Research

My research sits at the applied end of Applied Mathematics and is most recently driven by questions in psycholinguistics and cognitive psychology.

Understanding reading behaviour: eye-tracking and response times

Reading is a relatively recent development in human history, existing for only a few thousand years. However, it has become an essential life skill in modern society, one that is developed over many years of exposure, formal instruction and practice. Good reading skill underpins academic achievement and is a gateway to learning new vocabulary, more colloquial language and new grammatical constructions.

As Huey summarised over a hundred years ago, and which is still as true today, gaining a complete understanding of reading – how we learn to read, how we become fluent readers, how to best teach reading, etc. – is an important aspiration.

And so to completely analyse what we do when we read would almost be the acme of a psychologist’s dream for it would be to describe very many of the most intricate workings of the human mind, as well as to unravel the tangled story of the most remarkable performance that human civilization learned in all of its history. (Huey, 1908, p. 6)

Yet, our understanding of human reading is still incomplete. My research is concerned with elucidating and quantifying processes in two key experimental paradigms: visual word recognition and eye-tracking. In visual word recognition, my focus is on data from lexical decision tasks where participants decide whether a string on a screen is a word or not (e.g., goat vs. goart) and in eye-tracking I examine data elicited as participants’ eye movements are recorded while they read. For both experimental paradigms, I am interested in how individual participants behave, which means response times and accuracy in visual word recognition and fixation durations in eye-tracking. This requires moving beyond traditional statistical approaches and using modern techniques from distributional analysis and nonlinear time series.

Language change in networks

Language change is ubiquitous, and it is primarily driven by the input that we encounter. A key source for this is the conversations that we have with e.g. family, friends, peers, and colleagues. Many of the changes that occur in language begin with teens and young adults. As young people interact with others their own age, their language grows to include words, phrases, and constructions that are different from those of the older generation. Some of these changes have a short life span (have you heard ‘groovy’ lately?), but others stick around to affect the language as a whole. Studying language change makes it apparent that language evolution is a network phenomenon.

In my research, we strive to understand how different network properties (topologies, connection strengths) and different processes affect language change. This work combines extensive agent-based simulations with statistical analyses of the resultant patterns. A prime goal is to unravel the mechanisms that underpin language change as observed in the real word, e.g. in a historic context.

Methods development

The interdisciplinary nature of my work often requires the development of new mathematical, computational and statistical tools or the adaptation of existing approaches. Over the years, this has resulted in novel approaches to solve three-dimensional reaction-diffusion equations with spatially localised sources using Green’s functions (intracellular calcium dynamics), to analyse network synchrony of non-smooth oscillators using the Master Stability approach (networks of Morris-Lecar neurons and Franklin bells), to quantify emergent behaviour in excitable systems originating from stochastic thresholds (integrate-and-fire neurons and neural fields) and most recently to gauge the power (and drawbacks) of advanced statistical inference methodologies (linear mixed effects models vs generalised additive mixed model).