Analysing the timecourse of experimental effects in eyetracking data

Tracking language users’ eyes as they read or listen provides us with valuable insights into language cognition. By analysing how the eyes move as new input is received, we can make inferences about changes in cognitive processing. A common question in language cognition research is about the timing of these changes: for example, are non-native listeners slower to comprehend particular types of sentences than native speakers? If so, why? In this talk I will discuss how reading eye tracking and the visual world paradigm can be used for analysing the timecourse of language processing. I will introduce our own method for quantifying and comparing the onsets of experimental effects in visual world data, and show how the method allowed us to answer a question about why non-native speakers might be slower to process particular types of syntactic cues. Overall, the talk should give a broad overview of the issue of timecourse analysis in eye tracking, as well as introduce a specific timecourse analysis method with links to example code and data.

Neural machine translation for low-resource and variation-rich languages

Current machine translation architectures are based on deep neural networks and provide impressive translation quality for the major language pairs. I will start by giving a high-level overview of neural machine translation architectures and then focus on two challenging application scenarios.

The first scenario concerns low-resource languages. I will present our participation in the AmericasNLP shared task, which focuses on machine translation from Spanish to eleven indigenous languages of the Americas. I describe how a combination of techniques, ranging from data collection to knowledge distillation and post-processing, helps improve the translation quality.

In the second scenario, I investigate the suitability of neural machine translation techniques for the automatic normalization of phonetic transcriptions in multi-dialectal corpora. In this case, our focus does not lie in optimal normalization performance, but rather what the model learns about the different dialects and their relation with each other during the training process. In our case study with large Finnish and Norwegian dialect corpora, the model successfully identified the major dialect areas known from prior dialectological research.

New approaches to measuring the social meaning of language variation. The potential of the IAT and RRT for sociolinguistics

Analysing masculinities in spoken interaction: intersectionality, indexicality and “voice”

In this talk I introduce a selection of approaches that I have taken to the analysis of masculinities in my work on the self-recorded, spontaneous talk of a group of young men from ethnically and racially mixed working-class backgrounds in southeast London.

I will introduce an interactional sociolinguistic approach, informed by Bakhtin’s (1981; 1984; 1986) work on dialogicality and Tannen’s (1989; 2004) notion of constructed dialogue, which has allowed me to explore the young men’s use of voices in relation to a range of discourses which are shaped by and shape their intersectional masculinities. Intersections of race, ethnicity and social class inform many of the young men’s positions, for example when they talk about their own ‘overstanding’ of hip hop in relation to the mere ‘understanding’ of ‘white kids from Oxford’, or when they contrast their own experience of southeast London with that of ‘Hipsters moving into the hood’. Another example is their discussion of social inequality in the context of the ‘2011 London Riots’, where they consider their own position in relation to rioters who resorted to ‘stealing rice from a shop’ on one hand, and ‘people who don’t need shit […] they go skiing in the Alps at Christmas’ on the other.

I will argue that for the analysis of the indexical processes at the centre of the young men’s positioning we need to consider the ‘microcontext of interaction’ in relation to the ‘macrosociological order’ (Silverstein 2004: 638-639). Silverstein’s term ‘cultural concepts’ is useful in this endeavour, as it allows us to consider how speakers draw on words and expressions which index certain cultural knowledge and constitute important resources for their positioning. My presentation highlights the importance of approaching the analysis of these indexical processes from an intersectional approach and aims to provide insight into how ‘subjects mobilise (or choose not to mobilise) particular aspects of their identities in particular circumstances’ (Nash 2008: 11).

References:

Bakhtin, Mikhail (1981) The dialogic imagination (trans. Caryl Emerson and Michael Holquist). Austin, TX: University of Texas Press.

Bakhtin, Mikhail (1984) Problems of Dostoevsky’s poetics. Ed. and transl. Caryl Emerson. Minneapolis: University of Minnesota Press.

Bakhtin, Mikhail (1986) The problem of speech genres. In Speech genres & other late essays, ed. Caryl Emerson and Michael (transl by Vern McGee), pp 60-102. Austin: University of Texas Press

Nash, Jennifer (2008). Re-thinking intersectionality. Feminist Review, 89: 1–15. Silverstein, Michael (2004) Cultural concepts and the language-culture nexus. Current Anthropology 45 (5): 621-652.

Tannen, Deborah (1989) Talking voices. Repetition, dialogue, and imagery in conversational discourse. Cambridge: Cambridge University Press.

Tannen, Deborah (2004) Talking the dog: farming pets and interactional resources in family discourse. Research on Language and Social Interaction 37(4): 399-420.

Experience Sampling Methods

Quantitative Perspektiven zur dialektalen (Morpho-)Syntax in Österreich: Soziale und areale Faktoren

Introduction to Whisper