Intelligence, cognitive control
Measuring intelligence
Cognitive control / executive function
Measuring intelligence
Intelligence - the ability to reason, plan, solve problems, think abstractly comprehend complex ideas, learn quickly, and learn from experience
Not the same as book learning or test taking smarts
An ability crucial for catching on or making sense of things or figuring out what to do
Solving problems - knowing how to solve problems - they know stuff bc they have pipe knowledge
Some version that acknowledges solving problems
Crystallized intelligence - acquired knowledge, including verbal knowledge and expertise in a subject
Fluid intelligence - the ability to deal with new and unusual problems
There are other abilities that make ppl intelligent
Big difference in having knowledge and do the thinking (fluid intelligence part)
Intellectual development -
Late adulthood - have more crystallized intelligence, but less fluid intelligence (more knowledge collected over time like stronger vocabulary)
Early adulthood - less crystallized intelligence, but more fluid intelligence (faster at solving problems)
Large study - had ppl ages 20 to 80
U will know more words as you grow older
Mental speed, fluid intelligence, and working memory are all at a decline as you grow older (on average)
All these tasks - go together - this is not obvious
Origin of thinking about things - measuring intelligence - the beginnings
Universal elementary education - 1904 in france
Farmer kids and rich kids did not know the same information
How do we identify kids who need extra help?
Designed first intelligence tests (copy a drawing, repeat digits, recognize coins, explain why a statement did not make sense)
Used it to identify kids who needed help
Tests are there to see which kids are falling behind
We cannot ask ppl to draw coins
The kind of thing that is about fluid intelligence (solving novel problems)
To isolate fluid intelligence from crystallized - they have diff tasks
Ex - which pieces go together to finish this puzzle?
Usually normalized with a mean of 100 and a standard deviation of 15
Forced standard deviation of 15 on them
Took actual raw scores and applied normal distribution on them
If average person only gets 25 out of 35 right - they stretch out the distribution
What makes something a useful measure of individual differences?
Reliable - same score on two occasions (if i give u the same test on two occasions you will generally get the same score)
Valid - measures what it is supposed to measure (the ability to solve problems in abstract situations)
Finding ppl who are exceptional in a given area
Understanding measures of individuals
Personality measures - are trying to be stable individual measures - reliable and valid
IQ scores - reliability
IQ is a very reliable measure about as good as all personality and other individual difference measures
Test retest reliability very high (around .70 or .80)
SAT is very reliable
Does this reliably measure ppl around me?
Measurement error -
Fatigue, sickness, stress differences on the day of
Test taking environment (noise)
Item differences (ex - different questions)
Environmental changes during crucial periods of development (like teen years)
Practice effects
These tests are pretty reliable despite the errors
How do u measure validity?
Tells us something about how good you are at abstract reasoning
Valid for what? - IQ is not the worth of an individual
Valid for what? - validity of IQ tests as a measure of fluid intelligence amounts to whether we can use it to predict how well u solve abstract novel problems in other situations
Researchers are interested in g as a measure of other aspects of cognitive control
Measures of what this predicts - what makes it valid - is trying to be valid
Single largest sum in psychology
Validity as a measure of fluid intelligence
IQ predicts job performance
Intellectually complex professions - IQ predicts 36 to 64% of variance in success
Assembly line job performance - products only 4% of variance in success
Academic success - IQ predicts about 25% to 40% of variance in school (there are so many factors which is why the percentage variance is so high)
Just bc u are intelligent does not naturally mean u will get good grades
U have to show up to class and actually study the material
grades - we want to teach ppl how to learn and critically think
Later life outcome -
Even within the same middle class family - IQ still predicts 10% of variance in income in later life
Take all ppl in your sample and plot all incomes - u can predict where in that sample they fall
The flynn effect
Decade and IQ relative to 1920
If we gave u the same test in 1920 - more than 50% of population would test above the 50th percentile
Tests have to keep getting harder and harder
Has plateaued in some regions
What is this thing measuring? - in 1960s everyone in america on average - has changed as the world has evolved
Small gains - vocab, general knowledge, arithmetic
Large gains - abstract reasoning (ex - raven’s matrices)
We live in a more abstract world?
