Memory 2 Flashcards
Learning outcomes from last year:
- Understand and describe the reasons why we do not have perfect recall, making brief reference to the Seven sins of memory (Chapter 7; Schacter, 1999)
- Explain how these quirks of memory may affect us in everyday life
- Explain how the evidence supports the theory that memory is reconstructive
- Understand and explain the methods used to test memory
Learning outcomes for this year:
- Discuss the relationship between Semantic and Episodic memory
- Evaluate the evidence that consolidation is a two stage process
- Describe effects that occur as a result of memory being malleable
- Evaluate the evidence in support of suggestions that memory is malleable
- Evaluate the theory that memory is organised to facilitate thinking about the future
Question…
If normal memory function requires information to be stored in episodic memory and then transferred to semantic memory can people without episodic memories consolidate new information into semantic memory?
Two-stage model and systems consolidation –> information is initially encoded in episodic memory and then transferred to semantic memory. Therefore, according to the 2 stage theory, if information is not consolidated into episodic memory it is unlikely to make it into semantic memory
However, Vargha-Khadem et al., (1997) –> evidence that some people without episodic memory can form semantic memories (school kids with episodic memory issues as a result of lesions can still learn schooling information)
Are these theories wrong or the evidence from Vargna-Khadem?
Reminder - Theories of memory from last week…
> Two-stage model of memory (McClelland et al., 1995)
- Computer model
> System Consolidation Hypothesis (Born & Wilhelm, 2012)
- Neurological model
Both state that information is stored in Hippocampal systems (episodic memory) & transferred to cortex over time
The two theories, their predictions and evidence…
***Theory: Two stage model
Prediction:
- Information that conflicts with existing information requires time to be consolidated (think penguin example of catastrophic interference)
- Cortical learning is possible for information that is consistent with existing knowledge
Evidence:
- Rapid cortical learning in the absence of episodic memory (Sharon et al., 2011) –> possible that memory system architecture in children becomes reorganised in order to store information without having episodic memory, Sharon suggests this is possible in adults also
- Schema consistent information is learnt more quickly than schema inconsistent information (Tse et al., 2007)
***Theory: System consolidation hypothesis
Predictions:
- Information will be gradually abstracted and transferred from hippocampal regions to long term cortical regions (so if you have overlapping episodic memories where the same information is being repeated then core info will be abstracted into semantic memory and consolidated over time)
Evidence:
- Hippocampal activation lower for distant compare to recent memories (Takashima et al., 2009) –> episodic memory stored in hippocampal systems so we would expect not to see hippocampal activation for abstracted semantic memories
Conclusions we can draw from the theories, predictions and evidence
So we have evidence that supports the predictions the theories make. Rather than saying that the evidence suggests the theories are wrong, we can say it suggests the theories need some readjustment to accommodate the evidence
- Semantic learning can occur in the absence of episodic encoding (but it’s not optimal)
- Episodic memory facilitates semantic memory
- Interleaving of new with existing memory facilitates learning (new schema consistent information can link to existing information in memory, can strengthen learning of that new info)
What can we remember from last year?
Kahoot quiz
- Forgetting may be the result of…?
Lack of consolidation - Absentmindedness might be due to…?
Lack of encoding - Lack of attention may affect…?
Ability to encode - Lack of attention during encoding has a greater impact on memory than during recall
True - Spreading activation is a process that activates…
Semantically related information - spreading activation predicts that if lots of associations are active…
Memory will be better - The war of the ghosts story is an example of what?
Memory being malleable - The results of the DRM paradigm demonstrate how…
We misattribute the source of memories - the formation of false memories - Research conducted by Loftus demonstrates that memory is…
- malleable - Which of the following sins of memory demonstrate that memory is reconstructive?
- bias, misattribution and suggestibility
Last week we looked at Kan et al’s results that demonstrated the impact of semantic memory on episodic recall.
