PSYC2016 - Intelligence 2

The Two Disciplines Problem
Experimental Paradigm

- Standardisation of "situational variables" through treatments and experimental conditions (IV's)
- Investigate cause and effect relations (on DV's) ... but on incomplete information ...
- CONTROL is the important feature
- But, what is the role of individual differences in an approach dominated by "Comparison of Means"?
- "The goal is to control behaviour and variation within treatments is proof the experimentalist has not succeeded ... Individual variation is cast into the outer darkness known as 'error variance'"

Experimental Psychology

treat variance as 'noise'

rather than individual difference

Correlational Psychology

'... is in love with just those variables the experimenter left home to forget.'
- Goal is to predict variation within a treatment
- Driven by the development of Factor Analysis
- Focus on relationship between performances on multiple tests

"His sophistication in data analysis has not been matched by sophistication in theory. The correlational psychologist was led into temptation by his own success, losing himself in practical prediction, and then in a narcissistic program of studying his tests as an end in themselves"
Cronbach
^ lack of theory

Cronbach's (1957) Lament

"Psychology continues to this day to be limited by the dedication of its investigators to one or the other method of inquiry rather than to scientific psychology as a whole"

A suggested solution...
Aptitude x Treatment Interaction

Cronbach (1957): Aptitude x Treatment Interaction

Effectiveness of treatment associated with one's aptitude

Average effects still hold, but they are qualified by aptitude

1. Over simplification - Difficult to replicate
2. Practical and resource constraints on individualisation

Attempts at Integration since 1956

Aptitude-Treatment Interactions
- Snow (1966; also Cronbach; Lohman)

Componential analyses: Correlatoin with psychometric tests
- Sternberg (1977); Hunt (1980)

Differences in skill learning as a function of aptitude
- Ackerman (1987)

WM and control processes in intelligence
- Embretson (1995)

Working memory, controlled attention, PFC, Gf
- Engle, Kane and Tuholiski (1999)

Working memory and bindings
- Oberauer et al (2003)

Latent-Trait Theory (between vs within)
- Bosboom et al (2004)

Relational integration and Gf
- Birney et al (2002; 2009; 2019)

Have we succeeded?

As a discipline? NO...
- "Cognitive psychology promised to rescue differential psychology from psychometrics and return it to mainstream research. But this has not happened." (Lohman, 1994)
- "differential psychology was unable to achieve one of its central goals: The identification of the mental processes that underlie intelligent functioning. ... it was only recently that many in the field acknowledged that a research program dominated by factor analysis ... was incapable of producing and explanatory theory of human intelligence"

Identification of Mental Processes

Individual Differences as a pathway to Process
- Factor Analysis [Spearmen (1904) Thurstone (1947)]
- Factors were primitive/core psychological functions

- Guildford (1967): "It is only by correlating scores on a test with scores on other tests that we can test any hypothesis as to what it measures psychologically"
- "In the search of meaning of aptitude factors, we can take one more easy but very significant step to tie factors ... to psychological theory. This step is to say that such a factor is also a psychological function"

But this assumption was RIGHTLY challenged


"studies of individual differences never come to grips with the process or operation by which a given organism achieves an intellectual response. Indeed, it is difficult to see how the available individual differences data can be used even as a starting point for generating a theory as to the process nature of general intelligence or of any other specified ability"

Overview of the argument so far

1. Failure to identify processes of intelligence originated in the failure to consider aspects of tests that have an impact on performance during the test
2. Measurement models (structural process theory) of tests are needed; consider component processes of test performance
3. However, component scores have not been particularly useful, they don't explain individual differences
4. Tasks need to be developed that focus on qualitative differences between individuals

David Lohman & Martin Ippel (1993)

Main Claim:
- Cognitive approach to measurement is most useful when applied to tasks designed to elicit responses that reveal qualitative differences between individuals in knowledge or strategy

Differential Test vs Cognitive Task
- Differential tests are not the same as cognitive tasks and attempts ot model one from the other have not worked well

- Experimental tasks are founded on differences between processes/conditions (e.g. list memory)
- Differential tasks are focused on differences between individuals (e.g. Raven's Matrices)

Factor Analysis:
Common Processes

Each of these tests ARE based on a theory of 'processes'

It is just that the theory is often implicit

Not a sum
It is what they have in common

What is not in common is purer processes

Factor Analysis

a statistical procedure that identifies clusters of related items (called factors) on a test; used to identify different dimensions of performance that underlie a person's total score.

