Another group does it directly from fMRI: http://www.ncbi.nlm.nih.gov/pubmed/20460157 http://www.ncbi.nlm.nih.gov/pubmed/20460157 Among the caveats: really hig

Another group does it directly from fMRI: http://www.ncbi.nlm.nih.gov/pubmed/20460157 http://www.ncbi.nlm.nih.gov/pubmed/20460157 Among the caveats: really high-T magnet, not feasible for general use; the visual stimulus was actually a warped version that would be reproduced as the nicely shaped activation upon retinotopic projection (it's a log-polar mapping)... Still really cool. (ps: no-paywall version here: surfer.nmr.mgh.harvard.edu)