effects of working memory training in and older adults · t1 t2 4 weeks 17.11.2014 owmw, claudia...
TRANSCRIPT
Effects of Working Memory Training in Young and Older Adults
Claudia von BastianClaudia von Bastian
von Bastian, Langer, Jäncke, & Oberauer (2013), Memory & Cognition, doi: 10.3758/s13421‐012‐0280‐7
Cognitive TrainingCognitive Training
17.11.2014 OWMW, Claudia von Bastian 2
Training and Transfer EffectsTransfer is the generalisation of improvements in a
Training and Transfer EffectsTransfer is the generalisation of improvements in a practiced tasks to other non‐practiced tasks. Transfer is expected, if practiced and non‐practiced tasks shareexpected, if practiced and non practiced tasks share underlying processes (functional overlap).
OtherOther abilities
17.11.2014 OWMW, Claudia von Bastian 3
«functional overlap»
Training and Transfer EffectsA basic cognitive ability which correlates positively with
Training and Transfer EffectsA basic cognitive ability which correlates positively with multiple other abilities would be an ideal candidate to induce broad transfer.induce broad transfer.
W kiWorking memory
17.11.2014 OWMW, Claudia von Bastian 4
Training and Transfer EffectsTraining and Transfer Effects
P bl l iWorking Memory Capacity (WMC)
Intelligence
Problem solving
…
V b l l i A iWMCVocabulary learning AttentionWMC
Reading comprehensionText comprehension
Storytelling
( )
17.11.2014 OWMW, Claudia von Bastian 5
Barrett, Tugade, & Engle (2004)
Training and Transfer EffectsTraining and Transfer Effects
T i iTraining
Working memory training tasks
Untrained (structurally different) working memory tasks
Near Transfer
working memory tasks
Different (but related)Far Transfer
Different (but related) cognitive ability (e.g., reasoning)
17.11.2014 6OWMW, Claudia von Bastian
Empirical EvidenceCan we get any smarter through training?
Empirical EvidenceCan we get any smarter through training?
“Yes, we can!”e.g., Borella, Carretti, Riboldi, & De Beni (2010); Jaeggi et al. (2008); Jaeggi et al. (2010); Jausovec & Jausovec (2012); Karbach & Kray (2009); Klingberg et al. (2005); Stepankova et al. (2014); …( ); p ( );
“No, we can‘t.”,
e.g., Chooi & Thompson (2012); Colom et al. (2010); Harrison et al. (2013), Holmes, Gathercole, & Dunning (2009); Nouchi et al. (2012); Owen et al. (2010); Redick, Shipstead, et al. (2013); …
17.11.2014 OWMW, Claudia von Bastian 7
Reasons for InconsistenciesMethodological issues
Reasons for InconsistenciesMethodological issues• Lack of theory‐driven selection of training tasksSi l i di t t iti biliti• Single indicators to measure cognitive abilities
• Lack of active control groups
Does working WM training enhance fluid cognitive abilities with a methodological sound design?
17.11.2014 8OWMW, Claudia von Bastian
cf. Shipstead, Redick, & Engle (2012)
Theory‐Driven Task SelectionFacet model of working memory capacity
Theory‐Driven Task SelectionFacet model of working memory capacity
Storage and ProcessingSimultaneous
maintenance andStorage and Processing maintenance and manipulation
WMC Relational Integration
Shifting
Oberauer et al (2000; 2003)
17.11.2014 OWMW, Claudia von Bastian 9
Oberauer et al. (2000; 2003)
Theory‐Driven Task SelectionStorage and Processing
Theory‐Driven Task SelectionStorage and ProcessingComplex Span
896629Recall numbers:
17.11.2014 OWMW, Claudia von Bastian 10
Theory‐Driven Task SelectionFacet model of working memory capacity
Theory‐Driven Task SelectionFacet model of working memory capacity
Storage and ProcessingStorage and Processing
Coordination of multipleWMC Relational Integration
Coordination of multiple elements into new
structures
Shifting
Oberauer et al (2000; 2003)
17.11.2014 OWMW, Claudia von Bastian 11
Oberauer et al. (2000; 2003)
Theory‐Driven Task SelectionRelational Integration
Theory‐Driven Task SelectionRelational IntegrationTower of Fame
A BC D
56
B d Pi li fl
234Tom Cruise lives in
apartment 2A.
