Abstract

Speech perception and the evolution of languages have a long and rich history in the domain of linguistics research. I will, in this talk, present a physicist’s take on aspects of each of these two topics. In the first part, I’ll describe a minimal model of the way in which a listener deciphers a string of sounds and tries to reconstruct a word, both with and without mishearings. This will lead to a phase diagram with a separation between regions of easy and hard decipherability. In the second part of the talk, I’ll describe a model of the evolution of participles, and show that the inclusion of true dynamical competition does better than Ringe and Yang’s threshold principle (2016) in explaining the survival of particular forms.

The last part of this talk concerns the data analysis of a study of audiovisual cognition in a mixed-literacy population, where the use of fluctuations, rather than averages, is able to produce good data collapse from initially rather noisy data.