Lecture 1
Lecture 2
Lectures
3 & 4
Lectures
5 & 6
Lecture 7
Lecture 8
Lecture 9
Lecture 10
Lecture 11
& 12
Lecture 13
Lecture 14
Lecture 15
Lecture 16
Lecture 17
Lectures
18 & 19
Lecture 20
Lecture 21
Lecture
22 & 23
Lecture 24
End of class
Introduction to Cognitive Psychology
syllabus
Cognition and cognitive psychology
the scientific study of cognition
(cognition = thought)
an approach to psychology
(cognitive psychology - as a way of approaching human psychology)
What is cognitive psychology - the
scientific study of the mind
Five features
• 1. materialism - mind is
a function of the brain
• 2. nested approach to nature
- nature can be studied at different levels - can approach mind without
necessary knowledge of brain
• 3. use of models to understand
cognition
• 4. willingness to
invoke mental processes
• 5. experimental
History of cognitive psychology
1. Introspectionism
• 2. The rise and fall
of behaviorism
• 3. Antecedents of
cognitive
• Information processing
revolution
• the computer metaphor
• 4. Freudian psychology
- the importance of nonconscious cognition
• 5. Linguistics
Matlin's themes
• 1. cognition is active not passive
• 2. remarkably efficient
and accurate
• 3. positve info better
than negative info
• 4. cognitive processes
are interrelated
• 5. bottom-up and top down
processing
• bottom-up processing - or data-driven
processing - importance of the actual stimulus
• top-down processing - or conceptual
processing - influence of knowledge about world
Lecture 2
Exp 4605
September 1, 2005
Cognitive Neuroscience
Cognitive Neuroscience = Examines how cognitive processes
can be explained by the structure and function of the brain.
Gross anatomy of cerebral cortex.
Frontal lobe -- movement, emotion, speech, thinking, reality monitoring,
inhibition of impulse
Temporal lobe -- language comprehension, memory, and auditory system
Parietal lobe -- attention, reading, association, and touch (haptic)
Occipital -- vision
Ways of studying Cognitive Neuroscience:
Brain-damaged patients (brain lesions)
Neuroimaging.
Animal models (and single-cell recording)
Brain damage
Identify patients with brain damage.
Correlate brain damage to cognitive or behavioral deficits.
Assumption: if brain damage results in deficits in one type of
cognition (e.g., language), then it is thought that the area of the brain
is involved in that cognitive process.
Neuroimaging
PET(Positron emission tomography)
MRI and fMRI (magnetic resonance imagery; functional magnetic resonance
imagery)
EEG: electroencepholography
Positron Emission Tomography (PET)
measures blood flow
trace active areas in brain by injecting radioactive
oxygen or sugar into blood
active areas of brain require more blood, show more
radioactivity
good spatial resolution, poor temporals resolution
((new picture every 30 seconds)
MRI and fMRI
fMRIs measure the amount of blood in each area of the brain via magnetism.
measures deflection of oxygen atoms in magnetic
field
active areas have more oxygen
good spatial and temporal resolution (new picture
every .5 seconds)
EEG and Evoked-related Potential technique
EEG measures brief fluctuations in entire brain's electrical output
electrodes placed on scalp
localization occurs by comparing signal in different
regions of scalp
poor spatial but outstanding temporal resolution
(.001 second)
ERPs
ERP-- measure repeatedly after similar stimuli. average
across many EEG recordings to see "waves" or patterns
in electrical
output of brain.
Von Restorff effect -- oddball items are better
remembered (Fred, James, Michael, Greg, Joe, Mariana, Tom). "Mariana"will
be
remembered best.
the ERP shows a spike called a P300 (positive wave 300 msec.
after name) when the "oddball" item is given.
Lectures 3 & 4
September 6 & 8
EXP 4605
Attention
Attention: the concentration of mental activity.
Allows you to select or pick out the salient or important characteristics
of your sensory world. The information you seek out is attended information.
That which you ignored is unattended information.
Divided Attention:
Divided Attention: when one tries to pay attention to two or more sources
of information.
Typists who type and conduct conversations at same time.
With practice, one can learn to do it. Advantage: talk and
work at same time. However, one does not retain the meaning of one’s
typing.
Divided Attention and cell-phone use while driving: Strayer
et al. (2003) found that drivers were slower to apply breaks while speaking
on cell phone (hands free) than drivers not using cell phones.
Selective Attention
Selective Attention: when one attends to one channel of information
among many competing sources of information.
Example: among many voices, you can concentrate on the
voice of your conversational partner at a loud party.
Exception: cocktail party effect: sudden switch of
attention when you hear your name (or other relevant information) in unattended
conversation.
Test of Selective attention: Dichotic listening task.
Participants wear headphones
Each ear receives different message
Instructed to attend to one message; ignore other
Results show that memory is good for attended ear; poor for unattended
ear.
Results show that participants do not even pick up on simple perceptual
features in unattended ear, such as gender or language of speaker.
Exception: when own name is presented to unattended ear.
Test of Selective attention: Stroop effect
Instructed to name ink color of presented words.
Words such as “red” are presented but written in yellow ink.
Task calls for participant to say “yellow”
But competing semantic (meaning) slows participants down and
creates errors.
Stroop effect
RED
GREEN
WHITE
PINK
BLUE
YELLOW
PURPLE
Attention and eye movements
Attention generally associated with direction of gaze.
