An excellent article about the various theories concerning how we recognize words can be found here. Rather than blogging it myself, I e-mailed the link to Mark Liberman, who did the heavy lifting:
Larson explains that “[i]n my young career as a reading psychologist I had never encountered a model of reading that used word shape as perceptual units, and knew of no psychologists who were working on such a model. But it turns out that the model had a very long history that I was unfamiliar with.” His paper is an attempt “to review the history of why psychologists moved from a word shape model of word recognition to a letter recognition model, and to help others to come to the same conclusion”.
I myself had read many examples of the word shape model when I was teaching myself some basic typography skills, and have dutifully been passing on this misinformation to others, so I’m glad to be corrected.
As a bonus, Mark Liberman also links to Ray Girvan’s weblog about discredited scientific theories. (It looks like a great blog, but the Atom feed is not linked from the main site, and “auto discovery” is turned off in the template. However, it is a Blogger site, so I figured it had a feed somewhere. Sure enough, it is here.)