Decision-making is a complex interplay of neurophysiology, psychology and behavioural economics. The process involves mental models as internal representations of how things work, and heuristics for efficiency of thought. But these shortcuts come with baggage: unconscious biases, cognitive distortions and the pre-programmed, habitual white lies we tell ourselves. These flaws can flip clear thinking into poor judgment and cause big errors.
In any negotiation, beware of a particularly powerful and exploitable bias, the anchoring effect, where our brains lean heavily on an initial significant piece of information. Even an arbitrary or spurious number frames the scene, because the recipient of the anchor proposal involuntarily tries to rationalise why it could be reasonable, says Daniel Kahneman, Nobel economics laureate and the author of Thinking, Fast and Slow.
In retail, anchors are everywhere. A famous example is New York restaurant Serendipity 3’s launch of a $69 “haute” dog. Immediately below on the menu was a standard cheeseburger — for $17.95. The haute dog was great for publicity, and entered the Guinness Book of World Records as the most expensive version. But it was a decoy: its crazy price caused orders for the highly profitable cheeseburger to soar.
Military history gives sobering lessons on the possible cost of cognitive biases. The Allies’ Gallipoli campaign during World War 1 is a case study. Apart from First Lord of the Admiralty Winston Churchill’s hubris and tunnel vision, the British war cabinet exhibited confirmation bias in emphasising intelligence reports suggesting Ottoman weakness while discounting contradictory evidence. Premature closure bias — the tendency to settle rapidly on an initial interpretation — compounded the error.
Gallipoli was also an example of the sunk cost fallacy, the reluctance to abandon a course of action because of what has already been invested, even when this is illogical. The initial naval attack and subsequent April 1915 amphibious landings degenerated into trench warfare stalemate, with some 20,000 Allied deaths by July. But reinforcements continued to be sent. Then an August offensive was ordered. By the time forces were evacuated in January 1916, another 24,000 to 42,000 soldiers had died. Total casualties amounted to 300,000, almost 60% of those deployed. Gallipoli was “waging war with blinders on”, concludes military historian and risk strategist Martha VanDriel.
Daily life is replete with examples of the sunk cost fallacy: continuously repairing a rust-bucket car; paying for hated music lessons because we splashed out on a piano; sitting through a dud concert because we bought the ticket. Failing projects have a hold on us partly because of the human mind’s loss aversion. Neuroscience measures the psychological torment of loss at 1½ to 2½ times more powerful than the pleasure of an equivalent gain. So we have a cognitive instinct towards avoiding or making good on losses, even when this is irrational.
In a South African context, loss aversion and the sunk cost fallacy have meshed corruption, cadreship and political ego to waste hundreds of billions in multiple tranches of state-owned enterprise bailouts for Denel, Eskom, SAA, Sanral, and Transnet.
Financial markets also illustrate how rudimentary errors of judgment are made by even skilled analytical or legal minds. Despite numerous early warnings from private investors, and voluminous information, the US Securities & Exchange Commission (SEC) failed to check the reality of Bernie Madoff’s 30-year, $65bn Ponzi scheme, the biggest in history. The oddity of his funds’ consistently superb performance was ignored due to confirmation bias: his stellar reputation as a leading broker, former Nasdaq chair and service on SEC advisory committees spoke for itself.
Emotional drivers such as greed and overoptimism create bubbles, then fear causes the inevitable crash. The Madoff fraud’s $18bn wipeout was a tempest in a teapot compared with the wider 2008 crash, but it was symptomatic of multiple cognitive pitfalls epitomised by the financial crisis.
Recent global market-cap gains — a three-year surge that continues unabated on Wall Street — have been overwhelmingly driven by technology stocks, which now account for about half of the combined $69-trillion capitalisation of the New York Stock Exchange and the Nasdaq. Are analysts, by sticking to the bet that AI-supercharged technologies will surge ahead, exhibiting belief perseverance even in the face of growing global concerns about where AI is heading?
Perhaps they should heed signs of a herd mentality, when everybody follows the group because it feels safer to conform with what everyone else is doing. “Whenever more than four are gathered / You’ll find a band of fools”, sang Georges Brassens, France’s answer to Bob Dylan.