Ability tests from an item perspective

A major criterion is Reliability (repeatability of measurement)

The main problem with the individual differences approach to the study of intelligence was

that it did not have a clear conception of what a 'process' might be, or how particular aspects of a test might elicit those processes
Lohman & Ippel (1993, p43)

Inidividual differences as a way to understand processes

Heitz, Unsworth & Engle (2005, p74)
" The most puzzling realisation is that we have good reason to implicate attention in Gf, but we are devoid of a suitable explanation of how attention comes into play when performing tasks such as the Raven (matrices test)"

Within-Subject Process Design

- Borsboom, Mellenbergh & van Heerden (2004)
- Theories based on between-subject designs (factor analysis) cannot tell us about what is happening within a given individual
- Borsboom (2015)
- Birney, Beckmann. Beckmann (2019)

Mental Processes as Information Processes

Definition
- Information-Processing approach treats mental activities as different operations performed on symbols and symbol structures (mental representations)

Central Assumptions
1. Mental activities are decomposable into a series of relatively independent processes or operations
2. For each mental operation, a set of task conditions exist, that when varied, will exert a selective influence on these particular processes
3. These internal processes may produce externally observable effects on behaviour

In order to benefit from the information-processing perspective, we need to reconceptualize the measurement processes

- Not so much as, Measurement as applied statistics
- But more like, Measurement as substantive theory

Measurement as a Structural (Process) Theory

Measurement as applied statistics
- Goal to compare individuals
- Most often based on Classical Test Theory


Measurement as a structural theory
Guttman (1971)
- Goal: to assess the structure of relationships among 'observations'
- Substansive theory is needed to design a procedure for classifying observation
- This is the essence of Guttman's Structural Hypothesis
- Items are designed/selected to target underlying processes

What is a Structural (Process) Hypothesis?

Need to ask the question: What processes or operations are instantiated when people solve a cognitive task?

How do people make responses?

Mental Rotation Experiment (example)

Component Processes

Building a process theory and testing

Structural Theory: Component Processes

Performance =
average mental rotation ability + something to deal with encoding + something to deal with mental rotation + something to deal with comparison + something to deal with an interaction

Two implications to consider

1. Observational Design
- Judicial selection of items to measure
2. Assumption of behaviour consistency
- Differential psychology assumption that latent dimensions are stable

Observational Design

Random selection of items from the universe of possible items is not appropriate
a) Bias in measurement by specific items "ending up" in the test
b) Bias for some individuals if they have different profiles of strengths on the different processes

Suggests that different responses by the same person on different items may be systematic
- Within-subject variability (in part) due to different structural features of the task

TESTING COMPONENT PROCESSES

Measurement Design (what to score?)

Traditionally (psychometric), the object of measurement is simply the person
- e.g. some aggregate score (total, mean) from "randomly" selected items

BUT, process analysis allows multiple objects of measurement to be specified
- Using total score aggregates within-subject variance across items and relegates it to the residual

Assumption of behaviour consistency

Dimensional/Factor Analytic Theories
- Assumption: An individual's score/value on a factor (latent dimension) is a stable characteristic (of that person) that explains behaviour consistencies across relevant situations (in a test, items!)
- A structural hypothesis is not relevant because attribute is assumed constant (X = T + E)
- That is, the score is assumed identical across all observation conditions
- AND: this is implemented statistically in test theories (such as Classical Test Theory)

Assumption of Behaviour Consistency
Factor Analytic Theories

Assumption of Behaviour Consistency conflicts with the assumption made in research that aims to DESCRIBE HOW subjects achieve the responses they give (i.e. information-processing approaches)

Summary
Information processing approaches

Information-processing approaches rest on a view of measurement that is fundamentally different from that of dimensional theories of individual differences

Observational and Measurement Design

Observational Design
-The purpose of the observational design is to structure selection of items so that defensible inferences about the theoretical constructs or attributes can be made. Validity!

