Navigating the thin ice of science translation

17 Oct 2013 Posted by

The thin ice of truth – polarized, filtered, simplified…wrong

Last week, a colleague sent me this excellent piece by Christopher Chabris, in which he discusses why it matters that Malcolm Gladwell’s method of weaving the narrative and science together in his latest book (and all its predecessors, it must be said) bends boundaries of truth, and why it matters.  I highly recommend the article, because Chabris makes some excellent points, relevant no matter your field or approach to science, or story-telling, or simply how you learn about the world.

That article, combined with some recent thinking I’ve had to do for some upcoming presentations, inspired me to throw down a few quick thoughts on our Facebook page, trying to highlight some of the common ‘errors’ that I’ve seen in sports science, because ours is a field that lends itself to being ‘hijacked’ for popularity.  One of the great privileges of studying sport is that it (or health through exercise) is so relevant to so many, but the downside is that it can quickly become the figurative football beneath a pile of bodies scrambling to grab a fumble.

In the last few years in particular, barefoot running, 10,000hr mythology and the extreme carb crusade have all challenged thinking in some thought-provoking, but misleading and sometimes incorrect (with real consequences) ways.  They each highlight aspects of complexity and how the truth sometimes exists on thin ice.

As usual, my three quick thoughts turned into a lengthier article, and so here it is, reproduced on the site, with one or two additional thoughts woven in.

Adapted from Facebook, Oct 12 2013

Here are 3 quick tips on how to spot that you’re on the “thin ice” of truth when reading popular media translation of science (and some scientists, direct):

Top

1) Unnecessary polarization

A type of straw man error where the protagonist artificially creates an ‘either/or’ situation, polarizing a debate into A vs B. But not both.

This happened, in sports science, with the talent vs training debate. Was it 10,000 hours & training, or was it talent & genes? The correct response should’ve been “Both. Why must it be one without the other?” Yet books were written implying that the world had discounted hard work and training, which was ludicrous.

The polarization was “necessary”, because without first putting words in the mouths of people who weren’t actually saying anything, there was no argument, no sensational revelation and no world-changing theory for success that suddenly gave everyone hope (after purchasing the book).   So we might dub this the “Put words in their mouth error”, because only then do you have a battle to win.  Whether it really exists or not doesn’t matter.

So beware of foolish, unnecessary and non-existent polarization.

Top

2) Selective presentation of supporting evidence

I’d call this the “dark side of the moon” error because the protagonist only ever shows that side of the argument that supports the story, leaving the rest in darkness.

Gladwell did this in his latest book, citing a study that supported his theory, but not mentioning that a follow-up study, far larger, did not. Conveniently “shielded” from an inconvenient truth, the reader has no idea of the complexity of an issue, and so this is done because it doesn’t serve the agenda of the advocate to invite doubt – simple, powerful theories need definitive, unambiguous results.

In my field, I have seen this often in the nutrition debate, where the cause of obesity is reduced to an issue of carbs vs fat (note the polarization of this issue, by the way. Refer to point 1).

Here, protagonists for either extreme will show evidence that supports only their perspective, and potentially dozens of studies that do not are written off.  Because both sides do this, the result is that an individualized approach to nutrition is made very difficult, and experts appear as far apart on a topic as possible.  The public meanwhile, are ill-equipped to sift through the complexities, and are left confused by the two totally different views they are getting.

It also creates, and then re-inforces enormous confirmation bias, as people eagerly seek out those studies and anecdotes that support their view of their particular moon.

In science, we are supposed to be held to a higher standard, but too often this does not seem to apply.  What happens here is that the “pendulum of truth” swings wildly from one side to the next, becoming a wrecking ball that destroys the poor ‘victim’ in the middle.  There is also the problem that the focus on one aspect (in this case, nutrition) detracts from numerous others (physical activity and health, despite obesity), and overall progress is hindered by a detour that may be beneficial for some people only.

This is an irresponsible attitude to discovery, but one that most people do naturally, so it’s easy to excuse as an accident. It does sometimes have an upside, however, because it can pull both sides gradually towards the middle, where they may meet halfway and discover a truth, provided their egos don’t prevent it.

