Michael Shermer's latest book, Truth, is pretty good. If you missed it, my overall report on it is here. I found his chapter on "free will" especially interesting.
Shermer is a skeptic, and he's no fan of mind-body dualism, the idea that there's a little "mind" agent inside your body somewhere who makes decisions, causing the rest of your "body" to follow. That is, indeed, a supernatural concept.
Anyway: Shermer makes a big deal out of experiments that look at brain activity when certain kinds of decisions are made. His chapter begins (footnotes elided, emphasis, exclamation points, and sneer quotes in the original):
In 1985, the physiologist Benjamin Libet conducted a series of experiments that involved taking EEG readings of subjects' brains engaged in a task that required them to press a button at random intervals whenever they felt like it during the session. The results were revealing: Several seconds before the "decision" was consciously made by the subject, the brain's motor cortex was activated. Libet's research has held up well during the replication crisis in psychology. The neuroscientist John-Dylan Haynes, for example, employed fMRI brain scans in a 2011 study in which subjects situated inside the scanner observed a series of random letters and were instructed to press one of two buttons whenever they wanted. Participants were then told to verbally report which letter was on the screen when they "decided" to press the button. The results were equally striking: The time between brain activation and conscious awareness of a "choice" was several seconds and even a full seven seconds in some cases.
In these studies and others, scientists measuring subjects' brains knew which decision they would make before the subjects themselves knew it! For example, in a 2011 study, Itshak Fried and his UCLA colleagues recorded activity in the brains of subjects who were instructed to command their fingers to move. Astonishingly, the neuroscientists could detect neural activity related to the finger movement a full 15 seconds before subjects reported making the decision to move their finger! Narrowing down their search, they found activity in a tiny clump of 256 neurons in the medial frontal cortex that enabled them to predict with 80% accuracy which choice a subject would make a full seven seconds before the subjects themselves knew.
If these results don't sound spooky, then you're not thinking hard enough about them.…
To be fair, Shermer goes on in his chapter to criticize the conclusions anti-free willers like Robert Sapolsky and Sam Harris draw from such experiments. But I have further major quibbles! (Is "major quibbles" an oxymoron? Oh well.) Taking them one at a time:
I'm unsure Shermer is accurately reporting the Fried experiment. No biggie, but a 2011 Scientific American article (How Free Is Your Will?) reports that Fried's research found those neurons activating "as long as a second and a half before the decision, and as early as seven tenths of a second before it." That's pretty far off from Shermer's claim of "a full 15 seconds" (or, later, "a full seven seconds").
And, um, 80% accuracy? If you are a fan of Sapolsky/Harris determinism, that has to be disappointing. At best, that's a good bet, but not something to wager your life savings on.
There are recent claims that fMRI brain scans are routinely misinterpreted. Sabine Hossenfelder has a brief video about it:
I have no idea, however, whether this applies to the results reported by John-Dylan Haynes.
So what? If you read about brain stuff even a bit, you know that it's easy, and far from unusual, for humans to misperceive their own mental processes, even those processes that are outside the realm of free will. It's interesting that people might not get the timing of their finger-moving decisions exactly right. But does that somehow imply that people aren't making those decisions? I don't see how it does.
Baseball. I happened to be reading Shermer's book while concurrently watching the Boston Red Sox beat the Arizona Diamondbacks at Fenway on WSBK. And it occurred to me that what I was watching on the TV kind of demonstrated that the finger-moving experiments that so impressed Shermer don't really tell a complete story, and the application to the free will debate is problematic at best.
According to Google's AI: Major League fastballs average around 97 mph between the pitching mound and the strike zone, taking around 400 milliseconds for the trip. So in that 400 milliseconds: (1) the batter's optic nerves send what they detect about the ball's trajectory to the brain; (2) the brain "decides" whether to swing the bat, take the pitch, or sometimes (aieee! duck!) get out of harm's way; (3) if the brain decides to swing at the pitch, that command has to go down the nervous system to appropriate sets of muscles, sequenced in just the right order; and (4) those muscles must swing that two-pound bat into the desired path, with the desired velocity.
And (to repeat) all that has to happen in approximately 400 milliseconds. step (2) is probably the least time-consuming.
And so the typical batter's swing/take/duck decision takes far less time than what the brain-scanners are reporting in their finger-moving experiments. They may be right for the kind of the decisions their subjects are making. But that's not the whole story, as demonstrated hundreds of times every summer night in ballparks across America.
And speaking of robots… Oh, wait, I wasn't speaking of robots. But here you go: Could some smart folks build a batting robot with a couple video cameras, mechanical limbs, and a decently-trained AI?
Almost certainly. They have (specifically Sony has) already done it for ping-pong, as reported last year by Reuters: Ping-pong robot Ace makes history by beating top-level human players.
An autonomous robot ping-pong player dubbed Ace has achieved a milestone for AI and robotics in Tokyo by competing against and sometimes defeating top-level human players at table tennis, a feat that could presage an array of other applications for similarly adept robots.
There's video too:
There are also "baseball-playing" robots (Baseball-playing robot swings, hits and catches with pinpoint accuracy) but (reading the article) they aren't ready for the Big Show quite yet.
I'm not sure whether this is a point on the determinists' side, or the free will side.
So I'm not sure it matters. A brain's decision to swing at a fastball (or a ping-pong ball) need look nothing like that same brain's decision to go to college, solve a crossword puzzle, get married, or pick out a different color shirt from the L.L. Bean catalog.
The late psychologist Daniel Kahneman got semi-famous for writing a book Thinking, Fast and Slow which popularized dual process theory: two different modes of decision-making (usually called, unimaginatively, "System 1" and "System 2".) The important thing for the "free will" discussion: the brain "decides" things differently depending on circumstances. Specifically, the finger-moving experiments seem heavily "System 1", and they tell us next to nothing about the free will involved in "System 2"-style decision-making.
Disclaimer: I am not even a dilettante when it comes to psychology, neuroscience and sports physiology, so you might want to ignore everything I've said here.
![[The Blogger]](/ps/images/barred.jpg)
