Metamemory Flashcards
Nelson & Narens 1990 - Object + Meta Level processes
- object and meta level processes within memory
- object level > memory, reasoning, decision making
- meta level > monitor object level processes and exert control
- use of regulatory feedback loop
e.g. how well have I remembered this
Yaniv & Meyer 1987 - Retrieval failures - TOT - Infrequent Words
- gave people definitions for infrequent words
- people struggle to remember words for these things
- e.g. in old greek stories the food of the gods giving mortality to those who ate it (ambrosia)
TOT states - James 1890
- richness of subjective experience around memory> know enough to know its not another word > aware its in your memory but unsure of what it is
Brown 1991 > review of TOT - tot people will often be able to tell you first letter of word
Hart 1965 - Recall judge recognise (RJR)
- 3 stage experiment
1) - general knowledge memory test > ppts asked to recall answers
2) focus on questions where ppts cannot recall answers > asked to make FOK judgement (binary or on graded rating)
3) all questions given recognition memory test > can ppt recognise correct answer given set of options
Hart 1965 - Recall judge recognise (RJR) FINDINGS
- some people can recognise answers they couldn’t recall
- wanted to see if accuracy in recognition task matched predictions from FOK judgements
- found there is a relationship between judgements of FOK and recognition but it is not perfect
- ## systematic relationship between graded FOK judgements and recognition > higher FOk ratings correlate with higher recognition
FOK Measure - Calibration
- calibration between FOK and objective memory performance
-overconfidence> FOK judgement exceeds objective performance - underconfidence > objective performance exceeds FOK judgement
- ## calibration is not perfect
Squire 2009 - Memory distinct from metamemory
- memory is a neural function distinct from other cognitive functions
- relies on interactions among multiple systems
- hippocampus + neocortex + Medial temporal lobe
FOK measure - Consistency
- measure relative accuracy of metacognition
- internal consistency of judgements > when somebody says they are more likely to remember the correct answer are they more likely
- scale does not matter
- looks at correlation rather than deviation from objective performance
- Good resolution > stronger FOK = more likely to recognise answer > Gamma correlation
-refers to the stability or reliability of individuals’ judgments of their own knowledge or memory
Difference between consistency and calibration
To illustrate the difference between consistency and calibration, consider the following scenario: A person is presented with a list of words and is asked to rate their FOK for each word. Subsequently, they are asked to recall or recognize those words. If the person consistently rates some words as “known” across multiple trials, their FOK judgments demonstrate consistency. However, if their FOK judgments align closely with their actual memory performance (i.e., accurate judgments), it indicates good calibration
FOK - Prospective judgement
- how likely are you to recognise the correct answer when you see it
-making a judgement now on how well you will do in the future - opposes retrospective judgement
Hart 1965 - Direct access hypothesis
- memory system comes with built in indicator
- FOK serves as indicator of what is stored in memory when retrieval temporarily unsuccessful/interrupted
- id indicator signals item not in storage, system will not expend effort to retrieve it
- if indicator signal it is in storage > system will avoid redundantly inputting information it already posses
Hart 1965 - Direct access hypothesis - Built in indicator
- memory system comes with built in indicator
- FOK serves as indicator of what is stored in memory when retrieval temporarily unsuccessful/interrupted
- id indicator signals item not in storage, system will not expend effort to retrieve it
- if indicator signal it is in storage > system will avoid redundantly inputting information it already posses
Challenges to idea of direct access
- FOK not perfectly reliable (Hart 1965)
- FOKS can be very fast (Kolers & Palef 1976)
- Conceptual challenges > how memories are stored > connectionist models of knowledge > network of connections > cannot look at synapses to see how memories are stored > we can only understand these networks through inputs and outputs (see how well it retrieves)
Koriat 1993 - Answer accessibility
- RJR study with consonant strings (e.g FHJT)
- ppts given consonant string to remember > asked to recall > if cannot recall asked to judge if they can recognise among set of answers
- retrieving fragments indicates that the information is there in memory as opposed to a blank
- found partial answers predicted FOK judgements
- FOK varies with number of items retrieved
- partial answers used as cue
Reder 1987 - question familiarity
- effects of practice and priming on FOK
- manipulated ppts feelings of familiarity of question independent of what they knew
- priming phase > asked to judge e.g. how often do you see the words ‘wine’ and ‘grapes’ paired in texts
- Game show task
- some questions will be primed from earlier phase but others not > manipulates people familiarity of question independent of what is in their memory
- does FOK reflect familiarity ?
- ## ppts decide whether they want to answer question or not (based on FOK rather than asking for direct FOK judgment)
