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

 
Poster C39, Thursday, November 9, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

ERP effects for quantifier complexity, priming, and truth-value in an auditory/visual verification task

Aniello De Santo1, Jonathan Rawski1, John E. Drury1;1Stony Brook University

[INTRODUCTION] We examined the processing of quantified sentences in an auditory/visual verification task to probe: (i) truth-value/quantifier-type influences on the N400 ERP response, and (ii) ERP markers of quantifier complexity. Concerning (i): studies have demonstrated the N400 to be insensitive to truth-value/negation in verification paradigms (Fischler et al. 1983; Kounios & Holcomb 1992) while exhibiting modulations for subject/predicate relatedness (e.g., N400 for ROCK>BIRD in "A robin IS/IS-NOT a ROCK/BIRD). However, Nieuwland & Kuperberg (2008) argue such uses of negation are pragmatically unnatural, and when this is controlled, N400 amplitude can be modulated by truth-value (False>True). Concerning (ii): fMRI research (McMillan et al. 2005) has demonstrated that additional working memory resources are recruited in the processing of proportional quantifiers (MOST) which, unlike other quantifiers (ALL/NONE/SOME), requires maintenance/comparisons of the cardinalities of sets to evaluate truth. However, the time-course of complexity effects associated with MOST has not previously been investigated using ERPs. [METHODS] We presented quantified sentences (e.g., "All of the squares are blues") auditorily while participants simultaneously viewed arrays of colored shapes. Shape/color-combinations were used with four quantifier-types (ALL/NONE/MOST/SOME) to yield eight conditions varying quantifier/truth-value. Visual stimuli each consisted of fourteen colored shapes, with an even contrast ratio for ALL/NONE (7 yellow-circles/7 blue-squares) and opposing 2:5/5:2 ratios for MOST/SOME (e.g., 2 yellow-/5 blue-circles and 5 blue-/2 yellow-squares). False conditions used color/shape-predicates which were not present in the images (unprimed). These visual/auditory-pairs were presented to adult/native English-speakers (N=10) who provided (mis)match judgments following each trial. EEG was recorded continuously (32 channels, Biosemi-Active-2) and ERP mean amplitudes for successive 100 ms windows were examined for 1200 ms epochs (-200-0 ms baseline). Signals were time-locked to (i) the predicate onset to examine quantifier-type influences on truth-value and (ii) the onset of the quantifier to test for complexity effects for MOST. [RESULTS] Predicates showed opposite polarity N400 effects for ALL (False>True) relative to NONE (True>False), an earlier negativity for ALL (False>True) peaking ~200 ms, and subsequent P600s (False>True) for both ALL/NONE. MOST/SOME yielded a N400/P600 profile (False>True) and an early negativity (False>True; ~200 ms) obtained for SOME (but not MOST). Finally, N400 effects were larger for ALL/NONE than MOST/SOME conditions, while the opposite was true for the P600s (larger for MOST/SOME). ERPs time-locked to the onset of the quantifiers revealed a positivity for MOST relative to the three other quantifiers, beginning ~350-450 ms and sustaining for ~500 ms. [CONCLUSIONS] In pragmatically natural contexts N400s were driven by priming of the expected auditory continuation and were not modulated by truth-value, consistent with earlier findings (Fischler et al.). Consistent False>True effects were, however, obtained for the subsequent P600 response. We relate the early (~200 ms) negativity for ALL/SOME to Phonological Mismatch Negativities (PMMNs; Connolly & Phillips 1994), and discuss whether this should be regarded as continuous with the N400 or a perhaps a species of N200 (van den Brink et al. 2001). Our data also suggest complexity effects for MOST may reflect initial encoding, and may not arise downstream during verification.

Topic Area: Meaning: Combinatorial Semantics

Back to Poster Schedule