BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:On the interpretation of frequency effects in comprehension and pr
 oduction. - Harald Baayen (MPI for Psycholinguistics\, The Netherlands)
DTSTART:20061129T160000Z
DTEND:20061129T173000Z
UID:TALK5947@talks.cam.ac.uk
CONTACT:Susan Rolfe
DESCRIPTION:Harald Baayen\, Laura Balling\, Wieke Tabak\, Heidrun Bien and
  Hanke van Buren Radboud University Nijmegen and MPI for Psycholinguistics
 \, The Netherlands\n\n\n\nEver since Taft (1979)\, the word frequency effe
 ct has been used as a tool to gauge the representation and processing of m
 orphologically complex words. Two frequency counts have received special a
 ttention: the\nfrequency of the complex word as a whole (henceforth Comple
 x Frequency)\, and the frequency of its stem (henceforth Stem Frequency). 
 Stem Frequency effects have been advanced as the hallmark of constituent\n
 driven processing in procedural memory\, while Complex Frequency effects h
 ave been interpreted as evidence for item specific information available i
 n declarative memory (Pinker & Ullman\, 2002).\n\nIn this paper\, we argue
  that Complex Frequencies do not reflect the activation of or access to it
 em specific representations (similar to those of simple words). Instead\, 
 we think that they reflect memory traces of previous combinatorial process
 ing in procedural memory. We also argue that Stem Frequency is an inapprop
 riate measure for gauging constituent based processing. Instead\, Stem Fre
 quency seems to tap into a word's conceptual familiarity.\n\nOur arguments
  are based on two strands of evidence. The first strand concerns the signi
 ficance of Complex Frequency and the nonsignificance of Stem Frequency as 
 predictors of lexical decision latencies in\nregression studies of visual 
 and auditory comprehension of relatively low frequency words in English\, 
 Spanish\, and Danish. The traditional interpretation predicts Stem Frequen
 cy effects rather than Complex Frequency effects for such words\, the mirr
 or image of what we actually observe. Interestingly\, the absence of a sig
 nificant Stem Frequency effect is paired with the presence of various othe
 r predictors indicating constituent driven processing. This leads to the c
 onclusion that for comprehension\, Stem Frequency (as a simple unigram pro
 bability) is not the right measure for gauging parsing processes\, and to 
 the inference that Complex Frequency (as a joint probability) is a measure
  of experience with morphological processing.\n\nA second strand of eviden
 ce concerns speech production. According to the WEAVER (Levelt\, Roelofs\,
  Meyer\, 1999) model\, the Stem Frequency effect would arise during the en
 coding of a word's form. We used picture naming\nexperiments to probe the 
 early stages of speech production\, and position response naming to invest
 igate the later stages of word form encoding and articulatory preparation.
  For inflected words\, we observed a U\nshaped effect of Stem Frequency fo
 r picture naming\, but not for position response naming. This suggests tha
 t Stem Frequency effects arise during conceptualization and perhaps lemma 
 selection\, without\, however\, playing\na role during later stages of for
 m encoding. Instead\, word form encoding seems to be subject to lexical co
 mpetition\, as gauged by measures of phonological density and cohort entro
 py.\n\nComplex Frequency was never a good predictor in our production expe
 riments\, as predicted by decompositional models such as WEAVER. However\,
  measures gauging the complexity of a word's inflectional paradigm either 
 a type count of inflectional variants (for Spanish)\, or\nShannon's entrop
 y calculated over these inflectional variants (for Dutch) were predictive\
 , both in picture naming and in position response naming. This suggests th
 at throughout the production process\, the\nselection of the appropriate i
 nflected form from its inflectional paradigm is a computationally demandin
 g probabilistic process that is very different from straightforward select
 ion schemas such as\nimplemented in WEAVER. The effects of inflectional en
 tropy also bear witness\, albeit indirectly\, to the importance of the pro
 babilities of words' inflectional variants\, contrary to what fully decomp
 ositional\nmodels would suggest.\n\nConsidered jointly\, our results cauti
 on against linking frequency effects straightforwardly to hypothesized men
 tal representations of words and their constituents\, and call for more so
 phisticated distributional measures.
LOCATION:GR-06/07\, English Faculty Building
END:VEVENT
END:VCALENDAR