Jason Wells

How cool is this. MIT Open Courseware offers a class called Gödel, Escher, Bach: A Mental Space Odyssey. It is an entire course geared around Douglas Hofstadter’s book Gödel, Escher, Bach.

This is one of the greatest books I’ve ever read; it’s a true masterpiece. I first encountered it in high school. Ostensibly it’s a gorgeous web interconnecting art, mathematics and music, but that’s actually instrumental to the book’s true purpose. Quoth Wikipedia:

Hofstadter has emphasized that GEB is not about mathematics, art, and music but rather about how cognition and thinking emerge from well-hidden neurological mechanisms.

And now MIT offers a free class geared around this book, using it as a kind of textbook, bracketed with a syllabus and class notes in PDF. All this for free. Wow. Props to the instructors, Justin Curry and Curran Kelleher, and to MIT for providing this to the world gratis.

As I embark on intellectual adventures in cognitive science, as I intend to, I begin with the basics. The first question I tackled, and one I’m often asked, is: “What is cognitive science?” It doesn’t get much more basic than that. Even so, there is enough confusion about this to make it a very good question for anyone to ask.

I'm so confused.

Cognitive science is, of course, the science of cognition. Cognition is just another word for thinking–one that carries specific implications about how thinking happens. At this point it’s evident that we think with our brains, and presumably cognitive science is the study of this. So it purports to be, too, although there’s a whole lot more to it. You could see it as the study of mind and mental processes, in other words, of the mind and its operations; and since the mind is what the brain does, it’s not unreasonable to think that cognitive science would primarily study neural function.

Here’s where things get a little whack. There’s already another field for studying brains: neuroscience. Neuroscience is not cognitive science–although it is sometimes considered one of the “cognitive sciences.” In fact, not all neuroscientists seem to be even interested in cognitive science. So I wondered: how do they relate? For example, what is a question one discipline would address that the other wouldn’t?

I’ve asked several neuroscientists and a neurophilosopher this question directly; none could say! I even raised the question on Quora. No satisfactory answers there either. I began to wonder: given the billions spent on research into the brain, the mind and psychology every year, how could this be?

Then it gets more fun. Turns out there’s something else called cognitive neuroscience. When I discovered this, my initial reaction was, are you kidding me? But cognitive science is not cognitive neuroscience, despite the two having almost identical names. The latter is a subfield of neuroscience; the former isn’t. Yet the two definitely relate. For example, the Cognitive Neuroscience and Neuropsychology Lab at UCSD is part of the UCSD Cognitive Science department (which is considering my Ph.D. application right now).

And then it gets even better. There’s something else called computational neuroscience. Since cognitive science draws deeply from computer science (it had a major artificial intelligence research program for decades), you might at first think this is simply a synonym for cognitive science. But it’s not. It is the study of the brain as an information processor. It has multiple touchpoints to cognitive science, but again, one goes to places the other doesn’t, and vice versa.

So, what the fuck is going on? I’ve spent the past few months trying to clear this up for myself. Here’s what I’ve got so far.

Neuroscience is a life science. It’s about the nervous system, which includes the brain. It’s not really about thinking at all, although obviously it has huge relevance. A neuroscientist works with neurons, perhaps seeing how they can be affected by chemicals or sensing their electrical activity to get a grasp of neural functions. A neuroscientist might study what the brain does when a person thinks about adding two numbers, but not how a computer does the same thing, even though (I would argue) both are comparable forms of cognition, despite having very different realizations. And I doubt a neuroscientist would investigate anything cognitive that can’t be reduced to a neurological explanation. For example, the thing we call friendship is something mental that is maintained by neural functionality, but you can’t explain friendship in those terms for an individual brain. Friendship is social; it emerges from the interaction of at least two brains, and each may have very different things going on. Friendship can’t appear on an fMRI scan.

Here’s another clue. A cognitive scientist might very well study psychological and social relationships between different people without looking into neurology at all. I don’t believe a neuroscientist would do something like this unless it were to inform a specific inquiry into brain function. A cognitive scientist may even study how people and computers orchestrate themselves cognitively; this is in fact a topic of cognitive science known as distributed cognition.

So as best I can make out, cognitive science isn’t strictly reducible to neuroscience for two big reasons: one, it takes on phenomena that emerges beyond the bounds of the neurology of an individual brain; and two, it can encompass cognitive phenomena exhibited in intelligent machines, i.e. computers, which don’t have neurons.

