Executive Functions, g, Fluid Reasoning, and Working Memory

1
Executive Functions, g, Fluid Reasoning, and
Working Memory

• Theory and Measurement

2
Brief Background
Georgia State University
3
Purpose

• Collaborative research
• Intelligence (g, Gf), Working Memory, Executive
  Functions are all related and constitute core
  functions in intelligence
• IQ tests dont measure executive functions, thus
  dont measure intelligence
• Theoretical speculation

4
Overview

• School Psychology -overview
• What is intelligence and higher cognitive
  abilities
• 6,292,930 children in special ed.(Digest of
  Educational Statistics, 2002), many of which
  receive IQ tests
• Intelligence Testing
• Binet, Terman, Yerkes, Wechsler, Spearman,
  Cattell, etc.

5
History of Intelligence Theory

• Intelligence as a single general ability
  (Spearman)
• Intelligence as a pair of abilities
  (verbal-nonverbal)
• Intelligence as a limited set of multiple
  abilities
• Intelligence as a complete set of multiple
  abilities (Cattell-Horn)
• Intelligence as a hierarchy of narrow abilities
  underlying multiple broad abilities (Carroll)

6
CHC Model
Wheres Executive Functions?
7
g

• Empirically based
• Represents the shared commonality of all measures
  of cognitive ability.
• Difficult to interpret
• Why are all tests of cognitive ability positively
  correlated?
• Jensen
• Most important construct in psychology
• Nerve conduction velocity, based on RT studies

8
Interpreting g

• g-loaded tests involve complex cognitive
  operations such as reasoning and abstraction that
  is based on the complexity of the mental
  operations rather than specific knowledge in a
  given area (Arend et al., 2002 Jensen, 1998
  Spilsbury, 1992).
• Spearman
• Intelligence is induction of relationships/correla
  tes, mental energy

9
G and fluid intelligence

• Fluid reasoning
• Highest/indistinguishable loading on g
• inductive and deductive reasoning
• Ravens Progressive Matrices developed to
  measure g, rooted in biology.

10
(No Transcript)
11
Information ProcessingWM and Fluid Reasoning

• Reasoning ability and g is working memory
• Kyllonen, (1996) Kyllonen, Christal, R. E.
  (1990)
• Working memory and fluid reasoning are associated
  (Engle, Tuholski, Laughlin, Conway, 1999)
• g on a working memory factor to be as high as .96
  (Colom, Rebollo, Palacios, Juan-Espinosa,
  Kyllonen, 2004)

12
Executive Function (EF)

• capacities that enable a person to engage
  successfully in independent, purposive,
  self-serving behavior (Lezak)
• Duncan- executive functions deficits (goal
  neglect) are characteristic of individuals low on
  g.
• Not much research examining g, FR, EF

13
(No Transcript)
14
Study EFFR

• Purpose
• Examine dimensions
• N70, undergrads
• WCST
• Problem solving, feedback
• Category Test (Halstead-Reitan)
• Guess categories
• Trail Making A B
• Fluid Reasoning (WJ-R)
• Concept Formation, Analysis Synthesis, Spatial
  Relations, Verbal Analogies

15
Trails B
16
Variables

• Executive functions
• WCST- total number errors
• Category-total errors in category
• Trails B- time (higher worse performance)
• Fluid Reasoning (WJ-R)
• Concept Formation-describe rule for group
  inclusion
• Analysis Synthesis-reasoning
• Analogies
• Prediction ( dimensions)

17
Factor Analysis 1 factor
Total Variance Explained
Extraction Method Principal Component Analysis.
18
(No Transcript)
19
Figure 3.
Note WCSTP Wisconsin Card Sorting
Test-Perseveration CT Category Test TRIALSB
Trail Making Test B VBAN Verbal Analysis
SPRL Spatial Relations CNCPT Concept
Formation ANSN Analysis Synthesis.
20
(No Transcript)
21
Fit Statistics
22
EF and Fluid Reasoning

• G, Gf, WM, EF all highly related
• Single factor or Two Factor correlations fit best
• Theories are needed to clarify this relationship
  and merge psychometric theories with cognitive
  theories (Miyake, Friedman, Rettinger, Shah,
  Hegarty, 2001).

