|
|||||
Cognitive
Psychology PSY 504
VU
Lesson
22
PATTERN
RECOGNITION (CONTINUED)
Effects of
Sentence Context
Cognitive
psychologist wants to study
whether the sentence context
has effect. In this order
an
experiment
was conducted by Tulving,
Mandler & Baumal in
1964.
Tulving,
Mandler & Baumal (1964)
conducted an experiment to show
the effect at the
multiword
level.
They used material in their
experiment like
Countries
in the United Nations form a
military alliance.
The
huge slum was filled
with dirt and
disorder.
Each
sentence provides an eight-word
context preceding a critical
word. In various
conditions
subjects
would see the
following:
0
Context
disorder
4
Context
Filled
with dirt and
disorder
8
Context
The
huge slum was filled
with dirt and
disorder
The
critical word presented
after the context for a
very brief period. The
experimenters
manipulated
the duration of this
critical word from 0 to 140
milliseconds. They were
interested in
how
bottom-up information interacted
with context.
Results
Results
were showing that the
probability of a correct identification
increases both as amount
of
context
increases and as the
exposure duration
increases.
Conditions
flash
duration
percentage
of correct identification
0
context
0ms
0%
4
context
0ms
10%
8
context
0ms
16%
0
context
140ms
70%
4
context
140ms
80%
8
context
140ms
98%
0
context
60ms
30%
4
context
60ms
60%
8
context
60ms
70%
Maximum
effect of context is seen at
60ms exposure, although the
effect of context
doesn't
disappear
at 140ms exposure. The
effect diminishes somewhat
between 60 and 140
milliseconds
because
subjects in the eight-word
context condition are
performing almost perfectly
and show
little
benefit of further exposure,
whereas subjects in the zero
word condition continue to
benefit
from
the longer exposure. The
results indicate that
subjects can take advantage
of the context to
improve
their identification of the
words.
Implications
That
experiment shows that we can
use sentence context to help
identify words. With context
we
need
to extract less information
the word itself in order to
identify it. We can use
context to fill in
words
that didn't even
occur.
68
Cognitive
Psychology PSY 504
VU
We
are able to fill in the
missing from as we read the
sentence, perhaps we did not
even notice it
was
missing. PDP models can
help explain this effect of
context better than other
models while
also
accounting for feature
analysis.
Context
and Speech
Text
OK but does this work
with speech. Speech is
experienced sequentially in a more
linear
fashion
than text.
A
concept Phoneme
Restoration Effect demonstrated in an
experiment by Warren (1970).
He
had
subjects listen to the
sentence. The state
governors met with their
respective legislatures
convening
in the capital city. With a
120 ms pure tone replacing
the middle s
in legislatures.
Only
1
in 20 subjects reported hearing
the pure tone and
even he wasn't able to
locate it clearly.
Phoneme
Restoration Effect
A
nice extension of this first
study is an experiment by Warren &
Warren (1970). They
presented
subjects
with sentences such
as:
It
was found that the
eel was on the
axle
It
was found that the
eel was on the
shoe
It
was found that the
eel was on the
orange
It
was found that the
eel was on the
table
Wheel,
heel, peel, meal
In
each case, the denotes a
phoneme replaced by non-speech.
For the four sentences
above,
subjects
reported hearing wheel,
heel, peel, and meal,
depending on the
context.
Implications
The
implications of this experiment
are:
Context
fills in gaps and affects
our perception just as in
texts.
The
identification of the critical
word is determined by what
comes after the
critical
word.
Heel, peel, meal and
wheel are critical
words.
Thus
the identification of words
can depend on the perception
of subsequent
words.
In
a nutshell when you ever
face a problem you should
grasp the context. When
you grasp the
context
you are able to understand
and handle the
problem.
69
Table of Contents:
|
|||||