Brain Scanning: Could The Research Be Wrong?

Even with a large amount of data, brain scanning studies are still too small to reliably detect most links between brain function and behaviour.

Scientists and researchers have dedicated their life's work to investigating brain structures and activity of various functions. What if research indicates that most of it might be wrong?

It might not all be doom and gloom! Scientists believe that smaller and more detailed studies of the brain and behaviour can actually produce reliable findings.

Marek (2022) analysed MRI scans and behavioural data from over 50,000 participants (for example, from the UK’s Biobanks’s collection of brain scans). Most of the data was investigating the size of particular regions of the brain and others looked at the measurement of brain activity whilst individuals completed a task (to reveal how brain areas communicate). The researchers then used subsets of this data to produce smaller scale studies ranging from 25 to more than 32,000 people. They found associations in a study of 2000 participants only had a 25% chance of being replicated and studies with 500 or fewer participants had associations of around 5% of the time.

What does this all mean?

High 'r' values in brain scanning studies may be a fluke!

According to Marek, brain scanning studies needs to get much bigger, and this area tends to be rich in false positive data. Therefore, if you’re not looking at studies with thousands in a sample the likelihood is it is wrong or lucky in finding a reliable brain-behaviour correlation!

Marek, S. et al. Nature https://doi.org/10.1038/s41586... (2022).