Reworded…
How prior knowledge in a schema influences how your can recall an episode of experiencing that information…
Recap of exp. was a cognitive paradigm study using the prices of everyday items, congruent/ incongruent with irl prices –> observed the congruency effect - participants correctly ‘remember’ items that present the ‘true’ cost compared to items where the cost is ‘false’
Evidence from Kan et al suggests…
Declarative Memory
- Distinction between episodic & semantic memory
- Semantic memory appears to be more stable than Episodic memory (e.g., unaffected by lack of sleep)
- Episodic memory is malleable (e.g., affected by knowledge stored in semantic memory)
- Episodic memory requires semantic information to be bound with newly encoded information
Reconstructive memory
Binding of information between semantic and episodic memory in order to encode into episodic memory…
Reconstructive memory refers to the idea that retrieval of memories does not occur in some completely accurate form, as a video-recorder might replay a scene, but rather that recollection of memories is a process of trying to reconstruct (rather than replay) past events. The idea of reconstruction lends itself to error - of the 7 sins of memory 3 feed into this idea clearly;
- Bias
- Source misattribution
- Suggestibility e.g. Loftus
The process of combining aspects of memory results in misattribution of the source of the memory.
Reconstructive memory paradigm
DeMaster & Ghetti (2013)
(Information from semantic memory is bound with newly encoded information in episodic memory)
This exp. compared hippocampal activation during retrieval of episodic information in children aged 8 -11 against adults aged 18-25
Encoding:
“Remember the item and the colour of the border”
Test:
“What colour was the border for these items?”
Results:
Tested adults compared to children and presented evidence that the binding process is linked to the hippocampus and not fully developed in children
**THE THEORY
Episodic memory is constructed from existing information in semantic memory and newly encoded information in episodic memory
Episodic memory is therefore malleable (e.g. can be altered by new information)
What evidence is used to support DeMaster & Ghetti’s theory?
(Said: Episodic memory is constructed from existing information in semantic memory and newly encoded information in episodic memory)
Supplementary Exp:
Hupbach et al. (2008)
Hupbach et al. (2008)
Day 1 – All ppts
Name each of these objects in the basket
Day 1 – All ppts
Immediate recall of the objects
Day 2 – Condition 1
Consistent Experimenter
Same experimenter, different room
“Do you remember this basket? Can you describe the general procedure of what you did yesterday?”
Day 2 – Condition 2
Consistent Context
(different experimenter, same room)
“Do you remember this basket? Can you describe the general procedure of what you did yesterday?”
Day 2 – All ppts
New List items “name each of these objects”
(This is the information designed to update previous memory)
Day 2 – All ppts
immediate recall of new set of objects
Day 3 – All ppts
Recall as many objects as possible from day 1
- **Results:
- Having a consistent context is important, when it was consistent then lots of intrusions were present - resulted in episodic memory of first list was interfered with
- When the experimenter was the same there was little impact
***Conclusions - Suggested…
– Overlap in context (consistent) reactivates memory of original (day 1) list of words –> Reactivation makes the memory malleable, and so the malleable memory is modified by 2nd set (day 2) words
– Learning the second set of words (day 2) in a different room resulted in a new memory trace that was not mixed with the existing memory
Reconstructive Memory
Why is memory organised like this?
Suggestions
(Schachter & colleagues)
- Could be to facilitate up-dating memory (penguin example, if unable to update information would continue to conflict)
- The system needs to pull data from different aspects of memory to ‘imagine’ the future
Reconstructive Memory
Why is memory organised like this?
The Theory
“Memory is a reconstructive process to facilitate thinking about the future”
“Memory is a reconstructive process to facilitate thinking about the future”
Supporting Evidence
***Schacter & Addis 2007
- There is similar brain activity for remembering past & imagining future events
- Hippocampus involved in binding for both
***Addis, Wong & Schacter (2008)
- Participants recalled or imagined an event about which they generated as much detail as possible in 2 min
- Recordings of remembered and imagined events were transcribed
- Four events from each condition were scored based on the number of internal (episodic information) or external (non-episodic information) details provided.
- *Results:
- Similar pattern of response for thinking about the future vs thinking about the past (more internal episodic details than external non-episodic details for both past and future)
Evidence supporting Schacter, Wong and Addis
Remember they said
“There is similar brain activity for remembering past & imagining future events”
“Hippocampus involved in binding for both”
Basically saying that episodic memory is designed so we are able to think about the future, hippocampus involved in this
> The Evidence:
**Hassabis et al. (2007)
Hippocampal amnesic patients unable to imagine future events
**Irish & Piolino (2016)
Patients with semantic dementia with relatively preserved episodic memory are impaired when imagining future experiences (suggesting both episodic and semantic are involved in thinking about the future)
> The Conclusions:
- Evidence demonstrates that there are links between thinking about the past and the future that suggest these processes are similar
- Although episodic memory seems key to the process, there is also likely to be involvement from semantic memory
