Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
56 Cards in this Set
- Front
- Back
Asked people to assign probabilities to suspects and found the average per suspect remained constant at the expense of added ratings being over 100%; 3 suspects 113.6%, 5 159.1% |
Teigen (1983) |
|
Total probability of 5 were 2.1, but with some training this was avoided |
Robinson & Hastie (1985) |
|
1) Availability 2) Representiveness 3) Anchor-and-adjust |
Tversky & Kahneman (1974) 1) Judgements based on ease relevant instances come to mind 2) Based on extent an outcome is representative of the process or category in question 3) People adjust away from an initial 'anchor value' |
|
Estimate number of US deaths per year fro certain causes; over-estimate rare and underestimate common events (availability due to over-reporting of rare events) |
Lichtenstein et al. (1978) |
|
Participants guess topshop & IKEA prices; estimate for extremes were biased towards the mean (central-tendency effect) |
Matthews & Stewart (2009) |
|
19 famous men, 20 less famous women (and reverse); people recalled more famous names, 80/99 judged famous category more frequent (availability) |
Tversky & Kahneman (1973) |
|
Estimate -ing and -n- words - 13.4 -ing words, 4.7 -n- words. Availability of relevant instances |
Tversky & Kahneman (1983) |
|
People given character descriptions and asked probability they were a lawyer or engineer; ignored the base-rate they were told and judged purely on stereotype |
Kahneman & Tversky (1973) |
|
Gave character character descriptions and 9 subjects and found correlation between participants rating of likeliness and how much the character resembles the typical, but negative correlation between resemblance and estimated percentage chance |
Tversky & Kahneman (1973) |
|
Gave inverse fallacy problem to physicians and 95% gave 0.7-0.8 chance |
Eddy (1982) |
|
Reworded question with frequency and 8% success rate rose to 46% |
Gigerenzer et al. (1998) |
|
Roulette wheel, as run-length increases more likely to bet against it, after run of 6 85% bet against the run (Gambler's Fallacy) |
Croson & Sundali (2005) |
|
Probability of heads judged as less when there has been a streak |
Matthews (2013) |
|
Showed basketballers score probability judged as higher after a streak |
Gilovich et al. (1985) |
|
People believe their roulette bet more likely to win after win-streak |
Ayton & Fischer (2004) |
|
GF: Sampling without replacement generalised to random mechanical process HH: Generalisation of the fact that human action shows positive recency |
Ayton & Fischer (2004) |
|
Frequency people make inferences tested: MP 97%, DA 56%, AC 64%, MT 74% |
Schroyens et al. (2001) |
|
Card flipping task, 1/34 choose D & 7 |
Wason (1968) |
|
Suggest matching bias: 5, 9, G, 4 and 'if S then ~9' should lead to a selection of 4 but instead participants select 9 - select digits mentioned in question |
Evans & Lynch (1973) |
|
29/40 people respond correctly to 'if a person is drinking beer, they must be over 18' |
Griggs & Cox (1982) |
|
Cheater Detection: 75% success with task of detecting cheat over 25% with digits, 41% with familiar descriptive problem ('if you're going to Boston, you will take the train') |
Cosmides (1989) |
|
75% success with 'if you clear up spilt blood, you must wear gloves' |
Manktelow |
|
Show participants there is a rule on immunisation for travel to some countries - check whether people are following, success. Asked to pretend to be allergic to vaccine and check whether there is a rule that one must be immunised, 47% success. Asked to check if people had been following rule unnecessarily, failure. |
Girotto et al. (2001) |
|
88% success on task 'all A are B, all B are C, what follows', 8% on 'all B are A, all B are C' |
Roberts & Sykes (2005) |
|
Showed more than half said 'all A are C' to 'all B are A, all B are C' meaning we might take it to mean '...and vice versa'. More specific instructions gave a success rate of 94% |
Provitera et al. (1971) |
|
Gave participants syllogistic questions and 4 possible answers to select, 45 of the questions had no answer; strong correlation in mood (90% in particular) i.e. matching universal to universal, negative to negative... |
Begg & Denny (1969) |
|
Mental Models |
Philip Johnson-Laird: form model of state of world provided by premises, combine these models then search for alternative models consistent with them then validate. Difficult with multiple models |
|
Problems with more mental models solved slower and less accurately and those with better working memory did them faster |
Copeland & Radvansky (2004) |
|
Asked participants how many conclusion they considered in multiple-model question and found no more than in single |
Newstead et al. (1999) |
|
Found 71% of people accept invalid syllogism with believable conclusion |
Evans et al. (1983) |
|
Suggested that people try to construct only one mental model for capacity limits, and then if conclusion is believable try to validate, unbelievable refute. If this fails then there is a state of uncertainty, swayed by their belief about the probability conclusion is valid |
