Human Mental Abilities

Question 1

What are mental abilities?

Answer 1

Mental abilities: the capacity to perform the higher mental processes of reasoning, remembering, understanding and problem solving - sometimes subsumed under the term ‘intelligence’

We study the way they vary across individuals NOT what is common.

Question 2

What is a construct?

Answer 2

A construct is a theoretical entity that often cannot be measured directly, but can be assessed using a number of indicators or manifest variables.

Intelligence is considered a construct but is something that we can infer from observing behaviour.

Question 3

How might we measure a construct such as intelligence?

Answer 3

Step 1: define it i.e. what is intelligence?
‘Intelligence is a word with so many meanings that finally it has none’ - Charles Spearman (1927)
‘Intelligence is what intelligence tests test’ - Boring (1951) - problems as it is a very circular model

Step 2: operationalise it i.e. what does intelligent behaviour look like?

• Maybe even easier to ask what does unintelligent behaviour look like
• Need to devise a test to measure observable behaviour and operationalise the construct

Example: reaction time reflects ‘processing speed’ and number of items remembered reflects ‘working memory capacity.’

Theory and measurement are inextricably linked
You need a good theory in order to develop a good measure
But you also need a good test to accurately measure what you want to measure

Question 4

What are latent and manifest variables?

Answer 4

A manifest variable is one that CAN be directly measured or observed. It is the opposite of a latent variable, which is a factor that cannot be directly observed, and needs a manifest variable assigned to it as an indicator to test whether or not it is present.

Question 5

What are implicit theories of intelligence?

Answer 5

Informal definitions of intelligence, beliefs we all hold.

Blackwell, Trześniewski and Dweck (2007) looked at the learning trajectories of 7th graders.

• The belief that intelligence is malleable (incremental theory) predicted an upward trajectory in grades over the two years of high school, while a belief that intelligence is fixed (entity theory) predicted a flat trajectory
• Holding an incremental theory led to more effort and a more positive response to failure

Stenburg (1981) asked experts and general population what behaviours show good intelligence:

1. Verbal intelligence: good vocabulary, converses easily on lots of subjects
2. Problem solving: makes good decisions, poses problems in optimal way, plans ahead
3. Practical intelligence: sizes up situation well, determines how to achieve goals, displays an interest in the world at large

Question 6

What are explicit theories of intelligence?

Answer 6

Explicit theories use data collected from people performing tasks that require cognition.

Define the scope of the psychological construct they deal with

• Either whole domain (intelligence) or
• Specific subsets (e.g. verbal performance)

Theories are supported by (mostly indirect) evidence

• Internal consistency of the measure i.e. within-measure
• Correlate with other behaviour measures

Theory can be challenged because
- Doesn’t fit task data - theory is wrong and needs changing
- Measure is not good
- Tasks are too narrow, or lack ecological validity
- Tasks are too broad
No one theory can account for it all.

Question 7

What is Binet’s scale?

Answer 7

The youngest age at which a child of normal intelligence should be able to complete the task.
Determined by the age at which a majority of ‘normal’ children in the standardisation sample passed the task

Mental age - the age assigned to the most difficult task that you could complete
e.g. if you were 5 years old and able to solve tasks that were appropriate for 7 but not above, you had a mental age of 7

Question 8

What is Binet’s goal?

Answer 8

Identification and education - the scale was devised only to identify students in need of remedial education (i.e. to help and improve)
He believed intelligence can be augmented by good education; it is not a fixed and inborn quantity

Question 9

What are Binet’s stipulations?

Answer 9

1. The scores are a practical device
   - They do not buttress any theory of intellect (theory came later)
   - They do not define anything innate
   - We may not designate what they measure as ‘intelligence’ or any other latent variable
2. The scale is rough
   - It is an empirical guide to identifying mildly-retarded and learning disabled children
   - It is not a device for ranking normal children
3. Low scores shall not be used to mark children as innately incapable
   - Emphasis should be placed upon improvement through special training

Question 10

Contributions of H.H Goddard to the Binet

Answer 10

Brought Binet’s Test to the USA

• He used the test to prevent immigration and propagation of ‘morons’
• Testing of new immigrants at Ellis Island - by women trained in the detection of feeblemindedness (degenerate beings, responsible for many, if not most social problems)

Although Binet refused to define his scores as intelligence, Goddard regarded the scores as measures of a single, innate entity.

Classification (not derogatory at the time)
Idiot: Mental Age >2
Imbecile: Mental Age 3-7
Feebleminded: Mental Age 8-12
Moron: highest functioning mentally retarded

Question 11

Terman’s Contributions to Binet

Answer 11

• Revised and published Binet’s test as the Stanford-Binet
• Developed and published intelligence testing as “measurement”
• Became the standard against which others were validated

Even then there was a concern that people with low intelligence will be considered as congenitally and hopelessly inferior.

Question 12

What is the Intelligence Quotient (IQ)?

