Does irrelevant music cause an irrelevant sound effect for auditory items?

Abstract

The irrelevant sound effect (ISE) describes the significant reduction in verbal serial recall during irrelevant sounds with distinct temporal-spectral variations (changing-state sound). Whereas the ISE is well-documented for the serial recall of visual items accompanied by irrelevant speech and nonspeech sounds, an ISE caused by nonspeech sounds has not been reported for auditory items. Closing this empirical gap, Experiment 1 (n=90) verified that instrumental staccato-music reduces auditory serial recall compared to legato-music and silence. Its detrimental impact was not due to perceptual masking, disturbed encoding, or increased listening effort, as the employed experimental design and methods ensured. The found nonspeech ISE in auditory serial recall is corroborated by Experiment 1b (n=60), which, by using the same experimental design and methods, replicated the well-known ISE during irrelevant changing-state speech compared to steady-state speech, pink noise, and silence.

Acknowledgements

The authors would like to express their appreciation to Prof. Dr Alan Baddeley and two anonymous reviewers for helpful suggestions on an earlier draft of this paper and to Dr Alfred Zeitler for programming and servicing the experimental software. The experiments were funded by the Deutsche Forschungsgemeinschaft (DFG; German Science Foundation).

Notes

¹In Experiment 1 and 1b, significant effects on within-subject factors and their interactions are tested on homogeneity of variance with Mauchly's Test of Sphericity, using χ²-tests for estimation. In the case of a significant χ²-test, the degrees of freedom of the corresponding F-value are corrected with Greenhouse-Geisser's ϵ, if ϵ<.75 (according to common convention).

²Further results of the ANOVA follow: presentation modality, F(1, 38) = 0.24, MSE=4218.88, p=.626; serial position, F(3.37, 128.22) = 96.93, MSE=1014.79, p<.001, partial η²=.72; serial position×presentation modality, F(8, 304) = 8.67, MSE=428.00, p<.001, partial η²=.19; serial position×sound condition, F(11.58, 439.96) = 3.00, MSE=150.45, p=.001, partial η²=.07; sound condition×serial position×presentation modality, F(24, 912) = 2.43, MSE=72.58, p < .001, partial η ² =.06.

³Further results of the ANOVA follow: onset, F(1, 38) < 1, MSE=3112.27; serial position, F(3.55, 135.05) = 143.59, MSE=821.41, p<.001, partial η²=.79; onset×serial position, F(8, 304) = 1.54, MSE=364.91, p=.142; serial position×sound condition, F(12.33, 468.43) = 4.87, MSE=144.30, p<.001, partial η²=.11; onset×sound condition×serial position, F(24, 912) = 1.54, MSE=144.30, p=.048, partial η²=.04.