Review: Spark

Due to a series of unfortunate but probably-hilarious-someday circumstances, my final assignment of my undergraduate career at MIT is a book report on Spark: The Revolutionary New Science of Exercise and the Brain. A book report. Yes, I thought I was out of middle school too.

Since this will undoubtedly be An Experience (I am writing a book report — a book report — with a day-old wrist fracture to my dominant arm), I have decided to quasi-liveblog my reading of this book. I have a mug of hot chocolate, RainyMood in the background, and a lot of opinions about pop science. Let’s begin.

IMG_0871.JPG copy

9:00 pm: Introduction

This book has ten chapters. Among chapters dedicated to learning, anxiety, and ADHD, number eight is “Hormonal Changes: The Impact on Women’s Brain Health,” which makes me extremely optimistic that this book won’t be condescending at all.

Exercise “is simply one of the best treatments we have for most psychiatric problems.” I’m doing my best to stay open-minded, but that statement is going to take a heck of a lot of evidence to back up.

The book opens by describing the Naperville school district, which has apparently had remarkable academic success as a result of introducing an extremely thorough exercise program. “At a time when we’re bombarded with sad news about overweight, unmotivated, and underachieving adolescents, this example offers real hope.” As a recent adolescent myself, I’m pretty insulted on behalf of the under-20 crowd. This is a bleak and very dismissive view of a sizable chunk of the US population, and it sounds a lot like a barely-disguised Kids These Days.

9:21 pm: Chapter One (Welcome to the Revolution: A Case Study on Exercise and the Brain)

“It is 7:01 am, and for a small band of newly minted freshmen lounging half asleep on the exercise equipment, that means it’s time for gym.” I can think of very few things less pleasant as a high-schooler than 7am gym class. My first thought is that, given the research on the amount of sleep high schoolers need and how they don’t function especially well at 7am, a later start time (say, 9am gym class to begin the day) might be more optimal.

“This is Zero Hour PE…the objective of Zero Hour is to determine whether working out before school gives these kids a boost in reading ability and in the rest of their subjects.” OK, sounds reasonable so far.

“The kids in Zero Hour, hearty volunteers from a group of freshmen required to take a literacy class to bring their reading comprehension up to par, work out at a higher intensity than Central’s other PE students.” It’s good that they’re volunteers, and good on them for putting in that time. I would be curious to know how this Zero Hour scheduling fits into their class schedule, and how it impacts their homework/how much sleep they get each night.

“At the end of the semester, [the Zero Hour students] show a 17 percent improvement in reading and comprehension, compared with a 10.7 percent improvement in other students who opted to sleep in and take standard phys ed.” The book doesn’t say how many freshmen participated in this experimental program, but I’m already seeing potential for selection bias — it’s very likely that the kids willing to get up an hour early to work out for a school program are also more motivated in their other classes.

The administration decides that the Zero Hour program worked so well that it’s now a first-period (8am, I think?) class for all literacy students. “The literacy students are split into two classes: one second period, when they’re still feeling the effects of the exercise, and one eighth period. As expected, the second-period literacy class performs best.” As expected? As expected?! There should be no “as expected” in a rigorous experiment. Maybe this experiment can’t be thoroughly randomized and controlled — although, frankly, it absolutely could be if one were committed to good experimental practice — but if all of the teachers knew what outcome they expected to see, there’s a good chance that could impact their teaching. And did they try a third-period literacy class, or fourth-period?

Based on the results from the literacy classes, the “New PE” curriculum (which is never actually described in full) is used for all students at Naperville schools, and has been in place for the last 17 years. Students at Evanston’s New Trier High School scored an average of 26.8 on the ACT, two points higher than the average for students at Naperville Central High School (and 6.7 points higher than the state average of 20.1), but Naperville students do better on state standardized tests, “which are taken by every student, not just those applying to college.” That’s a very good point, although the impact of the statistic is slightly diluted because no information is given on exactly how much better Naperville students perform on those state tests.

Ratey next cites Naperville’s performance on the Trends in International Mathematics and Science Study, or TIMSS. In 1999, Naperville tested as an individual district to see how it compared to other states and countries, a very reasonable metric to use. It would be nice if Naperville also had its students take the TIMSS before instituting this new exercise program to allow for comparison, but hey, hindsight. And basic experimental protocol.

“On the science section of the TIMSS, Naperville’s students finished first, just ahead of Singapore, and then the North Shore consortium [a group of other districts from Illinois.]” They also placed sixth in the world for math. Now, this sounds extremely impressive, but first in the world with this PE program is not really significantly better than third in the world without this PE program, like the other Illinois district. Some hard numbers on scores here would also make comparison easier. (The book didn’t mention how the North Shore consortium scored for math, so I looked it up: they placed seventh. Sixth with PE and seventh without is unlikely to be a statistically notable difference.)

