diff options
Diffstat (limited to 'Master/texmf-dist/doc/latex/umthesis/intro.tex')
-rw-r--r-- | Master/texmf-dist/doc/latex/umthesis/intro.tex | 46 |
1 files changed, 46 insertions, 0 deletions
diff --git a/Master/texmf-dist/doc/latex/umthesis/intro.tex b/Master/texmf-dist/doc/latex/umthesis/intro.tex new file mode 100644 index 00000000000..f0e2ebd284c --- /dev/null +++ b/Master/texmf-dist/doc/latex/umthesis/intro.tex @@ -0,0 +1,46 @@ +\chapter{The introduction} +Modern research on lexical access began in the 1950's (though +\cite{ColeRudnicky1983} note very similar research performed in the 1890's by +William Chandler Bagley). Several statistical properties of the mental +lexicon have consistently been found to influence how humans process speech. +One of the earliest and most robust findings was that lexical frequency has a +strong influence on lexical access. Repeated research has shown that high +frequency words elicit quicker and more accurate responses than low +frequency words in a large variety of experimental conditions +(e.g.\ \cite{Broadbent1967, Taft1979, BenkiJASA}). Another factor which has +been reliably shown to affect lexical access is neighborhood density. +Neighborhood density is a metric of similarity, roughly defined as the degree +to which a word is similar to others (both phonological and orthographical +measures have been used). Words which have many similar words are said to be +in dense neighborhoods, whereas words which have few similar words are said to +be in sparse neighborhoods. In contrast to lexical frequency, which +facilitates the activation of a word in the brain, neighborhood density has +been found to inhibit activation (e.g.\ \cite{Luce1986, Luce1998, BenkiJASA, +Imai2005}). Of course these are not the only factors which affect language +processing, but they are the most frequently cited, and will be referred to +again in the following sections. +\begin{table*}[!htb] + \centering + \caption[Basic Predictions]{Basic Predictions: Predicted results are marked + with a checkmark, and a relative effect size is also given.} + \label{T:predictions} + \begin{tabularx}{\textwidth}{% + >{\setlength{\hsize}{1.5\hsize}\raggedright\arraybackslash}X% + *{2}{>{\setlength{\hsize}{.7\hsize}\raggedright\arraybackslash}X}% + *{2}{>{\setlength{\hsize}{1.05\hsize}\raggedright\arraybackslash}X}} + \hline\hline + \rule{0em}{1.1em}& English native listeners& German native listeners & + English non-native listeners & German non-native listeners\\[.3em] + \cline{2-5} + \rule{0em}{1.1em}lexical status & \checkmark robust& \checkmark + robust& \checkmark less than native listeners& \checkmark less than native + listeners\\ + morphology & marginal & more than English & less than L1& less than L1\\ + lexical frequency & \checkmark robust& \checkmark + robust& \checkmark less than native listeners& \checkmark less than native + listeners\\ + neighborhood density & \checkmark robust & \checkmark robust & \checkmark less + than L1& \checkmark less than L1\\[.3em] + \hline\hline + \end{tabularx} +\end{table*} |