We look with our fovea at attended location. When attention shifts,
saccadic eye movements direct us to new location.
Can we attend to locations in visual space other than the location
we are looking at?
The Posner paradigm.
Gaze is fixated at central marker. Then cue is given to direct
attention (but not eye movements) to another point in space. Cue
is usually accurate, but not always. Is response time faster when
cue is accurate than when not?
The Posner Paradigm
W X < Z
Expect light at location Z. When that occurs, trial is “valid.”
Expect light at location Z. When it occurs, at Y instead, trial
is “invalid”.
Eyes directed at X throughout.
Response times to W are slower than to Z.
Cognitive Neuroscience of Attention.
Two main neural areas:
anterior attention network: in frontal lobe and Posterior attention
network: in parietal lobes
Posterior Attention network: active in directing attention in
visual searches.
Damaged in hemi-field neglect.
Hemi-field neglect: patients ignore half (usually left) visual
world.
Anterior attention network: inhibiting responses, such as those
in Stroop (Posner & Fernandez-Duque, 1999). Top-down
control of attention.
Two topics in Attention and Consciousness
Thought suppression: consciously try to NOT think about something.
People are successful at first, but susceptible to rebound
effects later (that is, thinking about it more than they would ordinarily).
Blindsight: visual responses in the absence of conscious “seeing.”
Patients report blindness but can actually respond to visual
stimuli.
Competing explanations: islands of intact cortex vs. sub-cortical
routes from optic nerve.
Working memory
Lectures 5 & 6
Exp 4605
September 13, 15, 2005
Definition of working/short-term memory
Working memory: “is the brief immediate memory for material you
are currently processing.”
Equivalent to the older term: short-term memory
Classic research on working or short-term memory: Capacity
Immediate memory span (Miller, 1956) - 7+-2 items;
Measured with digit span
item = unit of information
Chunking - increase the amount of info in each item; increases the
number of recallable events.
immediate memory mnemonic - anything that improves performance
Digit span
Digit span task: repeat a string of digits immediately after hearing
them.
pronunciation rate also affects STM. Naveh-Benjamin and his colleagues
examined digit span in different languages. Found
differences in digit span that related to language used.
The more time it took to say digits in any particular language, the
shorter the digit span.
Duration
How long does information remain in STM
if rehearsal - indefinitely - (i.e. as long as it is activated)
if no rehearsal - estimates vary but 15-20 seconds is a good
bet
Serial position curve
The Serial Position Curve
free recall of a list of studied words
the primacy effect - words at beginning of list are well remembered
the recency effect - words at the end of list are well remembered
Forgetting
Brown-Peterson paradigm
subjects shown three items
engage in rehearsal prevention task for given interval (count
backwards)
retrieve three items
recall a function of rehearsal prevention interval
recall was better after a short period of rehearsal prevention
than a long one
semantic coding in STM --- release from proactive interference
Modern conception of working memory
Modern conceptions of working memory
Baddeley’s model: four major components: which can act largely
independently.
Four components
Phonological loop: auditory short-term memory
Visuo-spatial sketch pad: visual short-term memory
Episodic buffer: combines information from other system
Central executive: attentional mechanism which allocates attention
to different competing sources of input
Research on components
Core idea: we can use phonological loop and visuo-spatial sketch
pad simultaneously. With easy tasks, they do not interfere.
Visual and auditory tasks can be done simultaneously
Experiment on visuo-spatial sketch pad
Brandimonte’s exp: participants had to label mental images after
transformation; must use visual imagery; subject who
phonological loop was occupied actually did better
Visual/spatial imagery can interfere with working of visuo-spatial sketchpad
Lecture
7
Exp 4605
September 22, 2005
Long-term memory
Long term memory - large capacity
(unlimited) & long-lasting, but often difficult to access
- long-term
memory is not a warehouse, library, etc.
- memory is
multidimensional - cannot be represented by a single
"strength"
Long-term memory examples
Remembering where your car is parked
Recalling what you ate for breakfast
Remembering your first romantic
kiss
Recalling that Sicily is part of
Italy.
Remembering that 537 votes separated
George Bush from Al Gore in 2000 in Florida.
Special features
Flashbulb memories: salient
memories of personal context of emotional events
Where were you when you heard the
new of 9/11?
flashbulbs? Special
memory or just normal, but strong memory?
Source Monitoring
Source monitoring: determining
origin of one’s memory
Did I read it or did I just imagine
in?
What processes do we use to distinguish
the source of our memories?
Memory systems
episodic memory vs. semantic memory
episodic: memory for
individual events from our lives
semantic: knowledge
of the world
Procedural memory - skill memory
- not necessarily propositional
(Autobiographical memory -- memory
of one's life -- is composed of both episodic and semantic components).
Memory processes
Encoding (learning)
- putting information into memory
Storage (representation) - maintaining
the memory during a retention interval
retention
interval - time between encoding and retrieval
Retrieval activating the
stored information
Levels of processing
Assumptions of Levels of processing
(LOP) framework (Craik & Lockhart, 1972)
1. memory trace is byproduct
of perception (intentional and incidental learning)
2. perceptual systems are hierarchically
organized from low-level sensory (shallow) to high level semantic (deep)
analysis
3. durability of memory trace
- determined by depth of processing - deep processing leads to better memory
than does shallow processing.
