Nominal Modification

The Research Training Group on Nominal Modification at the Goethe University Frankfurt

Home » Colloquium: Shanley Allen

Colloquium: Shanley Allen

On Tuesday, February 4th GK Colloquium will be hosting a talk by Shanley Allen entitled ‘The role of information density and cross-linguistic influence in the L2 Processing of complex nominal compounds’. See below for the abstract. 

When: 4:15 pm-6:00 pm
Where: SH 0.107

The role of information density and cross-linguistic influence in the L2 processing of complex nominal compounds

According to the Uniform Information Density Hypothesis, speakers adjust the lexical and syntactic realization of their message in order to maximize uniformity in the amount of information per unit of signal, and to avoid peaks and troughs in the amount of information provided. However, very little is known about the role of information density for alternative encodings that are syntactically complex. In the studies reported here, we compare the processing of English complex nominal compounds (NCs) such as pharmaceutical market size increase, which present a peak of information density, to the processing of nouns modified by prepositional phrases (PPs) such as increase in the size of the pharmaceutical market, for which information density is more evenly distributed. We use eye tracking while reading to investigate whether these structures present more processing difficulty for L2 speakers than for native speakers, and whether there is evidence of cross-linguistic influence from the native language of L2 speakers (German vs. Spanish/Portuguese). Results show more processing difficulty for informationally dense structures across all groups, as well as some evidence of cross-linguistic influence from L1 structure to L2 processing. Overall, these results illustrate that speakers prefer to avoid peaks of information density in the processing of syntactically complex structures, and that what counts as ‘dense’ for L2 speakers is modulated by structures available in the L1.

Name of author

Name: Admin