Wow - add better sensors and you get more data - who would have thought. I suspect that the there is more to this than what the author of the press release managed to write

The article abstract (https://www.ncbi.nlm.nih.gov/pubmed/29176609) says that the old "nyquist density" sensor arrays were designed to capture the theoretical maximum amount of data present. But it turns out those models were wrong.

This seems like a no-shit sort of finding and hardly seems deserving of this article’s hyperbole. You add more electrodes, you get more information with diminishing returns. This seems like the CogSci version of adding a 5th blade to a razor.