Brad Pitt lives two floorsabove Tom Cruise in theapartment to the right
Tom Cruise lives in: ??? Brad Pitt lives in: ???
12 apartment to the right.
17.11.2014 OWMW, Claudia von Bastian 12
Theory‐Driven Task SelectionFacet model of working memory capacity
Theory‐Driven Task SelectionFacet model of working memory capacity
Storage and ProcessingStorage and Processing
WMC Relational Integration
ShiftingSelective activation of
relevant and inhibition of irrelevant representations
Oberauer et al (2000; 2003)
irrelevant representations
17.11.2014 OWMW, Claudia von Bastian 13
Oberauer et al. (2000; 2003)
Theory‐Driven Task SelectionShifting
Theory‐Driven Task SelectionShiftingTask switching
Green or blue?Green or blue?Round or angular?Round or angular?
17.11.2014 OWMW, Claudia von Bastian 14
Training WMC: Previous ResultsAre there distinct transfer effects of training different
Training WMC: Previous ResultsAre there distinct transfer effects of training different facets of WMC?
Storage and Processingg gComplex span
Relational Integration
Shifting
gCoordination tasks
gTask switching
Active Control
• N = 121 (M = 23 years)• 4 weeks, 20 sessions, each 30‐40 min.• randomized group assignment, double‐blinded
Visual matching tasks
von Bastian & Oberauer (2013)
g p g ,• adaptive task difficulty• experimental control via web‐based
monitoring with Tatool Online
17.11.2014 OWMW, Claudia von Bastian 15
von Bastian & Oberauer (2013)Tatool: von Bastian, Locher, & Ruflin (2013)
Training WMC: Previous ResultsAre there distinct transfer effects of training different
Training WMC: Previous ResultsAre there distinct transfer effects of training different facets of WMC?
Working Memory ↑g y↑Reasoning↑
No broad transfer
Shifting↑
(Binding ↑)
gReasoning↑
von Bastian & Oberauer (2013)
17.11.2014 OWMW, Claudia von Bastian 16
von Bastian & Oberauer (2013)Tatool: von Bastian, Locher, & Ruflin (2013)
Maximising Transfer?Maximising Transfer?
Storage and
Multifunctional TrainingWorking Memory ↑↑
↑Storage andProcessing Shifting↑
Binding ↑Reasoning↑Reasoning↑
i i i i iShiftingRelational
Integration
An intervention maximisingtransfer would be most usefulto counteract cognitive declineto counteract cognitive declinein older age.
17.11.2014 OWMW, Claudia von Bastian 17
Goal of the Present StudyInvestigating WM plasticity in young and old adults
Goal of the Present StudyInvestigating WM plasticity in young and old adults• Does training all three functional categories at oncelead to broad transfer effects?lead to broad transfer effects?
• Does training induce improved performance not onlyi b t l i ld d lt ?in young, but also in old adults?
17.11.2014 OWMW, Claudia von Bastian 18
DesignDesignPre Cognitive
performanceassessment Tr
aining Computer‐based
training Post Cognitive
performanceassessmentassessment
EEG (resting‐state, ERP)MRI and DTI
T assessmentEEG (resting‐state, ERP)MRI and DTI
Resting‐state fMRI Resting‐state fMRI
T1 T24 weeks
T1 T2
17.11.2014 OWMW, Claudia von Bastian 19
Method: Training InterventionMethod: Training Intervention
WM (young and old)Storage and processingRelational integrationRelational integration
Shifting
Active control (young and old)Visual searchVisual searchCounting
General knowledgen = 66 young (M = 23 years),
www tatool ch
y g ( y ),n = 57 older adults (M = 69 years)
17.11.2014 OWMW, Claudia von Bastian 20
www.tatool.ch
Tatool: von Bastian, Locher, & Ruflin (2013), Behavior Research Methods
Method: Active Control TrainingVisual search
Method: Active Control TrainingVisual search
17.11.2014 OWMW, Claudia von Bastian 21
Method: Active Control TrainingCounting
Method: Active Control TrainingCounting
333 22 666666 333 155555 2 4444 1 33355555 2 4444 1 3334444 1 666666
17.11.2014 OWMW, Claudia von Bastian 22
Method: Active Control TrainingGeneral Knowledge Quiz
Method: Active Control TrainingGeneral Knowledge Quiz
Which Australian animal feedsmostly on eucalyptus leaves?
