| dc.description.abstract | It has been found that some attributes of a word predict how easy
it will be to translate that word into a second language (Murray,
1986) . It is not clear whether these attributes are specific to
translation itself or reflect the ease of access to the lexicon
in the first language. The present study was designed to provide
a replication of Murray^s results as well as to determine whether
the word attributes that predict translation do so by
facilitating processing in the first language, or by facilitating
the production of a word in the second language. In a two phase
study, twenty-six bilinguals identified English words from
English non-words in a lexical decision task to provide a measure
of first language processing. In the second phase they translated
English words into French to replicate Murray (1986). The data
were analyzed in two parts. First, following Murray, an item
analysis which averaged response time over subjects was conducted
for translation and lexical decision. A multivariate regression
analysis of the scores revealed that word frequency was the best
predictor of both translation ease and lexical access; and that
number of synonyms, age of acquisition, and goodness correlated
highly with both processes; memorability, similarity, and
emotionality appear to be unique predictors of translation.
Second, a within subject comparison of translation time with
lexical decision times for words seen earlier in the experiment
revealed that lexical decision reaction time had a small but highly reliable correlation with translation time. There was no
pronounced effect of the lexical decision task on the speed of
translation. | |
