The study of language by psycholinguists

-interested in the mental processes that occur during the act of understanding language

-off-line versus on-line measures



off-line: we use tasks some time after language has been encountered to get insights into how that language was processed; these tasks often involve memory. In example below examine errors

-question: during comprehension do we store a "copy" of what we encounter?

Task: read a passage

later given a number of sentences; told that in each set one was encountered during reading; task choose the one that word for word as presented each set consisted of
-the target sentence: "Galileo, the great scientist, was driven from Padua to Modeno:

- a sentence with same meaning but syntactically different (eg., passive->active);"Galileo, the great scientist, drove from Padua to Modeno:

-sentence syntactically the same, but meaning different: "Galileo, the great scientist, was driven from Modeno to Padua":



Or consider this study: examines the processing of non-literal language
Question; when encountering a sentence in which the expressed meaning differs from the intended meaning, do we have to initially process the expressed (literal) meaning BEFORE we can figure out the intended nonliteral sense

we had Ss read passages in which a Proverb is used either in it's nonliteral (proverbial) sense or in a literal context; manipulated as well the familiarity of the proverb
--example; one can have a proverb "The grass is greener on the other side of the fence" placed in a context dealing with the colour of one's neighbours lawn (literal use) OR in a context dealing, in general, with envy (proverbial use)
--after reading a set of passages we used a cued recall technique (based on encoding specificity logic); ie cued by a phrase related to the literal meaning (used fertilizer) or nonliteral sense (was envious);; PREDICTIONS??



On-line studies of language;
In this type of task, one studies comprehension during the act of comprehension
-example; eye movements during reading
-example; moving windows procedure

Consider the following example:
is comprehension based on a modular model in which syntax dominates
-one syntax based comprehension model argues that people read word by word, building a mental structure as they go along; with each word encountered they add to the existing structure in a way that leads to the simplest phrase structure (principle of minimal attachment)



-can study this via GARDEN-PATH sentences
consider reading
"Fat people eat...

According to the theory, at this point the simplest structure is to assume what you have is a VP, and thus you should expect a NP (such as the continuation... "sauces"
But what happens if the next word is "accumulate"

according to the theory you have to change the structure to make sense of it (measure how long pause, where they look back to, etc.)



Issue: to what extent is syntactic parsing modular (that is, the syntax machine does it's work without taking into account semantic or pragmatic knowledge)

Thus: what would happen if we embed the sentence:
"Fat people eat accumulate" into a context which emphasizes that biological basis for weight gain.

Can we find evidence that syntactic processing is not modular but is interactive (resolves meaning by taking into account various sources of information at the earliest stages of comprehension)

Consider (McCrea et al) "The detective arrested the thief"
versus
"The thief arrested the detective"



Psych 235 Home Page