Data supporting LOP
semantic orienting task leads to
better memory than sensory or acoustic task (Craik & Tulving, 1975)
orienting task - experimental
control of LOP
shallow tasks might include
- counting "e's" in word, rhymes, sounds, etc
deep task - fit word into
sentence, classify into categories, etc.
better recall performance as a
product of deep processing
Results
LOP
LOP
Three variables that influence
depth of processing:
Distinctiveness
Elaboration
Organization
Principles of LTM 2
Lecture 8
Exp 4605
September 27, 2005
Retrieval
Retrieval: activating information stored in memory.
Retrieval is often very cue-dependent
Cue-dependency: need the right key to unlock the memory
Encoding specificity
Encoding Specificity: recall is better if retrieval context
is similar to encoding context
Examples
1) Baddeley and Godden (1978) looked at memory (recall) in scuba
divers - retention for info when learned underwater was best
when tested underwater; retention for info
when learned on land was best when tested for on land.
2) Eich (1998) - state-dependent learning; -- drug-dependent learning,
mood-dependent learning -- obey E.S. principle
ES results
ES
Availability/Accessibility
Encoding Specificity shows the difference between Accessibility and
Availability.
Accessibility -- that which can be retrieved now
Availability -- that which is stored in memory
Eyewitness memory
Classic application of episodic memory
Witness views crime -- only one chance to see event
Later much remember details of event for police, court, etc.
Misinformation effect
1. view event
2. condition 1 -- receive information about event some of it misleading
Condition 2 : no misleading event
3. memory test
Misinformation effect: Results
Finding: those who receive misleading information make more memory
errors than those who do not receive this information
False memories
Research shows that some memories can be actually false
False memories may occur because of suggestion or association with
real memories
False memory: association
Roediger/McDermott Procedure
-- present list of words all associated to a common word
-- free recall test
-- look for correct recall and recall of the critical intrusion
-- results: critical intrusion likely to be recall -- about same
as item from middle of serial
position; critical intrusion judged to be a confident
memory;
False memory: Suggestion
False memory induction paradigm
(Loftus & Koan, 1993, Hyman, Husband, & Billings, 1995)
ask individually repeatedly about event from childhood
"do you remember when you spilled the punch at your cousin's
wedding?"
enlist family members to help induce false memory
False memory induction
monitor the extent to which people "remember" (voluntarily report)
events that did not take place
Some people are highly suggestible and remember false events
Others do not ever succumb to false memory induction
Individual differences related to imagery and dissociation
Implications for repressed memory debate and psychotherapy.
Memory repression is a hypothetical mode of forgetting. It involves
the direct suppression of emotionally-difficult memories.
Cognitive Psychology -- repression is a hypothetical mechanism -- open
for investigation; there may be better explanations of
Some recovery of repressed memories are surely false memories
Memory strategies
Lecture 9
Exp 4605
September 29, 2005
Encoding into semantic memory
repetition rather than salience of single event (as in episodic memory)
Methods of improving encoding into semantic memory
Practice
Total time hypothesis: amount you learn is dependent on the amount
of time you devote to learning
Spacing effect
Spacing effect (distributed practice effect): learn more
if learning trials are spread over time rather than massed at once.
Equivalent to the advantage of spreading study over days rather than
“cramming.”
Baddeley and Longman -- typists learned faster when studying was distributed.
Mnemonics
Mnemonics: use of strategies to help memory
mostly based on imagery -- more useful for those who are better "imagers"
mostly effective when rote memory is required.
Mnemonic techniques
1. method of loci: people associate new to-be-learned items with
a series of well-learned physical locations. Groninger (1971)
found it effective for learning word lists.
2. keyword system for learning new vocabulary (also known
as linkword) -- a salient aspect of the to-be-learned word is used to
make a interactive visual image with
the known word (e.g., salamin eyeglasses picture a salmon wearing
eyeglasses).
Organizational strategies
1. chunking: combining smaller units into larger unit
2. hierarchy technique: arrange information in related classes
or categories
3. first-letter technique: apply meaning by using first
letters to come up with catchy slogan ? good for arbitrarily named
information (e.g., bones, arteries of body)
4. narrative technique: make up story that links unrelated items together
Prospective memory
Prospective memory: memory for things to do (in the future)
Important practical implications; taking medications, leaving appliances
on, etc.
Metacognition
Lecture 10
Exp 4605
October 4, 2005
Metacognition
Metacognition : our knowledge, awareness, and control of our
cognitive processes
Metamemory: our knowledge, awareness, and control of our
memory processes
Monitoring and Control
Monitoring: the ability to be aware of how our cognitive processes
are working.
Monitoring or metamemory accuracy ? when we know what we are
learning (or remembering) and know what we are not learning (or remembering)
Control: make strategic changes to improve one’s learning or
remembering
Examples of monitoring and control
Monitoring: a student realizes that s/he has learned all of the
definitions for a test
Control: A student decides that more study is necessary; so gets
some coffee and continues studying
Examples
Monitoring: A person realizes that s/he does not know enough about
auto mechanics to figure out why oil is leaking.
Control: The person brings the car to the mechanic instead of
doing the repairs at home.
Philosophical implications
People can monitor the accuracy and effectiveness of their own thoughts.
