Formulaic language is seen as a major hurdle for second language (L2) learners to achieve native-like fluency (Pawley and Syder 1983). The term ‘formulaic sequences’ (FSs), that is, fixed or semi-fixed word combinations that have ‘an especially strong relationship with each other in creating their meaning’ (Wray 2008: 9), has been used in the literature to mean anything from idioms (e.g. in a nutshell), through collocations (e.g. highly significant), to clusters or multi-word units (e.g. at the end of). Despite their differences in terms of idiomaticity, formal variability and frequency of occurrence, most FSs seem to be ‘stored and retrieved whole from memory at the time of use, rather than being subject to generation or analysis by the language grammar’ (Wray 2002: 9), making them particularly difficult for L2 learners to grasp. Over the last decade, corpus studies have revealed that FSs are particularly prevalent in academic discourse, providing an important means for discourse building (e.g. on the other hand, as a result, as can be seen) and stance taking (e.g. we argue that, to some extent), among other things (e.g. Biber, Conrad and Cortes 2004; Cortes 2004). Important as they are, most attention has been paid to academic writing and the challenges L2 or novice writers face thereof, whereas research on spoken academic genres has lagged behind in general (Basturkmen 2016)....