Re: Readability of scrambled text

Mark J. Reed wrote at 2003-09-18 15:33:10 (-0400)
 > On Thu, Sep 18, 2003 at 03:02:34PM -0400, Estel Telcontar wrote:
 > > >>Anidroccg to crad cniyrrag lcitsiugnis planoissefors at an
 > > >>uemannd utisreviny in Bsitirh Cibmuloa, and crartnoy to the
 > > >>duoibus cmials of the ueticnd rcraeseh, a slpmie, macinahcel
 > > >>ioisrevnn of ianretnl cretcarahs araepps sneiciffut to csufnoe
 > > >>the eadyrevy oekoolnr.
 >
 > Fair enough, but it's still interesting that random arrangements of
 > the internal letters, at least up to some limited distance from
 > their original position, causes almost no slowdown in reading.
 > Complete inversion of the above sort is another kettle of fish
 > entirely, and I'm not sure that its effects are really crartnoy to
 > the cmials of the ueticnd rcraeseh.  Maybe if someone actually
 > cetid that rcraeseh we could find out. :)
 >
 > -Mark
 >

This has been the subject of some investigation by the author and
readership of one of the weblogs I habitually read.  The "rsereach"
has been tentatively identified[1] as a 1976 University of Nottingham
PhD dissertation entitled _The significance of letter position in word
recognition_.  This conclusion is based on a similar text[2]
circulated previously, which turns out to be an excerpt of a letter to
_New Scientist_ by Graham Rawlinson, the author of said dissertation.
If the later message (which sparked this thread, and others across the
net) is the result of someone reworking the earlier one to disguise
its origins (which seems likely) then the research mentioned is
presumably Rawlinson's (other research is mentioned in the original
letter, by Saberi and Perrot[3], but neither was working in England).

[1] http://www.bisso.com/ujg_archives/000228.html
[2] http://www.bisso.com/ujg_archives/000060.html
[3] http://www.bisso.com/ujg_archives/000224.html