Klauer et al. (2000) |
|
Calculator jacket with walk for £5 off. 68% walk for first, 29% for latter |
Tversky & Kahneman (1981) |
|
Magazine subscriptions: 84% choose online and hard-copy when priced the same as hard-copy, without that option 68% choose only online |
Ariely (2009) |
|
80% of £4000, or certain £3000 - 80% choose the certain option |
Tversky & Kahneman (1979) |
|
Prospect Theory |
Focus on change in wealth with respect to a reference point from which there is a diminishing sensitivity to gains and a (steeper) diminishing sensitivity to loss (loss aversion) |
|
Endowment Effect; 50% preference between chocolate and mug, give participant one and ask if they want to switch 90% don't switch |
Knetsch (1989) |
|
Certainty effect: 82% choose certain £2400 over 0.6x£2400, 0.33x£2500, 0.01x£0. But 83% choose 0.33x£2500 over 0.34x£2400. |
Tversky & Kahneman |
|
showed 75% say an increase in chance from 5% to 10% is more significant than 30% to 35% |
Gonzalez & Wu (1999) |
|
Shown bets like 0.95x£2.5,0.05x-£0.75 vs 0.4x£8.5, 0.6x-£1.5 and asked to select which they prefered. Later told they have tickets for the bets and asked to value how mucht hey'd sell them for; 73% valued the latter on every bet which they had chosen the former |
Lichtenstein & Slovic (1971) |
|
Peanuts Effect: People less risk averse with low stakes. However, showed this effect was smaller with lower probabilities & prospect theory cannot deal with this |
Weber & Chapman (2005) |
|
A model where decisions are made by sampling and valuing money by pairing it with other amounts retrieved from memory |
Stewart et al. (2006) |
|
Categorical Theory of Emotions |
Darwin (1872), Ekman (1992); basic emotions with rapid onset, brief duration; cross-cultural facial expressions |
|
Dimensional view |
Emotions lie on orthogonal axes |
|
Injection of placebo, adrenaline and then put in room with happy or angry person. 0.98 informed, 1.78 ignorant in happy v 1.91 informed, 1.39 ignorant in angry, where score correlates to level of happiness |
Schachter & Singer (1962); cognitive labelling |
|
1) Bilateral amygdala damage meant impairment in 'fear' recognition 2) Lesions reduce/abolish acquisition of fearful response to neutral stimulus paired with aversive outcome |
1) Calder et al. (1996); DR and SE 2) Blancherd (1972) |
|
1) Controls show better memory for evocative images than others, Amygdala damage patients do not 2) Superior recall of emotionally arousing picture correlates with amygdala activation |
1) Adolphs et al. (1997) 2) Hamann et al. (1999) |
|
1) vmPFC lesion shows increased emotional reactivity but reduced emotionality 2) Show no SCR with socially-evocative image 3) More likely to pick Utilitarian response over emotionally charged sacrifice |
1) Anderson et al. (2006) 2) Damasio et al. (1990) 3) Koenigs et al. (2007) |
|
Iowa Gambling Task; vmPFC patients choose A&B more often. 2) Show lower anticipatory SCR 3) Same with amygdala patients but they also lose reward & punishment SCRs |
Bechara et al. (1994) 2) 1996 3) 1999 |
|
Ask questions after 20 trials about what participants think is going on. Pre-punishment, pre-hunch (controls have SCRs and no idea), hunch (anticipatory SCRs for A and B, patients don't) and conceptual (7/10 explain why A&B are worse, patients show no SCR and favour A&B) |
Bechara et al. (1997) |
|
Somatic Marker Hypothesis |
vmPFC take role in 're-living' emotional and somatic experiences associated with responses and these play a part in biasing decision process |
|
Patients with Pure Autonomic Failure (giving no SCRs) more likely to pick good decks |
Heims et al. (2005) |
|
Show that decks A&B also involve larger sums of money; performed IGT with more loss in A&B but higher sums in C&D and found SCR larger preceding C&D but still chose C&D |
Tombs et al. (2002) |
|
Suggest that as the first trials always win vmPFC patients form response tendencies which are difficult to overcome; if the failures are made clear earlier then 68% patients choose C&D and 72% controls |
Farah & Fellows (2005) |
|
Intuitive Reasoning Task (IRT); more pressing questions, found patients had good explicit knowledge so no need for unconscious knowledge |
Maia & McClelland (2005) |
|
Pick from a deck and guess if colour would match randomly appearing card; rigged so 60% guess correct with A&B and 40% correct with C&D; outcome magnitude not confounded with goodness of deck, questions show no overt knowledge - people pick from good decks more, larger anticipatory SCRs = better picking |
Dunn et al. (2010) |
|
Indicate food preferences weeks before then asked to imagine items on a menu and asked which they would choose. Found PET scans show activity in left amygdala & left mOFC for high incentive items; amygdala activity correlates with incentive, OFC with choice difficulty (as reported after) |
Arana et al. (2003) |