Understanding Figurative Language and Inferences in Cognitive Psychology, Slides of Cognitive Psychology

Download Understanding Figurative Language and Inferences in Cognitive Psychology and more Slides Cognitive Psychology in PDF only on Docsity!

Cognitive Psychology

Notes 12

Where We Are

 We’re continuing with higher cognition.

We still have:

 Language—Meaning

 Reasoning/Decision making

 Human factors

Plan of Attack

 The last unit was more about the

structure of language. This time we’ll

look at meaning. Our goal is to see how

cognitive psychology basics can be

brought to bear on an applied problem:

 Levels of representation of meaning (mostly

individual sentences).

 How to combine the meanings of sentences

into larger texts.

Levels of Meaning

 Literal: What the sentence actually says

(as close as possible to the exact words).

 Inference: Beyond literal to fill in missing

parts that aid comprehension.

 Figurative: The intended meaning is

different from the words used.

 Pragmatic: The words don’t convey the

meaning.

Literal Meaning

 One possibility is verbatim meaning (the

exact words). Unlikely:

 Think of examples of things that you have learned word for word. What do they mean? You usually have to repeat them to answer that.  There is evidence that verbatim representations are the fall-back strategy when other comprehension methods are not available.

Literal Meaning

 Mani & Johnson-Laird (1982):

 Provided determinate descriptions of

arrangements:

○ A is behind D ○ A is to the left of B ○ C is to the right of B

 Determinate descriptions were specific and

described an arrangement that could be

imagined (modeled).

Literal Meaning

 Mani & Johnson-Laird (1982):

 There were also indeterminate descriptions of arrangements: ○ A is behind D ○ A is to the left of B ○ C is to the right of A  Indeterminate descriptions were not specific and described an arrangement that would have to be represented with more than one possible model.

Literal Meaning

 Mani & Johnson-Laird (1982):

 Participants “remembered the meaning of

the determinate descriptions very much

better” (p. 183).

 Verbatim memory was only better for

indeterminate descriptions.

 It looks like, in the absence of a coherent

representation, participants fall back on

trying to remember the exact words.

Literal Meaning

 Evidence against verbatim:

 Sachs (1967): Participants heard passages

(e.g., about the telescope). At some point,

they were asked if a sentence was identical

to one in the passage.

Literal Meaning

 Evidence against verbatim:

 Sachs (1967):

○ He sent a letter about it to Galileo, the great Italian scientist. (original) ○ He sent Galileo, the great Italian scientist, a letter about it. ○ A letter about it was sent to Galileo, the great Italian scientist. ○ Galileo, the great Italian scientist, sent him a letter about it.

Literal Meaning

 Evidence against verbatim:

 Sachs (1967): Participants were asked

either 0, 80, or 160 syllables later in the

passage.

 The results were that the original form of the

sentence (verbatim) was only available long

enough to get the meaning.

 This argues against verbatim.

Literal Meaning

 Another way to think about literal

meaning is to use propositions.

 Kintsch, W. (1972). Notes on the structure of

semantic memory. In E. Tulving & W.

Donaldson (Eds.), Organization of memory

(pp. 247-308). New York: Academic Press.

Literal Meaning

 A proposition can be thought of as a single

idea from a segment of text.

 For example (p. 255) The old man drinks

mint juleps is really two sentences, one

embedded in the other:

 The man drinks mint juleps.  The man is old.

 This would produce two propositions.

Literal Meaning

 A proposition is a relation plus some

arguments.

 Kinds of relations (p. 254-255):

 Verbs ○ The dog barks. (BARK, DOG)  Adjectives ○ The old man. (OLD, MAN)  Conjunctions ○ The stars are bright because of the clear night. (BECAUSE, (BRIGHT, STARS), (CLEAR, NIGHT))

Literal Meaning

 Kinds of relations (p. 254-255):

 Nouns (nominal propositions) ○ A collie is a dog. (DOG, COLLIE)

 Arguments are usually nouns, but can be

whole propositions.

 Example:

 The old man drinks mint juleps.  (DRINK, MAN, MINT JULEPS)  (OLD, MAN)

Literal Meaning

 What propositions are in this sentence?

 The professor delivers the exciting lecture.

Literal Meaning

 What propositions are in this sentence?

 The professor delivers the exciting lecture.  (DELIVER, PROFESSOR, LECTURE)  (EXCITING, LECTURE)

Literal Meaning

 Evidence for propositions:

 Kintsch (1972) provided some data to

support propositions.

 Participants wrote all the clear implications

they could think of for sentences like Fred

was murdered. They did not write things that

were merely possible.

Literal Meaning

 Evidence for propositions:

 Kintsch (1972) found that inferences

supported aspects of the proposition theory.

 For example, for Fred was murdered , most

participants said the agent case was

necessary (e.g., someone murdered Fred).

 Change levels…

Inferences

 From Singer (1994):

 Androclus, the slave of a Roman consul stationed in Africa, ran away from his brutal master and after days of weary wandering in the desert, took refuge in a secluded cave. One day to his horror, he found a huge lion at the entrance to the cave. He noticed, however, that the beast had a foot wound and was limping and moaning. Androclus, recovering from his initial fright, plucked up enough courage to examine the lion’s paw, from which he prised out a large splinter (Gilbert, 1970) (p.479).

Inferences

 Singer (1994): How many inferences can

you find?

Inferences

 Singer (1994): How many inferences can

you find?

 Wound is an injury and not the past tense of wind.  Who is he?  Instrument used to remove the splinter.  Causal: Why moaning?

Inferences

 It is usually necessary for the

listener/reader to fill in missing text

information to make sense of what is

being presented.

 Diane wanted to lose some weight.

 She went to the garage to find her bike.

Inferences

 It is usually necessary for the

listener/reader to fill in missing text

information to make sense of what is

being presented.

 Diane wanted to lose some weight.

 She went to the garage to find her bike.

 Inference: Riding a bike is a way to lose

weight.