BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Using quantile regression and dynamic survival analysis to study t
 he time course of the lexical processing of complex words - Harald Baayen 
 (University of Tübingen)
DTSTART:20180201T160000Z
DTEND:20180201T173000Z
UID:TALK99925@talks.cam.ac.uk
CONTACT:Dr Calbert Graham
DESCRIPTION:This talk draws attention to two statistical methods that make
  it possible to assess whether the effect of predictors on a response vari
 able vary within the distribution\nof that response. Dynamic survival anal
 ysis is applicable to durational responses such as reaction times\, fixati
 on durations\, and acoustic durations. Quantile\nregression can be applied
  to any kind of measurement\, not only durations but also tongue positions
  or the amplitude of the brain's electrophysiological response to\nsome st
 imulus.\n\nSchmidtke et al. (2017) used nonparametric survival analysis (R
 einhold & Sheridan\, 2014) to show that whole-word frequency effects emerg
 e earlier in the distribution of reaction times in visual lexical decision
  and eye movement fixation durations compared to constituent frequency eff
 ects. However\, in the vast statistical literature on survival analysis\, 
 parametric methods are available (see Scheike and Martinussen\, 2007) that
  enable the analyst to take into account the different causes of exit time
 s: in lexical decision\, an exit time can be due to a word or a non-word d
 ecision\, and for fixation duration\, a saccade can be initiated to either
  a position within the word or to a position elsewhere. Dynamic survival a
 nalysis applied to auditory lexical decisions to English compounds reveale
 d early effects of compound frequency and late effects of modifier frequen
 cy\, replicating Schmidtke et al. (2017). However\, the competing risks se
 tting of dynamic survival analysis enables a further analysis of the nonwo
 rd responses to words. This analysis revealed that such error responses ar
 ise due to the intrusion of the\nmodifier.\n\nQuantile regression (Koenker
 \, 2005) is a regression technique that allows the analyst to move beyond 
 predicting the mean. How predictors work together can be scrutinized not o
 nly for the median\, but also for deciles such as 0.1\, or 0.9\, or\nany o
 ther quantile of interest. The qgam package (Fasiolo et al. 2017) integrat
 es quantile regression with the generalized additive model\, and thus make
 s it possible to study how nonlinear trends change across the distribution
 . Quantile gams applied to experimental data on morphological processing l
 ead to the same conclusions as dynamic survival analysis. Both argue again
 st decompositional theories of morphology\, and fit well with the discrimi
 native perspective on lexical\nprocessing (Milin et al.\, 2017) as well as
  with Word and Paradigm morphology (Blevins\, 2016).\n\nReferences\n\nBlev
 ins\, J. P. (2016). Word and paradigm morphology. Oxford: Oxford Universit
 y Press.\n\nFasiolo\, M.\, Goude\, Y.\, Nedellec\, R.\, and Wood\, S. (201
 7). Fast calibrated additive quantile regression. Manuscript\, University 
 of Bristol.\n\nKoenker\, R. (2005). Quantile regression. Number 38. Cambri
 dge university press.\n\nMilin\, Feldman\, Ramscar\, Hendrix\, and Baayen 
 (2017). Discrimination in lexical decision. PLOS-One\, 12 (2)\, e0171935.\
 n\nReingold\, E. M. and Sheridan\, H. (2014). Estimating the divergence po
 int: A novel distributional analysis procedure for determining the onset o
 f the influence of experimental variables. Frontiers in Psychology\, 5. ht
 tp://dx.doi.org/10.3389/fpsyg.2014.01432.\n\nScheike\, T. H. and Martinuss
 en\, T. (2007). Dynamic Regression models for survival data. Springer\, Ne
 w York.\n\nScheike\, T. H. and Zhang\, M.-J. (2008). Flexible competing ri
 sks regression modeling and goodness-of-fit. Lifetime Data Analysis\, 14(4
 ):464-483.\n\nScheike\, T. H. and Zhang\, M.-J. (2011). Analyzing competin
 g risk data using the R timereg package. Journal of Statistical Software\,
  38(2):1-15.\n\nSchmidtke\, D.\, Matsuki\, K.\, and Kuperman\, V. (2017). 
 Surviving blind decomposition: a distributional analysis of the time cours
 e of complex word recognition. Journal of Experimental Psychology: Learnin
 g\, Memory and Cognition.\n\nWood\, S. N. (2006). Generalized Additive Mod
 els. Chapman & Hall/CRC\, New York.
LOCATION:English Faculty Lecture Room GR-06/07\, 9 West Road\, Sidgwick Si
 te.
END:VEVENT
END:VCALENDAR