Fifteen subjects scaled monochromatic lights ranging from 460 to 640 nm by assigning percentages to the names blue, green, yellow, and red, representing the proportion of these hues perceived in the lights. The resulting hue-naming functions were compared to those predicted from the opponent-chromatic response functions of Werner and Wooten's (1979b) proposed average observer. The agreement between the two sets offunctions was reasonably good, which strengthens the validity of their average observer. Jameson and Hurvich (1955) were the first to measure quantitatively 'the spectral responsivity of the opponentchromatic channels, which were initially proposed by Hering (1920Hering ( /1964 in his opponent-colors theory. These quantitative measurements have since been replicated by others (Romeskie, 1978; Werner & Wooten, 1979a), using the same hue-cancellation technique as that used by Jameson and Hurvich.Werner and Wooten (1979b) have presented opponentchromatic response functions for the average observer. The functions are based on the data of seven subjects from three studies Romeskie, 1978; Werner & Wooten, 1979a). The importance of these functions, as Werner and Wooten (1979b) point out, is that they are useful for quantitative modeling when assumptions must be made about a standard or average psychophysical observer. Before such functions can be used with confidence, however, their validity must be tested. One way to do this is to examine the extent to which the opponent-ehromatic response functions for the average observer predict hue naming for a large group of subjects. Werner and Wooten (1979a) have shown individual opponent-chromatic response functions to be good predictors of the same individuals' hue-naming data. These same subjects' hue-naming functions have, in fact, been compared by Werner and Wooten (1979b) to the predicted functions of their average observer. But the average observer is based partly on these same subjects, so it is not surprising that the comparison was favorable. What is needed is a comparison of the predicted hue-naming functions from Werner and Wooten's average observer with those functions empirically obtained from a completely different population of subjects. That was the main purpose of the present study. Also, the subjects upon whom the average observer is based were not all experimentally naive and were all reasonably well practiced. Do their data generalize to naive and virtually unpracticed subjects? In the present study, this question was addressed as well.The authors' mailing address is: Department of Psychology, University of New Hampshire, Durham, NH 03824.
145
METHOD
SubjectsFifteen male, undergraduate students, enrolled in an introductory psychology course at the University of New Hampshire, served as subjects. Their color vision was assessed as normal on the basis of results from the administration of the Ishihara Pseudoisochromatic Test Plates and the Farnsworth Dichotomous Panel D-15 Test. The subjects ranged in age from 18 to 26 years.
ApparatusA one-channe...