Abstract
Response time–based dual-task paradigms are commonly adopted to measure behavioral
listening effort. Most extant studies used an all-response approach that included
secondary task responses under both correct and incorrect primary task responses during
analysis. However, evidence supporting this strategy is limited. Therefore, the current
study investigated the potential differences between including all responses versus
only including correct responses. Data from two previous studies were reanalyzed.
Experiment 1 included 16 listeners and used a dual-task paradigm to examine the effect
of introducing background noise on listening effort. Experiment 2 included 19 participants
and used a different dual-task paradigm to examine the effect of reverberation and
loudspeaker-to-listener distance on listening effort. ANOVA results obtained using
both analysis approaches were compared. The all-response and correct-only approaches
revealed similar results. However, larger effect sizes and an additional main effect
were found with the all-response approach. The current study supports the use of an
all-response approach due to its greater sensitivity to changes in behavioral listening
effort. However, a correct-only approach could be utilized to suit specific study
purposes.
Keywords
listening effort - dual-task paradigm