When we watch TV now there are 7 plots
Back then there was one very simple plot in a TV show
The world is now more complicated and complex
We live in more complicated worlds - working demands are harder
We are faced w more challenges and live further detached from things
We learn about animals and biological trees, etc
We have more abstract knowledge of the world
General gains in health
A lot of ppl had poor scores bc they were in poor health
Giving them all battery of health and mental tests
There is no selection bias - we are plotting general ability
Height is plotted here - they look very similar
The world got taller - general health - we have more access to healthcare and food and resources than back in the day
We have all gotten healthier and this makes us able to live in a more complex world
IQ scores - measures of more than just abstract reasoning
Higher IQ is associated w longer life expectancy, lower disease risk and better health behaviors
IQ predicts a lower risk of accidental death, possibly due to better decision making
Overall - IQ test scores correlate w GPA in HS and college, job success, salary, stability in marriage, staying out of jail, longevity
The higher someone’s IQ the less likely they are to get divorced
If IQ was causally related to access to resources (ppl came from more money)
Cognitive control - everything correlates
Inhibition, suppression of things
Can we think of fluid intelligence as a single thing? G and Gf as constructs
G - general intelligence
Gf - general fluid intelligence
Ppl who are good at one of these things tend to also be good at other of these things
How good r u at auditory perception - if u test good in auditory → will also test better in vocab tests
Ppl score better in school do better in sports - on average some ppl perform better than others
Anything cognitive - they are all correlated
There is some kind of g factor - this makes u more or less good at certain things
All of these tasks are related to one another - g factor - some ppl have more or less and this makes us better at certain things
Gf - general fluid intelligence
General vs specialized intelligence
Most modern IQ tests (ex - wechsler, stanford binet aim to measure g by including many subcomponents)
There are some general facts and aim to measure gf
The thing that is in common - gf and g - some are better at spatial or verbal intelligence
If u give ppl an intelligence test - some ppl are generally better at everything than others
Why are u better at all of these tests? - ppl have many hypotheses - processing speed - many ppl may have a faster processing speed (if u are faster at processing stuff u are just better overall)
Light on table and as soon as light flashes u press button (1,000 trials)
Ppl w faster reaction times score higher on IQ tests - this makes the person better at solving abstract problems
Infant habituation - the same stimulus is shown, infants who get bored faster means they have higher IQ - processing speed - some brains just process things faster than other brains and score higher on all these tests
Attention and attentional filtering - red squares and blue squares - some ppl remember the blue ones even tho they are told not to
Working memory capacity - ppl that can hold in mind more things
These correlate highly w intelligence measures
Internal buffers
Cognitive control - ex - ease of engaging system 2)
Some tasks demand system 2
Some ppl tend to be more reflexive
Inhibition - suppressing ur initial response to things
Not eating the marshmallow correlates to intelligence
Why is it that the same ppl are better at all the things? -
There does not have to be the same causal answers
Ppl like to think about it statistically
Working memory correlations
Gf - all correlate w working memory capacity
All of these tasks correlate w each other
Even tho these tasks are very different - they all correlate w each other
No evidence of intelligence improvement after working memory training: a randomized placebo controlled study
Cognitive control / executive functioning
Cognitive control - using system 2 at the right time
Central executive - what u want to hold in mind
We use tasks that gain reflexive training
A bat and a ball cost $1.10 in toral
The bat costs $1 more than the ball. How much does ball cost? - 5 cents
If it takes 5 machines 5 minutes to make 5 widgets, how long would it take 100 machines to make 100 widgets?
In a lake, there is a patch of lily padsl. Every day, the patch doubles in size. If it takes 48 days for the patch to cover the entire lake, how long would it take for the patch to cover half of the lake?
Everyone has an instinctual answer to everything
When u see a pic of a grandma - does this person have kids? - we dont know if this old lady has kids - we just intuitively say yes
Ability to stop intuitive responses
The word orange but it is written in purple ink
Respond only to the direction of the middle arrow, ignore the flanker arrows (hard to suppress intuitive thoughts)
Broad hearing of executive function and cognitive control
Inhibit - stopping and impulsive act
Being able to think about the consequences before acting
Shift - transitioning from one task to another
Preferred to non preferred activity
Letting go of a specific interest
All prefrontal cortex functions
Prefrontal cortex - all things we just talked abt relate to the prefrontal cortex
Humans are more intelligent than chimpanzees
Intelligence is a small part of life
Phineas gage - he built a railroad and the rod went straight through his skull - never loss consciousness & had huge long term issues -
Before the accident he was responsible, caring, and well behaved
But after the accident - he became erratic, profane, and impatient
Something about prefrontal cortex is related to intelligence
Tools of the mind
Having better health - makes u better equipped to learn
Preschool program improves cognitive control
Low income families - most malleable families
Statue game - everyone freezes and first person who moves loses
Scaffolding - inhibiting yourself - designed to make u not want to make the thing u want to do
Flanker task - direction of inside arrow
Reverse flanker task - direction of outside arrow
Kids were Playing statue game in class was better at flanker arrow task
Fluid intelligence - measured w IQ, has both reliability and validity
Hierarchical - both g and gf and specific abilities
Ability to do abstract reasoning problems (fluid intelligence) is related to speed of processing
Heritability, nature vs nurture
Underlying g and gf factors
Processing speed
Attention and attentional
Roots of intelligence
Behaviioral genetics - studying twins and adoption
Compare twins and see how intelligence effects intelligence
Monozygotic twins - identical twins - identical genes
Fraternal twins - half of the same genes
Can look at twins raised apart vs twins raised together
Sibs raised apart vs together
Heritability differences -
0 means no variance due to genetics
1 means all variance due to genetics
.30 means 30% is due to genetic differences
Same strains of corn in two bags
Plant all in same field in iowa - difference in environment is small - high heritability score
Half corn in desert in nevada and half in iowa -
Iowa corn looks great - large environmental differences make genetic differences a small part of the total variance
Heritable does not mean innate
Innate traits can have zero heritabiluty (ex -number of legs is close to 0 heritability - very little variance and most of it environmental)
Heritable traits can be environmentally determined
If u measure IQ
Identical twins have .9 correlation
Fraternal twins have .8 correlation
Sibs raised together a bit lower than twins
Unrelated individuals in diff households - no correlation
There is a very strong correlation w intelligence and genetics
Genes - 50%
Shared environment - 0-10%
Unique environment - 50%
True for intelligent, brain structure, schizophrenia, personality disorders, memory, success of family relationships, pain tolerance
Is intelligence fixed?
If everyone was raised in same house
In poverty the degree of IQ resemblance is the same for identical twins and fraternal - genes matter much less for shaping a person’s intelligence in this environment
Middle class and upper class ppl in these studies all have environments that are similarly
Environmental effects are easy to see in the negative direction - ex lead exposure decreases IQ
On average - most of variation in upper middle class america - most of it is genetic
Summary of intelligence
Approximately 50% heritable (not the same as 50% innate but definite environmental influences)