Research Hit: Surprise! Your Brain Is A Prediction Machine Not A Thinking Machine

In a first, human brain activity is shown to not see what is happening but see what it expects

Ok we’re good at predicting things, sometimes, but what do you mean by “brain activity not seeing what is happening” surely this is wrong?!

Yes, it looks like a mistake but that is what a group of scientists have observed in the human brain.

Of note is that this was in the human brain.

Yes, you often report on research into mice or rats!

Yes, that is because it is unethical to wire up human beings for the sake of research but there are some limited opportunities to do this. One of those situations is when patients with severe seizures have their brains wired up because traditional methods can’t see or predict from which part of the brain the seizures are coming from.

They have electrodes implanted in different brain regions and are then kept in hospital and monitored - until they have a seizure. Which could be days, or weeks. During this time their time could be used for research. Enter Chaoyi Qin et al. of the Netherlands Institute for Neuroscience (KNAW) in collaboration with the Jichi Medical University in Japan who conducted a number of video observations with 10 patients.

Ok, and what were they looking for?

Well, they wanted to see how the brain processes these actions. For this they used simple video sequences of people engaging in daily familiar activities such as making breakfast, by getting a bread roll, cutting it, putting on butter, and then spreading jam. So far so good. But they then used the familiar sequences in scrambled order.

And what should this do?

Well, we would all assume that the operating model of the brain goes from seeing to processing, to simulating action (and maybe then subsequently doing an action - that is observation learning).

In case you didn’t know. Watching an action also activates similar brain regions as actually doing an action. So when you watch someone walking your brain is also doing a little bit of walking!

Great!

But what was surprising is that this classic model of feedforward i.e. see and then activate other regions wasn’t how the brain activated!

What! How does it activate then?