ESR8: Discourse units and discourse alignment

PhD Fellow: Junfei Hu

Hello, my name is Junfei Hu. I am a Marie Curie PhD Fellow in UCLouvain (Belgium) under the supervision of Prof. Liesbeth Degand (UCLouvain), Prof. Martin Pickering (The University of Edinburgh) and Prof. John Du Bois (UC Santa Barbara). I am working on ESR8, Discourse Units and Discourse Alignment, in the Conversational Brain project. This project first aims to fill the gap in our knowledge about alignment at the discourse level by establishing whether (interactive) alignment applies to the building blocks of conversational turns through systematic examination of the relationship between syntactic, prosodic, and pragmatic units in conversational turn, based on a natural conversation corpus. According to the corpus-based findings, this project’s second aim is to investigate the extent and means by which discourse alignment contributes to turn-ending prediction using visual world eye-tracking.

Before my PhD candidate journey, I received my BA (2012) and MA (2013) in Chinese language and literature in Liaoning University (China) and City University of Hong Kong, respectively. After that, I worked as a visiting researcher (2015-2017) in University of California at Los Angeles to study the relationship between transitivity and subjectivity in Spoken Mandarin Chinese in terms of the argument structure of the verb nong (similar to make in English). In 2020, I received the Research Master (MPhil) in Linguistics and Communication Sciences in Radboud University Nijmegen and Max Planck Institute for Psycholinguistics (Netherlands) where I studied gesture production and comprehension by combining corpus analysis and eye-tracking experiment. I also taught Chinese as a foreign language in serval places in the Mainland of China between each academic endeavour.


ESR8’s project focuses on the interaction between (i) discourse units, (ii) discourse alignment, and (iii) discourse processing in conversation. It aims to establish whether alignment applies to the building blocks of conversational turns, beyond the level of words and phrases. Syntactic, prosodic and pragmatic components are known to be involved in the Turn-Constructional Unit (TCU), but the interaction between these three components has not been investigated in a systematic way. Recent empirical work at UCL has aimed at clarifying the syntax-prosody interaction through the concept of Basic Discourse Unit, and has revealed that about 40\% of BDUs are “congruent” (one-to-one syntactic-prosodic unit correspondence), whilst the syntax-prosody combination depends in the remainder of cases on the degree of discourse preparation, ranging from highly prepared (e.g., newsreading) to highly spontaneous. ESR8 will endeavor to better understand the role of BDUs in discourse alignment and discourse processing. The goals will be (1) to establish to what extent syntax+prosody units correspond to complete pragmatic units; (2) to determine to what extent speakers align their types of BDU (congruent vs non-congruent) in conversation; (3) to experimentally test the effect of different types of syntactic-prosodic-pragmatic completions on the process of discourse comprehension and turn-ending projection.

Expected results:

Provide a novel and detailed characterization of alignment at the basic discourse unit level, as a function of the communicative situation.

Based in Louvain-la-Neuve, Belgium

Full-time three-year contract, starting September 2020

PhD enrolment at: Université catholique de Louvain

Main supervisor’s institution: Université catholique de Louvain

Main supervisor: Prof Liesbeth Degand


  • Aix-Marseille University: modelling the overall sequence organization of activities within conversation, and development of an annotation protocol (5 months);
  • Orange Labs, Lannion: to carry out experiments with conversational systems; to identify common points and differences between the results obtained for finalized human-machine dialogues vs other types of dialogues; to determine the usability of the results for modelling the architecture of a conversational system (5,5 months).

Co-supervisors’ institutions:

  • Aix-Marseille University, France
  • Orange Labs, Lannion, France

Scroll to top