ISSN 0439-755X
CN 11-1911/B

›› 1981, Vol. 13 ›› Issue (01): 66-77.

Previous Articles     Next Articles

THE MEASUREMENT OF PSYCHOLOGICAL FUZZINESS

Ma Mou-chao(Institute of Psychology, Academia Sinica)   

  • Published:1981-03-25 Online:1981-03-25

Abstract: The present work attempts to obtain evidence for the psychologicalreality of fuzzy sets and to apply fuzzy sets to psychological research. Usually, there are two types of characteristic curves in reacting tosuch psychological conceptualization as the size of egg namely classicalsets and fuzzy sets. If people's reaction uses a threshold-decision criterion,resulting curve is characterized by classical sets. And their S curvedistribution will support the general hypotheses for the psychologicalreality of fuzzy sets. Data obtained both from tests of eighty-four Ss, and the same testrepeated eighty-four times on the one subject demonstrated consistently thepsychological reality of fuzzy sets. It is of interest too that children's curve about size of egg is analogicto that of adults in shape, but the former is clearly to be displacing towardsmaller side of the axis, and that is probably relevant to their experiences. In the second experiment the same Ss were shown 19 lines different inlength and were given a list of 9 names in such terms as "very near","near","not far","not very far","neither far nor near","not near","not verynear","far" and"very far". The instructions were to select the term thatrepresented best what one believed it to be. Preliminary trials presentedevidence for certain operators and indicated psychological reality of"Modificator", but no "concentrator". Furthermore it is important that theZadeh's operation about'x is not very discrepant with the psychologicalmeaning which was elucidated by empirical data. The third experiment dealt chiefly with such questions as theapplication of fuzzy entropy to measurement of inference process fordifferent stages of age. It is shown that adopting nonprobabilistic entropyof information may clarify successfully the striking differences amongvarious individual intelligent levels.

Key words: NULL