Language Comprehension

Question 1

How is language comprehension?

Answer 1

Comprehension is a rapid guessing game.
Comprehension is fast, highly incremental (process language as it comes in; on the fly), highly interactive

Question 2

Evidence showing spoken sentences are interpreted in real-time

Answer 2

Procedures:
The participants view a display of cards and then are told, “Now put the five of hearts that is below the eight of clubs above the three of diamonds.”

Eye movements show that, as the sentence is being uttered, it is being interpreted quickly and incrementally

Question 3

What is lexical access?

Answer 3

Process of accessing a word from your mental dictionary

Question 4

What are the theories of word recognition?

Answer 4

Parallel activation model
Cohort theory

Question 5

What is the parallel activation model?

Answer 5

We search in a serial parallel search

Predicts homophones activated simultaneously
The idea that both meanings of a homophone are activated until the context allows you to decide which meaning is being used

Question 6

What are homophones?

Answer 6

Words with different meanings but the same pronunciation

Question 7

Parallel Access of Meaning (Homophones) Study

Answer 7

Procedures: Cross-modal priming task
Hear a sentence ending with a homophone that either supports the noun meaning or verb meaning of the word watch.
While hearing the sentence, you are seeing a word on a screen. The word is either semantically related to the verb (look) or the noun (time) of the homophone watch or an unrelated word completely (dart).
The words are also presented at two different times; they are either presented immediately after the sentence ends or presented 250 msec after the sentence.
Judge if the word is real or not.

Findings:
They are faster to respond to both semantically related words compared to the one not semantically related when presented immediately after the sentence ends.

They are faster to respond to the semantically related word when the word is flashed 250 msec after the sentence ends.

Conclusion:
This supports the idea that people consider both meanings (parallel accessing for word recognition) before picking one meaning based on the context. There is parallel access when the semantically related words for both means are chosen right after the homophone is uttered, but when there is time (250 msec) to consider the context, then they are faster when looking at the word that is semantically related to that specific context

Suggests bottom-up activation followed by contextual selection
Suggests timing matters

Question 8

Cohort Theory of Spoken Word Recognition

Answer 8

Hearing the initial phonemes of a word will activate in parallel a cohort of words that share initial onsets (phonemes)
Further phonemes restrict the cohort until the uniqueness point is encountered when you know with certainty the word being said

Question 9

Evidence for Cohort Theory

Answer 9

Procedures: Visual World Paradigm
Track the eye movements of listeners as they hear instructions such as “Put the breaker above the square” with a chess setup present. There were clip-art images of a beaker (target word), a beetle (cohort competitor), and a carriage (unrelated word).
So, trying to see what people look at while the sentence is being uttered.

Findings:
All participants eventually reach the target word.
People are more likely to look at a cohort competitor than an unrelated word.
When the target word (beaker) is uttered, both the competitor and target are fixated around the same proportion. But by the offset of the target word then people diverge and more are proportionally looking at the target.

Question 10

How do computer algorithms parse grammar?

Answer 10

top-down parsers
bottom up parsers
Hybrid top-down/bottom up parsers

Question 11

What are top-down parsers for grammar?

Answer 11

Start with starting symbol, work down

Question 12

What are bottom-up parsers for grammar?

Answer 12

Start with words, find rules to generate them, work up

Question 13

What evidence shows that sentence processing is incremental and fast?

Answer 13

Evidence: shadowing task
Subject repeats words of a spoken sentence immediately after hearing it
High-performing subjects can lag at 250 msec
Thus showing how people actually analyze sentence structure on the fly

Evidence: garden path sentences
We garden path because we posit structure on the fly that later turns out wrong

If we didn’t incrementally comprehend sentences then we wouldn’t make mistakes like garden paths

Question 14

What is the modular view of comprehension?

Answer 14

Sound waves -> Phonological Processor (and lexical accessor) -> Syntactic processor (parser) -> Semantic processor -> pragmatic processor

Strictly feed forward

Question 15

What is the interactive view of comprehension?

Answer 15

Sound waves -> Phonological Processor (and lexical accessor) <-> Syntactic processor (parser) <-> Semantic processor <-> pragmatic processor

Information can flow forward and backward to earlier stages

Question 16

What is the interactive view of garden paths?

Answer 16

Garden paths can be avoided based on context

Question 17

Most reading studies suggested garden-paths couldn’t be avoided by semantics and context eye-tracking examples

Answer 17

The bully hit the girl with the wart and then..

We would experience a garden path when viewing the word wart regardless of meaning. Because we tend to connect the prepositional phrase (with the) to the verb (put).

The bully hit the girl with the stick and then..

We would not experience a garden path since the grammar made sense.
More consistent on the modular view.

Question 18

What is minimal attachment strategy?

Answer 18

Our initial parsing of sentences tends to stick to the simplest sentences

Question 19

But what is the difference between reading and listening?

Answer 19

The context during reading is memory-dependent, relatively impoverished, and just kind of hard

Question 20

The Visual Word Paradigm for Sentence Processing

Answer 20

Participants see a setup with a frog on a napkin (target), a frog by itself (competitor), a napkin, and a box

Participants were told, “Put the frog on the napkin into the box.”

Within 300 msec of hearing “napkin,” people stop looking at the competitor frog and look at the target.

Eyes don’t just go to any possible referent for a word, but instead, the eyes go to referents that are possible given the structure of the sentence -> No one looks to competitor’s napkin

Question 21

What is temporary syntactic ambiguity?

Answer 21

There are moments in a sentence that are ambiguous, but it is only temporary
Putting the frog on the napkin into the box is only temporarily ambiguous when referring to the frog on the napkin

Question 22

Study Using the Visual World Paradigm

Answer 22

Procedures: Visual World Paradigm
Participants were told, “Put the frog on the napkin into the box,” but with two different setups (conditions).

The 1-referent context has a frog on a napkin (target), a horse by itself, a napkin, and a box.
The context supports the idea that the phrase “on the napkin” refers to a goal interpretation of “put,” making it seem redundant.

The 2-referent context has a frog on a napkin (target), a frog by itself, a napkin, and a box. The context supports the scenario, making it the phrase on the napkin as a modifier to distinguish which frog they are addressing.

Predictions of Modular Parser:
People should pursue goal interpretation first, regardless of context. So in both conditions, the participant should look at an empty napkin.

Predictions of the Interactive view:
If context matters, only garden path in 1-referent context.

Findings:
Visual context matters, supporting the interactive view
Participants do not experience a garden path in the 2-referent context condition, and do garden path in the 1-referent context, with them looking at the empty napkin.

Question 23

What information sources matter during listening?

Answer 23

Referential/pragmatic constraints
Verb tendencies
Prosody

Question 24

What is prosody?

Answer 24

The patterns of stress and intonation in a language

Question 25

Prosody study

Answer 25

Procedures: Two participants: one classified as the speaker, who has to tell the listener to recreate the action of the experimenter, and the other as the listener, who is in front of the other but has a divider that prevents it unable to seeing what the other person is doing. The experimenter demonstrates either an instrument (tap frog with a feather) or a modifier action (tap a frog who has a feather) to the speaker. Findings: The speaker helps differentiate which action would differ in the length of pauses between words. So the listener would use the prosody of the speaker to decide the action they are being told.