Combining Multiple Psychophysiological Measures of Listening Effort: Challenges and Recommendations

Abstract

About one-third of all recently published studies on listening effort have used at least one physiological measure, providing evidence of the popularity of such measures in listening effort research. However, the specific measures employed, as well as the rationales used to justify their inclusion, vary greatly between studies, leading to a literature that is fragmented and difficult to integrate. A unified approach that assesses multiple psychophysiological measures justified by a single rationale would be preferable because it would advance our understanding of listening effort. However, such an approach comes with a number of challenges, including the need to develop a clear definition of listening effort that links to specific physiological measures, customized equipment that enables the simultaneous assessment of multiple measures, awareness of problems caused by the different timescales on which the measures operate, and statistical approaches that minimize the risk of type-I error inflation. This article discusses in detail the various obstacles for combining multiple physiological measures in listening effort research and provides recommendations on how to overcome them.

Keywords: listening demand; listening effort; psychophysiological measure.

Figure 1

Time characteristics of selected physiological measures. Dark gray lines indicate continuous measures; dark gray dots indicate noncontinuous measures. Dark gray dots with surrounding light gray boxes indicate noncontinuous measures that refer to time periods and not to specific points in time. The light gray boxes indicate the measurement epochs required to obtain the measure.