Fluid Intelligence and Working Memory Capacity

- A long-standing issue in fluid intelligence research is the need to define its nature at a deeper level than "novel problem-solving ability"
- Working-Memory (WM) Capacity has been seen as a candidate account
- Research into WM has concentrated on one of two complementary but different approaches

1. Experimental (Analytic) Approach
2. Psychometric Approach

- Each has strengths and weaknesses

Experimental Approach to working-memory

Definition: Most effort devoted to the "slave systems"

Tasks: Dual-task methodology (using fairly simple tasks)
- reasoning + span-task
- interference task (articulatory suppression)

Test of theory: Dissociation through Dual-Task or Neuropsychological evidence

Advantages: Can be applied to analyse the constituent processes of the slave systems

Disadvantages: has made less head-way in teasing apart the complexities of the central executive (partly due to less research on this topic)

Psychometric Approach to Working Memory Capacity

Definition: WM as a system necessary for concurrent storage and manipulation of information

Tasks: devised to combine storage and processing

Test of theory: Extent to which WM tasks predict a range of cognitive skills, such as reading, comprehension, and reasoning

Advantaged: Can (better?) tackle the central executive; work directly on problems of practical significance (reading comprehension; educational and clinical applications)

Disadvantages: Somewhat arbitrary construction of tasks that do not readily lend themselves to detailed analysis of component processes

Working Memory and Gf

Since the 1980s
- major theoretical and empirical developments of the construct of working memory as distinct from rote or span memory (short-term memory)
- several investigators began to argue that WM and intellectual abilities are highly related or even identical constructs

Daneman & Carpenter (1980)
- Relationship between individual differences in working memory and reading comprehension
- Using reading span tasks

Reading Span (Daneman & Carpenter, 1980)

"When at last his eyes opened, there was no gleam of triumph, no shade of anger"
"The taxi turned up Michigan Avenue where they had a clear view of the lake"

Assumption: It is a WM task because it requires processing and storage

Processing: Does the sentence make sense?

Storage: Recall the final words of each sentence

Daneman & Carpenter (1980)

Hypothesis
- Individual differences in reading comprehension reflect differences in working memory capacity
- Trade-off between processing and storage functions

Factor Analysis
- Latent variables ...
- Represent variance that is shared among all the tasks that are being used to identify the construct
- Task-specific variance that is unique to each task is removed, resulting in a relatively pure measure of the latent construct of interest

Results
- WM (reading) span correlated with three reading comprehension measures
- Similar correlations with a listening span task
- the correlation is not specific to reading (i.e. comprehension)

Traditional digit span and word span measures which DO NOT correlate with comprehension

Kyllonen & Christal (1990)
Reasoning ability is (little more than) working-memory capacity?!

Operationalising WM
- Needed a well formulated theory of the task requirements of different WMC tasks
- Theory available?
> Experimental psychology literature: Yes
> Differential psychology: Very little
- An important point!

- Used a 'varied-content and varied-process' approach to select WMC tasks
- Makes some sense to select a wide range of tasks (different content and process), and derive a latent variable from them all


Latent Variables
- Identifies what is common!
- There is a problem with this approach (i.e. selecting tasks using a weak theory):
1. Difficult to ascertain key cognitive mechanisms underlying performance on the tasks contributing to the latent variable
2. This makes it difficult to account for the covariation between WMC and g in terms of cognitive processes

Factor Analysis of WM & Gf (or g)

grouping mental processes

latent variables

- R
- GK
- WM
- PS
testing relationships between ^

Collected data on these different tasks



Correlations:
- Reasoning & WM r = 0.82
- Reasoning and General Knowledge r = 0.45
- WM and GenKnow r = 0.24
- Reasoning and ProcSpeed r = 0.25
- WM & ProcSpeed r = 0.47

Evidence for differentiation of the WM and Reasoning factors

Factor Analysis of WM & Gf (or g)
Structural component of model

Correlations:
- Reasoning & WM r = 0.82
- Reasoning and General Knowledge r = 0.45
- WM and GenKnow r = 0.24
- Reasoning and ProcSpeed r = 0.25
- WM & ProcSpeed r = 0.47

