The Social Cognition Parallax Interview Corpus (SCOPIC) provides naturalistic but cross-linguistically-matched corpus data with enriched annotations of grammatical categories relevant to social cognition. Development of the corpus is part of a project using an innovative stimulus-elicitation methodology in a language documentation context that results in exposition, cooperative conversation and narrative data in different task phases, the ‘Family Problems Picture Task’. Researchers then add layers of annotation from functional domains relevant to social cognition.

See our LD&C volume for more information and check out the recordings and some annotations in our PARADISEC collection.

The Family Problems Picture Task is organized around 16 cards that participants must describe and organize into a narrative.



This  methodology innovates in four ways:

(a)   It aims to increase the density of corpus attestation in a particular semantic domain; in this case it targets the broad spectrum of categories needed for social cognition

(b)   by allowing speakers to choose their own formulations for the same situations, ‘first-text-bias’ effects are eliminated, that is there is no bias coming from the original language of elicitation in what categories are expressed,

(c)   the card-sort format makes the protocol a narrative problem-solving task, eliciting high levels of speaker involvement, and

Clara Kusos Darr and Mingkui Aigid SocCog Matukar

(d)     the four task-phases (individual picture description, narrative problem-solving, third-person narrative, first-person narrative) induce different formulations and language choices revealing important dynamic variation in coding within each language, in addition to the cross-linguistic variation presented by the whole corpus.

Kadagoi Rawad Forepiso, Silim Pain and Sel Pain Wadom SocCog Matukar

At the core of SCOPIC is a series of ELAN (.EAF) files, a kind of xml structured text. These files are produced through the linguistic annotation software ELAN (EUDICO Linguistic Annotator) which is a multimedia annotator from the Max Planck Institute for Psycholinguistics, The Language Archive, Nijmegen, The Netherlands (cf. Sloetjes & Wittenburg 2008). ELAN is freely downloadable at http://tla.mpi.nl/tools/tla-tools/elan/. The ELAN files provide transcriptions, translations and annotation of data in media files and are linked to those associated media.

Screenshot 2018-01-30 21.28.16

Support Comes from: Australian Research Council (Language and
Social Cognition: The Design Resources of Grammatical Diversity; DP0878126), the Anneliese-Maier Forschungspreis awarded to Nicholas Evans by the Alexander
von Humboldt Foundation and the German Federal Ministry of Education and Research,
and the ARC Research Centre for the Dynamics of Language (CoEDL), funded
by the Australian Research Council (CE140100041).


DB                             humbolt