This month The Business of Pleasure is taking an unbiased (I promise!) look at biases. Why they’re there, the problems they bring with them, and how to manage them a bit more effectively (because frankly, we can’t live without them).
Selling, Fast and Slow.
This week saw the passing of Daniel Khaneman, the Nobel Prize-Winning psychologist and economist, whose 2011 book, ‘Thinking, Fast and Slow,’ probably did more to change the way we think about thinking than any book since Sigmund Freud’s ‘The Interpretation of Dreams,’ published 111 years previous.
The print run for ‘The Interpretation of Dreams’ ran to 600 copies, and took 8 years to sell out.
‘Thinking, Fast and Slow,’ sold over one million copies in its first year, partly due to my purchasing a dozen hardback copies for Christmas presents that year.
The main reason for the book’s success, however, was that Khaneman succeeded in shining a light into some dark corners of human behaviour in a way that was hugely engaging, entertaining and, most importantly, easily comprehensible for us non-psychologists and non-economists.
Now, a lot of the fun stuff came about during long strolls with his friend and colleague Amos Tversky and centered on a couple of catchy little numbers called heuristics and biases. In short, heuristics are the short-cuts necessary to enable fast, intuitive thinking, and biases are the short-comings these short-cuts can produce. Not that we should feel short changed. This mental short-hand may have given us an edge over our relatively bigger brained (and bigger-fisted) relatives, the Neanderthals, and, more prosaically, save us hours (if not dosh) at the supermarket. Aside from Natural Selection and grocery shopping, however, here are some thoughts as to why heuristics have remained an essential item for our evolutionary shopping basket...
For us, as a species, some things haven’t changed much since the time of dawn/dawn of time paradox.
It is still a big old scary world out there, and it keeps on changing at the speed of light/sound/smell/touch/taste.
It’s also changing at the speed of thought and at the significantly faster speed of sub-conscious impulses.
But what really adds complexity, is the fact that some individuals of our species are quite capable, like Dickens’ dark Mister Jaggers, of smiling with all their teeth. So we really need to know where we stand at all times. Both figuratively and literally.
In short, very short, there’s just so much going on out there, and so little processing power to figure it out with.
Oh yeah, and a lot of it appears to be bat-shit crazy, largely because our CPUs aren’t always up to figuring out why what’s happening is happening in the way it appears to be happening as it happens.
Faced with the above, in the mid 1950’s, and midway between Sigmund’s ‘Dreams’ and Daniel’s ‘Thinking’, the American Political Scientist, Herbert Simon, took a stab at making sense of how we make sense of things by suggesting that human rationality is ‘bounded’ and that, when faced with a complex problem, we don’t try to obtain the very best solution from all the possible options, instead we ‘search and satisfice, sequentially generating and trying out new options …until we find one that meets all essential criteria or as long as our outcomes are below aspirations. In other words, boundedly rational decision makers continually search for better options. This model of decision making represents the kind of behaviour that is compatible with the access of information and the computational capacity that are actually possessed by organisms.’
Simon was born slap bang in the middle of World War One, and his ground-breaking work on decision-making and complexity was recognized with both a Nobel and a Turing Prize. By the time of his death, in February 2001, this iterative explanation of human decision-making was being refined even further in line with advances in neuroscience. Instead of the idea of the mind as the deck of the Star Ship Enterprise, with Kirk and Spock (a.k.a. System 1 and System 2 in Khaneman and Tversky terms) interpreting whatever ‘sensors indicate,’ we were confronted with the far less reassuring notion of this (more-or-less) dynamic duo presiding over a bustling Betting Office, or frantic Dealing Floor, in which the primary role of the senses is, primarily, to confirm the accuracy, or otherwise, of the voyagers’ pre-configured predictions about what’s ‘out there,’ in the Great (and greatly alien) Unknown. And (Please God!) help us, over time, to make ever better bets and smarter trades. Now depending on the constellation of issues confronting us, and the available time and processing power, it appears that we are forever vacillating between Kirk’s gut instincts and Spock’s little grey cells. They both serve a purpose; the high speed evaluations that underpin snap decision-making are essential for dealing with rapidly changing issues and environments, but they necessarily reduce the scope for experimentation, for trial and error, through which more effective solutions may be advanced. And, as noted above, the other major drawback to intuitive decision-making is that it can trigger a proliferation of biases that can skew our judgement, limit our options, and send us, half-blinkered, down expensive and time-consuming blind alleys.
