| Popis: |
Speech-in-noise (SiN) research typically distinguishes two types of masking caused by background noise: energetic masking (direct interference between the target speech and background noise in the cochlea) and informational masking (interference higher in the auditory pathway) [1]. We have coined the term "informational interference" (inf-int) to refer specifically to high-level informational masking involving linguistic and cognitive factors and which is influenced by long-term knowledge. For example, unmanipulated, intelligible background speech is more distracting than an unintelligible, time-reversed version of that same speech [2]. In a series of recent studies (https://osf.io/csjya, https://osf.io/dkh7m, https://osf.io/yt8w9, https://osf.io/c46se), we attempted to isolate the locus of inf-int using word-list maskers. In their unmanipulated form, these maskers contain lexico-semantic information but no sentence-level syntax or prosody. In each study, we examined the difference in SiN performance for intelligible (forwards) word-list maskers compared to unintelligible but acoustically-matched (time reversed) versions of those maskers. Results showed that performance was no worse for the intelligible maskers. This is in direct contrast to the literature on informational masking, which would suggest that an intelligible speech masker should be more detrimental than an unintelligible equivalent [3]. These results may suggest that features associated with connected speech, such as sentence-level syntax or prosody, are required to give rise to inf-int. However, there are a number of methodological differences between our experimental set-up and previous studies which have demonstrated inf-int. To rule out the possibility that the null results obtained thus far with word-list maskers were due to an idiosyncratic feature of our experimental set-up, the current study will replicate the set-up of our previous studies but using sentence-based (i.e., connected speech) maskers. Specifically, it will compare an intelligible, forwards sentence-based masker (SENT condition) to an unintelligible, time-reversed version of that masker (rSENT condition). In all other respects – target materials and talker, signal-to-noise ratios, spectral matching etc. – the stimuli will be identical to those used previously (https://osf.io/dkh7m). [1] Mattys, S. L., Davis, M. H., Bradlow, A. R., & Scott, S. K. (2012). Speech recognition in adverse conditions: A review. Language and Cognitive Processes, 27(7-8), 953-978. [2] Freyman, R. L., Balakrishnan, U., & Helfer, K. S. (2001). Spatial release from informational masking in speech recognition. The Journal of the Acoustical Society of America, 109(5), 2112-2122. [3] Kidd Jr, G., Mason, C. R., Swaminathan, J., Roverud, E., Clayton, K. K., & Best, V. (2016). Determining the energetic and informational components of speech-on-speech masking. The Journal of the Acoustical Society of America, 140(1), 132-144. |
