Just for sharing an unpretentious but (IMO) interesting post from a blog I regularly read.

In commenting on an article about the results of an experiment aimed at "simulating" a specific case of traumatic brain injury and measuring its supposed effects on solving a particularly difficult problem, economist/game theorist Jeffrey Ely asked whether a successful intervention could ever be designed to give people certain unusual, circumstantially useful skills.

It could be that we have a system that takes in tons of sensory information all of which is potentially available to us at a >conscious level but in practice is finely filtered for just the most relevant details. While the optimal level of detail might >vary with the circumstances the fineness of the filter could have been selected for the average case. That’s the second >best optimum if it is too complex a problem to vary the level of detail according to circumstances. If so, then artificial >intervention could improve on the second-best by suppressing the filter at chosen times.

Any thoughts?