5 Comments »Posted on Wednesday 12 August 2009 at 12:25 pm by Jacob Aron
In Getting It Wrong, Health & Medicine, Inventions & Technology

Are a generation of children growing up with text messaging at risk of turning their brains to mush? The Daily Mail seems to think so.

“Predictive text messaging changes the way children’s brains work and makes them more likely to make mistakes generally, a study has found.”

The study in question was recently published by the journal Bioelectromagnetics, and did indeed look at mobile phone use and cognitive function in children. What Michael Abramson and colleagues did not find, however, was a causal link, despite what the Mail might think. Remember folks, correlation does not imply causation.

Researchers tested the mental abilities of 317 Australian 12 and 13-year-olds, and recorded their mobile phone usage. Results show children who had more calls and text messages were less accurate in memory tests, but completed them faster. The paper goes on to suggest that text messaging could be responsible, as predictive text “train[s] the user to favour speed over accuracy.” A quote from Abramson in the Mail article backs this up:

“We suspect that using mobile phones a lot, particularly tools like predictive text, is behind this.

“Their brains are still developing so if there are effects then potentially they could impact down the line, especially given that the exposure is now almost universal.

“The use of mobile phones is changing the way children learn and pushing them to become more impulsive in the way they behave.”

In a word: bollocks. The data gathered simply does not back this up. It may be completely true, but it’s not a statement that can be drawn from the evidence available in his paper. Makes a nice sound bite, though.

This line of argument is further undermined because the same correlation was seen with phone calls, not just texts, implying the underlying mechanism might be the same. Perhaps children who use their phones more often are just naturally more easily distracted, thus pay less attention? There is no way to tell from this study.

That doesn’t stop Baroness Susan “Facebook makes you fat” Greenfield weighing in, with her usual attacks on anything invented in the past couple of decades. In addition to suggesting “Generation Text” will cause the downfall of humanity, she has a go at Twitter:

When I was a child, if I wanted to tell someone about my day, I spoke to them face-to-face, I wrote them a letter or I walked to the phone box down the road.

Communication was far from instant and, although we were not aware of it at the time, it influenced what information we deemed worthwhile sharing.

Today, we can ‘tweet’ to the universe such inanities as: ‘I’ve just put my socks on.’ A friend can respond – ‘Congratulations!’ – within seconds.

A Twitter search for “I’ve just put my socks on” does admittedly turn up a single result, but this is just one of the thousands of message sent every day. Despite this common criticism of banality, I don’t think most people actually use Twitter in this way. No one cares what you had for breakfast, so tweeting about it probably results in a quick exodus of followers. Of course, I have no evidence to confirm this other than my own anecdotal experience, but at least I admit as much!

I agree with Greenfield that new technologies must be evaluated for potential harm. Where we differ is my requirement for causal links and solid evidence, rather than conclusions pulled out of thin air.


  1. 5 Comments

  2. Good post – what’s most bizarre about this study is, as you point out, that voice calls were also correlated with fast-but-inaccurate responding. Which directly undermines the theory that predictive texting is to blame.

    Also, they used self-reported phone use, which is obviously dodgy. I don’t have access to the original paper from home but – was this self-reported retrospectively (“How often did you phone in the last week”)? If so it’s even dodgier.

    By Neuroskeptic on Wednesday 12 August, 2009 at 7:22 pm

  3. Yes, didn’t quite manage to squeeze it in, but self-reporting, particularly from children, seems quite dodgy. The questions on usage were:

    What is the average number of calls you make
    per week? You can give me a range if that is
    easier.

    What is the average number of calls you receive
    on your mobile phone per week? You can give
    me a range if that is easier.

    What is the average number of text (SMS)
    messages you send and receive per week? You
    can give me a range if that is easier.

    By Jacob Aron on Wednesday 12 August, 2009 at 9:36 pm

  4. Well, there’s no way I could answer that accurately, let alone a 12 year old.

    By Neuroskeptic on Friday 14 August, 2009 at 10:05 am

  5. Yes, it’d be interesting to see if they could at least get an indication of the accuracy or error by obtaining (ethically) phone records for a sample of the respondents.

    It’s especially suss considering that some of their subjects “were less accurate in memory tests, but completed them faster.”

    Of course, it grabs headlines, and brings out the ‘get of my lawn’ crowd.

    By David on Friday 14 August, 2009 at 10:53 am

  6. I have only just seen your post, having been searching for “no evidence” and “predictive text”.

    My own suspicion regarding the study results is that funding tends to be tied to positive results of some sort. Having not found the radiation/brain effects that were initially being investigated, it looks as if the emphasis switched to a drive to find something (perhaps “anything”) publishable.

    By Mark Diamond on Sunday 11 October, 2009 at 10:29 am

Sorry, comments for this entry are closed at this time.