You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster C38, Thursday, November 9, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

Elementary composition in Language processing: an EEG study

Emilia Fló1, Álvaro Cabana1, Juan C Valle Lisboa1;1Facultad de Psicología, Universidad de la República

Combining words to represent a new concept is the basic combinatorial operation needed to generate and understand meaningful phrases. The elaboration of these complex structures of related concepts entail both syntactic and semantic composition processes which are difficult to disengage. Previous research in which stimuli were restricted to simple composition using MEG and fMRI have identified the left anterior temporal lobe to be involved in conceptual combinatory operations. In this work we aim at finding a simple composition marker in EEG using an adaptation of Bemis, D. K., & Pylkkänen, L. (2011) experimental design for Spanish. Contrary to English, Spanish noun phrases are constructed such that the adjective is preceded by the noun. Given this distinction and the original design, we decided to introduce a second task in order to establish the adequate control for the composition task. Stimuli for the three tasks were constructed randomly for each subject from a pool of 11 words denoting nouns, 11 words denoting colors, balanced for frequency and other psycholinguistic properties, and 11 consonant strings. The composition task consisted of 100 noun-adjective and 100 consonant string-adjective combinations, followed by an image congruent or incongruent to the verbal material. In one of the control tasks subjects were presented with combinations of noun-noun and consonant string-noun stimuli, as opposed to the second control task where participants were shown combinations of color-color and consonant string-color stimuli. In both control tasks the verbal material was followed by an image congruent to one of the presented words or incongruent to both. The purpose of these controls was to ensure that the difference in the composition task was not due to the amount of words presented in each condition. We performed a cluster permutation analysis to determine the spatial and temporal distribution of the neural activity related to composition. We show evidence of an activity specific to the composition task in a time window consistent with the literature.

Topic Area: Meaning: Combinatorial Semantics

Back to Poster Schedule