Evidence for differentiation of the WM and Reasoning factors

Kyllonen & Christal (1990) Conclusions

The central message: g = WM
- But also evidence for some dissociation

1. Reasoning had higher correlation with Knowledge than did the WM

2. WM had a higher correlation with Processing Speed than did Reasoning
- That is, differentiation on the basis of convergent and discriminant validity evidence with General Knowledge and Processing Speed

"...we do not see 'g' as a separate factor driving the correlation between WM and Reasoning. Rather, we believe 'g' can be identified with either WM or Reasoning"

"We concede to a certain degree of arbitrariness in creating tasks according to such a broad and vague definition of their requirements, and readers may find fault with the way we operationalised the working memory factor. But without well-developed models of information-processing requirements of the tasks, we can only proceed with what is available"

Baddeley (1992)

Individual differences approach use somewhat arbitrary construction of tasks that do not readily lend themselves to detailed analysis of component processes

Engle/Kane/Conway et al Studies

- Follow a similar method to Kyllonen & Christal, but on a tighter theoretical basis of WM
- Conclude that WM and 'g' are highly related, but not identical (~ 50-80% shared variance)
- Executive Attention view of WMC (controlled attention)

Differential Findings

STM -> not related to gf

WM -> related to gf

Alternative Methodological Approach

Engle, Tuholski et al (1999)

what is common?

How is this different?
postulating a common factor and see how this relates to WM, STM, gF

Critical Findings from Engle et al

1. STM storage and rehearsal factors cannot account for the relation between WMC and Gf
2. WMC and STM tasks yield correlated but separate factors
3. WMC factor is more strongly associated with Gf than is STM
- Empirical Question: If the storage, rehearsal, processing and strategic processes tapped by WMC tasks do not explain (all) the correlations with ability, then what does?
- The 'common-secret ingredient' is an attention-control capability that is elicited by WMC tasks to a greater degree than by STM tasks (controlled-attention)

N-back Task: Role of Inhibition

Does the current element match with the one N back?

False responses due to lure

Burgess, Gray, Conway and Braver (2011)

Lure activities
- (failed) Inhibition
- Gf and WM

Gray et al (2003) - 3-back task

Performance on Lure trials
1. Lure trials are more difficult than non-lure trials
2. Lure trials predict individual differences in performance on Ravens (Gf) over and above performance on other trials
3. Correlations between brain activity and Gf were greatest in high interference (lure) trials
4. Brain activity was more focused to specific areas in lure trials than in non-lure trials

Brain activity and lure activities

causal model

General Findings

Individual differences in performance on WMC tasks predict a variety of hugher order cognitive abilities (learning; comprehension; reasoning)
Even when:
- WMC and ability tests bear no surface similarity to each other
- So these factors cannot account for the relationship between WMC and ability

People with lower WMC show poorer control over thought and action than do those with higher WMC
1. Fail more often to prevent or recover from prepotent (automatic) responses
2. Show slower and less flexible allocation of attention to objects

Shipstead & Engle (2018)

- Tests of working memory capacity place primary demand on the maintenance related aspects of the working memory system and secondary demand on the disengagement aspects
- Fluid intelligence tests place demand in the reverse order

- Thus, the observed constructs of working memory capacity and fluid intelligence do not represent separate cognitive abilities, so much as they represent different and sometimes contradictory functions that are necessary for information processing

Executive Functions: Miyake, Emerson et al 2000

Shifting
- shifting between mental sets or operations by disengaging from an irrelevant mental set and actively engaging in a set relevant to the current task

Updating
- active process of monitoring incoming information and 'updating' items held in working memory by replacing irrelevant information with task-relevant information

Inhibition
- active and deliberate overriding of a dominant or automatic response in order to complete a task

Unity-Diversity EF Framework (Miyake and Friedman 2012)
- Unity is represented by the arrows from the distinct inhibition factor linking all tasks
- Diversity is represented as each factor is linked by arrows to their respective tasks (and only their tasks for updating and shifting)

Example of inhibition and updating tasks

+ Wisconsin card sorting task

Dynamic Model of General Intelligence
(Conway and Kovacs, 2018)

