Dan Peterson, probably my favourite blogger on sports science, has a recent piece in Science Daily on How Baseball Players Catch Fly Balls. He usually posts on his excellent blog, Sports Are 80 Percent Mental. His post, as usual, is excellent, but I wanted to take issue with the slightest of details (because that’s just how I am): why do novice outfielders often take a step forward when the crack of a bat and the launch of a ball indicates that a fly ball has just been hit in their direction?

As a former and largely inept outfielder for the Ascension Catholic Church ‘Steamrollers,’ 2nd grade and under team (I was more of a junior soccer player), I well remember our coach, Dr. Wickersham, telling us repeatedly, and to little effect, ‘don’t start running forward until you know the pop-up is going to fall in front of you.’ I also clearly remember the sinking feeling when, after failing to heed his advice, a fly ball flew over my head as I charged toward it, ultimately landing almost precisely where I had been standing the instant that ball was hit.

Peterson discusses a recent paper in the journal, Human Movement Science, ‘Catching fly balls: A simulation study of the Chapman strategy,’ by Dimant Kistemakera and colleagues. Kistemakera and his team set out to test the slight variations between the trajectories fielders took when running to intercept a fly ball, and the trajectories predicted by Seville Chapman’s ‘strategy’ of using the acceleration of the ball in one’s vertical field to control whether one was too close or too far from home plate to make the catch.

The Chapman strategy tested

Chapman, a physicist, pointed out that, if a catcher is standing at the right distance from the batter to make a catch, the acceleration of a parabolic hit in his or her visual field will be zero. If the ball appears to the catcher to be decelerating, the hit will fall short of him or her; apparent accelerating indicates that the ball is about to sail over the catcher’s head. If the catcher must move to get under the ball, running at the correct velocity to intercept produces the same effect. If the ball is accelerating in the visual field, the catcher better step on the accelerator as the run is not sufficiently fast to make the intercept. If the ball appears to be decelerating, the catcher should slow down as he or she is about to over-run the catch.

Chapman’s description of how a fielder might track the ball, and it’s general fit with observed behaviour, led to this ‘strategy’ for fielding being dubbed the ‘Optical Acceleration Cancellation’ strategy. I put ‘strategy’ in quotes only because the word might imply that baseball players and others who use it actually have any conscious awareness that they are doing this, which I don’t think they do. As far as most are concerned, they’re just running to where they think the ball will come down.

Empirical, virtual and theoretical research supports the general principle of the ‘Chapman strategy’ model, although there are some issues, such as the fact that a fly ball does not travel in a true parabolic arc because of drag. Some researchers suggest that it isn’t so much that the catcher ‘zeroes out’ the perceived acceleration as it is that the catcher uses a more general qualitative approach. That is, the running velocity toward the point of the catch may be constant, and the catcher knows the direction to run — forward or backward — and when to stop running, by the apparent direction of the ball’s apparent acceleration. In addition, some theorists point out that as the ball nears the ground, catchers seem to switch out of the Chapman approach to fine tune the position of their bodies and gloves.

Sam Peterson has a great post on Optical Acceleration Cancellation (and a later development, Linear Optical Trajectory theory) on his own website, Sports Are 80 Percent Mental: Baseball Brains – Fielding Into The World Series. Peterson does an especially good job sorting out how both approaches suggest an ecological psychology approach to the basic problem rather than a view of the brain as an ‘information processing’ organ.

That is, a model of the brain as an information processing and memory retrieving machine that manipulates information suggests catching a fly ball is a calculation and comparison problem; calculating the path and recalling previous experience to compare the current situation with previous experiences of catching (or failing to catch). In contrast, an ecological psychology approach ‘argues that the fielder observes the flight path of the ball and can react using the angle monitoring system.’ According to ecological psychologists, the fielder is not so much remember and calculating as it is monitoring sensory input and responding with patterned action to shifting perceptions. (If you’re interested, definitely check out the piece by Dan Peterson on the Sports Are 80 Percent Mental weblog.)

The new article by Kistemakera and colleagues doesn’t reopen the empirical question — they use the trajectories recorded in earlier video-based research of catching provided by McLeod and Dienes (1996). The Kistemakera research team took into account a number of variables, such as the delay in reaction from expert fielders and factored in a minimum threshold for perceiving the acceleration of the hit ball, to test the fit of Chapman strategy-based intercept pats to the routes actually run by expert fielders in the McLeod and Dienes study. Kistemakera et al. found that intercept paths to chase down fly balls predicted by their versions of Chapman’s model matched pretty closely to actual fielders’ trajectories, except for a few factors.

Among the exceptions, the delay in starting to move in actual fielding behaviour was slightly longer than anticipated by the model, up to 500 milliseconds, but this hardly undermines the overall pattern matching closely to what Chapman’s theory predicts. It turns out that the model suggests it will be easier to detect a trajectory destined to take the ball over the head of the fielder than falling short, but it’s also likely that running velocity forward and backward would not be identical, although that’s an empirical question (and probably answered somewhere else in the literature that I’m not aware of). In addition, I wonder what the role of sound perception is in the anticipation of where the ball will go, but that’s for another post.

Okay, so what’s your beef?

If I like Peterson’s discussion and the Kistemakera et al. article in general so much, what’s my issue? It’s actually with the very end of the Science Daily article. In particular, when I read, I was struck by this section:

Will those first few steps forward doom the Little Leaguer to years of fly ball nightmares? Actually, it might be our brain’s method of improving its viewpoint.

“For a fielder, making a step is a way of changing the magnitude of the optical acceleration, while preserving its informative value,” Kistemaker clarified. “A faster rise of the optical acceleration above the detection threshold may outweigh a possible initial step in the wrong direction. Making an initial step forwards is not only easier than making an initial step backwards, but might also be a better choice.”

