"In work with rats and zebra finches, neuroscientist Bence Ölveczky of Harvard University in Cambridge, Massachusetts, and his team found that stimulating one part of the brain to induce certain behaviours might cause other, unrelated parts to fire simultaneously, and so make it seem as if these circuits are also involved in the behaviour.
According to Ölveczky, the experiments suggest that although techniques such as optogenetics may show that a circuit can perform a function, they do not necessarily show that it normally performs that function. “I don’t want to say other studies have been wrong, but there is a danger to overinterpreting,” he says...
Ölveczky accidentally damaged one animal’s motor cortex while injecting the drug. He decided to use a toxin to permanently destroy that portion of the brain to see whether such a lesion would have the same effects as the temporary disruption. When the researchers tested this rat ten days later, they were surprised to find that it could still press the lever correctly, despite having not performed the task since the damage occurred. The observation suggested that the damaged circuit was never actually involved in the behaviour in the first place; without practice, the brain cannot simply switch to using a different circuit. The researchers concluded that their muscimol experiment had shut down multiple circuits, some of which were involved in the lever-pressing behaviour...
Ölveczky suggests that regions of the brain are so intertwined that suddenly changing one region — such as Nif — sends ripples through the rest of the system, which affects behaviours that are not normally dependent on the region.
This is really relevant for me to read right now, and it makes a lot of sense.
Similar experience that I had: I was using these specially designed synthetic receptor-drug pairs (DREADDs and CNO, if you are in neuroscience), which are supposed to be neuro magic because the receptor isn't supposed to respond to anything that is naturally in the brain and the drug isn't supposed to activate anything besides the synthetic receptor. So you get a mouse and you edit its genome so that it makes this synthetic receptor in the neurons you are interested in, and then you inject it with the drug and theoretically JUST those neurons are going to be activated or inhibited (depending on the type of synthetic receptor you use). But, we found that at really high doses of the drug the animal would start to behave different - even if it didn't have the synthetic receptor in its genome. So, obviously, the drug was doing extra stuff and also controls are full of information.
As I'm writing this, I'm starting to think really, actually, aaahhhh, about designing my graduate research and being in charge of projects and not just the big, fun theoretical projects you would come up with in undergrad but, like, the actual nitty-gritty of "how many different control groups am I going to need and if there might not be room in the mouse habitat right now, could I run just one of the control groups at first to see if I actually need the others?". And now I'm thinking, when I do these kinds of inducible gene expression techniques, should I also just test one where I do something permanent as an intense positive control? And how do I design that so it doesn't include all of the counfounds...
No comments:
Post a Comment