Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Automatic Processing of Morpheme Semantic Relations in Chinese Compounds: A Fast Periodic Visual Stimulation EEG Study
Poster Session A - Sandbox Series, Thursday, October 24, 10:00 - 11:30 am, Great Hall 3 and 4
This poster is part of the Sandbox Series.
Fengyun Hou1, Nina Kazanina1,2; 1University of Bristol, 2University of Geneva
Understanding a sentence means that the listener is able to extract information about the entities involved and relations between them. In language, relations are often encoded using a combination of explicit lexical means and syntactic configuration, such as prepositional phrase modifying a certain noun (e.g. “an apple in the basket” where the spatial relation is conveyed via the preposition “in”). However, relational encoding in language can also be implicit. A notable case is exemplified by compounds, in which diverse semantic relations between the constituent morphemes may be represented in the absence of any overt morphological makers. Chinese is a language in which approximately 72% of words are compounds consisting of two free morphemes (Lexicon of Common Words in Contemporary Chinese, 2008). In light of its rich and varied semantic relations between morphemes, Chinese provides a window for investigating relational encoding in language. For instance, in 酒瓶 (/jiǔpíng/, wine bottle), the morpheme 酒 (/jiǔ/, wine) modifies the function of the head noun 瓶 (/píng /, bottle), yielding the meaning ‘a bottle for wine’. In contrast, in 羊皮(/yángpí/, sheepskin) there is a possession relation between 羊 (/yáng/, sheep) and 皮 (/pí/, skin), i.e. ‘the skin of a sheep’. Previous studies using the relation-priming paradigm have shown that different relations influence the processing of compounds (e.g. Gagné, 2002; Spalding and Gagné, 2011). However, it remains an open question how these different semantic relations are processed by our brain, in particular, how automatically the brain extracts and processes relations. To address these questions, we investigated processing of relations within Chinese compounds in an EEG experiment using an oddball paradigm combined with fast periodic visual stimulation. We examined five types of Chinese compounds, including three types of noun compounds (1)-(3), a verb-noun compounds (4) and a pseudo-compound with no relation between morphemes: (1) modification-function (X for Y): 酒瓶 (/jiǔpíng/, wine bottle); (2) modification-possession (X of Y): 羊皮 (/yángpí/, sheepskin); (3) coordination (X and Y): 花草 (/huācǎo/, flowers and grass); (4) action (V+N): 弹琴 (/tánqín/, play instrument); (5) pseudo-compound (P-C): 头柴 (/tóuchái/, head firewood). Five stimulation sequences in which the stimuli appeared isochronously at the rate of 6 Hz were constructed. In all sequences, stimuli such as (1) were used as a frequent standard stimulus (i.e. base stimuli), with each of compound types (1)-(5) used as an infrequent stimulus (presented as every 5th stimulus, 1.2 Hz, i.e. oddballs). For example, the stimulation sequence in which (1) served as a base stimulus and (2) as an oddball was as follows: 酒瓶 (/jiǔpíng/, wine bottle) —衣架 (/yījià/, clothes hanger) — 笔筒 (/bǐtǒng/, pen holder) — 信箱 (/xìnxiāng/, mailbox) — 羊皮 (/yángpí/, sheepskin). We hypothesised that periodic EEG responses would be observed at oddball frequency and its harmonics in four sequences where the base and oddball compounds reflected different conditions, i.e. with (2)-(5) as the oddball, but not when they represented the same condition, i.e. with (1) as the oddball. Such findings would indicate that the brain automatically and spontaneously processes semantic relations between constituent morphemes of Chinese compounds.
Topic Areas: Meaning: Lexical Semantics, Syntax and Combinatorial Semantics