(1) Koala(2) Duckbill platypus(3) Kangaroo(4) Dingo
17.11.2014 OWMW, Claudia von Bastian 23
Method: Cognitive AssessmentMethod: Cognitive Assessment
Training
NearTraining tasks
Intermediate
transferStructurally similar, but different material
transfer Same construct (working memory)n‐back and binding task
Diff t t t ( i )Far transfer
Different construct (reasoning)Raven‘s APM (both young and old)BOMAT (only young)
17.11.2014 OWMW, Claudia von Bastian 24
Results: WM Training GainsExample: training gains for storage and processing
Results: WM Training GainsExample: training gains for storage and processing
13Young Old
9
11
ficulty
7
vel of d
iff
3
5Le
11 5 10 15 20
Session
17.11.2014 OWMW, Claudia von Bastian 25
Session
Results: Active Control Training GainsExample: training gains for quiz
Results: Active Control Training GainsExample: training gains for quiz
12
14Young Old
10
12
ficulty
6
8
evel of d
iff
2
4
Le
01 5 10 15 20
Session
17.11.2014 OWMW, Claudia von Bastian 26
Session
Results: Cognitive AssessmentResults: Cognitive Assessment2.5
Young WM Young control Old WM Old control *** = p < .001
1 5
2.0
)
* = p < .050
1.0
1.5
ct size (d
)
0 0
0.5Effe
‐0.5
0.0***
* * * (*)******
***Storage & P i
Relational I t ti
Shifting Storage & P i
Relational I t ti
Shifting WM Bi di
WM N B k
ReasoningRAPM
ReasoningBOMAT
Far TransferIntermediate Transfer
Near Transfer
Processing Integration Processing Integration Binding N‐Back RAPM BOMAT
Training Tasks
17.11.2014 OWMW, Claudia von Bastian 27
Summary• Does training all three functional categories at once
SummaryDoes training all three functional categories at oncelead to broader transfer effects? No transfer was less broad! No, transfer was less broad!
• Does training induce improved performance not onlyin younger, but also in older adults? Yes, older participants improved as
much as younger ones, except formuch as younger ones, except fortransfer on binding.
17.11.2014 OWMW, Claudia von Bastian 28
Conclusion• Transfer effects are small, vary between individuals, and are
Conclusion, y ,
difficult to replicate
• Intervention‐specific features and individual differences may impact training and transfer gains
• Who profits from WM training under which circumstances?Focus shift toward maximizing transfer by examining… who gains most?… which cognitive mechanisms are prone to change?
17.11.2014 29OWMW, Claudia von Bastian
A Roadmap for Training ResearchA Roadmap for Training ResearchINDIVIDUAL DIFFERENCESTRAINING
CONDITIONS INITIAL
TASKS BIOLOGICAL FACTORSAGE
PERSONALITY & CONDITIONS COGNITIVE ABILITY MOTIVATION
TRANSFERa) enhanced capacityb) h d ffi i
TRAINING PROGRESS b) enhanced efficiencyPROGRESS
OBSERVED EFFECTStransfer effects + noise
17.11.2014 OWMW, Claudia von Bastian 30von Bastian & Oberauer (2014), Psychological Research
Outlook: Ongoing WorkOutlook: Ongoing Work
C l D Si iCarla De SimoniPhD student at the Department of Psychology, Cognitive Psychology Unit, University of Zurich
Differential malleability of working memory updating processes: retrieval, y g y p g p ,transformation, and substitution of information elements
Sabrina GuyePhD student at the URPP “Dynamics of Healthy Aging”, University of Zurich
Predictors of working memory training success in older adults: Cognitive abilities, personality, affect, and everyday‐life activities
17.11.2014 OWMW, Claudia von Bastian 31
Thanks for Listening!Thanks for Listening!PhD Advisor and CollaboratorsPhD Advisor and Collaborators
Klaus Oberauer, Nicki Langer, Lutz Jäncke
Software Development André Locher and Michael Ruflin
Research AssistantsSabrina Guye, Veronica Heusser, Melanie Künzli, Eszter Montvai, Katharina Vogt, and Marc ZüstSabrina Guye, Veronica Heusser, Melanie Kün li, s ter Montvai, Katharina Vogt, and Marc üst
Funding SourcesgForschungskredit of the University of Zurich, Switzerland
Suzanne and Hans Biäsch Foundation for Applied Psychology
17.11.2014 32OWMW, Claudia von Bastian