Irrationality: the enemy within by Stuart Sutherland
[Published by Pinter & Martin Ltd.; 20th anniversary ed edition (30 July 2013)]
A very revealing read.The blue headings give the chapter titles, the text the points that struck me as most important.
Wrong impressions: Judging by the first thing that comes to mind. Publicity can influence jusgement; contrary to popular belief you are 40 time more likely to die in 1 hour cycling on main roads than in 1 hour of fairground rides. First encounters colour later assessments - first time you meet someone who is in a bad mood, you keep that impression even when later they are pleasant. Good looking people often thought to have better mind, sense of humour, athletic ability, but there is no actual correlation. Exam scripts tend to be marked higher with good handwriting and if examinee name shows they are male.
Obedience: USA Milgram experiment giving electric shocks showed ingrained obedience to authority even when this involves hurting other people. Junior doctors would not voice disagreement of treatment decisions by a more senior doctor. Massacres carried out by people with strong training in obedience and initial orders given by someone removed from the actual scene.
Conformity: In group situations we are more likely to agree with the opinions of others who speak first, even if we first had doubts. A decision publicly announced is more likely to be acted on - a factor in success of weightwatchers and addiction help groups.
Crowd panic: If all behave in an orderly way, more will be saved; there is less panic if someone in authority is present. Panic spreads - seeing others push makes you feel you can/should too. Systematic and pointless violence common in crowds (lynch mobs, football supporters). Also when several people in a situation that needs help, each feels less responsibility than if they were alone.
In groups and out groups: Group attitudes are more extreme than those of individuals. Leaders pick advisors and assistants of similar views, but not those of higher intelligence and speaking skills. Groups find ways to identify themselves (clothing, hairstyles) and may come to dislike or despise other groups. prejudice usually accompanied by use of stereotypes, noticing what supports already held views.
Organisational folly: Following traditional practice is easy, changing things is harder. Promotion may be by seniority and a job for life. Organisational structure may encourage overspending - the sum allocated for a year depends on the amount spent last year. Companies often hang on to loss making sections.
Misplaced loyalty: Look at several houses, choose one and buy it, then need to justify commitment so exagerate good points and minimise bad points, only acknowledging these is major defects or problems occur. Refusal to abandon a useless project in which a (often large) sum of money has been invested.
Misuse of rewards and punishments: Being rewarded for doing an activity may give less incentive to do the activity unless a reward is given. Those who win prizes may come to work less hard subsequently; those who lose experience failure. Pupils learn best when remarks are to help improve. Better results from surgery when people have participated in decisions.
Drive and emotion: Stress, rewards, punishments and strong emotions reduce flexibility in thinking. With piece work, concentrate on quality not quantity.
Ignoring the evidence: People tend to seek confirmation of a view or theory but should be trying to disprove it.
Distorting evidence: Two groups given results of studies that supported both views thought the studies favouring their view were more convincing and better conducted.
Making a wrong connection: Psycholanalysis is still used despite evidence it is useless; simply having someone listen to you does as well as therapy. In making a decision you need to know (1) A+B, (2) A but not B), (3) B but not A, and (4) neither A nor B.
Mistaken connections: Evidence shows telling patients what to expect (length of operation, pain they might experience) means they complain less about pain, need fewer sedatives and recover more quickly.
Mistaking the cause: Earlier correlations between (a) eating saturated fats and blood cholesterol levels and (b) personality types and heart attacks are not as strong as it was thought.
Misinterpreting data: Heavy smoking increases the risk of death from lung cancer by ten, but for every death from cancer, two will die from heart disease. The bigger the sample, the more likely it will reflect true frequency.
Inconsistent decisions and bad bets: To identify a car fault, a mechanic should know the probability of each fault and how long it takes to find and check high probability and low time faults first. When giving opinion on 4 or 5 point scale, most people favour the middle points.
Overconfidence: 95% of UK motorists thought they were better than average drivers. Most people think they will live longer than the average.
Risks: Failure to take into account the limitations of the human operator (e.g. poor design of dials and screens). In UK compulsory seat belt wearing reduced deaths of car occupants but increased risk to pedestrians and cyclists as drivers took more risks.
False inferences: Trainee pilots praised for very good flight did less well on next flight; those reprimanded for very bad flights did better next time. Extremes both exceptional - regression to the mean.
Failure of intuition: Listing pros and cons before making a decision allows you to take account of more evidence and alternatives. Specific tests are more useful than job interviews as determining suitability for jobs.
Utility: Growths in kidney are tested by either aspiration (low risk) tests for cysts or arteriography for tumours (2 days in hospital and risk of blood clots - doctors and patients agree it is 10 times as bad a procedure) . Negative result of either test does not rule out other condition, so aspiration should be done first unless likelihood of tumour is greater than 1/11. Many medical treatments have unpleasant side effects so Quality Adjusted Life Year (QALY) try to establsih how many years with a given disability are equivalent in people's preferences to one year of normal life. This figure is then compared with the probablities of death, complications or side effects of an operation or drug treatment.
Paranormal: People who believe in the paranormal almost certainly base this on too small a sample.
Causes, cures and costs: Under stress people tend to act on impulse. Conformity and embarrassment probably partially inborn as human survival depends on being a member of a group. However, group allegience in modern society can lead to inappropriate and irrational behaviour. There is evidence that learning one subject does not make it easier to learn a second unless content overlaps. A-level teaching in the UK emphasises rote learning with little stress on rational thinking. Multiple choice questions allow no scope for thought and depend entirely on memory. exams with a fixed time limit are likely to encourage impulsice and inflexible thinking.
From another source - not recorded
Keeping secrets: Attempts to keep information secret are not very successful. Outside commerce, ideas are to be drawn on and the inventor of an idea gains rather than loses if an idea is taken up by others.
END