At this point it seems like the best way to make sense of all this is as follows:

• Cognitive science is the study of the phenomena of cognition in the world, natural or artificial, leveraging insights from the following three fields and others, such as computer science, psychology, linguistics and even philosophy;
• Neuroscience is the study of the brain and nervous system, not artificial systems and not necessarily even cognition;
• Cognitive neuroscience is the study of the phenomena of cognition in the brain, that is, the biological underpinnings of thinking; and
• Computational neuroscience is the study of the brain and nervous system’s information processing abilities, cognitive or otherwise.

So there it you have it. The fields all interact in different ways and overlap considerably, and I didn’t even address other closely related fields like cognitive psychology or philosophy of mind. I take this complicated muddle to be the product of the deep confusion that exists about the mind today. All the same, the progress in this space is beyond warp speed, which creates new chaos and confusion at least as fast as discoveries and advances clear things up. In short, the “cognitive sciences” are in a state of total revolution… and revolutions get messy. There’s so much creativity in this space that these disciplines, with their distinct intellectual interests and traditions, either blend, separate or directly contradict one another in countless ways that constantly change.

I really like revolution. I think it’s a very exciting time to study cognition. There is clearly much to be done.

I don’t quite remember when I acquired the ambition of seeing Glacier National Park, but it certainly was years ago. There were plenty of reasons to go. It was reputed to be a vast and astonishingly beautiful wilderness. It was far away–in Montana, bordering Canada–yet still within range of a road trip in the van. We would be able to camp out of the van there. The hiking and photography were likely to be exceptional. But most of all, for me, it was about the glaciers themselves. I had never seen a glacier other than from a distance of miles or in a photograph, and here, I would be able to touch them. And these glaciers are special.

In 1850, Glacier National Park had 150 glaciers; today it has but 35. They are melting away, and not gradually. Climate modeling suggests that they will all be gone by 2020–just nine years from now. When they are gone, the entire local ecosystem will change, becoming more arid. This damnable fact hangs over Glacier like tragic inevitability. Glacier’s fate is sealed; it is doomed. I couldn’t do anything about it, but at least Heather and I could see how it is now, before it is gone forever.

A taste of the beauty to come. (Click to enlarge.)

So I was motivated to get out there sooner rather than later. As it turned out, this year was good timing. From what I’d been reading, it looked like early September would be the ideal time to make the trip; after the teeming hordes left, but before it got too cold. Heather and I sometimes make a pilgrimage out to Burning Man in that very timeframe, but we’d just been there last year. I was happy to do this instead.

We began to make serious preparations in late August. I’d resigned from Semantic, which jarred all of our future plans, but despite that, we decided to stick to this one. I’m applying to the UCSD Cognitive Science Ph.D. program, and taking the trip meant splitting my GRE preparations awkwardly in half, but I preferred that to waiting another year. Anyway I was stressed out about the test, so getting a break from it would be good.

San Diego to Glacier: 1416 miles, border to border.

We planned a two-week trip, nominally taking three days to drive there, three days back and spending the rest of the time in the park proper. The route I planned essentially followed the 15 from end to end, taking us through six states: California, Nevada, Arizona, Utah, Idaho and Montana. To that we would later add a brief day trip to the Canadian province of Alberta.

September 5th. The van was loaded; we pulled out and were off by mid-morning. This was later than I’d originally hoped, and now reaching Salt Lake by the end of the day looked unrealistic. But I’d make up as much time as possible.

As we took the 15 across Southern California, we hit the odd little rainstorm now and again. We also saw a lot of cops, presumably on the prowl for ebullient LA partiers heading to Vegas for Labor Day weekend. The timing was such that we’d just miss seeing playa-dust covered vehicles going the other way, returning from the playa. Heather noticed a cool teardrop trailer on a truck we passed, and that got us daydreaming about getting one to go with the van. They’re pretty neat. As we neared Primm, I got my first encounter with $5 gas. Not so neat.

Passing through Las Vegas was uneventful, we didn’t even stop. I actually didn’t notice passing through the tiny chunk of Arizona sliced by the 15. Soon enough we were in Utah. As we intersected a rainstorm and sunset, a double rainbow emerged, only the second I’ve seen.

I made very good time, driving 11 hours and ranging between 80-90mph much of the way. The van is seemingly invisible to the highway patrol. We made it to Provo, making up a lot of time we lost on the late departure. We crashed at a Spring Hill Suites, a Standard-like hotel–quite a thing to find in Provo, but Heather is resourceful.

