As I have recently been working to solve some headline typeface issues in the prototype project I went back to some notes I made earlier on the topic of word recognition and vision. From a usability aspect I have always been interested in the human eye and how we actually read letters and words, when we look at web pages we tend to scan trough the interface in somewhat similar manners, but when it comes to reading words and letters it’s a totally different matter.
In terms of web typography the media allows for a more individual way of reading with technical aids like zooming and contrast adjustments helping each user in the reading process, it’s a more dynamic way of reading compared to the analogue media. I have been however been curious to how the eye interprets words and sentences, do we read the letters one by one or do we compose words from the shapes of letters? There have been some theories on this subject trough the years and I found some interesting research on this in the book Basics typography 01. Virtual typography (Hillner M. 2009) that covers the most acknowledged word recognition theories.
The Bouma theory
Herman Bouma published his theories in 1973 in a paper called “Visual interference in the Parafovial Recognition of initial letters of words.” The Bouma theory assumes that text is perceived word by word rather than letter by letter. “The Bouma model evolved from evidence that was gathered in the field of cognitive psychology since the 1979s. Contemporary physiologists use the word Bouma synonymously with word shape in tribute to Herman Bouma.”
Bouma claims that we look at the centre of a word while looking at the surrounding letters using the peripheral vision. His theory is that we perceive the words as shapes and that we don’t read letter-by-letter. Much like when we look at a logo and the way it communicates through its overall shape and outline rather than through their precise spelling.
“The fovea is the location at the back of the eye that allows us to depict objects in detail. Surrounding the fovea is the parafoveal area that is responsible for our peripheral vision.”
Word by word
The first to suggest that we read words instead of just single letters was in 1886 the psychologist James Cattel. He did tests that supported this in conjunction with spelling mistakes. The spelling mistakes were higher when it was consistency in the word shape. So when shapes looked the same they were misread more often than if the word shapes were inconsistent and varied.
Serial letter recognition
“Other theories on this matter is the “Serial letter recognition” that claims that words are scanned letter by letter, starting with the one furthest left. It also remerged in the 70s but it was short lived due to the so called word superiority effect, which revealed the letters are more easily identified as part of a word than isolated.”
Parallel letter recognition
“The parallel letter recognition theory, developed in 1989, is the most recent and remains the most popular. This theory assumes that all the letters of a word are registered simultaneously and analysed in conjunction. The ability to perceive various letters simultaneously is apparently aided by the foveal recognition”.
This parallel letter recognition supports the fact that we see letters that’s in focus and not the ones that are in the corner of our eye, but we only register the things that are barely out of eye sight to the corners. The foveal area is particularly sensitive to motion, and that supports that it’s possible to guide the viewer’s focus of attention.
Whether the Bouma model deserves more credibility than the parallel letter recognition theory or visa versa, or if a combination of both models would be most appropriate, remains a highly contentious issue. Either way, people read word by word.