Descartes: I think, therefore I am.
Is this uniquely human?
What happens when metacognition is wrong.
We know what we don’t know (do you know V-P Cheney’s home phone #?)
Tip of the Tongue states
Tip of the Tongue: a feeling that you know the unrecalled word
and that you will retrieve it soon.
The same metaphor is used in many languages (Estonian, Arabic, Navajo,
almost all Indo-European languages)
Tip of the Tongue Examples
What is the last name of the first person to set foot on the moon?
What was the name of the little boy who was removed from his Uncle’s
house in Miami and returned to his father in Cuba?
What is the name of the large flightless bird from Africa?
What is the name of the religious group from Northern India whose men
where large turbans wrapped around their head?
Tip of the Tongue states
Reliably predict eventual recall and recognition
Reliably predict partial information about the target word.
Are very accurate but not perfect.
Appear to be caused by the familiarity of the topic and the accessibility
of partial information
Methodological tools
Feeling of knowing: how likely will you recognize information in the future/given at time of recall
Feeling of knowing example
What is the capital of Australia?
Answer ________
Feeling of knowing________
Feeling of knowing example
What is the capital of Australia
A)Sydney
B) Melbourne
C) Canberra
D) Perth
E) Cairns
F) Alicetown
Methodological tools
Judgment of learning: will I be able to remember this item later/given
at time of learning
Judgment of Learning Example
Learn the following pair
Captain - Carbon
Judgment of Learning Example
How likely will you remember the target word that went with “Captain”
25% 50% 75% 100%
Judgment of Learning Example
Captain ________
Results
Feeling of knowing and Judgments of learning tend to be accurate;
That is, when we judge we know, we tend to get the answer correct.
The correlations are high but not perfect.
Can we use these judgments to control our behavior?
Experiment: Can JOLs improve our learning?
Nelson and Leonesio (1988)
Given a list of French-English word pairs (e.g., pen - stylo)
Make judgment or learning
Make decision about how much time to study that pair: Allocation
of study time.
Given memory test
Findings
1. students allocated more study time to items which were judged to
be more difficult.
2.still remembered more of the easy ones (labor-in-vain). Did not compensate adequately for difficulty
Son and Metcalfe (2000)
Students given hard and easy passages to study (e.g., why we need to
take vitamins, post-modern interpretations of neo-classical fiction)
Some students were given only a limited amount of time to study; some
a great deal of time to study
Students allocated own study time.
Son and Metcalfe (2000)
Results
When time limited: students choose easy items; when time unlimited:
students choose difficult items.
Adaptive use of study time.
Original study with Ivy League students; replicated with Inner City
Public schools in NY (6th grade)
Summary
Monitoring
Control
Accuracy
Feelings of knowing
Judgments of learning
Allocation of study time
Imagery and Cognition
Lecture 11
Lecture 12
October 6, 11, 2005
EXP 4605
Imagery
Imagery --mental representation of things that are not physically present
Representation -- how items are stored in memory
Theories of representation
Analog representation -- mirrors physical world; stored in sensory
format (depictive format)
Propositional representation - abstract code; stored in language-like
code (descriptive format)
Shepard and Metzler (1971) -- Mental Rotation Experiment
- found that it took time to mentally rotate objects
- s's presented with two objects
- are the objects the same?
- need to rotate into same plane to determine
- time to answer question was a function of amount that would have
been
required to physically rotate them
Example of stimuli
Results
Shepard &Metzler
conclusion: mental rotation done in an analog fashion
Propositional View
Arguments for propositional view
-reporting of ambiguous figures (Chambers and Reisberg experiment)
Creating mental images
Finke, Pinker & Farah (1989) challenged notion that ambiguous figures cannot be perceived in mental images. They gave people problems like:
Finke et al. example
Imagine the letter B. Rotate it 90 degrees to the left. Put a triangle
directly below it having the same width and pointing down. Remove the
horizontal line.
What do you have?
The analog view
If Visual imagery is analog - it should share some features with vision
itself
- to the extent that a visual imagery and a visual task interfere
- supports analog system
The Perky effect
The Perky effect (Perky, 1911) -- when imaging an object, participants
failed to perceive an actual picture of the object.
Segal & Fusella (1970)
Updated by Segal and Fusella (1970)
Imagined either a visual object or an auditory object; then had to
detect a faint arrow or a faint sound. Visual imagery interfered with
visual detection; auditory
imagery interfered with auditory detection.
Brooks (1968) interference task
- inspect mental image of letter - make decision about each corner
"F"
- indicate response via voice, tapping or pointing
- people take longer when making pointing response
(pointing did not slow down other tasks)
- suggests the analog view and that imagery and visual perception
overlap
Neuroimaging research
Farah (2000) - neuropyschological evidence; imagery operates in the
some of the same brain regions as does vision itself.
PET scans show activity in V1 and other visual areas of brain during
imagery
People who are better imagers show more activity in V1
Cognitive Maps
Cognitive maps = internal representation of the way our spatial environment
is arranged
Heuristics = rules of thumb, used to govern our cognitive maps
Introduction to Language
Lecture 13
October 18, 2005
Cognitive Processes
The study of language: Linguistics
ways of studying language: traditional linguistics; anthropological
linguistics; historical linguistics (philology); psycholinguistics.