Process Overlap Theory
- Motivated by Gf-Gc theory
- Gf Working Memory relationship

WM Processes most likely reflect individual differences in executive attention

g is 'not a thing' but instead is the consequence of a set of overlapping cognitive processes sampled by a battery of tests

- Different intelligence tasks require multiple domain-general processes (explain positive manifold, and 'g')
- Overlap in these different processes has been demonstrated to lead to positive manifold

Therefore 'g' is:
- an index rather than a causal factor (cf Gottfredson)
- result of measurement

POT is influenced by the executive attention theory of WM

Executive Functions and CHC

Expertise

Ackerman
- Adult Intelligence
- Knowledge and expertise
- Conative dispositions are important
> Personality
> Interests

Ericsson
- Adult expertise
- Arrested Development
- Deliberate Practice

Learning Objectives
- Understand Ackerman's view on the importance of non-cognitive variables on the intelligence as knowledge, and evidence for this view
- Understand the concepts of automatisation, arrested development and developing expertise

A Broader Conceptualization of Individual Differences

Some Puzzling Facts
- Correlations between measures of intelligence and uni GPA rarely exceed 0.50 (i.e. 75% unexplained variance)
- Research in gerontology show that adults perform poorly on several tests of intellectual abilities, BUT function quite well in day-to-day activities
^ more to intelligence than just test results
- Teachers identification of over achievement and under achievement
- Expertise is not simply a function of g

Academic vs 'Real-World'

In contrast with
- Multiple Intelligences (Gardener)
- Creative and Practical Intelligence (Sternberg)
- Emotional Intelligence (Mayer & Salovey)

Phillip Ackerman: PPIK Theory

Summary of Ackerman's Arguments
- Under-appreciation of the importance of knowledge
- Inappropriate to use the methodology of child assessment for the assessment of adult intellect
- Intelligence Tests we use do not assess knowledge very well
- Broader conceptualization of knowledge is needed
- Personality and interests contribute to knowledge acquisition

PPIK

P: Intelligence as Process
P: Personality
I: Interests
K: Intelligence as Knowledge

Justifications for taking knowledge seriously

1. Lay definitions (our implicit/intuitive theories of intelligence)
- Person's intellect often defined as what things an individual can perform or achieve hence, performance on 'intelligence tests' is just a small part of implicit theories

In contrast,
- other competencies characterise broader aspects of adult intellect, such as being able to:
> figure out financial markets
> build a house
> write an article or novel
> perform scientific experiments
> and so on...

So, our implicit theories about intelligence suggest a broader conceptualisation

Second justification for taking knowledge seriously

2. Research
a) Investigation into knowledge-based and skill-based differences in experts and novices
- Declarative Knowledge: Knowing that/what (most Gc tasks measure this)
- Procedural Knowledge: Knowing how
- Tacit Knowledge and Practical Intelligence: Knowing with

b) Expertise is predicated on long study and practice to develop rich, specific knowledge structures

Influences on the development of PPIK: Gf/Gc

Fluid Intelligence (Gf)
- The ability to grasp relations between things
- Nonverbal abilities, inductive and deductive reasoning
- Culture free (????? in theory, but not in practice)

Crystallised Intelligence (Gc)
- Acquired knowledge and skills
- acculturated knowledge requires exposure to culture, formal/informal education
- may require some investment of fluid intelligence

Knowledge (Gc) develops out of an investment of Gf

Intelligence as Process: Maximal Performance

Intelligence-as-Process
- Speed of processing; memory; reasoning
- Classic Gf conceptualization has problems
- Some tasks are Gf for some people (Children) and Gc for others (Adults)

Note: General ability tests attempt to remove the benefits of specific expertise
- Sampling is very broad and content is not associated with expertise

Without the application of directed cognitive effort toward knowledge acquisition, performance on specific achievement tests will suffer

Personality

- Weakly related to intelligence-as-process
- Intelligence-as-knowledge is associated with two related personality traits
- Openness to experience (Big five factor)
- Need for Cognition

Need for Cognition Scale
- Like to solve complex problems
- Need things explained only once
- Love to think up new ways of doing things
- Love to read challenging material
x
- Have difficulty understanding abstract ideas
- Try to avoid complex people
- Avoid philosophical discussions

