zum Inhalt springen

Is low syntactic productivity associated with prediction error cost in reading? An eye-tracking study on Spanish

Mariia Baltais, Robert Hartsuiker

Ghent University

In eye-tracking during reading, unpredictable words required no additional processing cost in constraining contexts (when another word was highly predictable) compared to neutral contexts [1, 2]. But how about constraints imposed by the sentence construction rather than by semantic context? Constructions differ in the range of lexical items they occur with, i.e., in their syntactic productivity, which can be measured via the type/token ratio (TTR) [3]. For instance, the Spanish inchoative construction, which expresses the onset of an event (e.g., romper a llorar, lit. ‘break to cry’), is strikingly productive: many verbs can fill the inchoative slot, as attested in a corpus [4]. Furthermore, these inchoatives vary in productivity regarding the infinitive slot: some allow many infinitives, while others allow only few. In the latter case, does the existence of frequently occurring and thus predictable infinitives hinder processing of unpredictable but plausible ones?

We recorded eye movements of 66 native Spanish speakers during sentence reading (32 critical, 198 filler items). The neutral condition contained relatively productive inchoatives (mean TTR = 0.66, SD = 0.07) and the constraining condition contained unproductive inchoatives (mean TTR = 0.22, SD = 0.12). Importantly, infinitives in both conditions had low cloze probability, according to a norming study. There was no increase in the processing cost of the infinitive in the constraining compared to the neutral condition, despite longer total reading times of the inchoatives. Data suggest that low syntactic productivity of the inchoative does not make readers form specific lexical expectations regarding the upcoming infinitive.

Keywords: Syntactic productivity, sentence constructions, eye-tracking during reading, prediction cost, Spanish

  • 1. Frisson, S., Harvey, D. R., & Staub, A. (2017). No prediction error cost in reading: Evidence from eye movements. Journal of Memory and Language, 95, 200–214. https://doi.org/10.1016/j.jml.2017.04.007
  • 2. Luke, S. G., & Christianson, K. (2016). Limits on lexical prediction during reading. Cognitive Psychology, 88, 22–60. https://doi.org/10.1016/j.cogpsych.2016.06.002
  • 3. Baayen, R. H. (2009). Corpus linguistics in morphology: Morphological productivity. In Volume 2 (pp. 899–919). De Gruyter Mouton. https://www.degruyter.com/view/book/9783110213881/10.1515/9783110213881.2.899.xml
  • 4. Van Hulle, S., & Enghels, R. (in press). De Spaanse inchoatiefconstructie in beeld. Clusteranalyse als antwoord op het quasi-synonymie vraagstuk. Handelingen – Koninklijke Zuid-Nederlandse maatschappij voor taal-en letterkunde en geschiedenis.