Focus projection and prenuclear accents: evidence from lexical processing

ABSTRACT

Theories of Focus Projection claim that a single pitch accent on a verb's argument is sufficient to prosodically mark that verb as part of the focus, negating the need for a prenuclear accent on the verb itself. The present study employed online lexical processing to test this claim empirically. In three cross-modal associative priming experiments, listeners heard English SVO sentences with/without prenuclear accenting on the verb in both broad (VP) and narrow (object) focus contexts. Results showed that the absence of a prenuclear accent in broad focus contexts did not disrupt priming, but the presence of one in narrow focus contexts did. This disruption was found to be somewhat modulated by individual differences in “autistic traits”. Overall, the findings are interpreted as supporting a model that includes both (a) a Focus Projection mechanism and (b) an information structural function for prenuclear accents, with the latter possibly subject to cross-listener variation.

KEYWORDS:

• Focus projection
• prosodic prominence
• cross-modal priming
• lexical processing
• autistic traits

Acknowledgements

The author is especially grateful to Megha Sundara, whose methodological recommendations crucially enhanced the experiments presented here, and to the Associate Editor and two anonymous reviewers, whose comments greatly improved the resulting manuscript. I also thank Sun-Ah Jun, Patricia Keating, and other members of the UCLA Phonetics Laboratory for their comments at various stages; Henry Tehrani for writing the MATLAB script used to present stimuli; and Hannah Bower and Aiko Hieda for help running the experiment.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Throughout this paper, unless otherwise noted, pitch accent annotations correspond to the categories used in the ToBI conventions (Beckman & Hirschberg, 1994).

2. It should be emphasized here that the target constructions in the present study – English SVO sentences – represent only one of a wide range of structures for which Focus Projection theories aim to model focus-to-prosody mappings. Despite the remarkable empirical coverage that each of these theories enjoys, however (see Büring, 2006 and Gussenhoven, 1999 for overviews), it is this simple construction that continues to pose some of the most interesting challenges. Most pertinent to the present interests in prenuclear accent placement, SVO constructions are also an area where the two prominent models of Focus Projection do not agree.

3. For a recent and similar characterization of the function of pitch accents, based on predictability, see Beaver and Velleman (2011), as well as relevant discussion in Pan, McKeown, and Hirschberg (2002), Watson, Arnold, and Tanenhaus (2008), Calhoun (2006), and Turnbull (2016).

4. In connecting Birch and Clifton's study to the present discussion, one note of caution is necessary. According to the authors, verbs in their test sentences were not prenuclear accented, since an intermediate phrase boundary separated the accented verbs and their accented objects, and so both were nuclear accented in their stimuli. However, to the extent that they tested Focus Projection from nuclear accented arguments to relatively prominent versus relatively less prominent verbs, their results are relevant here. It is also unclear how salient the intervening phrase boundaries were in their stimuli, and thus whether listeners did not actually perceive the accented verbs in their experiments as prenuclear.

5. See also Peters, Hanssen, and Gussenhoven (2014) for dialects of Dutch and German, though the effects they report are small for focus size, possibly due in part to their conflating of focus type (contrastive/non-contrastive) and focus size (broad/narrow) in the structures they investigate. Additionally, they exclude some prosodic structures from their analysis, namely those with downstepped nuclear accents, though the authors report this to be a small number of exclusions overall.

6. Throughout the rest of this paper, examples of auditory primes will be shown in italics, visual targets in small caps.

7. For a recent and intriguing investigation into the mechanisms underlying priming of contrastive associates, see Husband and Ferreira (2016).

8. See Yu (2010) and Turnbull (2015) for discussion of a number of findings not discussed here.

9. All p-values shown in this study were generated using lmerTest (ver. 2.0–32; Kuznetsova, Brockhoff, & Christensen, 2016), and also checked against the method described in Baayen (2008, Chapter 7), which did not produce different results.

10. Although the focus × prime-type interaction term was not found to contribute to model fit (and was therefore discarded), a model was also constructed that included this interaction, as an additional test for its significance. Even in this model (which otherwise contained the same factors as the best-fitting model), the focus × prime-type interaction was not significant (B = −8.391, SE = 6.087, t = −1.38, p > .1)

11. As for Experiment 1, a model was also tested that included the interactions of the two linguistically interesting variables (here, prosody × prime type), even though it did not contribute to the fit of the model. This interaction was not found to be significant (B = −3.730, SE = 10.080, t = 0.37, p > .1).

12. A reviewer asks whether one of the prosodic conditions might have suffered an unnaturalness penalty, since only one was created through editing (i.e. cross-splicing was not used). However, this is clearly ruled out, since a priming difference between the two prosodic conditions was found in only one context (namely in narrow focus contexts, Experiment 3), and in this case it was not primes embedded in the edited sentences that failed to prime, but rather those embedded in the unedited (i.e. [+ Prenuclear Accent] sentences.