Any leftovers from a discarded prediction? Evidence from eye-movements during sentence comprehension

ABSTRACT

We investigated how listeners use gender-marked adjectives to adjust lexical predictions during sentence comprehension. Participants listened to sentence fragments in Spanish (e.g. “The witch flew to the village on her … ”) that created expectation for a specific noun (broomstick_fem), and were completed by an adjective and a noun. The adjective either agreed (new_fem), disagreed (new_masc), or was neutral (big_fem/masc) with respect to the expected noun’s gender. Using the visual-world paradigm, we monitored looks toward images of the expected noun versus an alternative of the opposite gender (helicopter_masc). While listening to the initial fragment, participants looked more towards the expected noun. Once the adjective was heard, looks shifted toward the noun that matched the adjective’s gender. Finally, upon hearing the noun, looks were affected by both previous context and adjective gender. We conclude that predictions are updated online based on gender cues, but sentence context still affects integration of the expected noun.

KEYWORDS:

• Spoken sentence processing
• prediction
• visual-world paradigm
• anticipatory eye-movements
• lexical integration

Acknowledgements

We thank Jordi Martorell and the Proactive group for helpful discussion throughout this work.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Arella E. Gussow http://orcid.org/0000-0002-6604-0481

Efthymia C. Kapnoula http://orcid.org/0000-0001-6640-1948

Nicola Molinaro http://orcid.org/0000-0002-7549-6042

Notes

1. Although we refer to processing during target presentation as “integration”, it should be noted that we cannot draw a clear distinction between recognition and integration of the word, as these processes highly overlap (Molinaro, Conrad, Barber, & Carreiras, 2010).

2. Adjective duration did not differ between gender-marked and neutral items (M = 637 ± 87 and M = 653 ± 207 respectively; t(90.01) = −.59, p = .55).

3. Noun duration did not differ between expected and alternative items (M = 755 ± 202 and M = 778 ± 205 respectively; t(133.98) = −.65, p = .52).

Additional information

Funding

This work was partially supported by the Ministerio de Economía, Industria y Competitividad, Gobierno de España, the Agencia Estatal de Investigación and the Fondo Europeo de Desarrollo Regional (grant PSI2015-65694-P, “Severo Ochoa” programme SEV-2015-490 for Centres of Excellence in R&D), and by the Eusko Jaurlaritza (grant PI_2016_1_0014). Further support derived from the AThEME project funded by the European Commission Seventh Framework Programme, the ERC-2011-ADG-295362 from the European Research Council. This project was also supported by the Ministerio de Economía, Industria y Competitividad, Gobierno de España through the convocatoria 2016 Subprograma Estatal Ayudas para contratos para la Formación Posdoctoral 2016, Programa Estatal de Promoción del Talento y su Empleabilidad del Plan Estatal de Investigación Científica y Técnica y de Innovación 2013–2016, reference FJCI-2016-28019.