Orthographic, Lexical and Semantic Processing

点击翻译按钮获取中文摘要

One principle of many present day theories of skilled reading is that word identification is a competitive process. Specifically, the basic set of assumptions is that when a word is fixated, an orthographic code is created representing the letters in the word and the order in which they occur. That code then activates the mental (“lexical”) representation of not only the presented word but also those of similarly spelled words. Those representations then compete with one another until, typically, the mental representation of the appropriate word eventually wins the competition (i.e., remains active) with the representations of the inappropriate words being inhibited. The purpose of this proposal is to continue my examination of the processes by which the appropriate word wins this competition, the processes allowing the reader to gain an understanding of the word being read. One major question is how to characterize the orthographic coding process. What is now clear is that the set of words with representations that are at least partially activated by any particular orthographic code is extensive. For example, although the orthographic codes of "avacitno" and "vacation" have no letters in the same positions, our data show that reading the letter string "avactino" creates an orthographic code that activates the lexical representation for the word "vacation", indicating that the two orthographic codes must be somewhat similar. (Note that "avacitno" and "vacation" do contain the same set of letters even though they are in different positions in the two letter strings.) One key issue, therefore, is how to characterize the nature of orthographic coding in a way that reflects this similarity. A second key issue is how to characterize the lexical selection process, that is, the process by which the correct word wins the competition. A third key issue is understanding how this activation/competition process affects the activation of semantic (i.e., meaning) information in memory, information that allows the word's meaning to be understood. The means of examining these issues involves presenting single words or nonwords ("targets") on a computer screen following word or nonword "primes" which are presented briefly and masked so that they do not reach consciousness. Readers are timed while they overtly respond to the targets either by classifying them as words or nonwords or by responding "yes" or "no" as to whether they match some pre-specified criterion (e.g., is the presented word the word WATCH?, or is the presented word the name of an animal?). By gaining a better understanding of the nature of orthographic coding and of the competition/activation processes among representations in the reader’s mental lexicon, we hope to obtain a better foundation for understanding: a) normal reading in adults, b) how best to teach children to read and c) how to remediate reading difficulties in reading disabled individuals.