Answer 12

Introduced in the Stanford-Binet test

Recall that mental age is based on the age level at which the majority of ‘normal’ children in the standardisation sample passed the test: this is problematic

Question 13

What is ratio IQ?

Answer 13

Was the original measure used in Stanford-Binet (now uses deviation IQ)

ratio IQ = mental age/chronological age x 100

Question 14

What are the problems with ratio IQ?

Answer 14

• Ratio IQ only works if mental age increases proportionally with chronological age and we know that’s not the case
• Difficult to make comparisons of intellectual performance across age levels e.g. who is smarter, a 13 yr old with a mental age of 15 or an 8 year old with a mental age of 10?
• Difficult to apply to adults
  What would be an ‘age-level-typical’ task for a 43 year old
  What does it mean if a 50 year old has a mental age of 25

Question 15

What are deviation scores?

Answer 15

Deviation scores (tells you where you sit relative to others like you)

From this can start to work out where a particular score might sit in the distribution

You can also measure how far one of the scores deviates from the mean - z score

Example: if you score 8, then you have scored at or above 96% of the population. Or you are 3 SDs higher than the average of 5.

Question 16

Why is it important to standardise?

Answer 16

Comparing Raw Scores across Groups
- Possible normative groups (subgroups):
14 yr old school children (general population)
18 yr old university students
8 year old school children

A score may have different percentile ranking for each group, and thus different psychological interpretations.

That is why it’s important to standardise these scores.

Standardising (norming) a test uses the ‘normal distribution’:
Raw scores → Z-scores → Deviation IQ scores

Question 17

What is deviation IQ?

Answer 17

IQ scores mean the same thing regardless of the comparison group

IQ=100 means you are average at any age (consistent with the ratio IQ familiar to psychologists)

All other scores reflect how far your score deviates from the average

• IQ of 70 means you are 2 SDs below the average
• IQ of 115 means you are 1 SD above the average

Question 18

What is the set up for the Standford-Binet IV test?

Answer 18

• 15 subsets in Four areas of cognitive ability
• Multiple separate tests, each in order of difficulty for 30-90 mins
• Special-purpose batteries available for certain populations e.g. blind, motor impaired, deaf etc

Standard materials: Four booklets of printed cards, blocks, form board, beads, large picture of a unisex, multi-ethnic doll, record booklet, guide

Examiner highly trained: recommendations on seating, precise wording, mode of query

Clinicians: not only standardized tests but also clinical observation. Will indicate features of an individual’s work methods, problem solving approach, and other qualitative aspects

Question 19

What is the general test process of the Stanford-Binet IV?

Answer 19

Establish a basal and ceiling level for each task - usually suggested by examinee’s age

• Basal level = four items passed in a row
• Ceiling level = three or more out of four consecutive items are failed (discontinue)

Scoring

• Each item response is recorded verbatim
• Items are either passed or failed according to strict guidelines

Raw score = Number of correctly answered items
- This score is later converted to a scaled score

Gives Standard Age Score (SAS), mean = 100, SD = 16

Question 20

What is group testing?

- Raven’s Progressive Matrices?

Answer 20

• Prototypical test of fluid intelligence (Gf)

- Broad ability concerned with basic processes that depends only minimally on learning and acculturation

Question 21

Distinguish between a single underlying factor and multiple abilities

Answer 21

Single Factor: All of the tests load on one intelligence factor (g)

Uncorrelated factors: a range of primary mental abilities that are unrelated to each other

Or you could have both:
Have factors that are relatively independent of each other but may be correlated through general factor (g)

• We try to infer by seeing how the individual scores correlate
• If they rise and fall together (are correlated) this suggests they are linked - different tests may depend on single ability
• If rise and fall independently (not correlated) this suggests they are not linked - different tests may depend on different abilities

Question 22

What are the different types of correlation?

Answer 22

Positive: as scores on one variable increase, scores on the other variable also tend to increase

Negative: as scores on one variable increase, scores on the other variable tend to decrease

Zero: no systematic relationship between scores on the two variables

Question 23

Spearman’s observations on how test scores correlated with each other?

Answer 23

Observation 1: all mental abilities correlate with each other to some extent

• There is something common underlying all the tests - positive manifold
• Some researchers interpret these results as meaning there is a General intelligence factor reflected - Spearman’s g

Observation 2: there seem to be multiple clusters of stronger correlation that are only weakly related to each other
- Whatever factor is ‘causing’ the correlations between tests 1-3, is different to the factor that is ‘causing’ the correlations between tests 4-6

Conclusions

1. There is something common underlying all the tests - positive manifold
2. There are separate factors underlying performance on tests 1-3 vs 4-6

Question 24

What is fluid and crystallized intelligence?

Answer 24

General Fluid Intelligence (Gf)

• The ability to grasp relations between things; solve novel reasoning problems
• Non-verbal abilities, inductive and deductive reasoning

General Crystallized Intelligence (Gc)

• Using acquired knowledge and skills
• Requires exposure to culture, formal/informal education
• May require some investment of fluid intelligence

Performance on a single task can (and is likely to) require both.