A brief smorgasbord of biases might include those I am about to share with you, but there are so many to choose from. Opponents of generative AI regularly lambast large language models for their biases, seemingly (unbiasedly) unaware of our 230 recorded human biases. And the 231,st which I’m putting on the record today. But rather than tell you what this newly-discovered bias is, I want to see if you can work it out. So switch on your bias-detectors and no fast-forwarding or scrolling to the end, please.
Negativity bias. We have a greater recall of unpleasant memories compared with positive memories.
Availability bias. We are much more likely to recall recent, nearby, and immediately available examples than remoter ones, and overstate their significance.
Illusory correlation. Perceiving a connection between two or more events when, in reality, the relationship is merely coincidental.
Illusory truth effect. We are more likely to believe statements to be true, regardless of their actual validity, simply because we have previously heard them. Even if we can’t really remember ever having heard them.
Hindsight bias. The tendency to see past events as predictable.
Confirmation bias. The tendency to search for, interpret, or recall information in a way that confirms one’s beliefs or hypotheses.
Choice-supportive bias. The tendency to remember one’s choices as better than they actually were.
Peak-end rule. Instead of perceiving the sum of an experience, we tend to register how it was at its peaks, as pleasurable or painful for example, and how it ended.
Humour effect. Humorous items are more easily remembered than non-humorous ones.
Bizareness effect. Bizarre material is better remembered than commonplace material.
Bias blind spot. The tendency to see oneself as less biased than other people, or to be able to identify more cognitive biases in others than in oneself.
So let’s take a few of these critters for a spin…
We’ve all been there. Monday morning. A crowded compartment on the pre-dawn commute. Only one person smiling. Lucky sod, smug bastard or grinning idiot? And is that a duelling scar, or a shaving cut, beneath the Peppa Pig Band-aid? We make an initial judgement on people around us in less than two one hundredths of a second, which is great when on the look-out for sabre-carrying psychos, but potentially perilous when stereotyping, misjudging, and underestimating, a competitor.
Same train, same train-of-thought. A family sitting opposite. The daughter, off to hockey, looks smart as a whip, something in the eyes. The son, similar age and school uniform, looks as daft as a brush …but then he asks his dad a question about questions, which his dad couldn’t answer. Daniel and Amos, Siggy and Bert, would probably have been stumped too. I certainly was. The kids are bookended by mum and dad. Both ‘lookers’, not a hair’s breadth between them in that department. The Matching Hypothesis, which was proposed in the 1950s, suggested that most humans gravitate towards partners who are of a similar level of attractiveness. As far as I’m aware, no subsequent experiments have replicated the results of the original research, but hey, here’s something you can try at home. Or on a train. Or perched on that well-worn bar-stool at your preferred people-watching place. Take a discreet peek at the couples around you. Grade them for attractiveness. Not as a pair, but individually. How close are the scores you’ve apportioned for each partner? A mile apart or a millimetre? My money (and rep) is on the latter. I’m not saying that you won’t find any pairings on a par with ‘Beauty and the Beast,’ but I’d be surprised if you’ll find many. And fair enough, right? There are perils in ‘punching above your weight.’ But when you break down physical attractiveness, basic physical fitness aside, all you’re generally comparing is regularity of features. Or, more succinctly, symmetry. That’s right, Natural Selection, 21st century-style, probably hasn’t moved half a degree up the dial since Adam’s grandson caught Eve’s grand-daughter’s eye through the moonlit orchids in the Hanging Gardens of Babylon. But should scoring equally for symmetry out-trump all and any more useful qualities, like caring, empathy, creativity, self-confidence, self-respect, sense of humour, sense of self, maturity, emotional intelligence, joie de vivre, positive attitudes towards risk-taking, or hey, the simple ability to love? No wonder there are so many divorces, relationship-splits and unhappy singles. And I know this is particularly difficult for my own listeners -there are just too few ‘tens’ out there for you to match with.
The family rise to leave at the next stop, and I note that the genius boy has another issue to potentially blight his life. He’s below average height. Research has repeatedly shown that taller people, of any gender, earn substantially more over a lifetime than shorter people, while slimmer women, other studies suggest, also earn considerably more than those with a higher BMI.
