Part I: Language Processing and First-Language Learning
Table 2.1.
Mean number of correct responses from Mazzocco (1997)
Type of key word
Age Group
Accurately used
Nonsense
Pseudohomonym
Preschool
5.81 (0.40)
4.16 (1.61)
0.75 (0.98)
Note: N
= 32; maximum correct = 6; standard deviation appears in parentheses.
Content made available by Georgetown University Press, DigitalGeorgetown,
and the Department of Linguistics.
It is true, however, that children do not seem to have the same difficulties when a
nonsense word is used to refer to the object discussed in the story (twenty-three pre-
schoolers got four or more nonsense word interpretations correct, compared with one
preschooler who got four or more homonym interpretations correct). The existence
of synonymy in the task leaves the interpretation of the processes involved somewhat
uncertain, however. Consider that the nonsense word condition could be regarded as
differing from the homonym condition in at least two ways that are relevant to lexical
acquisition. First, although subjects know that the object being discussed is a clown,
it is being given a new label in the nonsense condition; hence, the new label functions
as a synonym for “clown.” In the pseudohomonym condition, children must not only
accept two labels for the same object (synonymy); they must accept a label that al-
ready refers to another object (homophony) as well. If we can simply subtract the ef-
fect of synonymy from both tasks, we are left with the conclusion that the difference
in participants’ performance is related to homophony alone. This conclusion, how-
ever, requires one to assume that the effect of violating these language constraints is
additive. It may be, however, that concurrent violation of two constraints on language
acquisition has an effect that is more or less than additive. This nonadditive effect
may obtain especially if the difference between participants’ performance on the ac-
curately used words and the nonsense words did not differ significantly—an infer-
ence the reader is left to make because the comparison is not specifically reported by
Mazzocco (1997). Moreover, the study also does not provide an accurate baseline we
might use to quantify the relative difficulty in acquiring a novel word and meaning
pair compared with the difficulty in mapping a familiar word to a second (i.e.,
homophonous) meaning.
The study I describe below seeks to address the foregoing questions and prob-
lems. An experiment was prepared to make a direct comparison between the acquisi-
tion of nonsense and homophonous words. The experiment also was designed to
determine the extent to which the linguistic context and the visual context of a scene
affect the acquisition of homonyms in comparison to nonhomonyms. The expected
outcome is that children will be at least partially successful in mapping homonyms to
new objects but that they will be significantly more successful in mapping nonsense
words to new objects. The hypothesis is that differentiation provided by linguistic
and visual context will aid children’s acquisition of homophonous words.
HOMONYMS AND FUNCTIONAL MAPPINGS IN LANGUAGE ACQUISITION
23
Figure 2.2.
Comparison of many-to-one mapping with one-to-many mapping
Content made available by Georgetown University Press, DigitalGeorgetown,
and the Department of Linguistics.
Experiment One
Method
Do'stlaringiz bilan baham: |