The influence of context on word order processing – An fMRI study

Abstract

a b s t r a c tIn languages that have subject-before-object as their canonical word order, e.g. German, English and Danish, behavioral experiments have shown more processing difficulties for object-initial clauses (OCs) than for subject-initial clauses (SCs). For processing of OCs in such languages, neuroimaging experiments have shown more activation in the left inferior frontal gyrus (L-IFG) compared to SCs. The increased activation in L-IFG has been explained in terms of syntactic transformation demands, increased argument hierarchization demands, and increased load on working memory. Behavioral findings have indicated that context may facilitate syntactic processing, but it has not been investigated whether a supportive context can decrease the activity in L-IFG. With L-IFG as a region of interest (ROI), the present fMRI study of 21 Danish participants investigated how a supportive linguistic context would affect the processing of Danish main clauses with either an initial subject or an initial object. We found more activity in BA 44, BA 45 and BA 47 for OCs compared to SCs. The processing of Danish OCs is thereby seen to elicit effects in L-IFG comparable to previously investigated languages. The context manipulation showed reduced activity in BA 47 for SCs and OCs occurring after a supportive linguistic context, suggesting less pragmatic Journal of Neurolinguistics 26 (2013) 73-88 processing difficulties for sentence processing in a supportive context. Outside the ROI, the lack of context affected several regions in both the frontal, temporal and parietal lobes.

Lateralized task shift effects in Broca's and Wernicke's regions and in visual word form area are selective for conceptual content and reflect trial history

“…The regions observed for CS are overlapping with those observed in studies of other types of linguistic manipulations, such as semantic integration errors (Christensen and Wallentin, 2011), semantic competition (January et al, 2009;Schnur et al, 2009;Snyder et al, 2007), syntactic complexity (Christensen and Wallentin, 2011;Christensen et al, 2013;Kristensen et al, 2013;Tettamanti et al, 2009), syntactic working memory (Fiebach et al, 2005) and linguistic working memory in general (Kristensen and Wallentin, in press;Wallentin et al, 2006). Wallentin et al (2006) found that Broca's region activity correlated with response time in a working memory task where a visual scene was probed by linguistic cues.…”

“…Recently, it has been suggested that activation in this area reflects general linguistic processing costs (Kristensen and Wallentin, in press). In support of this are findings that syntactic manipulations yield increased Broca's region activation (Ben-Shachar et al, 2004;Christensen and Wallentin, 2011;Kristensen et al, 2013;Tettamanti et al, 2009) and findings showing that Broca's region responds to unpredicted word order, rather than syntactic manipulations per se, as defined by working memory demands (Fiebach et al, 2005), cloze probability (Obleser and Kotz, 2010) or context (Kristensen et al, in press). …”

Broca's region and Visual Word Form Area activation differ during a predictive Stroop task

“…One suggestion has been that activation in Broca's region reflects general linguistic processing costs (Kristensen & Wallentin, 2015), i.e., that whenever a linguistic process for some reason is difficult or challenged, it causes Broca's region to become more active. In support of this hypothesis are well-known findings that syntactic manipulations yield increased Broca's region activation (Ben-Shachar, Palti, & Grodzinsky, 2004;Christensen & Wallentin, 2011;Kristensen, Engberg-Pedersen, Nielsen, & Wallentin, 2013;Tettamanti et al, 2009) and findings showing that Broca's region responds to unpredicted word order, rather than syntactic manipulations per se, as defined by working memory demands (Fiebach, Schlesewsky, Lohmann, VonCramon, & Friederici, 2005;Wallentin, Roepstorff, Glover, & Burgess, 2006), cloze probability (Obleser & Kotz, 2010) or context (Kristensen, Engberg-Pedersen, & Wallentin, 2014). Broca's region activation is also routinely observed in the absence of word order manipulations (Fedorenko, Duncan, & Kanwisher, 2013;Novick et al, 2010).…”

Context Predicts Word Order Processing in Broca's Region

“…Both studies show that the discourse context affects sentence processing in the brain. However, so far, to our knowledge, no neuroimaging experiment has found an interaction between discourse context and word order, neither for Danish, nor for other languages (see Discussion section for a description of a recent null result by Kristensen et al, 2013).…”