Working memory, education, disability and clubbing

1
Working memory, education, disability and
clubbing

• Lorna Bourke

2
Investigating Memory

• Scientific Method
• science is attempting to understand and
  represent nature to understand human memory
  (Baddeley, 1994 p8).
• Theoretical Models
• Empirical testing
• Manipulate and control variables (extraneous).
• Observational techniques.
• Are they less scientific?
• Why are models important?
• Can we produce an overall model or theory of
  memory?

3
Historical Context of Human Memory Theory and
Research
Atkinson and Shiffrin (1968)
An Assembly Line?
4
Historical Context of Human Memory Theory and
Research

• Earlier models
• Atkinson and Shiffrin (1968) Modal Model
• Two component model of memory.
• Information feeds from one stage to another.
• Short-Term Memory plays a crucial role.
• Communicates with Long-Term memory
• Control Processes
• Rehearsal (Rundus, 1971)
• Problems with the Modal Model
• Neuropsychological Evidence (Shallice
  Warrington, 1970)
• Short-Term Storage and Long-Term Learning
  (Tulving, 1966 Morton, 1967 Bekerian and
  Baddeley, 1980).
• Recency and STS (Bjork Whitten, 1974 Baddeley
  Hitch, 1977)
• Coding (phonological and semantic) (Baddeley
  Levy, 1971)

5
Historical Context of Human Memory Theory and
Research

• New approaches
• Craik and Lockhart (1972) Levels of Processing
• Concentrated on mode of processing
• Shallow Deep Processing
• Rehearsal (maintenance elaborative)
• Short-Term Memory (Primary Memory)
• Separable component dichotomous view.
• Role of Coding in Long-Term Memory.

6
Working memory

• What would a working memory be required for?

7
Working MemoryBaddeley and Hitch (1974)

• Dual-Task Paradigm
• Tests the assumption that the short-term store
  acts as a temporary working memory.
• Task requires the participant to perform one task
  that absorbs most of the wm capacity.
• At the same time participant is required to
  perform another task thought to be dependent on
  wm (e.g. comprehension, reading, reasoning).
• Performing a concurrent STM task should lead to
  dramatic impairment in performance.

8
Dual-Task ExperimentConcurrent Digit Span
Reasoning

• 5-8-1-3-6-2-9

9
Dual-Task Experiment ResultsBaddeley (1986)

• Reasoning time increases with concurrent memory
  load.
• Effect is not striking, therefore possibility
  that working memory does not involve a unitary
  store whose limited capacity is likely to be
  totally absorbed when the limit of memory span is
  reached.

10
Working Memory ModelBaddeley and Hitch (1974,
1983, 1986)

Central Executive

Episodic Buffer
Visuo-Spatial Sketchpad
Phonological Loop
Visual Episodic
Language Semantics
LTM
Baddeley (2000) reports the addition of an
additional component Episodic Buffer
11
Phonological Loop

• Structure
• Phonological Store
• Articulatory Control Process
• Evidence
• Phonological Store
• Phonological Similarity Effect
• Conrad and hull (1964 Baddeley, 1966)
• PGVTVCD harder to remember than RHXKWY
• Store is based on a phonological code
• Unattended Speech Effect
• (Colle Welsh, 1976 Salame Baddeley, 1987,
  1989)
• Condition A Repeat visually presented digits
• Condition B Repeat visually presented digits
  background of spoken words
• NB. Not every noise causes this effect (Baddeley,
  Eldridge Lewis, 1981 Martin, Wolgalter
  Forlano, 1988 Salame Baddeley, 1989)

12
Phonological Loop

• Evidence
• Articulatory Control Process
• Word Length Effect
• Monosyllabic Words vs Polysyllabic Words
• wit-sum-harm-bag-top
• vs
• University-opportunity-aluminium-constitutional-au
  ditorium

13
Phonological LoopBaddeley, Thomson and Buchanan
(1975)
14
Phonological LoopWord Length Effect
Spoken Duration or No. of Syllables Ellis and
Hennelly (1980) Welsh Speaking Children
15
More EvidenceArticulatory Control Process

• Articulatory Suppression
• Operation of the Phonological Loop is disturbed
  if overt or covert articulation of an irrelevant
  item is required.
• Digit Span Utter Irrelevant Sound
• 5-9-8-2-1-4-3 the-the-the-the-the

16
Articulatory Control Process

• Does suppression impair performance because it
  demands attention (e.g. Parkin, 1988)
• Arguments against
• Non-articulatory secondary tasks (e.g. tapping)
  little or no effect on STM performance (e.g.
  Baddeley, Lewis, Vallar, 1984)
• PV doesnt use articulatory loop in STM tasks
  memory performance not impaired by suppression
  (Vallar Baddeley, 1984)
• Suppression specifically affects phonological and
  articulatory coding (Baddeley, et al., 1984)
• Cowan (1992)
• Word Length Effect may be due to delay in recall
  due to longer time taken to produce the long word
  sequences at recall.

