Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Human Eyes See Cosmic Bubbles that Algorithms Miss

  1. Mar 11, 2012 #1


    User Avatar
    Gold Member

    I thought this story was charming, since it hints that there still might be a few things people can do that computers cannot (yet) do. I have no idea of the significance of these bubbles, but I'd be interested to hear others comments.

    More than 5,000 space bubbles have been discovered in the disk of our Milky Way galaxy by a team of part-time citizen scientists.

    These bubbles are blown by young, hot stars into the surrounding gas and dust, and indicate areas of brand-new star formation, scientists say.

    "These findings make us suspect that the Milky Way is a much more active star-forming galaxy than previously thought," Eli Bressert, an astrophysics doctoral student at the European Southern Observatory, said in a statement. "The Milky Way's disk is like champagne with bubbles all over the place."

    Respectfully submitted,
  2. jcsd
  3. Mar 12, 2012 #2


    User Avatar
    Gold Member

    The human brain is absolutely astounding in its ability to do pattern recognition. To me the surprise is not at all that there are "still" things the brain can do that algorithms can't (yet) but rather than algorithms can do as much as they CAN at this relatively early stage of computer development.

    My favorite example is this: Have you ever seen a VERY simple line drawing that represents, say, Bob Hope, or Alfred Hitchcock? People have recognized those drawing on seeing them for the first time but it's pretty much unthinkable that a computer could come anywhere close to doing that.

    EDIT: I once read an article that explained in some detail how pattern recognition is a survival mechanism in humans (and other animals), and also how it causes the "false positives" that sometimes make us think we are seeing something that on closer inspection turns out to be a random pattern that we misinterpreted.
  4. Mar 12, 2012 #3


    User Avatar
    Gold Member

    It continually fascinates me how little information I need to be able to identify a familiar face in a crowd. A split second flash of 25% of their face, (2% of the whole person), from virtually any angle, in a database of thousands of people I know by sight.
  5. Mar 13, 2012 #4
    In January, the NASA Kepler team decided to let citizens take a look at data that their algorithms had already gone through. In form of graphs generated by the brightness of the stars. The point was to spot a pattern in the transitions, that could possibly prove to be a planet or a binary star. And again, human pattern recognition had its say. I think it actually led to the discovery of one or more planets. The one I heard of was Neptune sized..
  6. Mar 26, 2012 #5
    Yeah. Or how quickly we can identify a melody. Sometimes two or three notes will do it.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook