Search Abstracts | Symposia | Slide Sessions | Poster Sessions
When the order is relevant: Investigating the processing of temporal order violations in conjunctive sentences
Poster Session D, Saturday, October 26, 10:30 am - 12:00 pm, Great Hall 3 and 4
Maria Spychalska1; 1Max Planck Institute for Psycholinguistics
Conjunctive sentences reporting two past events, e.g., “She wrote a letter and read a book”, suggest that the events happened in the order of mentioning, even if there is no direct link between the events. In Gricean pragmatics, this phenomenon is described as “temporal implicature”, and is explained in relation to the conversational Maxim of Manner (“be orderly”). Other authors have proposed that the phenomenon may result from a more general property of discourse or narration structure. Yet, it is still an open question to what extend the temporal representation of events as observed in real life modulates the linguistic processing, in particular, whether the temporal order of events modulates predictive processing in language. I present results of a series of ERP experiments investigating these questions, focusing especially on the role of contextual relevance of the temporal order for the processing of reversed order sentences. The experimental paradigm resembles a memory game, in which participants assign points to a virtual player and read sentences describing the game events. In each trial, five cards were dealt and the player opened two of them. Afterwards, the participants assigned points depending on whether the cards belonged to the same category (animals vs. objects) or not (Experiment 2), or depending on the order of the cards’ categories (Experiment 1). Subsequently, a sentence was presented describing the game trial, e.g., “Julia has flipped a cat and a flower”. In the Correct-Order condition, sentences described the events in the order in which they happened; in the Reversed-Order condition, the events were described in the reversed order. In Experiment 3, the sentences used either “and” or connectives such as “before” and “after”, while the game task did not highlight the order. In experiments where only sentences with “and” were used, the violation of temporal order primarily triggered a P600 effect, although a modulation of the N400 was observed as well if the order was indirectly highlighted in the task. However, in the experiment in which sentences used mixed forms, order violations (on the first noun, so independent of the used connective) triggered a robust N400 effect as well as the P600. As the P600 is sensitive to the integration of both structural and semantic aspects of the linguistic input, the observed P600 effect for order violation supports the hypothesis that reversed-order sentences engage more effortful combinatorial processing that may be due to the violation of the expected discourse organization. However, the additional N400 modulation by order shows that, under the right contextual support, temporal implicatures can be processed incrementally and modulate meaning-related predictive processes. In addition, sentences with “before” and “after” that were semantically false with respect to the scenario’s order, showed an N400 effect on the second-mentioned event, so after the meaning of the connective could be integrated, relative to “and” sentences, where the order is derived pragmatically. (Comment: This is an update on the last year's abstract, which I could not present due to family situation)
Topic Areas: Meaning: Discourse and Pragmatics,