Interests

1. Realistic
2. Investigative
3. Artistic
4. Social
5. Enterprising
6. Conventional

(Intelligence as) Knowledge

1. Similar to Gc but broader and more encompassing (not just academic)
2. Gc is often determined 'historically'
3. Test of intellectual performance that is contextual
4. Developmental trajectory similar to Gc

- Argument is not that these (PPIK) are all measuring aspects of the same thing (which is one way to interpret correlations)

- These distinct attributes contribute interactively in the development of abilities, competencies and expertise

Development of Personality, Interests and Intelligence

Abilities and Interests develop in tandem
1. Ability levels determine the probability of success in a particular task domain
2. Personality and interest determine the motivation for attempting a task
3. With subsequent successful attempts at task performance, interest increases and knowledge in the task increases

With subsequent unsuccessful attempts at task, interest declines

During Childhood

Common (educational) experiences limit inter-individual differences in specialised knowledge
- The extent that hobbies and interest in other extra-curricular activities develop, and are motivating, expertise develops
- As one moves away from homogenous experience, differentiation in skills and knowledge increases - (inter)-individual differences grow

Therefore, in assessing ability, content is important

Evidence 1. trait complexes

clusters
- abilities
- personalities

Evidence 2

Relations between abilities, trait complexes, and knowledge

positive AND negative

Ackerman's Research Conclusions

1. Middle-aged adults more knowledgeable on several broad and specific domains compared with young adults
- Supports notion that focused cognitive investment over extended periods of time yields clear differences between individuals in depth and breadth of expertise

2. Measures of Gf, which declines as adults enter middle age, fail to fully account for individual differences in knowledge structures or that middle-age adults know more
- Gc is more predictive, but doesn't capture rich breadth and depth of domain-specific knowledge

3. Science/Math and Intellectual/Cultural trait complexes are supportive of domain-knowledge acquisition, whereas Social and Clerical/conventional impede acquisition

Cattell's Investment Theory: Gf and Gc

1. Historical and present-day Gc
- Ackerman: "For adults, historical abilities (or school version abilities) are those that are assessed by most standardised group tests used in schools - those that focus on the kind of knowledge and skills developed prior to adulthood"

2. Gc is experiential
- Influenced by time invested in intellectual pursuits over development, and by historical and current interests and memory

Open people are more likely to be attracted to more learning opportunities which positively affect Gf
- Gf positively affects Openness because the skills afforded by Gf are more likely to lead to success in novel, complex situations

Summary

Development of adult intellect

K.Anders Ericsson: Deliberate Practice

Modifiability of performances
- the biological mechanisms underlying performances can be improved substantially

Mechanisms
- Expert performance is mediated by complex modificable representations that allow experts to exhibit
> faster speed
> superior selection of actions, and
> more precise motor execution


Substantial Practice Effects
1. Short-term Memory Task
- Tested repeatedly every second day for many weeks and months
all improved 200%

Interpretation
- Acquired skill for storing and accessing LTM
- Expanded their functional working-memory

2. Physical Activity
push-ups

Mechanisms mediating superior performance of experts

the ability to
1. select superior actions
2. generate rapid reactions
3. control movement production

1. Select superior actions

Analysis of verbal think aloud reports
1. Performance of experts is mediated by increasingly complex control processes
- More extensive planning, reasoning, evaluation (expert chess players can follow chess games in their head)

2. Rapidly encode and store information so that it can be more effectively accessed and manipulated

Stages of Learning (for typical performance)

First stages of learning
- Try to understand what is going on and avoid making mistakes

Middle phase
- Gross errors become rare

Later phase
- Performance becomes 'automated'
- Further skill acquisition is minimised
- Risk of 'arrested development'

Typical Learning Process

Arrested Development and Deliberate Practice

- The principal challenge for attaining expert performance is that further improvements require continuously increased challenges that raise the performance beyond its current level
- Challenges create mistakes, mistakes are not enjoyable and creates a motivational challenge (pretty much the only place motivation comes in)
- Acquisition of expertise requires and orderly and deliberate approach
- Deliberate practice is structured, it requires full concentration

10 year rule?!

