Learning to learn

Question 1

What are metacognitive illusions?

Answer 1

misalignment between subjective assessment of how learning works (metacognition). Belief affects behaviour

e.g. thinks strategy A is better than strategy B

Question 2

Causes of metacognitive illusions?

Answer 2

most illusions traced to a fluency heuristic - easy/fluent learning leads to good memory and durable learning

Carpenter et al, 2013 - it makes so much sense, I don’t need to study them, I will remember it forever

Question 3

Reality of learning

Answer 3

desirable difficulties are useful for enduring learning- Bjork, 1994
Struggling is good and forgetting can be helpful to retain information. Surmountable obstacles are useful for durable memories

fluency heuristic and desirable difficulties are incompatible

Question 4

Carpenter et al, 2013

Answer 4

teaching evaluations and learning
-condition 1: fluent engaging
-condition 2: disfluent engaging
after the lecture they were asked to predict their performance on a test
- fluency breeds overconfidence
- fluent speaker was thought to of being able to learn more than disfluent speaker
- fluent speaker was rated higher in qualities such as organised, knowledgeable, prepared and effective.

Question 5

Yunker and Yunker, 2003

Answer 5

Higher ratings in introductory accounting were negatively correlated with grades in intermediate accounting

year 1 instructor compared to year 2 - if they liked the instructor would do better in year 2

popular instructors may adopt fluent teaching style - enjoyable for students but not better for learning.

Question 6

cause and effect issues

Answer 6

difficult to ascertain a relationship between rating and learning because of the correlation/relationship

some instructors may inflate marks to achieve higher ratings (Stroebe, 2016)

Question 7

active learning - Deslauriers, L et al, 2019

Answer 7

all students given handouts with the physics problems on the given topic

passive condition: instructors presented the slides that explained the concepts and solved the problems on the handout. Students completed the problems whilst listening

Active condition: students divided into small groups and worked on problems together and the instructor answered questions and showed solutions at the end

all students took a test at the end

student also asked which instructor they thought they had learnt more from - felt it was the passive instructor - metacognitive illusion

Question 8

e-Learning - Higham, P & Coria, K, 2018

Answer 8

40 minute lecture (10 x 4 min segments)
five groups:
- control (passive observation)
- annotation (lecture slide handout)
- regular notes (take notes on blank paper)
- key points (3 key points from previous lecture segment during lecture pauses)
- verbatim notes (copy everything off slides)

an immediate test and a test a week later
immediate test results (best-worst)
key points, verbatim, regular, annotation and control

delayed test results (best-worst)
key points, regular, verbatim, control, annotation

most forgetting: verbatim, annotation and control

least forgetting; regular, key points

no revision in any of the conditions

Question 9

Revision activities

Answer 9

compared control, annotation and regular note taking groups
40 minute lecture on MRSA
tested after 8 weeks
revise from copies of lecture handouts for as long as wanted
accuracy was similar across three conditions
time spent revising control, annotation and note taking
control spent longer trying to revise the content, longhand note taking meant students needed less time to revise to achieve the same mark

Question 10

Spaced learning Emeny W.G., Hartwig M.K., & Rohrer, D, 2021

Answer 10

spaced condition:
- tutorial - 4 problems - 1 week - 4 problems - 1 week - 4 problems, prediction 1 - 4 weeks - prediction 2 - test - post diction

massed condition:
tutorial - 12 problems - prediction 1 - 4 weeks - prediction 2 - test - post diction

spaced practice was better, but students believed that massed practice was better, at least at first when fluency was high.

Question 11

Kornell, N., & Bjork, R.A (2008)

Answer 11

Blocked units vs interleaved units

blocked learning - all one artist
Interleaved learning - all different artists

training: shown paintings with artists’ names below; blocked vs interleaved

test: attempt to classify new paintings by same painters (4 blocks)

more artists were remembers in the interleaving conditions

Question 12

Overall recommendations for Metacognitive learning and desirable difficulties

Answer 12

understand learner intuitions about learning are often wrong

fluency feels like it predicts learning but doesn’t

the better predictor of good learning is whether the process induced desirable difficulties

struggling during learning produces most durable learning

learner satisfaction doesn’t always mean good learning

Question 13

Backwards testing effect

Answer 13

taking a test on previously studied materials improves retrieval of the same information on a later test, usually relative to a restudy condition

the observation that retrieval practice produces superior memory to restudy

Question 13

Karpicke et al, 2009 - BTE

Answer 13

college students N=117,
students reported all the strategies that they used, and ranked them in order of frequency
84% said re-reading notes
55% ranked re-reading notes as top strategy
11% self tested
1% ranked self testing as the top strategy

Question 14

Kornell and Bjork, 2007 - BTE

Answer 14

asked ppts “if you quiz yourself while you study, why do you do so?

• to know how well I have learned the information I am studying
• to learn more than I would through re-reading
• I find quizzing more enjoyable than re-reading
  I usually do not quiz myself

Question 15

A trendy research area

Answer 15

Backwards testing effect has had more research from 1998-2018

Question 16

simple memory model

Answer 16

encoding - storage - retrieval

encoding- initial learning phase where ppts study information

storage- we can look at how long the information is stored for by manipulating the retention interval

retrieval- final test where ppts are asked to retrieve the information from the initial learning phase.

