Review Questions: Sounds & Sounds of Words
(1)Terms/concepts to know: contrastive sounds, phonemes, international phonetic
alphabet, categorical perception, within-category perception, across-category
perception, voice onset time, high amplitude sucking procedure, habituation,
dishabituation, head turn preference procedure, maintenance and loss theory,
functional reorganization theory, contrastive feature, unimodal distribution,
bimodal distribution, Switch Procedure, minimal pair, Stager-Werker task,
neighborhood hypothesis, visual choice task
International phonetic alphabet:
Categorical perception: Categorical perception occurs when a range of stimuli that
differ continuously are perceived as belonging to only a few categories, with no
degrees of difference within a given category.
Voice onset time:
High amplitude sucking procedure:
Head turn preference procedure:
Maintenance and loss theory: “Use it or Lose it” Infants maintain contrasts being
used in their language and lose all the others.
Functional reorganization theory: Unconscious filter imposed when sounds are
processed as language. Data distributions determine what the boundaries are in the
Visual choice task:
a. Contrastive sound – elements of a word that signal a change in
b. Phonemes – basic perceptual units of which speech is composed, units
that are used to build morphemes, languages have a finite inventory of
these units, they are not units of meaning EX: pig vs big
c. International phonetic alphabet – languages drawn from a common
set of sounds d. Categorical perception – occurs when a range of stimuli differ
continuously are perceived as belonging to only a few categories, with
no degrees of differences within a given category
e. Within-category perception – the ability to perceive the difference
between two sounds that are the same but are heard at different
f. Across-category perception – being able to perceive the difference
between two sounds that are different but are heard at different
g. Voice onset time – feature of the production of stop consonants, the
length of time that passes when a stop consonant is released and
when voicing, the vibration of the vocal folds, periodically begins
h. High amplitude sucking procedure – infant given a pacifier that
measures sucking rate
i. Habituation – infant sucks to hear sound until bored
j. dishabituation – infant will suck to hear sound if the sound is no
k. head turn preference procedure – infant sits on caretakers lap, wall in
front of infant has a green light mounted in the center of it, walls on
the sides of the infant have red lights mounted in the center of them,
and there are speakers hidden behind the red lights, sounds are
played from the two speakers mounted at eye-level to the left and
right of the infant, sounds start when the infant looks towards the
blinking side light, and end when the infant looks away for more than
two seconds, infant essentially controls how long he or she hears the
sound, differential preference for one type of sound over the other is
used as evidence that infants can detect a difference between the
types of sounds
l. maintenance and loss theory – infants maintain contrasts being used
in their language and lose all the others
m. functional reorganization theory – unconscious filter imposed when
sounds are processed as language…changes attest experimentally
reflect operation of postperceptual processes that activate for
language sounds, data distributions determine what the category
boundaries are in the filter, constructing this filter does not affect
base-level sound perception
n. contrastive feature – sounds that have “opposite” values for one
feature, EX: “k” and “g” differ only with respect to voicing
o. unimodal distribution – distribution with one single mode
p. bimodal distribution – distribution with two modes or distinct peaks
q. Switch procedure – measures looking time EX: look at the tam…switch
to look at the taam
r. Minimal pair –
s. Stager-Werker task –examine phonetic info used by infants in early
word learning, used recently developed procedure in which infants
are first taught word-object pairings, and then tested on their ability to detect a change in either the word, the object or both the word and
object, 8 mo olds did notice difference (know that “b” and “d” are
different), 12 mo olds did not, conclusions: 14 mo olds can
discriminate the minimally contrasting words, but they fail to notice
the minimal change in the sounds when they are paired with objects
ie: when they are words with associated meaning, they can perform
the task, when the words are more distinct, 14 mo olds used more
detail to represent sounds than they do to represent words
t. Neighborhood hypothesis – when a child knows two words that differ
only by a single phoneme (like “cat” and “bat”) more attention to
detail is required to distinguish them, content of children’s vocab
drives their ability to notice the difference between words that differ
u. Visual choice task – preferential looking, a two-alternative forced
choice looking task that compares visual fixations to target and
(2) Describe one way you can recognize if two sounds are phonemic in a language.
Identify phonemes (contrastive sounds that signal a change in meaning)
Kids of the world require knowledge of phonemes before they can figure out what
different words are - and when different meanings are signaled by different words.
The words sound the same but spelled and mean something different.
One way you can recognize if two sounds are phonemic in language is if the two
sounds signal a change in word meaning or if the two sounds are spelled differently
in two words that sound the same when spoken.
(3) “All languages use the same set of contrastive sounds.” Describe one way that
you could interpret this statement as false. Can you think of any way to interpret it
that might make it true?
Languages draw from a common set of sounds (which can be represented by the
International Phonetic Alphabet (IPA)), but only use a subset of that common set.
