Perceptual ratings of opposite spatial properties: Do they lie on the same dimension?

The issue of unidimensionality is dealt with in various research areas in the field of Psychology (e.g. conceptual spaces, semantic modeling, psychometrics) and always involves spatial modeling. An investigation of the dimensionality of opposite spatial scales (even basic) has however not yet been carried out. In this paper we look at whether opposite judgments of height, size, width and length (high/low, large/small, wide/narrow, and long/short) imply underlying unidimensional continua. In three experiments, independent ratings for the 8 above mentioned properties were elicited with participants looking at photographic representations of various objects (Study 1), real life objects (Study 2) and spatial extensions in object-independent conditions (Study 3). Explorative and confirmative factor analysis and Andrich Extended Rating Scale Models were applied in order to determine whether the ratings referred to opposite scales on the same linear continuum. Results from the three studies consistently revealed that this is not the case. A joint analysis of the data showed interesting interactions between the spatial properties analyzed suggesting a possible explanation for the lack of unidimensionality.

► We studied whether perceptual spatial opposites imply unidimensional continua.
► Independent ratings for opposite properties (e.g. high/low, large/small) were asked.
► Factor Analyses and Andrich Extended Rating Scale Models were applied.
► Results revealed that opposite ratings did not lie on the same continuum.