Computational neuroscience–frontal lobe model

Researchers have built a computer model which replicates our human ability to learn rules and use them to control ourselves—which scientists call “cognitive control”. Cognitive control is how we mold and channel our thoughts and actions to reach some goal. The prefrontal cortext (PFC) is a key player in cognitive control.

The computer model, described here, accurately mimics the behavior of normal people on typical tasks such as card sorting. It reproduces the process by which we learn new cognitive control rules, down to the level of biological properties specific to PFC neurons. The model shows why learning self-control takes so long—often into the teens, or even longer in some cases.

But then the scientists gave their model the equivalent of a frontal lobotomy—and found that the model failed to learn and apply rules in exactly the same way similarly damaged humans did. This is a big improvement on lobotomy techology: just use your mouse to uncheck the “frontal lobe” checkbox, instead of going in with an ice pick!

The paper was published at at PNAS; here is the abstract, from which I excerpt:

We show how the development of task-specific PFC representations can occur when a set of PFC-specific neural mechanisms interact with breadth of experience to self organize abstract rule-like PFC representations that support flexible generalization in novel tasks.

Neurotheology desperately needs a similar computer model, one which replicates the religious pathways in the brain. This is the subject of the field called “computational neurotheology”, which I referred to in earlier posts. (Of course, a model cannot prove that a particular mechanism is the one actually at work, but it can establish plausibility, show that existing conceptual models are underspecified, and provide insights for further research.)

Running such a model, the computer will, for the first time, literally experience God.

Leave a Reply