Language Groups: languages related by common ancestry
Psycholinguistics
Psycholinguistics-- focuses on automatic processes of speech perception,
language comprehension, word and reading comprehension.
Glossary of linguistic terminology
phonology: study of the sounds of spoken word (may vary across languages,
the basis of accents or dialects)
Phoneme -- the individual sound (b, p, sh, etc)
International phonetic alphabet gives a unique symbol for each sound.
prosody: the rhythms of language (also a basis for regional dialects)
Glossary of linguistic terminology
morphology: the study of words and word formation
cat = 1 morpheme
cats = cat/s/ = 2 morphemes
decatted = de/cat/ed = 3 morphemes (to remove cats)
lexical: related to words
lexicon = the mental dictionary
Glossary of linguistic terminology
orthography: related to appearance of written word
Alphabets: Latin-based, Cyrillic, Arabic, Hebrew, Hindi, Chinese:
each use a different set of symbols.
We will discuss the relation of orthography (written symbols) to phonology
(spoken sound)
Not all orthographies have the same relation to phonology.
semantics: related to meaning
Important cognitive topic: how do we derive semantics from utterances
we hear.
E.g. He went to the bank.
Glossary of linguistic terminology
syntax: word order and other grammatical features. this refers to
descriptive grammar not presciptive grammar.
All languages, dialects, have specific implicitly learned syntax that
governs how we speak.
I am working/I be working.
Descriptive vs. presciptive grammar
Prescriptive: specific rules that allow one to follow formal
rules of writing or shape speech into “upper class” speech; e.g., Don’t
use “ain’t”
Descriptive: implicit rules adopted by members of a community; e.g.,
one Jedi; many Jedi; one Micky Mouse, many Micky Mouses (analogous to Andy
Mann and the whole family of Manns).
Glossary of linguistic terminology
pragmatics: language use (e.g., irony, metaphors, etc)
Pass me the salt, NOW, you fool!
Pass me the salt, dude!
Please pass me the salt.
Can you pass the salt when you get a chance, if you please?
Language and Thought: the Whorfian hypothesis
language and thought: most contemporary theorists think that language
constrains thought, but that individual languages (English, Spanish, etc)
do not constrain thought.
Cassell & Casagrande (1958)
Navajo verb forms
Navajo uses different verb tenses to refer to rigid (rocks) and non-rigid
objects (ropes)
would Navajo children use different classification as a consequence?
Three groups: Navajo-speaking, English-speaking Navajo; English-
speaking Bostonians -- children asked to classify toys
No difference between Navajo-speakers and Bostonians; E-S Navajos
did use color more than rigidity
The innateness hypothesis
Language: Innateness hypothesis: that the ability to learn human language
is coded in our genetics.
No particular language is genetically determined; just the human ability
to acquire language is at least partially genetically determined
Lecture 14
Exp 4605
October 20, 2005
Language comprehension
how are we able to understand spoken language so rapidly?
Speech perception
1. information about sounds is transmitted in parallel
2. context allows listeners to fill in missing sounds
3. visual cues aid spoken language
4. listeners impose boundaries (top-down processing)
The Phonemic restoration effect
subjects filled in a missing sound with the sound that most made sense
given the context. However, subjects actually report hearing the
sound (Warren & Warren, 1970)
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 *eal was on the table.
Visual Cues as an aid to speech perception:
The McGurk effect
- the sound presented is one sound (ba), but the sound you see
the lips making
is another (da) --the person reports hearing the sound "da."
Word boundaries
the boundaries between words are not actually present in the stimuli
-- they
are imposed top-down by listeners who expect certain word boundaries.
Cognitive factors
Negatives: people understand positive statements faster than
negative ones
*
+
The star is above the plus
The star is below the plus
We say yes to the first sentence faster than no to the second
Cognitive factors
Passive Voice
The man bit the dog
The man was bit by the dog
First is verified faster
Cognitive Factors
Nested Structures: lead to confusion
The man, who was mean to animals, bit the dog, which was tied to tree,
when he went crazy.
The crazy man was mean to animals. He bit a dog tied to a tree.
Ambiguity
Ambiguity also slows down comprehension
Example:
The chicken was too hot to eat.
Visiting Relatives can be boring.
And newspaper headlines:
See page 306 of text. My favorite:
Local high school dropouts cut in half.
Transformational grammar
Cognitive processes convert surface structures to deep structures to
derive underlying semantics
The chicken is too hot to eat.
Depends on whether you convert the chicken as an implied object to
an unseen eater. OR
The chicken is the actual subject of the sentence.
Lecture 15
Exp 4605
October 25, 2005
Reading
From Matlin
1. Reading is visual. Speech perception is auditory
2. Reading is spatial. Speech perception is temporal.
3. Reading is self-paced
4. Speech perception augmented by non-verbal cues.
Others
1. Reading is less recently evolved and not all people
are literate
2. children learn reading with relative difficulty (and through
specific instruction), but speech with relative ease
English writing system
Orthography to phonology
Complex indirect relations
Examples: pint, lint
tough, through, cough, bough
Mood, good, blood
ghoti
More Direct orthography-phonology correlations exist in the writing
systems of Spanish, Italian, Serbian. Less direct systems exist in
Chinese, Japanese, & Hebrew.