Dynamic Assessment

Binet (1909); Thorndike (1924); Andre Rey (1934)
- Necessity to measure 'ability to learn'
- Suggested testing educability

Lev Vygotsky (1943/1962)
- Zone of Proximal Development
- An interaction between social context and individual
- What can be achieved with scaffolding instruction

Rueven Feuerstein (1970's)
- Mediated Learning Experience
- "Cognitive modifiability' is achieved through mediation of the complexities of the environment (task) to the individual

Lecture 9 Overview

Learning potential as cognitive flexibility
- Working-memory and Gf
- Structural process hypothesis
- Conative determinants of learning

Dynamic Assessment of Learning Potential

Ability as competencies in development
- Assessment as a snapshot in time

Dynamic Assessment as Intervention
- What can be achieved with feedback?
- How much is feedback needed to achieve?


begins unitary, becomes much more diverse

A paradigm for studying dynamic processes

Structural Process Hypothesis
1. Historically, understanding intelligence has been a between-person endeavor
2. Psychometrics are unable to identify the mental processes that underlie intelligent functioning
3. Within-person level processing must be incorporated in between-person measurement models
4. Need theoretically substantive structural process hypotheses and experimental manipulations

Cognitive Flexibility

Cognitive flexibility as an executive function assessed through shifting and set-switching tasks
- BUT .. this is an incomplete perspective, and maybe too narrow
(Wisconsin Card Sorting Test)

The extant literature suggests understanding cognitive flexibility as
- A cognitive approach to novelty processing
- A conative disposition towards novelty

Wisconsin Card Sorting Task Scoring

The task
Sort 64 cards to match either
- colour rule (red, blue, yellow or green)
- form rule (crosses, circles, triangles or stars)
- number rule (one, two, three, or four figures)
The sorting rule changes without notice

Requirement
- Monitor for rule changes
- Shift sets accordingly
- Sort cards following the new sorting rule

Scoring (most common)
- Set shifting (flexibility) difficulties indicated by total number of preservative errors

Cognitive flexibility definition

Cognitive flexibility is reflected in adaptive performance as a strategic response to novelty in dynamic environments

We can operationalise this
- Correlates of problem solving that almost exclusively relies on between-subjects designs, is limited in its capability to serve as the foundation for understanding dynamic decision making
- So use within-subjects designs

Lecture 9

Advantages of...
- Investigation of repeated observations
- under systematic conditions (experimental control; structural process hypothesis)

Entails...
- Within-subject designs and analytics

Hypothesis Testing ...
- Experimental manipulations hypothesized a priori to make differential demands on intelligence

Why ... ?
- To better understand the dynamic nature of processes underlying intelligence

Case Study 1
- Latin Square Task

Case Study 2
- Raven's (Psychometric Assessment)

Case Study 3
- Complex Problem Solving (Microworld)

Moderators: Gf, Personality, Emotional Intelligence, Conative Mindsets

Case 1: Manipulations of Working Memory in the Latin Square Task

Simultaneous storage and processing

Case 1: Results

Complexity Effect
- Is the relationship between Gf and performance moderated by task manipulation (i.e., RC and Steps)?

Case 2: Learning Trajectories

Can we explain the individual variance?

- Cognitive Difficulty - arousal?
- Learning from experience - worry?

Neuroticism

Learning Trajectories

Case 3: Complex Problem Solving (CPS) & Microworlds

Simulation variables

Input variables
- Decision variables
- Those for which the problem solver or learner sets the values

Output variables
- consequences of the input decisions plus effect due to intervening relationships within the model

Complex Problem Solving (cps)

Microworlds

Microworlds: Examples in Business Education

Dynamic Performance Parameters

Microworlds (Dynamic Systems)

Input variables
- Decisions

Output variables
- Cumulative penalty

Dynamic Performance Parameters
variables

- Penalties
- Attempts

Case 3: Results

Summary
Cognitive Flexibility

Cognitive flexibility as learning potential
- Limitations of the use of correlations and factor analysis alone
- Structural process hypothesis
- Working memory and Gf
- Conative determinants of learning