Since this is the science forum, I was trying not to make that step (the philosophical one). Rather, I'm asking the prior question: where does the philosophical concept of a "mistake" come from? Can we pin it to a scientific (or dare I say "natural") phenomenon?
Philosophy comes before science when you ask a basic enough question. Science is a philosophy after all. A methodological philosophy of controlled testing for the uncovering of knowledge.
But what you asked, It's more of a philosophical question in my mind.
The answer in my estimation is that a mistake is something that doesn’t fit, something that is inconsistent with data, something that is not useful. Leads one to be mislead or otherwise non functional.
So, it is going to be a question of value which is never going to be clear as it is based in both individual perception and shared experience.
So, science can not necessarily answer the question for you "what is a mistake" it can however help you find out what is true.
I guess what I keyed on in that study was that the subject was taught to identify what was "correct." The study notes that "correct" equates to "desired." Hmm. I see a circle of logic forming here.
But it is true that value is what we desire, and desire can be arbitrary, therefore
things labeled correct can be incorrect. They are "correct" in the arbitrary context but when applied as a standard in objective reality they conflict, so they are incorrect.
But what you've noticed is that: You can teach false things!
You can get a dog to salivate on command with a dinner bell. There’s no objective link between the dinner bell and the food that generally comes with it they are just two experiences that the dogs brain has fit together. Humans can be taught to do this too and their brain's neurotransmitters will react accordingly to the scheme you give them.
Abstraction is free to be arbitrary and still consistent. We do it all the time with games.
I understand that they are discussing the patterns that develop in the neurons, but those patterns fail to explain why something that doesn't fit is a mistake. It seems they have adopted #3 from my list, and it's a reasonable thing to do. But does that make the others in the list unreasonable? I don't think so.
It's a "mistake" because of the rules. Just like moving your piece four spaces when there is a six on the dice. The game may be arbitrary, but the rules within it can still be consistent.
A mistake is always going to be an inconsistancy between an outcome or an answer and some standard.
The study you posted is deliberately arbitrary.
After animals could perform the association well (after at least 30 correct responses and 90% correct trials over the previous ten trials for each cue), the association was reversed with no signal, and they had to relearn the new association. Perhaps because there was no explicit signal, so that any error might signal a reversal, the animals relearned the association slowly after reversal (Figure 1). Each recording session consisted of three to eight reversals (four to nine trial blocks). By requiring animals to repeatedly relearn the associations, we could dissociate learning-related effects from artifactual effects that resulted from slow shifts over the course of a session, related, i.e., to motivational changes or changes in the position of the electrode relative to a neuron.