Theories of Word Recognition
Direct Access --reader can recognize written words without access to
sound
representations
Indirect Access (Phonologically mediated) -- readers must access sound
representations in order to recognize words
Dual -Route -- readers may use either or/both.
Theories of Word Recognition
Direct Access
Theories of Word Recognition
Indirect Access
Homophones
She has blond hare
The none says her prayers
In Yellowstone Park, they saw a huge grizzly bare.
The young child hurt her knee and let loose an enormous whale.
Van Orden (1987) "A rows is a rose"
Category judgment task
stimuli: "is a rose/rows a flower?
is a daffodil/crowbar a flower
Rose and Rows are pronounced the same, but mean something different.
Thus indirect access predicts interference for rose and
rows.
Van Orden (1987)
look at speeded judgments -- looked at error rate (also reaction time)
compare category member and a homophone of that category
member (rose/rows), and control non-homophone(daffodil/crowbar)
find that more errors were made to homophones than to
non-homophones
Which view does this experiment support?
Lecture 16
Cognitive Processes
Exp 4605
October 27, 2005
Neuropsychology of Language
Key brain regions (left hemisphere):
Broca's area -- production of language
Wernicke's area -- comprehension of language
Aphasia
Aphasia -- the the catch-all term meaning any disorder in speech, comprehension,
reading and writing caused by brain damage.
Broca's aphasia -- non-fluent aphasia -- problem with speaking
Wernicke's aphasia -- fluent aphasia -- problem with comprehension
Neuroimaging
PET localization studies
method of subtraction -- brain is always active -- must establish baseline
to localize
Allows you to pinpoint selective areas of activation during particular
cognitive task
1. initial control task --- simply look at Fixation
2. see or hear word -- do nothing -- "sensory" task.
(2 - 1 yields additional activity in auditory
or visual cortex) *but different pattern of activity for words
and non-words
3. see or hear word -- then repeat it "output task"
(3 - 2 yields activity in motor & sensory
areas and insular cortex)
4. see or hear word -- generate assocation "association task"
(4 - 3 yields activity in Broca's
area, posterior temporal cortex, anterior cingulate cortex, and the
cerebellum).
Hemispheric specialization
Broca’s and Wernicke’s are localized to the left hemisphere;
Except 1% righthanders; 10% left-handers
(note these numbers vary from those given in Matlin)
New research shows that most women and left-handed men have language
representation in both hemispheres
Split-brain patients
Corpus collosum connects left and right hemispheres.
As a treatment for incurable epilepsy; corpus collosum was surgically
disconnected in patients during 1960s - 1980s.
Miraculous recovery from epilepsy; very few side effect (wandering
left hand)
Reason: information goes to both hemispheres from sensory organs
Split-brain patients
When information is presented to only one hemisphere; other hemisphere
does not get information.
Showed that in most patients, the right hemisphere is silent, even
when it reads words and uses them appropriately, but cannot drive speech.
Lecture 17
Cognitive Processes
Exp 4605
November 1, 2005
Language Development
Developmental issues
How infants acquire the rudiments of language
How children learn language skills
Innateness view
Humans (particularly infants and young children) have innately determined
mechanisms that allow them to acquire language rapidly.
Evidence for innateness in children
No need for explicit instruction; all children learn to speak
Critical period for learning speech
Children deprived of true language will create own (e.g., Nicaraguan
sign language)
Speech perception in infants
Speech perception starts to develop early
By six months of age, infants respond more quickly to phonemes
from their native language; by ten months, may stop responding to differences
not recognized in their native language.
Non-nutritive sucking, gaze direction, and habituation are used
to measure these variables.
Language production in infants
Beginning to associate some words with objects
Cooing - making vowel sounds ( 2 months)
Babbling - begin practicing consonants by eight months
Motherese - child directed speech; easy words, easy grammar emphasize
borders between words
Around first birthday - first word is uttered
Children’s language learning
Fast mapping - using context to make reasonable guess about word’s
meaning
Taxonomic assumption - label can apply to other objects of the
same category
Overextension - use of word for objects not considered appropriate
Underextension - using a word in a narrow sense than adults do
Telegraphic speech: language that includes only content words;
omits words which serve only syntactical function.
Examples
Telegraphic speech: “Elmo now.” “more cookie.”
Overextension - “apple” applies to all fruit
Fido applies to all dogs.
Underextension: “fruit” only refers to oranges; “dog” only applies
to Fido.
Advanced language learning
By age 4-5 most children are learning the social rules of speech
Politeness, turn-taking, code-switching, conversational formats,
etc.
Lectures 18, 19
Cognitive Processes
November 8 and 15 , 2005
Problem-solving & Creativity
Ordinary and extrordinary problems to solve
Getting home safely after drinking
Planning a trip to California
Surfing big waves in Hawaii
Getting into graduate school
Figuring out the plans of terrorists
The Die Hard 3 problem
3 gallon jug
5 gallon jug
Unlimited supply of water
Need exactly 4 gallons to defuse the bomb
Problem-Solving Approaches
Algorithms -- method that will always produce the correct answer
advantage: get the correct answer
disadvantage: may be costly in terms of time and effort
Problem-Solving Approaches
Heuristics -- "rule of thumb" -- using strategies that will usually
generate the answer, but are not guaranteed
advantage: usually get the answer with minimal time and effort
disadvantage: don't always get answer -- may be systematic biases.