17
Visual-Spatial Sketchpad

• Evidence
• Separable slave system for images - Brooks (1967)

18
Visuo-Spatial Sketchpad

• Evidence
• VSSP Two Components (Logie, 1995)
• Visual Cache
• Stores information about visual form and colour
• Inner Scribe
• Rehearses information in visual cache
• Transfers information from visual cache to
  Central Executive
• Beschin, Cocchini, Della Sala and Logie (1997)
• NL (damage to visual cache only had
  impoverished mental representations)

19
Central ExecutiveBaddeley (1996) Baddeley
Logie (1999)

• Mechanism for control processes in working
  memory
• Coordinates activity in the slave systems
• Controls encoding and retrieval strategies
• Switches attention
• Mental manipulation of material in the slave
  systems.
• Does not have a storage capacity

20
Central ExecutiveBaddeley (1996) Baddeley
Logie (1999)

• Functions of the Central Executive
• Ability to sustain attention
• Hasher Zachs (1988) Robertson, Manly, Andrade,
  Baddeley Yiend (1997)
• Inhibit distracting stimuli and concentrate on
  the goals relating to the task.
• Dual-Task Performance
• Spinnler, Della Sala, Bandera and Baddeley
  (1988) Baddeley, Della Sala, Papagno Spinnler,
  1997.
• Capacity to coordinate information from the two
  slave systems.
• Verbal Fluency
• Baddeley (1996)
• Ability to search and retrieve information from
  Long-Term Memory
• NB Baddeley (1996) Baddeley et al (1999) do
  not suggest these are the only functions that
  could be performed by CE.

21
How do we use working memory?

• What is the difference in size between the UK and
  USA?
• Retrieve information from long-term memory (CE)
• Use CE to co-ordinate activity between VSSP and
  Phonological Loop
• Use VSSP to store relevant visual images as well
  as mental manipulation in comparing sizes.
• Use phonological loop to store verbal information
  (visual images recoded for verbal output as well
  as verbal internal representations) prior to
  output (speaking or writing)

22
Daneman and Carpenter (1980)

• Provided an alternative conceptualisation of the
  cognitive mechanism underpinning working memory.
• Integrated perspective
• Capacity of WM dependent on the efficacy with
  which people store and process information.
• If more resources are used to process information
  then less available for storage.
• Developed method of testing this assumption
• Reading Span Task (Complex Span)
• Traditional digit span tasks not always/less
  correlated with comprehension.

23
Engle, Kane, Tuholski (1999)
24
Alternative ExplanationsTowse, Hitch Hutton
(1998)

• Integrated Models
• Trade Off Notion
• May be due to time delay increased delay
  associated with storage decrements (Towse, Hitch
  Hutton, 1998)

25
Neuroanatomical Evidence
26
Applications of Working Memory

• Reading
• Spoken Language
• Comprehension
• Writing
• Dyslexia
• Mathematics
• Downs Syndrome
• Ecstasy

27
Working memory and spoken language

• Vocabulary acquisition
• Comprehension
• Length and complexity of utterances
• Adams Gathercole (1994, 1995, 2000)Adams,
  Bourke Willis (1999)Oakhill, Yuill Parkin
  (1988)

28
Working memory and writingBourke Adams (2003)
29
I play with my friends. I open the door. I played
out. I played hopscotch. I played with my
friends.
The boy and the cat went to the park and they
climbed up a tree.
I took my cat for a walk. He was good. When we
got home my cat ran up the stairs. I ran after
her. Then I went into the garden. I and my cat
went on the swing. I went up in the sky. So did
my cat. Mum shouted at the cat.
4-5 years
30
Like Ben, many children find independent writing
a struggle because they are faced with too many
hard things to do at once. He has to plan what he
will write, think of which words to choose and
how to order them into sentences, work out the
spellings for each and transcribe them all onto
the page. For experienced writers, much of this
is automatic, and only occasionally requires
conscious control. There are parallels with other
complex skills. Experienced drivers do not need
to think about the controls, which simply become
an extension of themselves, leaving them free to
concentrate on where they are going. The learner
driver often needs someone else to take care of
the pedals, so he or she can concentrate on the
steering and vice versa. (DfEE, 2001 p.10)
31
Assessments

