The Right Choice-II

 

Kahneman and Tversky once rigged a wheel of fortune so that it stopped only at numbers 10 and 65. Students in an experiment were asked to rotate the wheel and write down the number on which the wheel stopped. They were then asked two questions.

Is the percentage of African nations among UN members larger or smaller than the number you just wrote?

What is your best guess of the percentage of African nations in UN?

Spin of the wheel gives no clue to the answer of the second question. But students who saw 10 and 65 answered 25% and 45% respectively – clustered around the number they got in the wheel of fortune. Our estimates for an unknown quantity remain close to the first piece of information offered. This is anchoring effect. It has been incontrovertibly proved in many psychology experiments. Salesmen exploit it to boost profits. People bought 7 cans of soup when a nearby sign read, ‘LIMIT OF 12 PER PERSON.’ Sales dropped to half when the sign read, ‘NO LIMIT PER PERSON’. In negotiations for property, final buying price remains anchored to the initial cost quoted by the seller.

Today, drugs used in anaesthesia are markedly safe when administered by experts and with due care. But most have potentially serious side-effects. In a busy practice it is not rare to come across these instants occasionally. Every such instance influences my anaesthesia practice for a few days. I avoid the particular drug if there is an alternative, or employ it tentatively and in much reduced dose. All the time being aware that there is no scientific basis for my fear. We assess the frequency of a class or the probability of an event by the ease with which instances or occurrences can be brought to mind. This is availability heuristic. Salient and dramatic events are readily retrieved from memory. People judge deaths by accidents more likely vis-à-vis stroke. Stroke-deaths never make it to newspapers. People overestimate frequency of divorces in Bollywood celebrities and corruption among politicians – events noticeable by their salience – compared to general public. Air travel reduces after a major air accident. Personal experiences are more available than incidents that happened to others.

Market news slips my ken like water off a duck’s back. I cursorily glance at the economy headlines as I turn the pages of the paper. These always have an explanation for the dramatic ups and downs in the market: economy survey which predicts low growth rate, rise in US Fed rates, party at the centre having lost a state election, increase in interest rates by the RBI, forecast of good monsoons, good indirect-tax collection, lowered petroleum prices. Reading these, it appears that markets are not dependent on uncertain, unpredictable influences. A fact astonishes me. Experts never forecast an extreme change in the immediate future, while they are never short of reasons for yesterday’s coup or a wild-run at the stock exchange. At work is one of the strongest cognitive illusions, the hindsight bias. It is the tendency to overestimate one’s ability to have foreseen the event, upon learning the outcome of the event. It results from the illusion that we understand past. The fact that we don’t, forces mind to construct a plausible story to accommodate all situations that make the unlikely event highly probable – giving the bias the epithet ‘I-knew-it-all-along’ effect. This is our narrative-building mind at its most prolific.

People in an experiment were asked to forecast an outcome in national politics. After the event they were again asked to recall the probability, they had assigned the event. If the possible event did not come to pass, those who had predicted that it would, stated that they had always considered it most unlikely. Those who had predicted its likelihood overstated their accuracy of prediction, once the event happened. We not only build a narrative of the past to justify a surprising event, we even change the predictions about the event we made in past, to accommodate new facts – unconsciously, without being aware of this sleight of our minds.

Hindsight bias can have dangerous effects on the evaluations of decision makers. It prompts observers to assess the quality of a decision by the outcome and not by the processes of judgement employed to arrive at the decision. Serious complications can rarely occur even during a low-risk surgery, in spite of all precautions by the treating surgeons. After the fact, jury will be inclined to believe that the operation was actually risky and the surgical team that advised it should have known better.

Paul Meehl was an acclaimed psychologist of last century. In his most popular work, he analysed whether clinical predictions based on subjective impressions of trained professionals were more accurate than statistical predictions made by combining a few scores or grades according to a rule. Trained counsellors predicted the grades of freshmen at the end of school year, after an interview with the freshman. An algorithm was also used for the same prediction which used only a fraction of information available to experts, viz., high school grades and one aptitude test. Formula was more accurate than 11 of the 14 counsellors.

Applicants for admission to an academic institute or for a job are commonly interviewed by experts to assess suitability. Long-term predictions, about interviewee's performance in future, that interviews are required to project, are extremely difficult to make. Experts do not standardize the questions they ask. Neither have they evolved a uniform grid of responses on which to assess performance of every applicant. They rely on their subjective judgements in grading a candidate. They do not validate their assessments by comparing these with candidates' later performance in the job or academics. Subjective and intuitive judgements of experts mar the efficacy of interview. My college conducted interviews across the country to select students for undergraduate course. On selection, students were given an identification number which corresponded to their rank in the composite assessment (entrance examination and interview). In my years in college, I did not notice that our performance in the course had strong correlation with this rank. I do not know if an analysis was ever done to assess the validity of this system of selection. Our intuitive judgements do not broach any suspicion about their validity. Their evolutionary advantage is the unshakable belief they engender, producing instant response.