Unscramble these words
Kmli
Graus
Tewes
Recma
Fofce
irknd
Problem-Solving Heuristics
1. Hill-climbing: move directly towards goal
2. Means-end analysis: break into smaller sub-problems - figure
out how to do each sub-problem
problem - most people think that solution of sub-problem must
"look" closer to solution of actual problem. In fact, the solution may
involve some backtracking (e.g., hobbits and orcs)
3. Analogy heuristic --solve a problem by applying a solution to an
earlier problem.
Two definitions
Mental set: keep using the same solution
Functional fixedness: the roles we assign objects tend to remain fixed
Metacognition and problem-solving
insight problems -- sudden solutions (sudden reorganization of representation
of problem)
non-insight problems -- step by step solutions
Insight and non-insight problems
Non-insight
3x + 7 = 28
X = ?
Insight
Arrange 4 trees so that each tree is the same distance from all other
trees
Metacognition and problem-solving
Metcalfe & Weibe (1987) hypothesis: people's metacognition would
be stronger and better for non-insight than insight problem-solving
Metcalfe and Wiebe (1987)
metacognition -- knowledge or feelings concerning one's cognitive processes.
M & W asked problem-solvers to make warmth judgments on their proximity
to solution as they worked on the problems
Metcalfe and Wiebe (1987)
Found that warmth judgments predicted solution to non-insight, but
mispredicted solution to insight problems.
- implies that people may not have conscious access to solving insight
problems, but do have access to solving non-insight problems.
Creativity
Difficulties in studying creativity
what is creative cognition? who are creative people?
Approaches to creativity
-- psychometric approach -- intelligence and multiple intelligences
Creative cognition stresses
Originality and relevance
-- spontaneous restructuring
-- expertise vs. creativity
Incubation effect
- if a problem is difficult, eventual solution may benefit from a delay
between solving sessions
Smith and Blankenship (1991)
Gave "tricky" problems to people to solve, such as RAT (rough, cold,
beer: draft)
Solve riddles
lines reading lines
aholene
Remote Associates
Smith and Blankenship (1991)
If people solved, go onto to next problem
If unsolved, give clue, either helpful, misleading, or no clue
Then either try to solve 5 minutes later or 48 hours later
Smith and Blankenship (1991)
Misleading clues hurt performance in immediate solution condition
Misleading clues did not hurt performance in delayed
Smith and Blankenship argue for mental sets.
Implications for creativity
Creativity may be the ability to overcome mental sets.
possible explanations: unconscious processes at work solving problem
people need to assume new "mental sets"
(i.e. similar to functional fixedness)
Lecture 20
Cognitive Processes
Exp 4605
November 17, 2005
Reasoning
Reasoning -- transforming the given information in order to reach
conclusions
Logical reasoning -- using established rules to draw clear-cut conclusions
Decision-making -- assessing and choosing among several alternatives
-- involves uncertainty
Reasoning
If A then B rule in the world
Facts in the world
A
Not A
B
Not B
Conditional Reasoning -- the relations between condition
1. Affirming the antecedent -- leads to valid conclusions
2. Denying the consequent -- leads to valid conclusions
3. Affirming the consequent -- leads to invalid conclusions
4. Denying the antecedent -- leads to invalid conclusions
Example
If the professor returns the exam, then I will be happy.
The professor did not return the exam.
Therefore, I am not happy.
Valid or invalid??
Errors in Conditional Reasoning
1. not examining all possible models of statements (people may be
happy for lots of reasons)
2. illicit conversions
(if you are a basketball player, you are likely to be tall * if you
are tall, you are likely to be a basketball player).
Errors in Conditional reasoning
3. confirming hypothesis rather than disproving it
The Wason task -- illustrates a common error in human reasoning
subjects are given four cards, and a rule:
if a card has a vowel one one side, then it has an even number on
the other side.
The cards shown have only one side visible.
E K 4 7
which cards do you turn over to test the rule?
Wason task
criticism: real life situations
if a person has a beer, then they should be over 21
BEER COLA 22 17
In this problem (logically equivalent) there are never errors made
Errors in conditional reasoning
4. belief biases
If a person is tall, then he plays basketball.
Keith Boykins plays basketball.
Therefore, he is tall.
(seems to be common sense, but violates logical reasoning; in this
case, affirming the consequent).
Lecture 21
Cognitive Processes
Exp 4605
November 22, 2005
Heuristics
The HIV question
2 million people in Miami-Dade county
40,000 people HIV+
That’s a 2% rate.
Standard test shows positive 99% of time when person is positive
Shows positive when negative 1% of time
Person has a positive result-- how likely is it that they are HIV positive
HIV question
Decision-making and heuristics
Decision -Making and Heuristic Judgments
Major problem in human cognition -- low intuitive understanding of
the laws of probability
Intuitive understanding and misunderstanding of probability
important seminal work -- Kahneman and Tversky
important criticism -- Gigerenzer. Heuristics are suited for the
kinds of situations they evolved for.
Representative Heuristic
judging a sample on the basis of its similarity (and random-looking
appearance) to the population from which it was taken
leads to certain problems with judging single samples?
Example
e.g. -- which is more likely outcome of flipping a coin eight times?
