The case of scalar implicature processingPenka Stateva
, Anne Cheylus
, Jean-Baptiste van der Henst
, Mélody Darblade
, Chiara Caretta
, Anne Colette Reboul
, Greta Mazzaggio
, 2019, published scientific conference contribution abstract
Abstract: Implicatures like ‘Some politicians are smart’ (interpreted as ‘Some but not all politicians are smart’) are defined scalar implicatures. A heated linguistic debate has focused on how we derive those implicatures: some authors consider the computational process as linguistic in nature (Levinson, 2000), others as pragmatic in nature (Sperber & Wilson, 1995). A growing body of research, prompted by pioneering work by Bott and Noveck (2004), focused on the computational cost related with the computation of scalar implicatures. The present study addresses such topic through the use of different experimental techniques. With Experiment 1 (N = 57) we replicated the third experiment of Bott and Noveck (2004), the first study that identified a cost related to a pragmatic response. With Experiment 2 (N = 58), using a pseudo-word paradigm, we excluded the possibility that the computational cost is due to an experimental artifact, such as an increased difficulty in moving up in the conceptual hierarchy (e.g., ‘Some elephants are mammals’) than in moving down (e.g. ‘Some mammals are elephants’). In Experiment 3 (N = 54), with a Sentence Evaluation Task, we collected reading times, reaction times and eye gaze data. Results showed that the cost of the computation disappears when there is contextual support. Overall, our results seem to support the idea that scalar implicatures are not automatically computed with context playing an important role.
Found in: osebi
Keywords: scalar implicatures, eye-tracking, experimental pragmatics, reaction times
Published: 22.09.2021; Views: 799; Downloads: 3
Fulltext (0,00 KB)
This document has many files! More...