The authors aimed to determine whether hearing impairment affects sentence comprehension beyond phoneme or word recognition (i.e., on the sentence level), and to distinguish grammatically induced processing difficulties in structurally complex sentences from perceptual difficulties associated with listening to degraded speech. Effects of hearing impairment or speech in noise were expected to reflect hearer-specific speech recognition difficulties. Any additional processing time caused by the sustained perceptual challenges across the sentence may either be independent of or interact with top-down processing mechanisms associated with grammatical sentence structure.
Forty-nine participants listened to canonical subject-initial or noncanonical object-initial sentences that were presented either in quiet or in noise. Twenty-four participants had mild-to-moderate hearing impairment and received hearing-loss-specific amplification. Twenty-five participants were age-matched peers with normal hearing status. Reaction times were measured on-line at syntactically critical processing points as well as two control points to capture differences in processing mechanisms. An off-line comprehension task served as an additional indicator of sentence (mis)interpretation, and enforced syntactic processing.
The authors found general effects of hearing impairment and speech in noise that negatively affected perceptual processing, and an effect of word order, where complex grammar locally caused processing difficulties for the noncanonical sentence structure. Listeners with hearing impairment were hardly affected by noise at the beginning of the sentence, but were affected markedly toward the end of the sentence, indicating a sustained perceptual effect of speech recognition. Comprehension of sentences with noncanonical word order was negatively affected by degraded signals even after sentence presentation.
Hearing impairment adds perceptual processing load during sentence processing, but affects grammatical processing beyond the word level to the same degree as in normal hearing, with minor differences in processing mechanisms. The data contribute to our understanding of individual differences in speech perception and language understanding. The authors interpret their results within the ease of language understanding model.
1Cluster of Excellence “Hearing4all”, 2Institute of Dutch Studies, and 3Medizinische Physik, University of Oldenburg, Oldenburg, Germany.
Portions of the data presented in this article were presented at the annual meeting of the German Association of Audiology, Erlangen, March 08–12, 2012..
The authors declare no further conflicts of interest.
Received November 8, 2014; accepted April 28, 2016.
Address for correspondence: Rebecca Carroll, Institute of Dutch Studies, University of Oldenburg, Ammländer Heerstraße 114–118, 26111 Oldenburg, Germany. E-mail: firstname.lastname@example.org
Received November 8, 2014
Accepted April 28, 2016