Science ‘journalism’ skills

A good night’s sleep improves ‘guitar’ skills” says The Telegraph, adding “Forget the hell-raising image – the key to being a top rock star is a good night’s sleep, a new study suggests“. Notice the scare quotes around the word ‘guitar’? That’s because this study didn’t look at guitar skills at all. It looked at Guitar Hero skills.

The Telegraph claims that the research was published in the most recent issue of the journal Sleep, though it’s not in there. In fact, this is data taken from an abstract of a conference presentation that won’t be given until later today, according to the slightly more in-depth e! Science News. While conference papers are of course interesting, they are not peer reviewed, nor do they always even involve completed studies – the conference cycle is a few months out of phase with the publishing cycle, and conferences often serve as a way for researchers to test the water on how well their research will be received.

You can read the abstract for yourself in the conference brochure (PDF, pp. A37-38). The problem is that it’s written in quite dense scientific language, and may take a couple of reads to understand. What actually happened was that 15 students played a single song on Guitar Hero III until they could hit 50-75% of the notes (they don’t say which songs were used, nor what difficulty, sadly, though I do hope one of them was “Through The Fire and The Flames” on Expert. All in the name of science of course!), then played it again 12 hours later. Subjects who had a period of sleep between the first and second playthrough experienced a greater improvement in note accuracy, and from the look of the statistics presented in the abstract this is at least worthy of further study, though it could hardly be called definitive.

Still, with all this talk of 12 hour intervals and sleep-wakefulness conditions, it would be easy to grasp the wrong end of the stick. And so The Telegraph has.

Scientists used the video game “Guitar Hero III” to look at the link between sleep and the ability to coordinate.

Students were asked to stick to strict sleeping and waking patterns, and then tasked with performing songs on the game.

Those who hit the most right notes, at 68 per cent accuracy, had enjoyed twelve hours of sleep. In contrast, those who had been awake for the same period had an accuracy of just 63 per cent.

First of all, it was a test of learning, not coordination. “Tiredness makes you clumsy” is a hypothesis so rigorously proven that it wouldn’t be news ever.

Secondly, there is no mention in the actual abstract that subjects were forced to sleep for 12 hours every night. In fact, the abstract specifically mentions that sleep duration varied across the group. While they do claim to have found a correlation (r(13) = 0.52, P = 0.046) between sleep duration and improvement, it is a small one, especially given a sample of 15 people (or maybe 13 – perhaps they removed the two male subjects when collating the data); according to Wikipedia, it barely scrapes through the 0.05 significance test – the conventional test of correlation, which indicates there is a 1-in-20 chance that the correlation is just a fluke. And at that correlation level, it’s likely that even a short, disrupted night’s sleep will improve learning.

Thirdly, by “awake for the same period”, they actually mean “awake for 12 daylight hours”. The students weren’t kept up all night, but simply performed their tests in the evening. They slept the same amount as everyone else, and as far as I can tell, they followed their normal sleep patterns.

There are 1133 abstracts listed in that brochure – “Manipulating finger temperature to promote sleep” (0073, p. A28) sounds particularly interesting. If The Telegraph really want to support some argument about sleep deprivation, why not check the chapter on sleep deprivation – 65 abstracts, all on the effects of losing sleep – rather than taking a study on learning and cognition and twisting it out of shape? I suppose that in a brochure filled with articles like “Gammahydroxybutyrate and R-baclofen promote NREM sleep in hypocretin/ataxin-3 and wild-type mice” and “Predictive capabilities of the four-variable screening tool, the STOP questionnaire, and Epworth sleepiness scale for sleep disordered breathing in the sleep heart health study (SHHS)“, a study called “So you wanna be a rock star? Sleep on it” practically leaps off the page by comparison.

This is probably all nitpicky stuff. I mean, this a column filler article about a completely harmless misinterpretation of an unpublished scientific study of little relevance to anything. Perhaps a few parents will tell their children to go to bed earlier – not necessarily a bad thing – and the rest of the world will carry on with life uninterrupted. What does it matter, really?

Well, if this is how they cover the simplest science, how much faith does it give you in their coverage of more weighty matters?

  1. #1 by ukenagashi on Wednesday, 9th June 2010 - 13:39 GMT+0100

    Manipulating finger temperature is… yeah, what is that?

    Also it probably says a lot that some people blindly believe the media when it comes to science, and I blindly believe you XD It’d be nice if schools would teach the general critical thinking/reading skills that would benefit everyone in the battle to get the media to tell the straight truth.

    • #2 by atomicspin on Wednesday, 9th June 2010 - 13:55 GMT+0100

      Yeah, to be fair, you probably shouldn’t be believing me blindly either. Lord knows I don’t.

      Apparently “distal finger temperature” – the temperature of your fingertips – has been connected with falling asleep faster. They didn’t actually change finger temperature, but shone red lights on people so they thought their fingers were getting warmer and… yeah.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

<span>%d</span> bloggers like this: