Human semantic associations

Word association test

Rather early on, it was noted that words in the human mind are linked. American clinical psychologists G. Kent and A.J. Rosanoff [KEN 10] perceived the diagnostic usefulness of an analysis of the links between words. In 1910, the duo created and conducted a test of the free association of words. They conducted research on 1,000 people of varied educational backgrounds and professions, asking their research subjects to give the first word that came into their minds as the result of a stimulus word. The study included 100 stimulus words (principally nouns and adjectives). The Kent- Rosanoff list of words was translated into several languages, in which this experiment was repeated, thereby enabling comparative research to be carried out. Word association research was continued in [PAL 64], [POS 70], [KIS 73], [MOS 96], [NEL 98], and the repeatability of results allowed the number of research subjects to be reduced, while at the same time increasing the number of word stimuli to be employed, for example 500 kids and 1,000 mature research subjects and 200 words [PAL 64] or 100 research subjects and 8,400 words [KIS 73]. Research on the free association of words has also been conducted in Poland [KUR 67], the results of which are the basis for the experiment described below.

Computational linguistics also became involved in research on the free association of words, though at times these experiments did not employ the rigors used by psychologists when conducting experiments, for example, those that permitted the possibility of providing several responses to an individual stimulus word [SCH 12] or those that used word pairs as a stimulus [RAP 08].

There exist some algorithms which generate an association list on the basis of text corpora. However, automatically generated associations were rather reluctantly compared with the results of psycholinguistic experiments. The situation is changing; Rapp’s results [RAP 02] were really encouraging.

Finally, association norms are useful for different tasks, for example information extraction [BOR 09] or dictionary expansion [SIN 04, BUD 06].

< Prev   CONTENTS   Source   Next >