3 ms·

The actual article is not nearly as shocking as the hn title. They have not been able to implant any thougths yet. What they did was merely put people in a ment

• mistercow 14y ago
  No, they didn't even do that. What they did, in a nutshell, is show strong evidence that the adult visual cortex can be trained. There's a lot of discussion about "induction" here, but the only intervention here was presenting them with images. The experiment goes like this (as far as I can tell): 1. Have a bunch of subjects perform a multiple choice visual identification task and scan their visual cortex with an fMRI while they do it. 2. Build a statistical model that lets you predict, based on an fMRI, which of the three answers the respondent gave. 3. Have the subjects do more of the same kind of test, repeatedly, but without answering; just have them stare for six seconds. Use the statistical model to classify which answer they would give, and show them a circle that gets larger depending on how strongly the model classifies the answer as correct. Ask subjects to maximize the size of this circle without telling them what it means. 4. Repeat step 1 and measure the change in accuracy. What did they find? The subjects improved at the task, even though they didn't know what they were supposed to be doing. Incredibly cool, and completely unrelated to "direct to brain" learning.