The Modularity Hypothesis
Fodor’s
proposal that some cognitive processes, in particular language and perception,
operate on only certain kinds of inputs and operate independent of the beliefs
and other information available to the cognitive processor or other cognitive
processes.
A
proposal from the philosopher Jerry Fodor (1983, 1985) made a quite different
argument about the relationship of language to other aspects of cognition.
Fodor
argued that some cognitive processes—in particular, perception and language—are
modular. What does it mean for a process to be a module? First, it means the
process is domain-specific: It operates specifically with certain kinds of
input and not others. With regard to language, for example, Fodor argued that
sentence parsing involves processes that are specific to the division of
phrases and words into constituents. Such processes are meant only for parsing
and are of little use in other cognitive tasks.
Modularity
of a process also implies that it is an informationally
encapsulated process: It operates independently of the beliefs and the
other information available to the processor.
The
modularity hypothesis, then, argues
that certain perceptual and language processes are modules. (In the case of
language, one such process is that which parses input utterances.) These
processes are thought to be set apart from other cognitive processes, such as
memory, attention, thinking, and problem solving, that are thought to be
nonmodular. Modular processes operate automatically and independently (at least
at the first stages of processing) of other cognitive processes, such as
thought. Modular processes are domain specific, which means that they are
specialized to work with only certain kinds of input. The syntactic parsing
aspects of language are not used in other kinds of cognitive processing. In
this sense, then, language really is a special and very independent cognitive
process.
The
experiment by Swinney (1979) on lexical ambiguity resolution offers findings
that support the modularity hypothesis.
No comments:
Post a Comment