One way that we can prove that all languages do not use the same set of contrastive
sounds is with the example of r/l. R/l is a phonemic contrast in English but not in
Japanese. Lisa = Risa for Japanese speakers. One way that we can prove this
statement to be true is …
(4) Will the acoustic signal of the vowel sound “oo” (as in the English word boot,
represented as [u] in IPA) vary between speakers of the same language? If so, give
an example that shows this variation. If not, explain why not.
The acoustic signal of the vowel sound “oo” will contrast between speakers of the
same language because vowels combine acoustic energy at a number of different
frequencies, listeners must perform a “frequency analysis” of vowels in order to
identify them EX? (5) Give an example of contrastive sounds in English that differ only in their voice
onset time (VOT). Which has a shorter VOT and which has a longer one?
Dae vs tae
An example of contrastive sounds in English that differ only in their voice onset time
is “dae” and “tae”. “dae” has a shorter VOT and “tae” has a longer VOT.
(6) Consider the graphs below. The lefthand one shows subjects’ responses to a task
while the righthand one shows their reaction time.
(a) Briefly describe what decision a subject must make for a discrimination task.
Would be to see if the two sounds are the same of different.
(b)Briefly describe what decision a subject must make for an identification task.
Would be able to decide if the one sound is da or ta.
In an identification task one must identify what the sound is.
(c) Is the task the subjects were asked to perform a discrimination task or an
identification task? Explain how you know. (Hint: Look at the x axis labels for
both graphs, and the y axis label for the lefthand graph.)
In this task subjects were asked to perform an identification task because
subjects were tested on how long in milliseconds to hear the “dae” sound.
(d)Why does reaction time rise sharply in the righthand graph? Your answer
should make reference to both when the sharp rise occurs (when compared
to the lefthand graph) and what it means for the subject to have the reaction
time rise in this kind of task.
For the reaction time to rise sharply that means there was more confusion.
Unsubjects is unsure about what they are hearing. Reaction time rises sharply in the righthand graph at about 25ms whereas on
the lefthand graph there is a steep drop in the amount of people who heard
the dae sound at 25 ms. The lefthand graph shows more uncertainty and/or
error at the category boundary. The righthand graph shows a longer
decision time at the category boundary.
(7) Why is reaction time longer at a category boundary for an identification task?
Because the subject is more uncertainty causing a longer decision time.
Reaction time is longer at a category boundary for an identification task because the
subject is unsure about hearing the difference between the two sounds and
identifying exactly when the two sounds switch. “dae” and “tae” sound the most
similar at a category boundary.
(8) Why is reaction time shorter when stimuli cross a category boundary than when
stimuli are within the same category for a discrimination task?
Because it is easier to discriminate between two sounds than to identify one.
Reaction time is shorter when stimuli cross a category boundary than when stimuli
are within the same category for a discrimination task because the difference
between the two sounds such as r and l can be perceived more easily. Sounds
within the same category are often perceived to be similar so they are harder to
(9) Why did Werker et al. (1981) test both adult Hindi speakers and adult English
speakers in their experiment when they were trying to determine when English
infants lost the ability to hear Hindi contrastive sounds?
Hindi adults can easily distinguish sounds that are used contrastively in their
language. English adults are terrible (below chance), though there is some variation
depending on which sounds are being compared. English infants between the ages
of 6-8 months aren’t quite as good as Hindi adults - but they’re certainly much
better than English adults! They haven’t yet learned to ignore these non-native
Werker et al. Tested both adult Hindi speakers and adult English speakers in their
experiment when they were trying to determine when English infants lost the
ability to hear Hindi contrastive sounds because adults can’t distinguish sound
contrasts that are non-native for both vowels and consonants even if the difference
is acoustically salient. They wanted to find out what changes from childhood to
adulthood. They f ound out that Hindi adults can easily distinguish sounds that are
used contrastively in their language while English adults are terrible, though there is
some variation. English infants aren’t as good as Hindi adults, but they are much
better than English adults. They have not learned to ignore the non-native
(10) Why were Salish and Hindi children tested in the experiment by Werker & Tees (1984) that examined English infants’ ability to perceive non-native (Salish/Hindi)
Control groups (make sure experiment is doable by infants): Hindi and Salish
infants do perfectly
Salish and Hindi children were tested in the experiment by Werker & Tees that
examined English infants’ ability to perceive non-native contrasts because they
were the control group to show that the task was possible/doable by infants.
(11) Is the Maintenance & Loss Theory a structure-changing theory or a structure-
adding theory? How do you know?
Use it or lose it! Infants maintain contrasts being used in their language and lose all
the others. Because information isn’t added but it is lost and replaced with new.
The maintenance and loss theory is a structure changing theory because infants
maintain contrasts b