On Tuesday, Claudia Pons Moll will present on rhotic metathesis in Algherese Catalan. With any time left over, I’ll present some work on iterated gradual learning (handout available on syllabus page).
On Thursday, I’ll discuss the Coetzee and Pater paper on OCP-Place in Muna, as well as Anttila and Wilson and Obdeyn’s responses to it (papers available on syllabus page).
On Tuesday, Presley will present on Wilson’s (2006) artificial language learning/MaxEnt grammar paper, and will tell us about her ongoing work on Turkish coda clusters
On Thursday, Clint will present on Ryan’s (2010) paper on Tagalog morphotactics, and will tell us about his ongoing work on modeling of learning Chamorro /um-/ infixation.
No stats this week – see the post immediately below if you’re missing them…
I didn’t get a chance to show you the logistic regression for the Kager/Pater data. As discussed in Crawley ch. 14, since these are proportion data, a logistic model is appropriate. Crawley (p. 255) in fact discusses an analogous case with two categorical explanatory variables. Here are the KP data formatted appropriately. Try to construct the analysis yourself based on the Crawley example, or run mine. This model does not include an error term for repeated measures – it looks like we’ll have to follow fashion and use a mixed effects model…
On the subject of stats, I may not have told you that I posted a nice explanation of planned and unplanned contrasts that Wendell sent me a couple of weeks ago – see the page to the left. Also, I’ve just posted a flow chart for choosing a statistical test that Kevin sent – see the stats resources page.
On Tuesday, we’ll talk about scalar constraints and constraints in a stringency relation, in OT and HG. There is no reading – I’ll bring in materials from the Potts et al. paper that didn’t make the final cut. On Thursday, Brian will present Zuraw’s 2009 paper on frequency effects on Tagalog phonological processes. With whatever time is left over, we’ll discuss linear and logistic regression (see syllabus for Crawley page listing).
On Tuesday, we’ll talk about ANOVAs, revisiting some data from Kager and Pater that illustrate repeated measures and an interaction in a two-way ANOVA. Time permitting, I’ll also return to the OT-Help-based serial HG/OT comparison that we didn’t get to earlier in the semester. On Thursday, Kevin will present on category learning in mixture of Gaussians models. If you have limited time, concentrate on the Dillon et al. paper (and if you have time for only one of the other two, read Toscano and McMurray).
On Tuesday, we’ll finally hear from Claire on the three wug-testing papers.
Thursday, I’ll present the Pater and Tessier paper on artificial language learning and the connection between phonotactics and alternations. I’ll also show “Perceptron” in action in modeling some artificial language learning results (not Pater and Tessier’s).
The slides from the week 8 presentations on perceptual illusions are now linked from the syllabus page. On the software page you’ll now find an R script “Perceptron” that implements “HG-GLA” with noisy and exponential HG as described in Boersma and Pater (2008).
In week 9, we’ll devote Tuesday’s class to discussion of sections of chs. 5 and 6 of Crawley – the topics I plan to cover are listed in the syllabus page.
Thursday Claire will present a series of wug-test studies on the learning of “natural” vs. “unnatural” alternations.
On Tuesday we’ll prepare for James Kirby’s visit by reading his paper on the modeling of phonologization. Clint and Kevin will be presenting, but since this material is challenging, everyone should make their best effort to understand it before class.
On Thursday, Brian will present Goldrick’s overview paper, which discusses general issues about the relationship between phonology and processing, and uses phonotactic judgments as an example. We’ll use the remaining time that day to talk about normality and deviations from it (Crawley ch. 5).
As usual, links to readings are on the syllabus page to the left.
On Tuesday, Robert will present on Finite State Machines in phonology, focusing on the Wilson 2006 handout linked from the syllabus page, but also discussing some broader applications of FSMs for constraint formalization. With any leftover time, we’ll discuss stats, focusing on the chapter on variance (ch. 4) in the Crawley text.
On Thursday, I’ll present the experimental work Kager and I have done on phonotactic knowledge. Both the paper and the reviews are linked from the syllabus page – discussing the reviews will allow us to talk about general issues about how to deal with them.
See syllabus for description and links