HHHHHHHH
THHTHTTH
Dimensions of representativeness
1. random-looking outcomes (see above)
2. Sample size --- the law of large numbers vs. the law of "small
numbers"
law of large numbers -- percentage differences will be closer, absolute
numbers differences will be greater
small numbers -- percentage differences may be of greater
small sample fallacy: expect small samples to confirm to expected
outcomes.
Representative heuristic
3. base rate & Baye's theorem -- need to take into account base
rates.
The HIV example.
4. Conjunction fallacy -- joint probabilities are judged to be more
likely than single probabilities
which is a more likely to occur?
a. FIU's basketball team will win the NCAA championiship
b. it will rain this week and FIU basketball team will win NCAA
champs
Representative heuristic
5. Gambler's fallacy -- "this machine is due"
Expected values vs. expectations of success
Lecture 22 and 23
November 29 and December 1, 2005
Decision-making 2
Availability Heuristic -- people judge frequency or probability in terms of how easy it is to think of examples of something. The ease of generating examples influences our decisions about the entire category.
Dimensions influencing availability
1. recency -- Recent events are judged to be more common
People more afraid of flying after recent plane crash
Dimensions influencing availability
familiarity -- familiar events are judged to be more common
Empirical examples: 1) estimating the populations of countries
familiar countries are overestimated
unfamiliar countries are underestimated
2) effects of media coverage -- inflate estimations of likelihood
of crimes, terrorism, etc.
Estimate the following populations
Nepal
Nicaragua
Israel
Nigeria
Bangaladesh
Indonesia
Cuba
Iraq
Bulgaria
Actual populations
Nepal 26 million
Nicaragua 5 million
Israel 6 million
Nigeria 100 million
Bangaladesh 120 million
Indonesia 100 million
Cuba 10 million
Iraq 24 million
Bulgaria 8 million
Dimensions influencing availability
. illusory correlations
a) when people see relations when none actually exist because particular
dimension is salient.
b) critical in stereotypes -- the stereotypical example is more familiar,
hence more available, hence overestimated
Halo effect -- positive traits go together
Stereotypes
All X are Y
Whenever we see an X who is Y it activates and confirms the stereotype
Whenever we see and X who is not Y it fails to activate and disprove
the stereotype
Anchoring and Adjustment
Anchoring and Adjustment heuristic -- use anchor and add variability.
8 X 7 X 6 X 5 X 4 X3 X 2 X 1
vs. 1 X 2 X 3 X 4 X 5 X 6 X 7 X 8
“the foot in the door.”
Framing effects
Framing Effects -- outcome of a decision can be influenced by
1) the background context of the choice
2) the way in which a question is worded
Framing effects
Scenario A
Outbreak of disease will kill 600 people
A: 200 peoples lives will be saved
B: 1/3 600 saved; 2/3 nobody saved
Scenario B
C: 400 people will die
D: 1/3 nobody will die; 2/3 all 600 will die
Framing effects
A: 72%
B: 28%
C: 22%
D: 78%
Tversky & Kahneman (1981) -- risk-averse to save gains; risk-seeking
to avoid losses. .
Overconfidence in decisions
Overconfidence in Decisions -- people overestimate the likelihood that
any individual decision (or answer) is correct.
What is the altitude (feet above sea level) of Mt. Everest (the highest
mountain on Earth)?
Reasons for overconfidence: my-side bias: easier to think about why
you would be right than wrong.
contrasts with global underconfidence
Hindsight Bias
Hindsight Bias -- refers to overconfidence in reporting our likelihood
of accurately predicting an outcome (after the fact). "I knew it all along."
Conclusions
Optimists vs. pessimists: Perspectives on Decision-making
Optimists: good for what decision-making is intended for (evolved
for)
Pessimists: easily misunderstood and mislead
Evolution and Cognition
Evolutionary Theory
Logic of natural selection:
1. Variation within species: individuals vary within a population
(whether people or bacteria) along many traits.
2. Heritability: some traits are related to an individual’s ability to contribute to the next generation. Parents’ genetic makeup is related to children’s genetic makeup. This heritability is coded in the DNA of genes. All life on earth has DNA (or RNA) in genes. From viruses to elephants and people.
3. Adaptation: individuals must adjust or adapt their behavior to the environmental conditions. Individuals that do so survive and contribute to the next generation. If these adaptations are genetic, they will be passed to the next generation.
Natural selection: traits that are heritable and advantageous
in terms of fitness will increase across generations.
Natural selection is a fact. It has been observed in the field
and in experiments repeatedly.
Important Note: (contribution of the late great Stephen
Gould among others): Natural selection is NOT the only factor that
drives evolutionary changes. E.g. correlated traits (spandrels)
and sexual selection.
Important data from last 25 years: behavior appears to have inheritable
component in humans.
Thus, it is likely that cognition which underlies behavior also has
a heritable component.
Thus, selective pressures may cause (or have caused) the natural selection of some cognitive processes.
The single most studied example is the human language ability.
Arguments for an evolved capacity for language
1. present in all human groups.
2. same areas in brain in all people.
3. Genetic disorders of language.
4. rapid development of language in children.
5. advanced cognition but not language in non-human apes
Experimental psychologists, however, usually avoid evolutionary hypotheses
because you cannot do the experiments to test evolution on humans (for
obvious ethical reasons). However, unfortunately history has
done these experiments.
People brought into communities where native languages are not
shared develop creoles within a generation.
Well-studied case Nicaraguan sign language.