy Clark and David Chalmers)

Wait, you might say, what about when I use a pencil and paper? Or a calculator? Does that count as my cognition? David Chalmers and Andy Clark put forward this case in their 1998 paper, “The Extended Mind.” Whenever we utilize an external source for processing power, including other people, writing things down, typing on a calculator, or writing on a computer, all of these become a part of our mind. We cannot, they argue, draw clear boundaries between our minds and the world. In this way, Andy Clark argues, we are already naturally cyborgs (cf. Naturally-Born Cyborgs, 2003). With the rise in smartphones, which include memory storage, notes apps, calculators, reminders, and more, our minds become increasingly extended into technology. He seems to have a point there.

The hybrid way (having our cake and eating it too)

Of course, many philosophers acknowledge the way both views are helpful to understanding the mind, taking a “hybrid” approach. Some things, they admit, don’t require representation or computer-like processing in the brain. Embodied skillfulness, like sculpting and playing music, for example, does count as mindful activity without involving representations or computational thinking. But, they insist, much of the mind does work like a computer. This seems imminently reasonable to me.

For more on embodied cognition, I’ve included a further reading list of papers I enjoyed at the end of this essay. For now, let’s touch on its relationship to the broader field of phenomenology.

Perception and phenomenology

Phenomenology studies experience and perception. I’ve written an introduction to one of the most famous phenomenologists, Martin Heidegger. Phenomenology crucially “brackets” theoretical considerations, especially scientific ones, by setting them aside when studying experience. In other words, it examines how we experience color—forget what science says about light wave lengths and retinas. That’s the idea, anyway.

Phenomenology isn’t anti-science; it runs parallel to it. That said, some take phenomenology to be prior to, and therefore preeminent to, the abstractions of science. Therefore, if someone argues from empirical science that conscious experience is illusory (cf. the late Daniel Dennett, fourth horseman of the atheist apocalypse, and the married neuroscientist/philosopher power couple, Patricia and Paul Churchland), we should reject their arguments on the more foundational evidence of phenomenology. This is because science couldn’t exist without conscious experience.

Édouard Manet, “The Spanish Singer,” 1860.

Here’s the kind of reasoning phenomenologists do. The process of experiencing a cup is not like adding “container + handle = new item, cup;” our experience is something more like, “thing for which drinking from is appropriate.” The gives itself as a drinking thing. Our existence as drinking beings, beings for whom drinking is important, is intimately involved in our perception of the world. Cups particularly. Indeed, we cannot imagine the world as anything but one where we care—where we engage in practical, lived activities. However, if we wanted to catch a spider, the “cup” might offer itself as a “spider catcher” to a brave person. Phenomenology generally connects well to embodied cognition. They are bedfellows. Embodied cognitivists often draw inspiration from Husserl, Heidegger, and Merleau-Ponty.

As you might guess from this cursory introduction, phenomenology doesn’t necessarily force us into relativism or subjectivism, but it highly qualifies rationalism. Which brings me to a few thoughts on prayer.

Prayer and embodiment

Protestants, like myself, often suffer from bad cases of rationalism. The reformed churchmember looks down his common-grace-given nose at his liturgical brothers and sisters for their reliance on the physical, embodied participation in the sacred. We begrudgingly take communion. We get by on technicalities in this regard; cheap grape juice and tiny wadded-up pieces of paper peppered with flour suffices as the bread and wine.

I thrive, says the Protestant, on theological truths, black and whites, alone. Forget sola scriptura, try sola theologiae on for size. Limited Atonement is a logical, rational conclusion of Calvin’s theology—let’s adopt it despite it being plainly unbiblical. (See my Dad’s recent article for some insight here. Also, I’m a Protestant—so please no one take offense. I’m just poking fun in good spirits. Here’s the comment button in case you want to overreact anyway—it’s good for the algorithm.)

“I can pray wherever, in whatever stance,” says the Protestant. “I can pray in the car, sipping a cup of coffee. I can pray while falling asleep. I can pray anytime.” Yes we can, and praise God. We can go boldly before his throne, like a young son or daughter. But might we miss something from the liturgical brothers and sisters? Do we miss something by not incarnating the sacred? Are we missing out on incense? Prayer on our knees? The procession of the golden cross? The uniforms of our shepherds? Has the pendulum swung too far? Do we miss something without an embodied state to our prayers?

The embodiment of companionship: two chairs

If you have trouble focusing on your meditation on scripture and prayer, consider speaking out loud. Consider bowing, or dropping to your knees. I’ve been throwing plenty of big words your way. So let’s consider an accessible tool. Bob Beaudine wrote a tiny book called 2 Chairs. For its size and simplicity (it scores about a fifth-grade reading level throughout), it’s remarkably deep. It’s deep not in its analysis or theory, but in application. As someone who can overthink things (surprising, I know), my ritual of 2 chairs has been immensely gratifying and sanctifying.

The practice is simple. Take two chairs, set them in front of each other in a quiet place and time, and sit in one. Imagine, sitting in the other chair, God. Speak out loud. Your perception, if you’re a Christian, will quickly change. You’ll realize God had called you to the meeting, not vice versa.

Let’s be wary of the “complexity bias,” the idea that complex solutions appear better than simple ones. If you struggle to connect to God, especially relationally, as a friend, pull up a chair.

And maybe we should do something about our crappy communion bread.

Soli Deo Glora

A few papers on embodied cognition:

Anderson, Michael L. “Embodied Cognition: A Field Guide.” Artificial Intelligence 149, no. 1 (2003): 91–130. https://doi.org/10.1016/S0004-3702(03)00054-7.

Foglia, Lucia, and Robert A. Wilson. “Embodied Cognition.” WIREs Cognitive Science 4, no. 3 (2013): 319–25. https://doi.org/10.1002/wcs.1226

Chemero, Anthony. “Radical Embodied Cognitive Science.” Review of General Psychology 17, no. 2 (2013): 145–50. https://doi.org/10.1037/a0032923.

Garzón, Francisco Calvo. “Towards a General Theory of Antirepresentationalism.” The British Journal for the Philosophy of Science 59, no. 3 (2008): 259–92.

Leitan, Nuwan D., and Lucian Chaffey. “Embodied Cognition and Its Applications: A Brief Review.” Sensoria: A Journal of Mind, Brain & Culture (Australia) 10, no. 1 (2014): 3–10. https://doi.org/10.7790/sa.v10i1.384.

Turing, A.M. (1950). Computing machinery and intelligence. Mind, 59, 433-460.

Van Gelder, Tim. “What Might Cognition Be, If Not Computation?” The Journal of Philosophy 92, no. 7 (1995): 345–81. https://doi.org/10.2307/2941061.