Phonological Organization + Development
What is the difference between a phoneme and a morpheme?
A phoneme is the smallest unit of sound used in language. A morpheme is the smallest unit of sound used in language that carries meaning.
What does “lexical access” mean?
It means recognizing a word as a word that you know (e.g., hearing “leam bacon” but pulling “lean” out of your lexicon as the word that you probably heard).
Suppose a child uses the word “spoon” to refers to a soup ladle. Which of the following word learning biases would lead the child to think make this overgeneralization?
a. whole object assumption
b. taxonomic assumption
c. mutual exclusivity
d. joint attention
b. taxonomic assumption
Describe the hypothesis-testing account of cross-situational word learning.
Making a single guess at the word's meaning, then seeing if that guess is "confirmed" in a later situation.
When English speakers are asked to provide the plural form of “gub” (a nonsense word in English), they tend to pronounce it /gʌbz/ with a /z/ sound at the end to indicate plural. This is an example of:
a. assimilation
b. dissimilation
c. deletion
d. epenthesis
e. metathesis
a. assimilation
Consider the sentence: The cow jumped over the moon.
“-ed” is
a. a prefix
b. a suffix
c. an infix
c. an infix
A noun in one language tends to mean the same thing in another language (e.g., “dog” and “perro” both refer to a canine pet). However, what verbs encode in one language are not necessarily encoded in the verbs of another language (e.g., verbs in English tend to specify the manner of motion (e.g., “run”), while verbs in Spanish tend to specify the goal location of the motion (e.g., “exit”). This fact supports the
a. natural partitions hypothesis
b. linking expectation hypothesis
c. idea that qualities of the specific linguistic input affect which types of words are learned first
a. natural partitions hypothesis
What 3 things do we as learners rely on when utilizing cross-situational word learning?
Social cues, object presence, and linguistic framing
In the word kæts “cats”, /s/ is a
a. phoneme
b. morpheme
c. word
d. all of the above
e. a and b only
e. a and b only
How does the logogen model fail to account for successful interpretation of /lim/ in “lean bacon”?
The set of logogens consists of known words. “Leam” is not in the list, and “lean” does not match the pronunciation the listener heard. Context doesn’t help because it also doesn’t support a word that sounds like “leam”.
Why do we even think that to learn the meaning of a word, a learner would need multiple exposures to that word in various contexts?
There are an infinite number of possible referents for a word in any given situation, but multiple exposures allow the compare across situations and see if a word’s potential meaning from one instance holds up in the next instance.
What happens when a child uses conjectured mapping that is not confirmed?
They attempt to guess again and provide another opportunity for conjectured mapping.
True or False: In English, /s/ and /z/ are allophones when used at the beginning of a word.
True. /sɪŋ/ (sing) and /zɪŋ/ (zing) differ by a single sound and that difference does signal a change in meaning. Here /s/ and /z/ are phonemic, not allophonic.
A parent is concerned that their child isn’t speaking correctly, and instead makes up words such as “goed” and “wented”. Explain to the parent why the child’s speech actually suggests an important development in their language acquisition, rather than something to be concerned about.
The child is showing more morphological complexity than a child who just says “go” – it indicates that they are learning the ways in which the past tense is formed and working on generalizing it. But some words are irregular and it takes time to learn that the general rule doesn’t apply to these irregular verbs.
What is the reason, according to the natural partitions hypothesis, that children’s early vocabularies consist mostly of nouns?
It’s because nouns refer to physical things that don’t require you to take a perspective, whereas verbs label events that you need further linguistic or social context to identify.
What did Yu and Smith conclude about word learning through their experiment?
They concluded that learners keep track of the co-occurrences of labels and objects in order to figure out which object occurs to which label.
In Werker et al. (2002) (covered in class), 14-month-old English-learning infants didn’t, on average, look longer at a picture paired with the sound /dI/ (“dih”) if previously they had seen that same picture repeatedly paired with the sound /bI/ (“bih”). However, when the researchers looked specifically at just the 14-month-olds with a comprehension vocabulary size of greater than 200 words (as compared to the 14-month-olds whose vocab size was < 200 words), they found that they did look longer on the “switch” trials than they did on the “same” trials. The researchers considered this suggestive that the more words a child knows,
a. the more practice they have at paying attention to minimal phonetic distinctions, thereby allowing them to notice the difference between /dI/ and /bI/ in the experiment.
b. the more likely the child is to have encountered the words /dI/ and /bI/ already in their day-to-day life, thereby making them more likely to pay attention to the switch from /dI/ to /bI/ in this experiment.
c. the better they probably are at matching a sound to a picture, thereby making them more likely to notice when a picture previously paired with one sound is suddenly paired with another similar sound.
d. the older they probably are, making them better able to sit through the experiment and pay attention to the repetitive stimuli.
a. the more practice they have at paying attention to minimal phonetic distinctions, thereby allowing them to notice the difference between /dI/ and /bI/ in the experiment.
How does the distributed cohort model account for successful interpretation of /lim/ as /lin/ in “lean bacon”?
The distributed cohort model includes a context layer that keeps track of recent context and is therefore able to take into consideration the place of articulation of the following consonant: /b/ is +bilabial. Given this context, the model interprets that /m/ was only this particular nasal because of coarticulation of the upcoming bilabial, but if the word had been uttered on its own, then probably the nasal was /n/ because “lean” is in the list of known words and the context supports that interpretation.
Children learn words best when the speaker does “follow in labeling”. This occurs when the speaker
a. uses words to label the objects, events, and relationships that the child is paying attention to in that moment.
b. uses the same word repeatedly to provide the child with multiple opportunities for figuring out the meaning of that word.
c. follows up an unclear usage of a word with a better and more transparent use of the word.
d. follows the child around, letting the child do most of the talking in order for the child to get practice labeling things.
a. uses words to label the objects, events, and relationships that the child is paying attention to in that moment.
Suppose a learner hears a novel word (such as “WUGGIES”) in two different situations. In the first situation, the learner believes it to refer to a pair of rain boots, but in the second situation, there is no rain boots around and the conversational context does not seem like it could have anything to do with boots of any kind. According to the hypothesis testing model of cross-situational word learning (Trueswell et al., 2013), after having guessed incorrectly the first time, the learner has no better chance than before of now correctly guessing the word’s meaning the second time he hears it. Explain why.
According to Trueswell et al. (2013) the learner did not keep track of the other things the word WUGGIES could have been referring to, so they are essentially starting over – trying to figure out the meaning of WUGGIES from the current situation only.