Brown bag lunch (Jan 2013)

Posters

Barry Devereaux (Centre for Speech, Language and the Brain)

Parsing sentences are unlikely: corpus-based analyses of the neural processing of verbs

Verb subcategorization frame (SCF) preferences affect sentence processing by placing constraints on how potential arguments can be incorporated into the emerging syntactic representation (e.g. McDonald 1994). Most studies have used norm-based measures of syntactic preference based on subjective responses, but these may not accurately reflect verbs’ SCF usage frequencies. Here we take a different approach, asking whether verb SCF data derived from large cross-domain corpora determine patterns of neural activation as listeners process verbs in sentences.  In an fMRI study, subjects listened to sentences containing locally syntactically ambiguous phrases (e.g. “insulting neighbours”). The verb participle functioned as either a gerund (“insulting neighbours is not encouraged”) or adjective (“insulting neighbours are not respected”).  From the VALEX lexicon (Korhonen et al. 2006), we obtained SCF frequency distributions for each verb. In a Principal Components Analysis (PCA) we extracted 6 components.Component 1 accounted for most variance (29%), and reflected the likelihood of the verb taking a direct object complement. We expected verbs with low values on Component 1 (e.g. “yawn”) to show a preference for the adjectival reading, since the following noun is unlikely to function as the verb’s theme. Verbs with these properties produced greater activation in bilateral IFG, LpMTG and LIPL,consistent with this prediction. This indicates a role for SCF preference in sentences where the syntactic role of the verb is ambiguous. Since the VALEX measure reflects properties of verbs alone (independent of the other words in the sentence), the results demonstrate that access of verbs’ lexico-syntactic representations specifically is an effective heuristic in ambiguity resolution.

Click here to download the poster.

Michelle Sheehan (DTAL)

Partial control, inflected infinitives and defective intervention

This is a poster which was presented at the North East Linguistic Society, New York, 2012. It presents work carried out as part of the ERC project Rethinking Comparative Syntax. European Portuguese allows inflected infinitives to be controlled in certain restricted contexts, very similar to those which permit case independence in languages such as Russian. One such context is in instances of partial control, such as the following:

(1)%O Manel           preferia                    reunirem=se

the Manel         preferred.3SG        meet.INF.3PL=SE.3

'Manel would prefer to meet tomorrow.'

The infinitive in (1) is inflected as 3rd person plural, even though the controlling subject is 3rd person singular. Building on previous work (by Raposo 1989, Rabelo 2004, 2010 and Modesto 2007, 2010), this poster examines this phenomenon in European Portuguese and proposes an analysis in terms of defective intervention.

Click here to download the poster.

Elisabeth Fonteneau, Li Su, Mirjana Bozic, William Marslen-Wilson (Dept. of Psychology, MRC Cognition and Brain Sciences Unit)

Brain network connectivity during morpholexical processing: an MEG/EEG study

The goal of this research is to characterize the neural circuitry that responds to linguistic complexity as indexed by a potential inflectional morpheme (play-ed). Within the neural language system, linguistic complexity engages primarily left lateralised processes, whereas general processing complexity - as indexed by lexical competition - engages a more bilateral network (Bozic et al. 2010). To track the dynamics of this network we investigate cortical oscillatory synchrony in a combined magneto- and electro-encephalography (EMEG) auditory study. Participants listened to lists of words that varied on these two core processing dimensions. Linguistic processing complexity was manipulated by the presence or absence of a potential inflectional morpheme (played). General processing complexity was engaged by the presence of onset-embedded lexical competitors (claim).

Phase-locking values (PLV; Lachaux et al. 1999) based on source estimates (MNE, Hämäläinen et al. 1993) were computed to determine trial-by-trial covariance of the posterior superior temporal gyrus (pSTG) and Heschl’s Gyrus (HG) with other regions of interest. Results time locked to the onset of an inflectional morpheme revealed that phase synchrony increased in the ?-band (30-50 Hz) between L-pSTG and left pars opercularis (L-BA44). In contrast, processing an onset-embedded competitor induced synchrony in a network linking the L-HG and left pars orbitalis (L-BA47). These findings suggest that enhanced cross-cortical interactions between left temporo-frontal areas are necessary and highly specific for morpho-syntactic computations. Synchronisation in oscillatory dynamics reflects the transient coupling of functional networks related to specific computational processes in language comprehension.

Click here to download the poster

Caroline Whiting (Department of Psychology)

Spatiotemporal dynamics of speech comprehension

Extensive evidence has revealed a bilateral fronto-temporal network supporting speech comprehension (Marslen-Wilson & Tyler, 2007) and a left-lateralised sub-system specialised for processing of grammatical complexity (e.g. walk-ed; Bozic et al., 2010). Using combined MEG-EEG, we addressed how neural activity in this network is modulated by properties of the speech input. We manipulated the presence of an inflectional affix (–s and past-tense –ed) to investigate what regions are sensitive to morphological complexity. Stimuli consisted of phrases presented in 3 contexts (I walk, He walks, He walked) and acoustically-matched stimuli (musical rain), which share the auditory properties of speech but are not interpretable. Analyses were performed using representational similarity analysis (Kriegeskorte et al., 2008), which examines the information carried by patterns of neural activity in order to determine the stimulus distinctions that are emphasised in a given region at a given time. The first dissociation was between pronouns (I vs. He) and between words and musical rain at 80 ms in bilateral fronto-temporal regions. Left inferior frontal cortex (BA 44) was the only region to show sensitivity to differences between affixed and non-affixed words, from -50 ms to 100 ms after affix-onset. These results suggest speech-specificity emerges by 80 ms and is modulated by acoustic properties of the input (differences between I and He). Processing within left BA44 was sensitive to the presence of an affix, consistent with neuroimaging results showing this region plays a key role in morphological processing.