So, if you’re now coaching Little Leaguers, be patient. Their brains may still be learning the math.

Okay, so I know that the last paragraph is a bit of hyperbole — Person’s a clever writer, and it’s a good ending. Because I’m so hypersensitive about the difference between ecological psych approaches and ‘thinking machine’ approaches to the brain, I probably over-react to some metaphors. Peterson is much more agnostic about whether ecological psychology or information processing theory is more plausible, but I’m pretty convinced that, in the case of fast actions and motor control, the brain is better modeled by ecological psych (though not necessarily for some other functions). The ‘learning the math’ joke, however well placed, is not an eco psych metaphor.

But that’s not really my issue. My issue is with the explanation of the odd first step, toward the batter even when the ball might be going the other way. Kistemakera and colleagues write:

A second characteristic that is irreconcilable with a straightforward interpretation of the Chapman strategy is the initial forward motion of a catcher irrespective of landing position. However, this appears not to be a general feature as McLeod and Dienes noted that not all fielders stepped forwards. Furthermore, Oudejans, Michaels, and Bakker (1996) found that primarily non-experts moved initially in the wrong direction, suggesting these initial movements were based on a wrong judgment. However, an alternative explanation would be that novices make a step to change the magnitude of the optical acceleration thereby facilitating its detection. At ball release, the ball is relatively far away rendering the optical acceleration relatively small. (Kistemakera, Faberc and Beek 2009)

Peterson presents both but focuses on this later interpretation, that by moving, athletes exaggerate the apparent motion so that, in some sense, they can get a clearer read of where the ball is going. That’s possible, but I think that the first explanation, that the first step forward regardless of trajectory, is a ‘rookie mistake,’ is more likely for a number of reasons, and holding onto the possibility of systematic, patterned error is important for thinking about neuroanthropology.

Explaining the forward step as an adaptive strategy seems improbable, first, because if the relative acceleration in the visual field has to reach a specific threshold before the fielder can detect it, moving forward would increase the acceleration more for balls traveling in a trajectory taking them over the fielder than those falling short (at least, as best as I understand the perception of the ballistics in the visual field). The problem is that catchers have a harder time detecting the short fly ball than the long one, meaning that taking a step forward would skew their accuracy in the direction in which they are already more accurate. If stepping is a good strategy, perhaps stepping backward would make more sense, or experts would demonstrate the step more than novices.

Second, and more importantly, is that the forward step might be a formulaic behaviour ‘released’ by the sound of the hit before the catcher can perceive the direction. I think that this was what Dr. Wickersham was trying to teach the Ascension Steamrollers; he was coaching us not to do a relatively instinctive motor pattern when, after standing for a long time in the field, the sudden crack of bat on ball caused us to react.

It would be interesting to learn if, in other settings of whole body reaction, there was also a tendency to step forward in novice practitioners, before sufficient inhibitory counter-responses developed. I suspect that there might be from reflection on my own teaching of martial arts and other sports-related experiences; that is, I think that there might be a tendency, when a person is learning relevant stimuli in a sports setting, to move forward or toward the stimuli until a person learns to inhibit this behaviour.

The pattern of stepping forward would not necessarily be a ‘mental’ response; it might rather be a motor pattern, an ease, or just a tendency, to move forward to initiate movement (rather than moving sideways or backward, which might be marginally more difficult). For a neuroanthropology of movement, we need to consider the way that enculturations of various sorts — including learning to inhibit formulaic motor patterns — play a role in developing skills. In the case of learning to catch a fly ball, gaining skill might not simply be filling an empty vessel with skillful actions but might also include gaining the ability to inhibit or modify basic movement patterns into more useful, refined reactions.

The disagreement with Peterson is pretty insignificant, a matter of emphasis more than substantial disagreement, but it also reflects a potential for neuroanthropological analysis. If we find forward stepping errors in a range of activities, even when the motion does not offer any potential advantage (such as in learning to avoid being kicked in capoeira, where I’ve seen the reaction), then this cross-cultural or cross-activity pattern would suggest that it was not necessarily an adaptation to the current perceptual problem. I certainly would like to believe that my steps toward the batter in Little League were a well-considered strategy, but with my recollections of having to run after the hit after it flew over my head, bouncing toward the outfield fence, I kind of doubt it. I’m sure Dr. Wickersham would probably agree.

Enculturation is not just impressing expertise on a ‘blank slate’ of a human being. Rather, skilled enculturation is also shaping existing patterns of motor response, some of them likely quite neurological ‘primitive,’ even existing from very early in development, into new actions for specific contexts. As the late Esther Thelen’s studies of infant walking have shown, children learn to work, in part, by harnessing the dynamic properties of their own body, building steps in part out of basic, rudimentary motions of their bodies present long before they take their first steps.

References
Chapman, Seville. 1968. Catching a baseball. American Journal of Physics 36(10): 868–870. (abstract)

Kistemakera, D. A., H. Faberc and P. J. Beek. 2009. Catching fly balls: A simulation study of the Chapman strategy. Human Movement Science 28(2): 236-249. doi:10.1016/j.humov.2008.11.001

McLeod, Peter, and Zoltan Dienes. 1996. Do fielders know where to go to catch the ball or only how to get there? Journal of Experimental Psychology: Human Perception and performance 22 (3): 531–543. (abstract)

R.R. Oudejans, R. R., C. F. Michaels and F. C. Bakker. 1997. The effects of baseball experience on movement initiation in catching fly balls. Journal of Sports Science 15(6): 587–595. doi:10.1080/026404197367029 (abstract)