The Science of Word Recognition

"The Science of Word Recognition" is a comprehensive rebuttal of the word shape theory of word recognition, as first posited by James Cattell in 1886.  At the time, this theory, which explains that words are recognized by their shape rather than by individual letters, was based on 4 experimental findings:  1.  the Word Superiority Effect (letters are recognized better in the context of words), 2. the fact that we read lowercase letters at a faster rate than uppercase letters, 3. misspellings are frequently overlooked when the shape of the overall word is not changed by the mistake (tesf vs. tesc), and 4. text that is a mix of upper and lowercase letters is difficult to read.  However, these 4 findings have other logical explanations.  For instance, the context provided by the word itself is what allows the reader to remember the letters in a word.  Take, for example, acronyms.  When acronyms are used to study, like KPCOFGS (King Philip Came Over From Geneva, Switzerland:  Kingdom, Phylum, Class. Order, Family, Species), it is the meaning associated with the information that allows it to be retained, rather than the shape of the word.  Finding 2 is explained by "practice effect"--the idea that we are more practiced, due to our grammatical conventions, of reading mostly lowercase letters.  If we were to practice reading in uppercase letters, we would get faster at that over time.  Finding 3 is explained by the fact that the effects of letter shape and word shape can't be separated.  In other words, in "tesf" the "f" is very similar to "t" (which makes the word have the same shape as the target word "test").  However, this similarity also calls to mind the missed letter, "t."  The ability of our minds to make cognitive leaps when reading calls to mind what I do when reading phonetic spellings--I can figure it out when it is close enough, and if it is pretty close, it doesn't slow me down at all, though I do notice the error.  Finally, in relation to the 4th finding, I have a friend who makes her facebook status updates in alternating caps, and I often don't read them.  They are difficult to read and difficult to look at.  I equate this to messy handwriting--unfamiliar variations are a strain. 

Larson explains the other major theories of word recognition, one of which I had never heard of before: serial word recognition.  This theory (briefly) posited that each letter was read and processed one at a time, from left to right.  However, eye movement research put an end to this idea, as researchers realized that the eye actually fixates and saccades in a side to side movement, supporting the current theory of word recognition, Parallel Letter Recognition.  This theory, supported by both Perfetti's Verbal Efficiency Thoery(1985) and Adams's (1990) model of reading, explains that the recognition of individual letters occurs in parallel as the eye fixates on several letters (and words) at a time.  I am familiar with this information (lengths of fixations, the ways our eyes move when we move, and so forth), but I had not read the research it came out of, so it was interesting to learn about the tests used:  The Moving Window Study and the Boundary Study.

I know from the mass email I get that says something like "If yuo cna raed tihs" and is an argument for the idea that we don't read each letter of a word that the idea that we don't read each letter is still alive and well.  (This concept is further discussed here.)  Spear-Swerling and Sternberg's (1996) stage theory of reading explains that visual-cue readers, who are pre-alphabetic, do somewhat recognize words in this manner; for example, small children will recognize the McDonald's logo.  Perhaps this developmental stage and the teaching of sight words to children propagate the myth that we skip letters and even words.  However, as Adams (1990) asks, "...Can they learn 50,000 [words]?" (p. 1229). 

One question I have:  What is the etymology of the phrase "Bouma" used to represent the idea of word shape?

Another question I had regards the functor words Larson mentions.  He says that these are sometimes skipped, although they are at least partially read with shorter words in fixations.  I wonder if this is one reason these tend to be the words most often missed by (at least older) struggling readers?  These "small" words slay my students' accuracy scores.  I know part of what makes them difficult to remember is that, although they can carry a lot of meaning in a sentence, these words have less concrete meaning than content words, making them harder to remember.

Finally, if we know that students don't learn words by their shapes, why do some teachers encourage children to learn words this way, as evidenced by the popular worksheets found at abcteach.com and enchantedlearning.com.  It infuriates me that there are so many misguided notions regarding learning to read, and that so often those who teach are kids to read don't have proper training.  It is our kids who suffer the most.