Sometimes, to dampen the ego factor underpinning our faulty cognition, it’s necessary to laugh at how we think
Errors are woven within the drama of elite sports, and critical mistakes made by supremely gifted athletes are inevitably attributable to judgment, not a lack of skill. Consider the intense pressure of a penalty in football. Goalkeepers have a psychological advantage, as they are not expected to make a save. Yet many surrender this edge by employing an action bias — the cognitive impulse to do something, because agency creates a perception of control.
So, instead of staying alert and poised in the middle of the goal, which would give them a higher save probability, they decide, in advance of the kick, which side to dive. Counterintuitively, in this year’s Africa Cup of Nations final, Senegal’s goalkeeper, Édouard Mendy, stood still as Morocco’s Brahim Díaz, anticipating the keeper’s usual lateral movement, tapped the ball straight down the middle. By avoiding the cognitive trap of action bias Mendy made an astonishing but embarrassingly straightforward save.
Díaz is due some sympathy. Statistically, he would have scored the penalty 85% of the time and overconfidence in his chosen tactic was less to blame than his mind playing loss-aversion tricks.
Contextually, cognitive mistakes are not disastrous in sport. But in other ambits they can have catastrophic consequences.
On March 23 2005, a huge explosion occurred at BP’s Texas City refinery. Fifteen people died, 180 were injured, and damage was estimated at $2bn. The US Chemical Safety & Hazard Investigation Board (CSB) determined that one of the world’s worst chemical disasters was rooted in systemic organisational and safety failures.
Warning signs had been evident for years, it was ascertained, but management assumed that information, systems and decision-making processes were sound simply because no major incident had yet occurred — a classic case of outcome bias. Where safety incidents had been flagged, they were dismissed as minor and insignificant, a cognitive flaw known as the normalisation of deviance bias, which is a ticking time bomb in safety-critical industries.
Do we learn from our mistakes? Hindsight bias is the distortion of memory which induces the belief that we knew an event or breakdown was coming. This soothes failure and allows us to retain confidence in the apparent logic and efficacy of our decisions despite evidence to the contrary. Ego is often involved, with one withering study concluding that, especially when we feel threatened, “we actually tune out rather than tune in, [and] there is no learning whatsoever from failures”.
This partly explains why, after the Texas calamity, BP’s corporate culture continued to emphasise production speed and cost-cutting, including in critical safety systems. And why, just five years later, another disaster struck on the company’s watch. The blowout of its Deepwater Horizon rig in the Gulf of Mexico killed 11 workers and ranks as the worst maritime oil spill in history. There was an “eerie resemblance” between the two explosions, the CSB concluded.
Sometimes, to dampen the ego factor underpinning our faulty cognition, it’s necessary to laugh at how we think.
Edited by French psychologist Jean-François Marmion, The Psychology of Stupidity is a masterful takedown of leaders, politicians, so-called experts and psychology academics themselves. “Two intellectuals, sitting on their duffs, won’t get as far as a brute who walks,” introduces one chapter.
Another discusses the strange heuristic of anthropomorphism, which attributes human feelings and characteristics to animals. To some creatures, anyway, but not those we eat. Cognitive dissonance files away what happens at an abattoir. Cows are beef, buck becomes venison, pigs transform into pork. The nomenclature’s bizarre side effect is that a surprisingly high proportion of Americans believe that brown cows provide their chocolate milk, and almost half the children in France do not know where ham or steak comes from.
People’s apparent irrationality is not entirely a defect. Many biases are inbuilt evolutionary adaptations that enhanced our survival odds. For this, in the modern world, there is a price to pay.
Soon, however, cognitive flaws and bad decision-making may not be overly important. AI’s trajectory suggests that we could reach the point at which machines make all our decisions. If this sounds like an all-in existential crisis for humankind, we can take comfort from neuroscientists Antonio Damasio and South Africa’s Mark Solms, who argue that it isn’t thought that is the origin of consciousness, but emotion. So it may work out well for our species if we don’t have to exercise our minds — as long as we keep feeling.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.