I’m excited to say Akiva Bacovcin and I will be presenting our paper The logistic language learning curve? at FWAV. We’ll be presenting on the second day (Thursday, May 19th). Looking forward to it!

The logistic function, which has long served as the basic model of discrete language change (cf. Altmann et al. 1983; Kroch 1989), has both deep connections to change in biological populations as well as a practical implementation in terms of logistic regression. However, no particular cognitive mechanism has been proposed to underlie the logistic (Kroch 1989, 4). Yang’s (2002) variational learning model offers a cognitive basis for change insofar as its mean dynamics yield S-shaped curves. Indeed, using simulations, we show that given sufficiently large datasets, it is possible to gain insight into whether the learning or logistic model generated an empirical S-shaped curve.