• Nonverbal Cognitive Ability
• WISC (Block Design) (M 13 SD 7.8)
• Working Memory Assessments
• Fractionated Model (Baddeley, 1986 1996)
• Phonological Memory
• Nonword Repetition (M 33.13 SD 5.43) rubid
  diller
• Word Span (M 3.77 SD 0.85)
  pig-duck-bag-sun
• Digit Span (M 5.03 SD 0.99)
  3-9-4-5
• Visual Memory
• Corsi Blocks (M 3.60 SD 0.90)
• Visual Pattern Span (M 3.28 SD 0.80)

32
Assessments

• Central Executive
• Dual-Task Co-ordination 2-9-3-5
• (M 85.87 SD 21.30)
• Verbal Fluency
• (M 9.12 SD 2.08)
• Sustained Attention to Response
• (M 2.73 SD 1.39)

tortoise
dog
33
Alternative cognitive explanations

• Vocabulary knowledge
• Speed of processing
• Phonological awareness

34
Downs Syndrome

• Features of cognitive impairment
• Poor short-term memory (Mackenzie Hulme, 1987)
• Poor phonological memory compared to visual
  memory (Marcell Armstrong, 1982)
• Poor articulation skills (Dodd, 1976 Gibson,
  1987)
• What does this mean for WM?
• Suggests that phonological loop component of
  working memory may be impaired

35
Children with Downs Syndrome Information Sheet
Children with Downs Syndrome - Information Sheet

• Short-term auditory memory
• How does short-term auditory memory work?
• Is there a relationship between Downs Syndrome
  and working memory?
• Why do many children with Downs Syndrome have
  poor working memory?

36
Ecstasy UseWareing, Fisk Murphy (2000)
http//www.erowid.org/references/refs_view.php?AS
howDoc1ID409
BBC News HEALTH Ecstasy 'damages memory'
37
Ecstasy UseWareing, Fisk Murphy (2000)

• Ecstasy Amphetamine based stimulant
• Immediate effects elation and enhanced
  subjective arousal.
• Long-term effects mood, sleep appetite
  disturbance

38
Ecstasy UseWareing, Fisk Murphy (2000)

• Morgan (1999) found memory deficits affecting
  both immediate and delayed recall.
• Previous research focused on episodic memory
  recall.
• Wareing, Fisk Murphy (2000) aimed to establish
  differences in users and previous users on
  working memory functioning ( information
  processing speed, anxiety arousal).

39
Ecstasy UseWareing, Fisk Murphy (2000)

• Participants
• 30 individuals (10 non-users, 10 current users,
  10 previous users)
• Equal numbers of males and females
• Measures
• Word Span
• Brooks Spatial Matrix
• Visual Memory Task
• Verbal Fluency
• Random Generation Task
• Information Processing Speed
• State anxiety and arousal questionnaire

40
Ecstasy UseWareing, Fisk Murphy (2000)

• Results
• Random Generation Score
• 1-s rate letters produced
• Previous users (M 59.50 SD 24.09)
• Current users (M 65.20 SD 23.69)
• Non-users (M 94.40 SD 7.73)
• Significant plt.01 (F2, 25) 8.57
• Information processing speed
• Number of stimuli classified
• Three letter
• Previous users (M 20.00 SD 2.00)
• Current users (M 18.90 SD 5.61)
• Non-users (M 22.7 SD 3.02)
• Not significant pgt.05 F (2, 25) 2.57
• Nine letter
• Significant plt.01 F (2, 25) 15.94

41
Conclusions

• Users generated fewer letters
• Consistent with Morgans (1998) observation
• Previous users are as affected as current users
• Suggesting that the effects are long-term
• Criticisms
• Confounding variables (Wareing et al, 2000)
• What else?
• Validity of working memory tests
• Test retest reliability not reported
• Not all central executive measures explain
  differences (verbal fluency) what does this
  mean?
• Findings need to be treated with caution
• Good points
• Indicates that further research on central
  executive and ecstasy is appropriate

42
Lecture Summary

• It appears that the relatively simple model
  proposed by Baddeley and Hitch (1974) can account
  for neuropsychological evidence
• Models that were previously proposed as
  representing a dichotomy within working memory
  research are converging, particularly, when
  considering the functions of the central
  executive
• Although there is a convergence in theoretical
  conceptualsiation, the tasks used to carry out
  research vary
• Working memory is currently applied within a
  number of domains and is now becoming encompassed
  within remediation frameworks in educational
  settings