Oh dear lord: “Naperville 203 is a demographically advantaged school district: 83 percent white, with only 2.3 percent in the low income range, compared with 40 percent in that range for Illinois as a whole. Its two high schools boast a 97 percent graduation rate. And the town’s major employers are Argonne, Fermilab, and Lucent Technologies, which suggests that the parents of many Naperville kids are highly educated. The deck — in terms of both environment and genetics — is stacked in Naperville’s favor. On the other hand, when we look at Naperville, two factors really stand out: its unusual brand of physical education and its test scores.” What actually stands out is that this is a wealthy, mostly-white district full of students whose parents are literally particle physicists. It would genuinely be more surprising if these kids didn’t test well in math and science.

The book then goes into more detail about Naperville’s New PE, which honestly sounds like a very thoughtful and well-designed program, if not as world-changing as Ratey wishes it were. The head PE teacher “decided to shift the focus to cardiovascular fitness” rather than team sports, and had students run the mile once a week. He introduced new equipment like exercise bikes so “nonathletes” could earn extra credit to make up for their low grades on activities like the mile.

After testing a heart rate monitor on a student “who was thin but not the least bit athletic,” the PE teacher discovered her heart rate was about 90% of maximum while running despite her comparatively slow pace. He then “started thinking back to all the kids we must have turned off to exercise because we weren’t able to give them credit” and completely reworked the PE program as a result, with students “graded on how much time they spend in their target heart rate zones during any given activity.”

I find this really admirable, especially the teacher’s willingness to adapt his program based on new data. I’m a huge proponent of evidence-based policymaking, which involes identifying the outcomes one wants from society — be that higher test scores on math and science, better college completion rates, reduced cost-per-user for healthcare, you name it — and systematically testing programs and collecting data to see what produces improved outcomes. I think this PE teacher and I would get along very well.

Equally admirable is the story Ratey relates next, about how the PE teacher and PE coordinator “learned by…attending sports physiology seminars, reading neuroscience research papers, and constantly e-mailing their findings to each other.” Between their enthusiasm for research and their willingness to adapt existing programs to accommodate new evidence, these teachers would make some pretty fabulous scientists.

Current gym classes in the district involve a variety of options. On any given day in gym class, students are “lining up to get on the climbing wall, arguing about who was going to get to use a new exercise bike attached to a video-game monitor, running wildly on treadmills, playing a video game called Dance Dance Revolution [and] all wearing heart rate monitors.” This actually sounds awesome. Giving students a variety of options for how to exercise and making them enjoyable options is a great way to ensure that they both get a decent amount of exercise and don’t totally hate it.

With that positive interlude, we will now return to our regularly scheduled bad data.

“Among California’s 279,000 ninth graders…those who scored a six on the FitnessGram [the highest score] ranked, on average, in the sixty-seventh percentile in math and the forty-fifth percentile in reading on the Stanford Achievement Test. If these scores seem less than stellar, consider those of the students who passed only one of the six areas: they ranked in the thirty-fifth and twenty-first percentiles, respectively.” I’m on campus at MIT right now. If I scream “correlation is not causation” loudly enough, do you think Ratey could hear it over at Harvard?

“When the CDE [California Department of Education] repeated the study in 2002, it factored in socio-economic status…Within the lower-income students, fitter kids scored better [on the academic tests] than unfit kids.” Parental involvement is another confounding variable here, and one that definitely needs to be accounted for.

“In my experience, PE wasn’t really about exercise. Quite the opposite — it discouraged exercise. The cruel irony was that the shy, the clumsy, the out of shape — some of the kids who could most benefit from exercise — were pushed aside to sit on the bleachers.” This is a very valid point, and one I’m glad to see the author bring up.

“One of the most innovative changes he made at Central was to add a mandatory square-dancing class for freshmen.” This is another idea I’m extremely impressed with — it’s unlikely anyone comes in with a ton of square-dancing experience, so it levels the playing field for all of the students and gives them a chance to interact and get exercise in an extremely low-pressure way.

“Since the program [a version of the PE program used in Naperville schools] started in 2000, the standardized test scores of Titusville’s students have risen from below the state average to 17 percent above it in reading and 18 percent above it in math.” Titusville is an economically depressed city with an median family income of $37,000. This is a far more remarkable outcome than Children of Physicists Do Well On Science Tests — why not lead with this?

And on that note, we are finally done with chapter one. On to chapter two!