Why bother? What our eyes tell about psych verbs' (non) causative constructions"

PARIS, LUIS; SHALOM, DIEGO EDGAR; ÁLVAREZ FEDERICO; WAINSELBOIM, ALEJANDRO JAVIER; GATTEI, CAROLINA ANDREA

Buenos Aires

Workshop; II Escuela de Lingüística de Buenos Aires; 2017

Departamento de Letras, Universidad de Buenos Aires

Evidence in psycholinguistics has shown that the language comprehension system resorts to different types of linguistic information in order set predictions about the incoming input. Prior findings have shown that the predictive principles that drive the language parser are not infallible, thus resulting in higher reading times, higher error rates and differential neural activity when there is a mismatch between the predicted and the actual inputs (Altmann and Kamide 1999; Arai and Keller, 2013; DeLong et al., 2014; Hagoort et al., 1993; Kutas et al., 2011; Kutas and Hillyard, 1983, among others). We report an eyetracking experiment that examined argument linking and the role of arguments prominence in Spanish sentence comprehension. In order to do so, we tested the interplay between word order and morphological case assignment in sentences with psych verbs that allow alternative case marking. We also investigated whether the aspectual characteristics of this group of verbs have a cognitive correlate during reading. More precisely, we examined the contrast between stative and nonstative subclasses of psych predicates like ?aburrir? (?to bore?) and ?enojar? (?to annoy?), exemplified in Marin and McNally (2005, 2011). We predicted differences in the time course of eye-movements according to the sentences? word order and case marking, showing increasing reading times when the sentence required a reversal of arguments hierarchy. We also expected a higher cognitive cost for comprehending sentences in which case marking was incongruent with the aspectualsubclass of the verb (e.g. ?aburrir? + accusative object / ?enojar? + dative object). Result showed an interaction between word order and case marking for late eye movement measures, thus confirming that linking and arguments prominence also play a role in the incremental comprehension of sentences that allow alternative case marking. The interaction between case assignment and verb subclass exerted its influence also in early eye movement measures. Furthermore, an off-line comprehension task revealed that thE interaction of these three types of information engender differential reading times and error rates. We interpret these findings under the light of a neurocognitive model that takes into account both syntactic and semantic dimensions for an explanation of language comprehension (Bornkessel and Schlesewsky, 2006).