That truth will usually emerge when two views of the moon are combined to create a fuller picture.  Unfortunately, the antagonism created by extremism often prevents this, and it takes hard work to a) halt the antagonistic cycle, set aside conflict and invite ‘opposition’ to change your belief (when did it become a fight to begin with? Aren’t we supposed to searching for the same truth?); and b) get round the back to that different view, but it doesn’t come with quite the same limelight to acknowledge complexity and present two sides of a debate. Frustratingly, everyone loses.

Top

3) Over-simplification

Related to 1 & 2, I suspect this is symptomatic of modern thinking – “If I can’t understand it in a tweet, then it’s not worth knowing”, but the desire to simplify is destructive to truth when dealing with complex issues.

In exercise and health, unfortunately, there are so many connections that when you pull a particular string, there are multiple possible outcomes, and they are rarely predictable. Simple causality of A leads to B is the message people want, but it rarely applies.

What is more, in things like health and performance, the outcome is often severely delayed from the action – it can take years for weight to be lost or health status to change or performance to become world class. People are terrible at acting when the result is so delayed, and so many confounding events can affect the outcome that the message sounds tepid and suffers from poor uptake. I am led to believe, thanks to a stimulating conversation recently, that nature conservation efforts are hampered by identical challenges.

It’s little wonder then that the market embraces simplicity, even when it is foolish. Case study of this is the barefoot running debate, where a simple argument of “we are born without shoes, it’s natural and thus better” has spawned an industry.

Even science weighed in, when a Harvard lab published a paper in Nature (gasp, it must be infallible) lower loading rates when barefoot. What they failed to mention was a) the loading rate-injury link is tenuous (and non-existent for many injuries) and b) not all barefoot running is created equal and some people, for reasons as yet unknown, go the other way and may be worse off.

The simple story, in other words, had a fair amount of fine-print. But hey, who reads that stuff anyway? There is no fine-print in a 140 character tweet.

(If you are interested in the complexity, by the way, I’ve written a collection of articles on barefoot running, with the three most recent ones tackling this study, and other applications.  Barefoot running is also our Featured Series on the home page this week)

Simplification is of course necessary – imagine you are asked for directions from your local convenience store to the high school.  If you describe every single building, landmark, intersection that the person will see, as well as all the ones they won’t, their chances of success are slim.  So we filter out that which we believe is unnecessary (sometimes making Error 2) and then simplify that which we believe is important (“Left at the lights, third right, past the train station, it’s on your right”).  I’ve almost certainly been guilty of this, as has everyone else, but it’s irresponsible to do so for complex issues where frankly, we don’t completely understand the directions we are giving anyway.

The trick is to recognize how to simply explain complexity, because in sports science, whether it’s high performance, barefoot running, diet, or just about any subject, if it is simple, it is unlikely to be correct.  This is what Einstein was onto when he encouraged people to simplify things as much as possible, but no more.  And if you don’t know the directions, then just say so…

Top

The solution? Disprove yourself, & own the translation

I’d say the solution is simple, but that violates my whole argument, and it’s not anyway!  However, part of it is to remember one of the fundamental requirements in science – disprove it, don’t prove it. If you want to advocate for barefoot running, set out to understand why some people fail, not why they succeed (we are attempting this research – more on that another time). If you promote an extreme diet, then find people who do not achieve results with that diet, don’t look for informercial type feedback from those who do.

By doing this, you quickly discover complexity, because exceptions disprove rules, demanding explanations that you cannot shoehorn into a polarized, filtered and simplified reality.  You’ll quickly discover how much you can’t explain, and that’s humbling and helpful in equal measure.

Failure to do this will make you blissfully unaware, possibly even happy and content, until you lead an army of followers off a cliff you should’ve seen coming.

Finally, science needs to own the translation of its findings better. There are amazing examples of this, but too often, it is left to pied piper storytellers to change the world. Gladwell is right about the inaccessibility of science, and we should address it. But without falling into the same traps of polarizing, oversimplifying and presenting the convenient side of the story.  We can do that if we own our own stories, and backed by slightly more rigorous training and thinking, we deliver a message that is accurate and credible.

Over and out

Ross

PS: Thank you so much to those who kindly gave donations in response to the launch of the site and my previous article.  I won’t overuse the request, and your readership and the dialogue we have is far more valued than the donations, but if you do enjoy the content and wish to support it, please consider a small donation – the button is at the top right of each page!  Thanks again!

%d bloggers like this: