Episodic and Semantic Memory Flashcards
episodic memory
memory for specific events located at a specific point in time
- mental time travel
- backward to relive earlier episodes
-forward to anticipate and plan future events
semantic memory
memory for facts
- no mental time travel
- e.g., world knowledge; vocabulary;rules etc
- short delay: information is recalled in episodes
- long delay: the same information is integrated into semantic memory
amnesia and memory type study
Spiers, Maguire, and Burgess (2001) - 147 cases of amnesia
- substantial or even dramatic loss of episodic memory
- semantic memory effects more variable and generally smaller
- damage to the hippocampus (and the MTL) affects episodic memory far more than semantic memory
- BUT: Hippocampal amnesia may affect acquisition of new semantic memories more than retrieval of old ones (Clark & Maguire, 2016)
Semantic dementia patients
- severe loss of concept knowledge but intact episodic memory (and intact cognitive abilities)
- damage anterior frontal and anterior temporal lobes
areas of the brain that contribute to semantic deficit
- anterior frontal lobe
- anterior temporal lobe
areas of the brain that contribute to episodic deficit
- amygdala
- hippocampus
- entorhinal cortex
- perirhinal cortex
- parahippocampal cortex
episodic vs semantic memory
- independent systems
- many long term memories comprise a mixture of episodic and semantic aspects
- they dynamically interact and affect each other
Schemas
- structured representation of knowledge about the world, events, people or actions
- can be used to make sense of new material, to store and later recall them
- are influenced/determined by social and cultural factors
Bartlett: Meaning and Schemas
- recall of complex materials
- examined recall errors
- unlike Ebbinghaus, he stressed participants’ effort after meaning
Native American folk tales:
- people commited many errors and distortions when asked to recall
- made the story more coherent and omitted details
- distortions more consistent with their own semantic knowledge
- ‘westernised’
- criticism: vague instructions
Role of Schemas: Sulin and Dooling, 1974
story about dictator - “Gerald Martin” (unknown) or “Adolf Hitler”
- test sentence: “He hated the Jews..”
- short delay (5 mins): no difference between the groups
- long delay (1 week): participants who read about Hitler were more likley to incorrectly agree
schematic knowledge may affect memory especially at longer intervals
role of meaning
ascribing to meaning to stimuli affect encoding and storage
Carmichael et al., 1932
- used shapes that looked similar to words to prove this
Paivio’s Dual-coding hypothesis
more imageable words are more memorable
high imageability (2 routes) : visual appearance + verbal meaning (apple)
low imageability (1 route) : visual appearance OR verbal meaning (hope)
multiple encoding routes improve the chance of successful recall
Levels of processing hypothesis: Craik & Lockhart (1972)
Why does meaning facilitate LT memory?
multiple levels of processing from input to LTM (hierarchical)
- visual (structure)- least deep processing
- phonological (acoustic)
- semantic (meaning) - deepest processing
deeper is not always more memorable
levels are processed simultaneously
Levels of processing task: Craik & Tulving, 1975
task:
words studied and participants asked to make 3 judgements:
- visual processing (e.g. ‘is TABLE in upper case?’ Y/N)
- phonological (e.g. ‘does DOG rhyme with LOG’ Y/N)
- semantic (e.g., ‘does … fit in the sentence …’ Y/N)
Test: recognise the words (old and new)
- deep processing - better recognition - particularly for YES responses (semantic best)
replicated in numerous studies
affects both recognition and recall
Transfer-appropriate processing (TAP)
Memory retrieval is best when the cues available at testing are similar to those available at encoding
can explain levels of processing effect
Transfer appropriate processing study: Morris, Bransford, and Franks (1977)
task:
- incidental learning: participants were not told that they would be tested later
- phonological or semantic judgements about words
test:
- standard recognition test for the encoded words
- rhyming recognition test for the encoded words
results:
- standard test: same as LOP theory
- rhyming recognition test: Phonological led to better performance
conc:
- learning more efficient when tested the same way was learned
why is deeper coding better?
- Craik and Tulving 1975: Richer and more elaborate encoding leads to better memory
- elaborative rehearsal enhances delayed long-term learning more than maintenance rehearsal
maintenance rehearsal: as something was learned
elaborative rehearsal: linking it to other material
Hierarchical organization
- Bower et al. 1969: recall is better when words are organised than when presented in scrambled order
- Tulving, 1962: memory is benefited by subjective organization - chunking together separate words for recall, even if not encoded together
items often chunked together if they are: - linked to a common associate
- come from the same semantic category
- form a logical hierarchical structure or matrix
Summary of factors that aid encoding
- create connections (imagery, meaning)
- organisation (recall by groups, present in an organised way)
- LOP/TAP (deeper processing, similar encoding - retrieval procedures)
- active creation (generate, test)
Hierarchical network model (Collins & Quallian, 1969)
Semantic memory organised into a series of hierarchical networks
- major concepts are represented as nodes
- properties/features are associated with each concept
cognitive economy: properties are stored higher up to minimise redundancy
this model was verified using the sentence verification task:
- “decide as quickly as possible whether sentences are true or false”
problems with the hierarchical network model
familiarity:
- “a canary has skin” is not a familiar sentence
- when controlled reduces the hierarchical distance effect
typicality:
- verification is faster for more representative member c ategories, independent of hierarchical/semantic distance
spreading activation model: Collins and Loftus, 1975
- semantic memory is organised by semantic relatedness
- length of links indicates the degree of semantic relatedness
- activity at one node causes activation at other nodes via links
- spreading activation decreases as it gets further away from the original point of activation
support for the spreading activation model
semantic priming tasks - McNamara, 1992
- presenting one stimulus that is semantically related makes subsequent processing faster
Deese-Roediger-McDermott (DRM) paradigm
- activation spreads from studied words to related words
spreading activation model: evaluation
it is more flexible than hierarchical network model
- can account for more empirical findings
- reduces specificity of model’s predictions
- more difficult to test
limitations:
- oversimplified (each concept represented by single node)
- does each concept have a fixed mental representation?
- no consensus on best way to measure semantic distance
situated simulation theory
- concepts are processed in different settings
- their processing is influenced by the current context/setting
- concepts incorporate perceptual properties and motor- or action-related properties
e.g., activated aspects of “bicycle” concept reflect current goals (Barsalou, 2009)
situated simulation theory: evaluation
processing of concepts depends on the situation and the perceptual + motor processes in a given task
limitations:
- how variable are concepts (stable core + context-dependent elements) across situations?
- are these properties secondary - after concept meaning has been accessed?