Value of wine increases with age. But there are many factors that influence future quality of a wine, even those harvested from the same vineyard. Investors invest in wine like art. It is critical to know future value of a wine. Experts are paid high fees for this. Princeton economist and wine-lover Orley Ashenfelter devised a formula that predicts the price of a wine based on three features of the wine: the average temperature over the growing season, the amount of rain at harvest time, and the total rainfall during the previous winter. This formula has performed consistently better than the highly paid experts.

I routinely evaluate patients for suitability of surgical interventions. Many patients have co-incidental illnesses with the surgical condition which affect the risk of surgery. Each illness influences outcome of surgery in unique ways. A composite risk of surgery for such a patient helps in deciding the relative harm vs benefit of surgery, and to be prepared for a complication. Evidence based medicine – application of best available medical research in the care of a patient – recommends algorithms for assessment of surgical risk. These are not unlike formulas for the future value of vintage wine. And are simple, yes-or-no variety, indices: Does the patient suffer from coronary artery disease, does the patient have evidence of kidney dysfunction, etc.  They have been validated in large multicentre controlled trials. But many anaesthetists are loathe to use these indices. Aversion for formulas is a propensity of system 1. These doctors continue to pronounce the risk of surgery in vague terms like moderate, low, and high. This depends on their anecdotal assessment. Dr Virginia Apgar, in 1953, devised a simple scoring system based on five variables (heart rate, respiration, reflex, muscle tone, and colour) to assess the wellbeing of a new born child and evaluate need for resuscitation. This simple formula, used in delivery rooms across the world, is credited with significantly reducing infant mortality rate.

It has been shown in numerous experiments that expert judgements are usually inferior to formulae and algorithms where many factors influence outcome. Experts try to be too clever by half. They feel the simple formula cannot combine complex features which only they can understand. Experiments prove that complexity reduces validity. Expert judgements are also vitiated by the inconsistency of experts. Experts give different weightage to the same factors on different occasions.

Kahneman and Tversky’s theories have had a huge impact on economics. Richard Thaler, an economist at the University of Chicago and winner of 2017 Nobel prize in economics, says that Kahneman and Tversky’s 1974 paper on Heuristics and Biases, changed the way he understood economics. He is the founding father of behavioural economics. Standard model of economics believes that everyone in economics is rational and selfish, diligently, ceaselessly working for their self-interest. Behavioural economics postulates that the agents making economic decisions are humans and their decisions are influenced by their emotions and feelings. Kahneman has devoted a section in the book on choices in economics. These are novel concepts for me, thrilling beyond imagination. This piece is sagging under its weight. I must save the subject of behavioural economics for another post.

In what are some of the most alluring chapters in the book, Kahneman talks about his research on how we evaluate life – our own and others'.

We experience and remember our lives, it seems, through two different selves, the experiencing self and the remembering self. Colonoscopy is a painful procedure. Anaesthesia was not the norm for the procedure when Kahneman did his experiments in early 1990s. In an experiment, patients undergoing colonoscopy were prompted every 60 seconds to indicate the level of pain they were experiencing. After the procedure patients were asked to rate the ‘the total amount of pain’ they had experienced. Analysis of results revealed that patients did not rate total amount of pain as the sum of experienced pain at various moments in the procedure. A pattern was clear in the results. Global retrospective rating of pain was the average of the pain reported at the worst moment and at its end – the peak-end rule. Duration of the procedure had no effect on the ratings of total pain – the duration neglect feature of the remembering self. Patient whose colonoscopy lasted for triple the duration of others but who had much reduced pain at the end, rated the total pain significantly less.

We think of life as a story and want it to end well. Although it is the experiencing self which experiences the travails and the joys of each passing moment, story is woven by the remembering self, neglecting the duration and remembering only the peaks and the end. We do not evaluate life by integrating pain and pleasure of its every moment. Life of a person who has led a long happy life but with few years of suffering in the end is deemed less happy compared to another who died instantly and painlessly in an automobile accident in his youth. We feel sad for a friend when after his death it is revealed that his children’s love for father was fake and was only for his wealth, although he had died believing that his children genuinely cared for him. A few seconds of poor recording at the end can ‘spoil the entire experience’ of a long captivating musical piece. Peak-end rule and duration neglect is observed in people’s evaluation of holidays. They do not want to repeat the experience of a holiday that was joyful in its entire course but ended badly.

 

Such, such are the strange ways of our mind: two systems weighing same propositions differently, irrational emotions triumphing over rational choices, two selves weaving a different story of the same life. Mind evolved to aid the body in eking out a biologically successful existence - producing maximum copies of its genes - in a rapidly changing world. It fabulously performs its functions. A child of five can effortlessly walk in a dense wood, can gather from the expressions on mother’s face and the tone of her voice whether she is in a happy mood – tasks in which a most advanced computer comes a cropper. We have known for long that ‘to err is human’. Psychology today has revealed the mechanisms of these errors and the situations that prompt them. This revelation provides a glimpse of the unfathomable powers of human mind and its few failings. Kahneman’s book is a gripping story of this discovery.