Activities and Seminars

Sentence processing workshop
Date: Jan 12, 2012

Invited Speakers: Adrian Staub and Chuck Clifton, University of Massachusetts at Amherst, US.

Invited Speakers: Adrian Staub and Chuck Clifton, University of Massachusetts at Amherst, US.

10:00 – 11:00 ADRIAN STAUB (Univ. of Massachusetts at Amherst, US) : “Lexical and contextual effects on eye movements in reading: Puzzles and solutions”. The starting point of this talk is the observation that in eye movements in reading, effects of lexical variables (e.g., frequency) and effects of contextual variables (e.g., predictability, plausibility, syntactic fit) sometimes appear in the same early measures, i.e., first fixation duration and gaze duration on the critical word. This pattern presents a challenge for ‘autonomous’ models of visual word recognition, as it suggests that lexical and contextual manipulations may be influencing the same stage(s) of processing. However, drawing on a wide range of experimental results (both my own and others’), I argue that in fact the evidence is consistent with a model according to which lexical and contextual manipulations influence distinct stages. Specifically, I argue that predictability (i.e., cloze probability) influences a pre-lexical, perceptual stage of processing, and manipulations of plausibility and syntactic fit influence a post-lexical, integrative stage. Finally, I suggest that these experimental data are not consistent with models according to which a word’s conditional probability is the primary influence on processing difficulty.

11:00 – 11:30 COFFEE

11:30 – 11:55 NICOLA MOLINARO (BCBL): “Building up expectations based on what we know” In this talk, I will present an EEG study about expectation/pre-activation processes during the comprehension of fixed multi-word expressions in Spanish. I dissociated pre-activation EEG phenomena related either to semantic knowledge (classically associated to N400 modulations) and/or to lexical associations during sentence comprehension. In line with previous finings I found evidence of a P300 component related to the closure of an expectation developed while reading a fragment of a multi-word expression. In addition, I will discuss some oscillatory brain activity (in the theta and gamma band) that would correlate with the recognition of the multi-word expression.

11:55 – 12:20 ADRIANA HANULIKOVA (BCBL): “The effect of speaker’s identity on syntactic processing” An important property of speech is that it explicitly conveys features of the speaker’s identity. Although previous studies have shown that speaker’s identity affects semantic processing (Van Berkum et al. 2008), it is not clear how a speaker’s identity will influence grammatical processing. Here we report a study of subject-verb agreement in Slovak, when the agreement depends on the speaker’s gender (as cued by his/her voice) compared to when it depends on the formal grammatical gender of the subject.

12:20 – 12:45 SIMONA MANCINI (BCBL): “Tracking the time-course of agreement processing: ERP and eye-movement evidence” In this talk I will present convergent data from ERP and eye-movement experiments on subject-verb agreement comprehension in Spanish. Specifically, I will compare the processing of truly mismatching sentences (person violations, e.g. * El periodista. escribiste. un articulo /*The journalist wrote an article ) with apparently mismatching ones (Spanish Unagreement patterns, as in Los periodistas. escribimos. un articulo , interpreted as We journalists wrote an article ). I will show that i) initial processing stages are devoted to the analysis of subject-verb featural consistency: in presence of a feature mismatch, regardless of whether the sentence is truly or apparently mismatching, a similar response is indeed elicited; ii) Later stages of processing are instead devoted “to solve” agreement mismatches, with this operation having a different outcome in Unagreement as compared to true person anomalies.

14:15 – 14:40 MANTE NIEUWLAND (BCBL): “Processing Sentence Truth-value” I will present the results of several ERP studies on the role of propositional truth-value in sentence processing. These studies examined the role of truth-value in comprehension of so-called ‘difficult’ sentences: sentences with negation, sentences with quantifiers, attitude statements and counterfactuals. The results suggest that also in these sentences, in contrast to claims in the literature, propositional truth-value can impact semantic processing without a delay.

14:40 – 15:05 DOUG DAVIDSON (BCBL): “A dynamic causal model for the effect of syntactic complexity during spoken sentence processing” This talk will present the results of an MEG experiment comparing the responses to canonical and non-canonical constituent orders in spoken German sentences. Using a model of effective connectivity, several different connectivity profiles will be compared to model the difference in the evoked response.

15:05 – 15:30 ANDREA MARTIN (BCBL): “Memory architectures and mechanisms for language comprehension: SAT and ERP evidence” In this talk I will outline behavioral and electrophysiological data from long-distance dependency formation during online sentence comprehension. I will argue that (1) the cues at retrieval provide direct access to relevant representations in memory, eliciting them without a search, and that (2) interference at retrieval is a function of how diagnostic a cue is to a unique linguistic representation in memory.

15:30 – 16:00 COFFEE

16:00 – 17:00 CHUCK CLIFTON (Univ. of Massachusetts at Amherst, US) : “Situational Context Affects Definiteness Preferences: Effects of Reading Strategy”. Four experiments used self-paced reading and eyetracking to demonstrate that readers are, under some conditions, sensitive to the presuppositions of definite vs. indefinite DPs (determiner phrases). Reading was faster when the context stereotypically provided a single possible referent for a definite DP or multiple possible referents for an indefinite DP than when context and DP definiteness were mismatched. This finding goes beyond previous evidence that definite DPs are processed more rapidly than indefinite DPs when there is a unique or familiar referent in the context, showing that readers are sensitive to the pragmatics of (in)definiteness. However, the finding was obtained only when readers had to perform a simple arithmetic task between reading a sentence and seeing a question about it. The intervening task may have encouraged them to process the sentence more deeply in order to form a representation that would persist while doing the arithmetic. The methodological implications of this observation are discussed.