Question 17

Roediger & Karpicke, 2006 - BTE

Answer 17

phase 1- encoding - ppts studied two prose passages, they studied one passage twice and the other passage once and then completed an initial free recall test
phase 2- final test- ppts completed final free recall tests for each passage either 5 mins, 2 days or 1 week later

the time between the encoding and final test is called the retention interval

results;
after 5 days, the test group recalled as much as the restudy group did after 2 days
if we want to remember information for longer than 5 min, test-taking is the better strategy

Question 18

transfer appropriate processing (TAP) - BTE

Answer 18

final test performance should be better when cognitive processes required at encoding and retrieval are the same/similar

prediction - magnitude of testing effect should be greater when the initial and final tests have the same format

Question 19

Retrieval effort hypothesis - BTE

Answer 19

retrieval is more difficult than passive study
retrieval practice serves as a desirable difficulty
prediction- the greater the retrieval effort, the greater the benefit of testing:
- more difficult initial test should produce larger effects than easier tests
- retrieving after a long delay should produces larger effects than retrieving after short delays

Question 20

Mediator effectiveness hypothesis - BTE

Answer 20

• prediction: people activate related information (mediators) during initial test
• semantic elaboration during initial test boosts subsequent memory

Question 21

McDaniel et al, 2007 - BTE

Answer 21

• students took weekly quizzes or given information to read only
• quizzes were multiple choice or short answer questions with feedback
• quizzing but not restyling enhanced final test performance relative to control material
• short answer quizzes were more beneficial than mcq

Question 22

Agarwal et al, 2021 - BTE

Answer 22

• metanalysis of 50 studies
• examined a wide range of studies - different education levels, content area, number of retrieval practice opportunities, timing of retrieval practice
• 94% of studies revealed positive effects of retrieval practice
• majority of effect sizes (57%) were medium or large
• note potential for publication bias but overall results are promising

Question 23

Kromman et al, 2009 - BTE

Answer 23

• medical students completed a resuscitation course following stimulated cardiac arrest
• intervention group - 3.5 hours of teaching and 30 mins low-stakes testing
• control group - 3.5 hours of teaching and 30 mins of scenarios
• 2 weeks later- final practice test of learning outcomes
  and final test performance is better in the intervention group

Question 24

Stress impairs memory - Smith et al, 2016 - BTE

Answer 24

• ppts studied a list of words and images
• restudy or free-recall retrieval practice (no feedback) x 3
• 24 hours later- stress induction (answering maths questions in front of a crowd) OR non-stressful control task solving maths questions alone)
• memory for half the material tested 5 mins into stress/control task
• memory tested 20 mins later

results;
- overall retrieval practice enhanced recall
- study group stress impaired recall
- retrieval practice group., similar recall for stressed and non-stressed ppts

Question 25

Tests reduce anxiety - BTE

Answer 25

71% of undergrads and postgrads feel anxious about their classes and schoolwork

72% of precollege students said retrieval practice made them less anxious

Question 26

forwards testing effect

Answer 26

the phenomenon in which taking a test on previously studied material improves learning of new information, usually relative to restudy control condition

Question 27

spunzar er al, 2008 FTE

Question 28

encoding vs retrieval mechanisms

Answer 28

encoding - storage - retrieval

not mutually exclusive

Question 29

release from proactive interference

Answer 29

• test provides a change in context that restudy does not
• context change allows ppts to segregate the lists, and discriminate between them, more effectively
• change in context prevents build up of proactive interference (PI) form one list or another

Question 30

spunzar et al, 2013

Question 31

quality mind-wandering - Jing et al, 2016

Question 32

traditional learning theorists - learning from errors

Answer 32

• thought errors were bad
• if someone wants to avoid errors on a test that counts, they should avoid errors during the learning process
• erroneous answers will become associated with the question and then be repeated
• lead to influential errorless learning approaches such as step by step guidance, positive reinforcement for correct answers and incorrect answers ignored

Question 33

proactive interference

Answer 33

when old memories interfere with the acquisition and retrieval of newer memories
errors might interfere with memory for the correct answer
would expect errors to be repeated on the final test

Question 34

an alternative perspective to learning from errors

Answer 34

carefully controlled experiments from cognitive psychology challenge the errorless learning view

engaging in errors can promote active and deep learning

feedback is crucial

Question 35

the generation effect

Answer 35

a phenomenon in which people typically remember information that they have generated themselves better than information generated for them

Question 36

the protesting effect - kornell et al, 2009

Answer 36

• ppts studied weakly related word pairs
• no exposure to the answers before the protest
• ppts guess mostly incorrect
• any correct guesses removed from the dataset
• testing improved learning even when all the answers were wrong
• errors fostered learning
  -many replications of this effect
• ppts often be,ie. that they will remember more from simple reading than generating errors - metacognitive illusion

Question 37

search set theory

Answer 37

• producing a guess causes activation of related concepts
• eg pond (water, tadpole and frog)
• partial activation of the correct answer (frog)
• prediction: presenting should only be beneficial for related word pairs, where the target will be part of the search set.

Question 38

set search theory - grimaldi & karpicke, 2012

Answer 38

• same experiment as kornell et al 2009
• weakly related pairs and unrelated pairs
• pre testing improved recall of related word pairs, but not unrelated pairs
• shot in the dark guesses don’t improve recall

Question 39

increased motivation to learn, Seabrooke et al, 2019

Answer 39

• ppts gave higher motivation ratings to learn facts that they had guessed than not-guessed
• names paired with faces
• had to guess or told their fav food, name. occupation and hobby

Question 40

confidence in errors: the hypercorrection effect

Question 41

butterfield and metcalfe, 2001

Answer 41

• ppts answered questions and rate their confidence for each
• feedback provided after each answer
• 5 min retention intervals
• cued recall final test
• high confidence errors more likely to be corrected than low-confidence errors

Question 42

surprise and attention- butterfield and metcalfe, 2006

Answer 42

• ppts completed a hypercorrection task
• ppts also asked to detect soft tones
• ppts missed more tones that were presented with feedback following high-confidence errors than low-confidence errors
• suggests that ppts attention was captured by the feedback