NEUROBOLLOCKS AND FALSE POSITIVES

 
Remember way back when I posted about how there’s no such thing as a “male brain” or a “female brain”?  Well, today I was sent an article that:

a) Illustrates one of the ways in which brain scans can mislead if not viewed in context, and with full information of the stated assumptions;

b) Is flippin’ hilarious.

It’s this one: What a dead fish can teach you about neuroscience and statistics, sent to me via my friend Claire Witch File.

That’s right, they put a dead salmon in an MRI scanner! But all in the name of science. This has got to be the most creative way of proving a point about statistics that I have ever seen (and this is coming from the person who did their ‘A’ Level stats project on Smarties in order to prove that you can both demonstrate intellectual rigour, and fill your face with chocolate at the same time. I got an A on that coursework, and the world’s biggest sugar crash. Joke’s on you, Mr. P!).

As silly as it sounds, it demonstrated the care that must be taken in such experiments to set the appropriate significance level, and to be sure you’re interpreting the results correctly.

Three things I learnt as a result of reading this:

      1. I knew that these scans show blood flow to regions of the brain, but I didn’t know that what you’re seeing on the scan is the blood flow resulting from what happened up to 6 seconds ago.
      2. The images from an fMRI scan show a statistic – which must be viewed in context and measured against specific criteria with robust controls.  Basically, it shows which brain regions are likely to have activity – but it tells us nothing about that activity – yet.  There’s a lot about the brain and the mind that we don’t know, but we’re amassing new knowledge quickly.  We may one day be able to discern that sort of information with brain scans; we already know quite broad information on what’s going on in there – we need to refine it.
      3. Studies that sound really outrageous will get you a lot of media exposure.  They even won an Ig Nobel for their work!

 

1 thought on “NEUROBOLLOCKS AND FALSE POSITIVES

Leave a Comment