simulateNDRa: Generate simulated naming latencies and pronunciations

In PeterHendrix13/NDRa: Generate Simulated Naming Latencies and Pronunciations in the NDRa Model

Description

Generate simulated pronunciation for a set of words or nonwords. lexicon A dataframe with the colums "Word" and "Gestures". "Gestures" are demi-syllables (see Klatt, 1979).

Usage

simulateNDRa(lexicon = lex, weightsSem = weights_sem,
  weightsPhon = weights_phon, parameters = list(wSem = 0.2, wPhon1 =
  0.05, wPhon2 = 0.098, wH = 0.152, wCompl = 1.27, backoff = 0.01, wlex =
  4.7, N = 20, wAct = 0.055, rtConst = 450), parallel = TRUE,
  numCores = detectCores(), verbose = TRUE)

Arguments

weightsSem	An orthography-to-semantics weight matrix with letter unigrams and bigrams as cues and words as outcomes. The default, "weights_sem" uses the weight matrix from Hendrix et al. (2018).
weightsPhon	A phonology-to-semantic weight matrix. The default, "weigths_phon" uses the weight matrix from Hendrix (2018).
parameters	A list with the model parameters "wSem", "wPhon1", "wPhon2", "wH", "wCompl", "backoff", "wlex", "N", "wAct", and "rtConst". The default values are the values used by Hendrix (2018). For more information, also see Hendrix (2018).
parallel	Should computations be carried out in parallel? Defaults to TRUE.
numCores	The number of cores to use for parallel computation. By default all available cores are used.

References

Hendrix, P, Ramscar, M., & Baayen, R. H. (2019). NDRa: a single route model of response times in the reading aloud task based on discriminative learning. Manuscript.

Klatt, D. H. (1979). Speech perception: a model of acoustic-phonetic analysis and lexical access. Journal of Phonetics, 7, 279-312.

Examples

# Load data for the ELP simulations in Hendrix (2018)
data(elp)

# Generate simulated naming latencies and pronunciations
elp = simulateNDRa(elp)

PeterHendrix13/NDRa documentation built on May 7, 2019, 6:05 a.m.