ABSTRACT
This chapter reviews in some methodological detail two studies concerned with the psychophysics of the perception of one component in a mixture of odours; the first involves binary and the second ternary mixtures. The data of psychophysical experiments are frequently degenerate in that subjects only use a few responses out of a wider available choice. There are two ways of analyzing the dynamic relationships in these time series, each is informative and they can be mutually supportive. Information transmission in the time-independent sense of a stimulus-response plot as measured by a linear regression is virtually all destroyed in the AB group data. The intensity of one target component within an odour mixture of three qualitatively distinct component substances is a problem that has been spasmodically attacked since early work of Zwaardemaker and von Skramlich. The qualitative picture is consistent, and is compatible with a claim by von Skramlich that the qualities of components can get phenomenologically lost in a mixture.