23
Brain based research

• Frontal Lobes important for EF
• (Luria, 1973, 1980)
• EF deficits from non-frontal areas (Reitan
  Wolfson, 1994)
• Fluid Reasoning
• Left posterior activity on Ravens (Haier, White,
  Alkire, 2003)
• Dorsolateral and cingulate important (J. Duncan
  Owen, 2000)
• Ventromedial-emotional aspects of reasoning
  (Damasio, 1994)

24
Review

• All measures of cognitive ability have a common
  variance, g.
• Constructs of g, working memory, fluid reasoning,
  and executive functions are highly related.
• No consensus exists for defining any of the above
  mentioned constructs. However, there is
  overlapping characteristics of these different
  constructs in definition and measurement. It is
  assumed the shared definitions and measurement of
  these constructs accounts for the empirical
  association.
• No consensus exists for defining the information
  processing demands for the above constructs, but
  of these constructs working memory is best
  defined by information models followed by
  executive functions. Activation of
  representations that can be manipulated is core
  to most models.
• In general, tasks that are considered indicators
  of the constructs listed above, are highly
  related to the frontal lobes, specifically the
  left dorsolateral, of the brain but also on the
  interaction of the frontal lobes with other parts
  of the brain, specifically a left posterior
  region involved deemed important in guiding motor
  plans and attention.

25
Future Research

• Theoretical understanding of these constructs
  will come from brain based research, informed by
  evolutionary psychology

26
Help from primate models of Frontal-Parietal
Interaction (Fuster Petrides)
27
One other problem Content

• Theories must be related to adaptive problem
  solving
• Reasoning is influenced by information content
  (Cosmides, 1989)
• Evolutionary theories of social exchange and
  cheater detection (Wason task)
• Where does information content fit into
  intelligence? Gc?
• Unresolved

28
Problem Solving

• Initiation of frontal lobes
• Frontal lobes defines problems and urgency
• Parietal lobes-stored perceptions creates model
  (simulation vs. attention) or problem to guide
  motor behavior
• Frontal lobes evaluates/selects best simulation

29
(No Transcript)
30
(No Transcript)
31
(No Transcript)
32
(No Transcript)
33
Dorso-frontal and poly-sensory associative
parietal cortex areas
34
Re-defining intelligence

• Executive functions, working memory, fluid
  intelligence, g, are highly related and represent
  variations of intelligence
• G is the degree to which a task involves
  processes in model (selective attention,
  dorsofrontal to parietal interaction,
  evaluation).
• Global scores are meaningless (Lezak)
• Mixture of many different constructs
• IQ tests dont measure intelligence (Ardilia)
• Cognitive-perceptual simulation rather than just
  information processing

35
Assessment

• Further devaluation of FSIQ
• More specific definition of intelligence
  constructbetter measures
• IRT models for item level analysis to relate
  difficulty to brain regions
• Justify EF type measures in IQ tests (may clarify
  ability-ach discrep)
• Reduce processing type tests in IQ estimation
• More open ended type questions to reduce
  structure (Goldberg Damasio)
• New directions in social-emotional assessment

36
Assessment Outline

• Intelligence
• Top-down
• Fluid reasoning
• Working memory
• Executive functions
• Processing
• Bottom-up
• Primary Sensory
• Sensory
• Motor
• Auditory
• Visual
• Secondary Processing
• Motor (complex motor patterns)
• Auditory
• Visual
• Learning
• Rate
• Retention

37
Clinical Applications

• Disorder linked to functional deficit
• Separating anterior disorders (EF) from posterior
  disorders.
• Goal neglect-knowing what to do but not doing it.
• Internally directed attention (CPT, Trails B)
• Working Memory span
• Conceptual manipulation

38
Hierarchy of abilities Evolution of cognitive
abilities

• Stimulus-response (environment-motor)
• Associative Memory (stored representation)
• Self-initiated memory retrieval
• Cognitive simulation -goal-oriented active
  manipulation of perceptual representations to
  simulate behavioral alternatives (modeling
  social motives?)

39
Questions

• Contact info sdecker_at_gsu.edu