Are all the triangles blue? – ERP evidence for the incremental processing of German quantifier restriction

Abstract

The present ERP study investigates the neural correlates of pictorial context effects on compositional-semantic processing. We examined whether the incremental processing of questions involving quantifier restriction is modulated by the reliability of pictorial information. Contexts either allowed for an unambiguous meaning evaluation at an early sentential position or were ambiguous with respect to whether a further restrictive cue could trigger later meaning revisions. Attention was either guided towards (Experiment 1) or away from (Experiment 2) the picture–question mapping. In both experiments, negative answers elicited a broadly distributed negativity opposed to affirmative answers as soon as an unambiguous truth evaluation was possible. In the presence of ambiguous context information, the truth evaluation initially remained underspecified, as an early commitment would have resulted in the risk of a semantic reanalysis. The negativity was followed by a late positivity in Experiment 1, but not in Experiment 2, suggesting that attention towards the mismatch affected semantic processing, but only at a later time window. The current results are consistent with the notion that an incremental meaning evaluation is dependent on the reliability of contextual information.

Gricean Expectations in Online Sentence Comprehension: An ERP Study on the Processing of Scalar Inferences

“…According to Nieuwland et al (), low autism values are indicative of a pronounced pragmatic sensitivity, and comprehenders with low AQ values are more likely to exhibit a pragmatic N400 effect (p. 329). Finally, in some studies, the N400 was accompanied by a late positive component, which has been suggested to reflect task‐related processing strategies associated with the verification process in general (Augurzky et al, ; Nieuwland, ).…”

“…Event‐related potentials from three sentential positions (quantifier, color adjective, shape noun) were analyzed statistically by repeated‐measures anova s on mean amplitude values. Four regions of interest ( Rois ) were introduced (see Augurzky et al, ): left anterior ( Roi 1 : F3, F7, FC1, FC5), right anterior ( Roi 2 : F4, F8, FC2, FC6), left posterior ( Roi 3 : CP1, CP5, C3, P3), and right posterior ( Roi 4 : CP2, CP6, C4, P4). All analyses were calculated within an early N400 time window (300–400 ms) and a time window capturing late positive effects (450–800 ms) following Augurzky et al ().…”

“…To date, ERP studies on implicature processing are still scarce and do not reveal a consistent pattern of results. One component that has been associated with the processing of quantified sentences in general is the N400 (Augurzky, Bott, Sternefeld, & Ulrich, ; Freunberger & Nieuwland, ; Urbach, DeLong, & Kutas, ; Urbach & Kutas, ). The N400 has been observed as a processing reflex of the so‐called underinformative sentences, that is, sentences with some used in a situation where the speaker could have used all instead (e.g., Some elephants are mammals ; see Hunt, Politzer‐Ahles, Gibson, Minai, & Fiorentino, ; Nieuwland, Ditman, & Kuperberg, ; Zhao, Liu, Chen, & Chen, ; but see Noveck & Posada, , for an effect going in the opposite direction).…”

“…To this end, we employed an experimental design comparable to the one from a recent ERP study on the processing of the quantifier all (Augurzky et al, ). In that study, different pictorial contexts comparable to those in Fig.…”

“…Alternatively, the observed effect could be explained by considering the relative predictability of the adjective, which varies as a function of the pictorial context and the already processed sentence information prior to the adjective. Since Augurzky et al () examined the processing of questions, it is difficult to distinguish between these approaches (see Augurzky et al, , for discussion). Therefore, this study uses declarative clauses and tests whether ERPs at several sentence positions can be explained in terms of comprehenders’ expectations of incoming lexical material.…”