ADDITIONAL. For your convenience, here are links to the rest of the seven-part Glacier National Park travel journal:

• Sept. 6-7th: Provo to the Park, West Entrance.
• Sept. 8th: Avalanche Lake, Lake McDonald Lodge.
• Sept. 9th: A Day on Going-to-the-Sun Road.
• Sept. 10th: The Hike to Grinnell Glacier, Many Glacier Hotel.
• Sept. 11th: Waterton Lakes National Park in the Red Jammer.
• Sept. 12-16th: Departure, Bigfork, Salt Lake and the Return Home.

Maybe it was something about the cooler, wetter autumnal weather of late. Maybe it’s the flexibility my schedule now has, being unemployed. Or maybe it was the intermittent encouragement/prodding I’ve been getting from Pilot Mike… whatever the reason(s), I got out my old mountain bike this morning and took it for a little spin.

It has been hung on the back wall of the garage, behind a mountain of other outdoor gear, for well over a year, unused and neglected. It’s my old Cannondale, a very light bike for its size, easily carried in one arm. Its responsiveness, durability and elegant design make it the best bike I’ve ever had. And I do love mountain biking. Why, I’ve been asking myself more and more often, is it just hanging on the wall?

So at last I took it down off its mounting hooks. It was covered in dust, both mundane and playa kinds. I wiped it down, peeling off the electrical tape I had used to cover its logos. This was to discourage theft on the playa when I took it out there in 2003. (Now I have a much more suitable industrial trike for that.) As expected, the inner tubes were completely flat. I gently reinflated them to 50 PSI, giving each time to adjust to the pressure and temperature differentials. Old inner tubes, even new ones can be touchy. I flipped it over in the grass to spin the wheels, seeing whether it needed lube. Surprisingly, it really didn’t.

So I took it out for a little spin, just through the neighborhood. Street by street I quasi-orbited the house, unwilling to go far in case an inner tube leaked or exploded. And the ride was fine; the bike is in good shape overall, I’m happy to say. The rear derailleur is stuck, but that’s not a big deal. The inner tubes continue to endure, despite their great age.

And I remembered the feel of cool air on my face, the joy of powering up a hill and leaning into a sharp turn. Having a chance to look at what was there, rather than having it fly by in an undifferentiated blue. The wholesome feeling of using my own muscles for propulsion, rather than a two-ton gasoline consuming box. It was all coming back to me, and it felt good.

So I’m getting back into it. Yes, there will indeed be bike ass, but it will pass. A bit later, I’ll take it out to Mission Trails and have a go in the dirt and rocks on a rough single track–where it belongs. Where I feel the most connected to the experience of the ride and the beauty of nature. Then I’ll take it to more ambitious places. I’ve daydreamed about Moab for years. And perhaps I’ll also ride it to and from UCSD, should I be accepted into the Cognitive Science Ph.D. program as I hope to. It’s only five or six miles from home. Anyway, it feels very good to bring this back into my life. It’s a return to something I’ve missed, a part of my life I let fade away for no good reason.

Side note. Some of you know I went backpacking through Australia for a month back in 1999. While there I took many photos (in 35mm! ah, the old days) and wrote a travel journal. After I got back, I made it into a section of an old web site I used to maintain. Since I left Semantic, I took it as a side project to migrate it to this blog. The entire journal is now available, and if you’d like to read about my adventures, start here. Each following page is the link at the very bottom left. (Note: I wrote it 11 years ago. Some of it makes me feel silly to reread, but I want to preserve it as is. Just keep that in mind.)

Kinda wish I’d included this in my last post… an article from Discovery News, “The Web is Your New Memory.” FTA:

The study authors tested this hypothesis by giving subjects a list of facts they were unlikely to know, such as “The space shuttle Columbia disintegrated during reentry over Texas in February 2003,” then had them type them into a computer. As the subjects typed in a fact, they would receive a message on the computer screen telling them that the item had been saved in one of five folders, each with a different name. When quizzed afterward, most subjects were unable to remember the details of the facts, but about 30 percent recalled the name of the folder where the information was stored.

Seems that the Internet changes the way we remember facts. Rather than just remembering a fact, we remember how to get to the fact.

The study authors propose that the Internet now fills this role, becoming the ultimate know-it-all friend whom we can access anytime.

That’s one way to put it, which keeps the phenomena comfortably external. But the truth is that it performs a cognitive function in place of your brain, not in place of your friend. It may not be internal to your body, but it’s internal to you.