I’m looking for a person who knows about, or research papers on the subject of, human perception of simple shapes, and distinguishing between them. I’d like to produce a set of, say, 64 shapes or glyphs which are all as different from one another as possible.
If anyone knows of something I should read or someone I should talk to, please let me know.
You might want to search google for “bongard problems”. But you’re likely to know about them already.
this might be useful, generated distinct shapes for use as icons:
http://www.idiom.com/~zilla/Work/VisualIDs/visualids.pdf
Probably totally unrelated to your topic. But I’ve once read a research paper which claimed that people can more easily distinguish words than icons. Ever since I’ve set my toolbars in mozilla to “text only”. And I must admit: I love it.
Just as an interesting side note (but not specifically what you want), this article here details the research done by Xerox when inventing the original GUI. It has a section describing in good detail user reaction tests regarding several sets of icons they used and a graph of speed / recognition.
I remembered seeing years ago (SIGGRAPH?) papers on decomposition of 3d shapes into a shape alphabet, went digging… you want to google for ‘geon’, ‘gsd’ (geon structural decomposition), and ‘Biederman’. His research was in the late 80s/early 90s and most of the recent papers I found seemed skeptical. But its a set of 24 shapes, sounds like the kind of thing you are after.
If you’re really interested in going whole hog into this thing, you may want to look at the research of James Gibson, dating back to the late 50’s and 60’s. Particularly his thorough “The Senses Considered As Perceptual Systems.” Among many intereseting points, Gibson argues that GRADIENTS OF TEXTURE are critical to recognizing things like contour and depth.
Not that I have anything useful to point you to, but if you’re thinking of applying this to web sites as an anti-phishing strategy, it feels a lot like file manager emblems: http://www.gnome.org/learn/users-guide/2.6/ch07s04.html
If you know someone who has the Oriely book ‘brain hacks’ they have a large section on there about human vision. There is a wonderful list of references including many research papers about human visual recognition of shapes. I would say 1/3 of the book covers how the brain handles vision.
Hi Gerv,
Long time no speak. Clearly one person who knows lots about shapes is Roger Penrose (of Oxford Maths department fame – tessalations are his speciality). On the perception side it is a very complex field. JJ Gibson’s work is useful but old. Professor Richard Gregory at Bristol might have something useful to say. Depending upon how simple you want the glyphs to be 64 may be a bit ambitious. Depending on what you wanted to use them for I may be a bit more useful.