Metacognition, D&M Ch 4 Notes -- "Feeling of Knowing and Tip of the Tongue States"

D&M Chapter 4 Notes -- Feelings of Knowing and Tip of the Tongue States

History -- early metacognition studies
Feeling of Knowing

• First metacognitive judgment subjected to rigorous experimental scrutiny
• Joseph Hart in 1965
• RJR method -- Recall -> Judge -> Recognition
• Recall: ask a trivia question
• Judge: If can't answer, ask for a FOK
• Recognition: Prompt with a multiple-choice

Recognition tests keep bothering me -- they seem ripe for distortion by the selection of answers. What if the answer is made obvious by the alternatives presented, or what if the answers are all so clever and close that they introduce new possibilities to the mind of the subject? For example, if we ask, 'Who was the first person to orbit the moon?', and the answers were ridiculous ("Joe Banana, Suzy Spaceship"), surely everyone would get it right -- and if the test was composed primarily of ludicrous answers, eventually their FOK would climb rather high in response. If the experimenters then switched to very difficult items, the FOK would suddenly appear overconfident (and so on). If the answers presented were all very plausible and some introduced new interpretations (such as including Russian cosmonauts as well as Americans, perhaps introducing confusion to anyone who had been thinking about the topic from an Americans-only point of view), or if each question was a close variation of the one that proceeded it (for example, if there's a series of questions about capital cities, a person might score more highly on later ones because they are increasingly dialed in and focused on that type of knowledge).

Theories about Feeling-of-Knowing FOK Judgments

Target Strength Account (historical theory) Proposed a 'strength' assessment -- if the strength of the target is higher than the recall threshold, you recall it, if the strength is below the recall but above the recognition, you recognize, if below both then you're just guessing or maybe seeing it triggers your memory??
Not well-supported in studies. Ex: definitions task from Yaniv and Meyer -- prompted to recall a word from a definition, and then to recognize word among non-words. But this is essentially too generalized, because recognizing a word among non-words might pull from your area knowledge, not your specific knowledge of that trivia question This is partially what I was wondering about above Conclusion: FOK is not directly targeted but rather more general to a topic. Heuristic-Based/Contemporary Theory Proposed an 'inference' assessment -- your FOK is based on familiarity with the cue. Example, you see a face, and you remember that face, but not the name to go with it. Or you think about a book title, and are trying to recall the author.
Two types of cues:

• familiarity with the domain (you read a lot)
• perceptual or conceptual familiarity with the cue (you know you've read the book, you've seen the person a lot)

Evidence seems to favor this hypothesis. Also, FOK judgements are made faster than other types.

Target Accessibility -- another heuristic basis for FOK judgments
Idea here is that the act of making an FOK may actually improve our overall recall -- you might pull up partial data, like a first initial, or an image, etc. The more you pull up, the more you infer that you will get it correct in a multiple-choice situation. This is similar to what I was thinking about how proximity dials you in -- although this is for a single question rather than domain knowledge

Which has a stronger effect on FOK -- cue familiarity, or accessibility?
Or, more importantly, how do people monitor the varying cues they receive from FOK activity?
Nelson, Gerler, Narens made a list of 12 potential influences on FOK -- social desirability, actuarial information (question looks easy), etc. Research needed!

Tip of the Tongue States

William James described this state of consciousness as "peculiar. There is a gap therein; but no mere gap. It is a gap that is intensely active."
We may have a blocker -- an incorrect alternative that keeps popping up even though it's wrong. Diary studies indicate the TOTs occur 1-2 times/week for younger adults, almost twice as often for older adults. Also possible to generate a blocker externally (experimenter suggests an incorrect answer), but may function differently. Incubating (come back later) is often evoked as a way to escape an insight problem.
Widely believed that blockers damage recall. But experiments with blocked and unblocked TOTs seem to be recalled at the same level, so perhaps blockers are not so blocking.
Experimentally tend to be assessed using rare words -- which successfully elicited TOT states. Participants might be able to recall some letters, number of syllables, etc. Seems to indicate that the TOT state generated this way is accurate -- they know it, but can't quite retrieve it.
Experiments seem to indicate a feeling of TOT is reliable -- more likely to actually retrieve it if you are TOT about it.

What causes TOT states? How can we explain the accuracy of TOT feelings? as compared to . . . FOK . . . maybe? Or relative to what?

TOT may give you complete access to conceptual meaning but not sound/phonemes. Lexical access is thought to entail two levels of processing -- Stage 1, semantic/syntax representation is accessed -- Stage 2, phonological representation is accessed. TOT speculated to occur because you can access the semantic/syntax level but not the phonological. This is thought to explain the higher rate of TOT in older adults. To test this experimentally, researchers tried prompting people with words that sound alike ("abstract" if the word looked-for is "abdicate"). Italian speakers were able to access word gender when in TOT (syntax). Cue familiarity also seems to stand up experimentally. If you retrieve anything, you infer you know it -- and the more you retrieve, the stronger your sense of knowing.

How to validate the inference from the accessibility theory? Let's play TOTimals! Idea: made-up animals, facts, pictures. Result 1: the more you recall (minus the wanted detail), the more TOT state you will report. Result 2: The more information on the TOTimal card, the more frequent the TOT state. Seems to validate the 'accessibility' theory.

Ideas on how to cure a TOT state:

• Make sure your hands are free, gestures may break it.
• Answer other kinds of questions about the topic
• Walk through letters of the alphabet
• Ask someone
• Don't get frustrated, TOT states are typically resolved

Brain Bases of FOK Judgments

• Some ideas that frontal lobe is significant here
• Amnesiacs have normal relative FOK -- low FOK, low recall
• By contrast, Korsakoff patients -- memory impaired and FOK is also impaired.
• Indications that the 'novelty' section of the brain is significant -- high novelty, low FOK

Functions of Feeling of Knowing

• Helps you decide how long to search your memory for the definition of an unknown word -- first, strategy selection (let me think, instead of looking it up), and then ongoing monitoring (hm, am I feeling like I'm on the right track?).
• Actually same speed to think if you know it, then answer, as if you just answer! So even those being asked to just answer were surely making some kind of FOK.

Since Hart's initial FOK work, many core questions under investigation -- why does FOK exist? What's the nature of it/how does it work? It's not Hart's original theory about the strength of your actual knowing, it's a series of heuristics.