Start-end : 2015-2019

 

Summary:

The project involves characterizing laughter with adults and then considering its emergence among infants and subsequently. So far for adults : (i) we have developed a semantic analysis, (ii) worked on developing detailed means of classification of form, function, and positioning (relative to the laughable), with results that refute much of the accepted wisdom, using a French/Chinese/German corpus we compiled. We have also begun exploring the neurocorrelates of laughter perception when embedded in natural dialogue using functional Near-Infrared Spectroscopy (fNIRS).

 

Results:

Many of the achievements of OLA14 arose through the PhD work of Chiara Mazzocconi (to be defended in spring 2019) and the postdoc of Ye Tian (2014-2017, within the ANR-DFG project DUEL) as well as international collaborations with David Schlangen (Bielefeld), Julian Hough (now at QM-U London), and Sophie Scott (UC-London). One important achievement of OLA 14 , has been the compilation of the DUEL corpus (Hough et al., 2016). This corpus consists of 24 hours of natural, face-to-face, loosely task-directed dialogue in German, French and Mandarin Chinese. It includes audio, video and body tracking data and is transcribed and annotated for disfluency, laughter and exclamations. Data from the DUEL corpus was used to refute existing views on the placement of laughter and published in leading venues on formal and computational dialog (SigDial/SemDial conferences). Work in the project provides extensive evidence (contrary to the received view for millenia) that laughter has content that interacts with speech and we have offered the first formal semantic account integrating laughter and dialogue processing. This material was the basis for a course at the European summer school in Logic, Language, and Information.