Visuele woordherkenning bij het lezen

One of the most important things we do durlog reading is recognizing words. This process, in wbich elementary visual perception and linguistic knowledge combine harmoniously, bas intrigued experimental psychologists from the start, causing word recognition to be investigated thoroughly as early as the 1880's. Though initially the general consensus was that reading consisted of spelling out the letters consecutively, it soon became clear that a word is percei-.:ed as a whole, but there remained the problem whether or not within this whole the constituent letters played an important role. Although Cattell, and Erdmann and Dodge supposed that words are only perceived as wboles, a closer inspeetion of their experimental conditions reveals that their explanations are neither necessary nor sufficient for their data. But in retrospect, these workers, together with their contemporanes who advocated a more analytic approach basedon letter recognition, had already obtained a comparatively coherent view of the perceptual aspects of reading. Here the work of Huey was of great value by its inlegrating power. It is remarkable that changes have been continuously carried through in reading education from the beginning of the 19th century, that were oot at all or hardly related to empirica! research in word recognition. After a long period of experimental and educational neglect word recognition was adopted in the· common theory of learning and perception: the New Look, in the fifties. Newly discovered properties of iconic memory made clearer why words are so much better perceived than unrelated strings of letters. However, it was oot until the end of the sixties that word recognition was explicitly studied again in the context of reading; in the Netherlands particularly by Bouma. During this period a number of quantitative word recognition models evolved, having considerable descriptive power. All these models have in common that elementary features or letters of words are identified, independently of each other, and are combined by the reader's word knowledge. In this way the analytic aspects of recognition are integrated with tbe perception of the whole. But oot only the identity of the lettersappears to be important; their location within the wordseemsas importantastheir identity. Basically, this is the simplest form of spelling: which letters are where, irrespective of their neighbours. Considering this tunetion of the place of letters in word recognition perceptual arguments in spelling reforms, heretofore largely neglected, seem of paramount importance. Finally, to elucidate the role of word recognition in reading further investigations are needed on the comprehension of continuous text durlog individual eye-fixations.