Distal prosody influences lexical interpretation in online sentence processing

Distal prosody influences lexical interpretation in online sentence processing Meredith Brown (mbrown@bcs.rochester.edu) Department of Brain & Cognitive Sciences, University of Rochester Meliora Hall, Box 270268, Rochester, NY 14627-0268 Anne Pier Salverda (asalverda@bcs.rochester.edu) Department of Brain & Cognitive Sciences, University of Rochester Meliora Hall, Box 270268, Rochester, NY 14627-0268 Laura C. Dilley (ldilley@msu.edu) Department of Communicative Sciences & Disorders, Michigan State University 116 Oyer, East Lansing, MI 48824 Michael K. Tanenhaus (mtan@bcs.rochester.edu) Department of Brain & Cognitive Sciences, University of Rochester Meliora Hall, Box 270268, Rochester, NY 14627-0268 Abstract Previous work examining prosodic cues in online spoken word recognition has focused primarily on local cues to word identity. However, recent studies have suggested that sentence-level prosodic patterns can also influence the downstream interpretation of lexically ambiguous syllables (Dilley & McAuley, 2008; Dilley, Mattys, & Vinke, 2010). To test the hypothesis that these distal prosody effects are based on expectations about the organization of upcoming material, we conducted a visual world experiment using fixations to competing alternatives such as pan and panda, which differ in the presence or absence of a prosodic boundary after pan(-). As predicted, the acoustic properties of distal sentence material affected the proportion of fixations to the monosyllabic competitor beginning 200 ms after the onset of the target word. These findings support the hypothesis that expectations based on perceived prosodic patterns in the distal context influence lexical segmentation and recognition. Keywords: Prosody; spoken word recognition; lexical competition; perceptual organization; visual world paradigm Introduction Expectation and prediction are increasingly playing a major role in models of language processing (e.g., Levy, 2008; Ju- rafsky, 1996; Hale, 2001), and effects of expectations based on lexical and structural properties of sentences are well doc- umented (reviewed in Kamide, 2008). However, another pos- sible source of expectations that may be useful in language processing are global prosodic patterns, including prosodic phenomena that are distal to (i.e., several syllables removed from) the locus of processing. Such patterns may also in- fluence the interpretation of proximal sentence material by contributing to listeners’ expectations about the lexical struc- ture of upcoming material. Most work at the interface of prosody and spoken word recognition to date has focused on the effects of local prosodic cues to word identity, such as stress, duration, and the strength of local prosodic bound- aries (Cutler & van Donselaar, 2001; Salverda, Dahan, & McQueen, 2003; Christophe, Peperkamp, Pallier, Block, & Mehler, 2004). In this paper, we provide evidence that dis- tal prosody generates expectations based on perceived global prosodic patterns, and that these expectations constrain word segmentation and recognition. Background Previous work using non-linguistic auditory stimuli ranging from simple tone sequences to musical passages has shown that listeners detect pitch, temporal, and/or amplitude pat- terning in distal (i.e., non-local) auditory context. Such pat- terning influences the metrical and grouping structures that listeners perceive downstream (reviewed in Handel, 1989; Huron, 2006). For example, listeners tend to perceptually organize a continuous stream of tones of equal duration, tem- poral separation and amplitude that differ only in terms of pitch height (as high or low pitched) into repeating sequences of either high-low or low-high perceived groups, with the first element in each group perceived as accented (Woodrow, 1911; Thomassen, 1982). Although high tones are more fre- quently perceived as more prominent than low tones, the per- ceived relative prominence of high and low tones within each bitonal group depends on the context. To describe such per- ceptual proclivities in the domain of music processing, Ler- dahl and Jackendoff (1983) proposed parallelism preference rules, which state that when a musical passage contains seg- ments that are perceived as similar or repetitive, parallel parts of segments will be construed as having similar metrical and grouping structure. These studies suggest that distal regu- larities in the acoustic characteristics of auditory stimuli can influence the processing of proximal material. Reports from linguistics suggest that prosodic context of- ten exhibits regularities in pitch, duration, and/or ampli- tude. In particular, speech intonation and rhythm often show characteristics that listeners perceive as patterning (Couper-