You hit the nail on the head. I am definitely referring to correlates of boring tasks. But IMO with freely behaving subjects, the neuronal mass observed with ephys might not be enough to truly capture the dynamics of the circuit supporting the complex behaviour.

I guess it depends on the signal. For spikes that might hold true. I was more referring to the fact that just because it explains the most variance doesn't automatically mean it's behaviourally relevant.

I agree with you. They challenge but don't negate. PCA (tSNE or whatever) is a useful tool as well, and people should look beyond the first few components, putting more emphasis on task relevance than variance explained.

Are you sure about that?

It's the controversy.

I find it controversial that people find it controversial.

At this point it's entirely your fault 😁

I feel like the blame for such thinking lies not only on the MP. There's a disconnect between the funding and the output in the public eye.

I only heard about extraordinary clams.