Empirical findings about the perceived interaction of the components of qualitatively complex taste stimuli are reviewed and summarized. The revision of multidimensional scaling equations to incorporate a simulation of a mutually inhibitory or facilitatory interaction between components is discussed. Data from a previously published study on the perception of similarity among taste mixtures are re‐analysed for illustrative purposes and a description of them is shown to be less adequate when provision is made for one form of interaction than when this is omitted. A reclassification of stimulus‐response paradigms for studying taste mixture perception is outlined; and it is suggested that the interaction of taste components may be an artifact of the response measures u