When I told the mathematician Philip Welch that I was writing a book about what goes on in the mind of an expert in action, he looked at me a bit askance and asked, “wouldn’t an fMRI answer such questions about whether experts think?” This comment surprised me, not because of its content—it is a reasonable question to ask—but rather because of its timing: he was just sitting down to a dinner that my husband and I had cooked, and thus I would have expected him to merely nod his head in polite dinner-conversation manner and commence eating before the soup got cold. As he is usually is the picture of decorum, this issue must have really struck him. And since it has likely struck some of you as well, let me recount more or less what I said that night in response.
There is no doubt that we have learned a great deal about the mind of the expert in action from empirical research that has been carried out by neuroscientists and psychologists, as well as by physiatrists, psychiatrists, and sports medicine physicians. Such researchers conduct experiments in controlled settings, and controlled settings are important since (at least ideally) they enable us to manipulate one variable of a situation, such as an expert’s self-monitoring of her actions, to see how it affects another variable, such as performance success. However, the nature of expertise makes it extremely challenging to study in controlled laboratory settings.
One reason for this is simply that experts are typically not willing to spend an afternoon as test subjects; they have better things to do and have become experts precisely because they make good use of their time. In this respect, the scientific study of conscious experience, which is sometimes thought of as an empirically recalcitrant phenomenon, has an advantage over the scientific study of expertise. Consciousness, it is said, confounds scientific investigation since science can reveal only the neural or behavior correlates of consciousness but never consciousness itself. Whether or not this is true, however, it is a trivial matter to find conscious college students—the blood-alcohol level of the typical student notwithstanding—who are willing to participate in psychology experiments in exchange for a measly $10 or course credit; however, finding experts who are willing to participate in experiments is near impossible. If the question of how to understand consciousness scientifically is, as it is often called, “the hard problem,” studying expertise is the really hard problem.2
Getting experts into the lab, though near impossible, is not impossible if one has a large enough grant or, more likely, if one is in the right place at the right time. For example, the psychologist and Master-level chess player, Adrian de Groot, who has been referred to as one of the founders of cognitive science, carried out much of his groundbreaking work on chess while he was ship-bound along with other members of the Netherlands’ National Chess team on their way the 1939 Chess Olympiad in Buenos Aires. Lashed to the deck (or nearly so), the chess players were willing to co-operate with de Groot; typically, though, psychologists do not have at their disposal such means to induce participation (de Groot 1978).
On those rare occasions when experts do make it into the lab (as well as when you bring the lab to them, as de Groot did) a distinct obstacle researchers face when trying to understand the expert mind in action via controlled experiments, is that such experiments cannot capture the urgency of a Trauma Center, the excitement of opening night at the Paris Opera House, or the overwhelming aspiration, hunger, and yearning to beat your opponent at chess. Being asked to repeatedly dribble soccer balls through a slalom course is not going to elicit quite the same drive as being asked to play for your country at the World Cup. It is somewhat analogous to studying sleep deprivation in the lab: it’s so boring in the lab, participants can’t stay awake, yet on a mission a fighter-jet pilot will stay awake for days on end. One theme of this book, to put it baldly, is that controlled settings destroy expertise.
The psychologist Andres Ericsson suggests that since experts can perform in a wide range of conditions, they can perform their skills in the laboratory as well. And to a degree this is correct. Depending on their respective expertise, experts can, for example, putt a ball, play a sonata, or choose the next move in a chess game in a laboratory setting. But they will not be on their mettle; that is, they will not be roused to perform at their best. Of course, it might be possible for them to perform at their best in a laboratory setting—a million dollar reward might do the trick, for example—yet the typical incentives do not suffice to inspire excellence. For some investigations into expert performance, this might not matter. For example, if you are interested in where basketball players focus their eyes right before they shoot, perhaps whether they are highly motivated or not will not change this. However, if you are interested, as I am, in whether experts exert significant amounts of effort during a performance or a game, a lackadaisical approach to their tasks in the laboratory does matter.
Another problem with some experimental research into expertise is that such experiments often ask experts to engage in actions that are quite different from those that occur during an actual game or performance. For example, dribbling a soccer ball through a slalom course and then reporting, at the sound of randomly generated tones, which side of the foot had just touched the ball (Ford et. Al 2005, and Beilock et al. 2002), or, as is asked of baseball players in another study, performing a batting task without a ball and to say at the sound of randomly generated tones whether at that instant the bat is moving up or down, are things experts would typically not do, even if they are thinking about their own performance in other ways. Research that fails to capture real-life settings is sometimes criticized by psychologists as not being “ecologically valid.” What exactly ecological validity amounts to is an open question; however, experiments looking at performance that differs to such a large degree from that of the target of investigation should make us question whether the results can be generalized to apply to expert performance out of the lab. As we will see, researchers can sometimes trade in some element of control that exists in a laboratory setting for more ecological validity. And I rely on some of these studies to support my position that experts think about, monitor, and exert effort in performing their actions. Yet finding the right balance between control and ecological validity is a challenge.
Moreover, in direct response to Philip—who was naturally ready to pitch into the dinner at this point—quite apart from the difficulty of performing a brain scan on a high-diver mid-air, brain imaging technology is not always useful in the study of expert action since, at least as things stand now, we do not adequately understand what areas of the brain underlie thought, effort, and attention. As Yarrow and colleagues (2009) concluded in a recent review article on neurological research investigating elite athletes, “clearly expert and novice athletes use their brains differently, but precisely interpreting these differences in terms of their functional roles seems some way off at present” (p. 589). And so, as I discuss in later chapters, although there are a number of neurological studies that are suggestive, there are none that are conclusive, and even what exactly they are suggestive of is frequently open to debate.
Finally (and I noticed Philip reach for the spoon upon hearing this word), when we add to these challenges the fact that there is very little agreement among the expertise scientists about who exactly counts as an expert and what exactly expert action is, we are left with an area that is, indeed, difficult to study scientifically, as well as one that has at least the potential to benefit from philosophical reflection. Philosophy, no doubt, faces obstacles as well, rather large ones. And I assume that the final answer—or, if there is no final answer, then something closer to the final answer—to the question of what goes on in the mind of an expert in action will be revealed in scientific inquiry, for I am very much of the opinion, as the philosopher of science Bas van Fraassen (1996) once put it, that “there are no science stoppers” (p. 80). Accordingly, I believe that if controlled experiments tend to destroy expertise, what we need to do is not give up, but find a way to change the controls. Nonetheless, I think that at this point in the study of expertise it might be worthwhile to take a brief pause in the scientific investigation into the question of what goes on in the mind of the expert in action—if only long enough to read this book—and ponder it from, if not a strictly philosophical, then at least a philosopher’s point of view.