Question 25

Differentiate between fluid and crystallized intelligence

Answer 25

Although conceptually different, Gf and Gc correlate to varying extents

Strongest evidence that they are different is that they show different developmental trends

• Fluid rises to young adulthood, then falls off in old age
• Crystallized rises and plateaus, roughly speaking

Correlation is strong in the beginning but then gradually decreases.

Question 26

What are primary mental abilities?

Answer 26

7 separate areas of mental ability revealed on tests
- Verbal comprehension
- Inductive reasoning
- Numerical fluency
- Word fluency
- Spatial ability
- Memory
- Perceptual speed
Relative levels differ among individuals

Question 27

What is Classical Test theory?

Answer 27

Any observed score has two components:

• The True Score (the real level of ability)
• And some Error component (random variance)

X=T+E

Question 28

What are sources of error in testing?

Answer 28

Test construction e.g. choice of items/stimuli; content of the test

Test Administration e.g. variability in examiner, variability in examinee

Errors in Scoring e.g. failure to use ‘rubric’ consistently

Interpretation Subjectivity e.g. evaluation of response

Question 29

What is systemic error variance in testing?

Answer 29

• Systemic error will NOT decrease the estimated reliability
• But it will reduce the validity of the test - we are not measuring what we think we are measuring
• If a set of scales consistently gives readings 1 kg too light
• If an assessor consistently gives an extra (unwarranted) mark on an assignment

Question 30

What is reliability?

Answer 30

Reliability - is our test accurate

• If a test measures a consistent trait in a person, then it should consistently produce the same answer
• It should not really be affected by random fluctuations
• If a test is reliable, it should be able to distinguish between people who differ on the construct
  Because any variation in performance is due to true differences in that ability, not random error

Question 31

How can we estimate reliability?

Answer 31

1. Test-Retest Reliability: same group of people measured twice on the same test
   However:
   - in theory, changes from X from time 1 to time 2 are due to measurement error
   - so we assume true score doesn’t change
2. Equivalent (alternative) Forms
   - Measure the same phenomenon using two different forms of the test
   - Correlation between Form 1 and Form 2 is the reliability of the test
   - Or compare two halves of the test (split-half correlation)
3. Internal consistency - Cronbach’s alpha
   - If every possible split-half correlation was computed, their average is called Cronbach’s alpha (α)
   - Reflects the extent to which all items measure the same thing

Question 32

What is validity?

Answer 32

Is our test measuring what we think it is measuring?

Is the test used appropriately, for its intended use? I.e. if the test was developed for adults, is it being used for children?

Question 33

What is content validity?

Answer 33

Coverage of the ‘domain’

• Does the test assess behaviour that is representative of the domain of behaviour that we want to measure?
• Typically discussed in educational and achievement testing situations but also relevant in other areas

Define the boundaries and structure of the domain

• Boundary: what is considered part of the domain and what is not
• Structure: test content reflects the structure of the domain

Question 34

What is construct validity?

Answer 34

How well defined is the construct measured by this test?

• Convergent validity: is the construct related to other theoretically similar constructs/tests?
  Expect high correlation with similar constructs/tests
• Discriminant validity: is the construct independent of other, unrelated, psychological constructs?
  Expect low correlation with unrelated constructs

Question 35

In summary, how do we measure intelligence?

Answer 35

Step 1: work out what it is we want to measure (pragmatic and theoretical)
Step 2: work out what it looks like (signs, manifest variables)
Step 3: Devise tests and scores
Step 4: work out how it is structured (single factor vs multifactor)
Step 5: assess whether our test is valid and reliable

Question 36

What is a heritability estimate?

Answer 36

Heritability (H) is the proportion of phenotypic variance (Pv) that is due to genetic influences (Gv)
H ≈ Gv/Pv
The value of H is always between 0 and 1 (it is a proportion)

Whatever remains of the obtained heritability reflects phenotypic variance that is attributable to environmental and residual effects

Thus 1-H represents the combined effects of environmental and residual factors.

Question 37

What does heritability depend on?

Answer 37

‘H’ depends on:

• The genetic variability of that population
• The degree of variation in its environment

Change variability in the environment

• Reduce variability e.g. uniform environment, schooling
• Result: increase the relative contribution of genetic influences; behaviours will appear to be more heritable

Change variability in genetic contribution

• Reduce variability in genes
• Result: environment will have a more significant contribution

Question 38

What is evidence that the IQ is substantially heritable?

Answer 38

Bouchard & McGue (1981) put together an index of similarity in IQ
Size of correlation in IQ increases as heritability increases.

IQ correlation of same person taking the test twice and MZ twins is almost the exact same.

Question 39

What are the basic designs in heritability and environment studies?

Answer 39

1. Family resemblance studies