“…Once the fragments were selected, they were imported into the tool MMA-Video for their annotation and analysis. Following previous explorations of the use of semiotic resources in lectures (Bernad-Mechó, 2018;Bernad-Mechó & Fortanet-Gómez, 2019;Crawford Camiciottoli, 2015, 2016Tan et al, 2016), the modes that were taken into account in the analysis were: the verbal mode (exploration of metadiscourse vs content language, as well as the percentage of pauses over the total duration of the clips), the use of gestures -iconic, metaphoric, deictic or beats (McNeill, 1992), gaze direction, head movements -iconic or novel (Norris, 2004), posture, proxemics, and facial expressions. The mode of paralanguage, although undoubtedly relevant in multimodal analysis, may become quite subjective when trying to quantify it (Bernad-Mechó, 2018) and, therefore, it was excluded from the analyses.…”