The Measurement of Emotion Intensity: a Psychophysical Approach✽

Despite its phenomenal salience, the intensity of emotions remains a neglected issue of emotion research. The very notion has long been shown subtle and elusive, never precisely matched by physiological arousal or, on the subjective side, by the activation dimension so common among emotion space models. This paper argues in favor of reframing the whole matter into a psychophysical measurement approach. An experiment focusing on emotion words related to seven basic discrete emotions is reported, that relies on a cross-modal matching paradigm to check the validity and quantitative properties of the scaling achieved. Considerations that put the method’s contribution in perspective are also offered. A surprising feature of emotion studies is the enduring neglect of intensity issues. As underlined by Frijda et. al. (1992), the fact is all the more surprising as it contends with the well-acknowledge relevance of felt intensity to the overall experience of emotion. Moreover, it also collides with the central role assigned to an intensity component, on more or less a priori grounds, by several influential theories of emotions (Plutchick, 1982; Reisenzein, 1994). Two severe consequences of this neglect are (1) the risk of a widespread confounding of intensity effects all across the field of emotion studies (see Levenson, 1988, for a description of the problem within the scope of ANS physiological research on emotions) and (2) the constraints imposed by resorting to an unmeasured independent variable, whenever the subject matter turns out to be the impact of emotions on other variables. The experiment here reported was meant as a first exploratory step towards a quantified approach of emotion intensity. Focusing on emotion words, it asks the following question: Can subjects validly and consistently rate overall intensity of a given expressed emotion? This question serves two functions: (1) it addresses the meaningfulness of the very concept of expressed overall emotional intensity, hereby recasted as a measurement problem, and (2) it embodies the view that the measurement of overall intensity stands as an empirical issue to be addressed on its own right, not to be dismissed by a priori assumptions regarding the «loose» multidimensionality of emotions. ✽ This research was partially supported by the Foundation for Science and Technology of Portugal. Grant BD 19487 to Francisco M. Cardoso.