作者: Dwight W. Curtis , Lois C. Mullin
DOI: 10.3758/BF03199378
关键词:
摘要: Subjects made magnitude estimates of the average loudness pairs 1,000-Hz tones varying in sound pressure. A test fit an averaging model employing analysis variance suggested that judgments were internally consistent. However, parameters a two-stage based on assumption power transformations imposed both input and output implied nonlinear function, inconsistent with model. Additional analyses nonmetric scaling solution also was nonlinear, indicating this implication not artifact strong assumptions Large differences found among functions individual subjects, it these may have inflated error term variance, reducing its to detect violations additive Similar performed data from grayness collected by Weiss (1972).