Now that was just one myopic people-watcher traveling six stops on the District Line. It goes without saying that decision making in organizations is also prone to the biases inherent in intuitive thinking, and many deploy debiasing interventions to try and remove automatic biases from organisational choices. However, recent research seems to suggest that reversing biases by periodically adopting the opposite automatic preference can be a more effective strategy. An interesting approach, which may in time produce results, but change is challenging and only one quarter of change initiatives are successful in the long term. As one paper succinctly puts it: “From the evolutionary neurosciences point of view, the obstacles to change may come from three characteristics: the human brain tends to (1) save energy, (2) avoid uncertainty, and (3) imitate others. The authors of that paper propose deploying change management techniques to help disrupt the powerful grip of biases such as those proposed by the STREAP-Be model: ‘Safety, Trigger, Reward, Emotion, Alignment, People, and Behaviour,’ including strategies for re-shaping behaviours such as:
(1) Setting short term wins in which a goal is split into smaller tasks, micro-deadlines, and mini-rewards;
(2) Intentionally breaking a behaviour pattern and acting “as-if” it has been achieved (‘fake it until you become it’);
(3) Identifying the desired behaviours and rewarding them consistently until they become a habit;
(4) Aiming for “good enough” decisions, rather than striving for “optimal choices” and focusing on progress rather than perfection.
(5) Using an action oriented approach to decision-making. ‘We are what we repeatedly do,’ …said Aristotle.
As you might imagine, a lot of this change management approach to biases overlaps with that other famous ‘H’ of System 1 ‘habits.’ And, as with heuristics, there are pluses gained from habit-centred behaviours as well as minuses… otherwise we’d have deleted them from that same evolutionary shopping basket long ago. Converting activities into habits conserves energy, which can free up some of that precious processing power for more complex System 2 stuff, and, just as with heuristics, the rapid response time of a habit reactions may save our skin in sudden and stressful physical (and commercial) encounters. Indeed, extensive research with our rodent relatives has shown that any strong emotional arousal, and particularly acute and chronic stress, automatically sends decision-making down the System 1 ‘habit’ chute.
Besides helping to deal with actual physical (or competitive) threats, the mere act of avoiding uncertainty can pay dividends, because in so many situations we find ambiguity to be stressful. And often debilitatingly disruptive. But there’s a further advantage that goes right to the historical heart of habit forming; habits are a huge part of how we learn. Monkey see, monkey do, or ‘Ecoutez et repetez,’ be it a phrase in a foreign language, a tennis serve or a golf-swing. Over and over again, hour after hour after hour, we consciously set our mental road crews to work on building ‘habit highways’ that will permit new skills to be performed at speed. However, a cute phrase or three hundred yard drive are merely nice-to-haves. The ability to acquire and ape the behaviours of our social groupings, and follow the unwritten rules underwriting our conduct (and safe conduct) among the geographically cramped cohorts of the most potentially lethal species in the history of the planet, that, my friends, is survival stuff.
But for all the advantages of habit behaviours, the down-side can range from mildly disadvantageous to downright fatal. Once they’ve gone to the great expense of sending neural road crews out to construct these high-speed behavioural motorways, our fuel-conscious brains naturally tend to make use of them, even when they might be taking us in the wrong direction, or, more spectacularly, off the end of a half-built bridge. ‘Because’, as Daniel and Amos observed on one of their walks, ‘change costs energy, which means past investment is lost.’ This ‘switch cost’ of building new roads, or new neural pathways, might, for many everyday behaviours, be prohibitive. But when there is stress or trauma supercharging our reactions, habits may become uncontrollable circuits, as can be all too frequently seen in pathologies such OCD and addictions.
But we can have our cake and eat it. As individuals, within or without organisations, we can choose (System 2) to put in the work building behaviours that can then benefit by becoming positive, energy-efficient ‘go-to’ habits (System 1) to improve our problem solving and decision-making. And while an entire sub-industry has been created to supply exactly this, it’s label ‘Mindfulness’ rather obscures the relatively mind-less, System 1 habit component so vital to the process of incorporating beneficial changes.
The train has reached the terminus. The passengers have left the carriage. Did anyone spot the 231st bias that I have recorded, today, for the first time in the history of our species? I’ll give you a clue. Last week, at a conference, we learned how newly-minted large language models are set tasks to test their efficacy. Apparently, ‘looking for a needle in a haystack’, the ability to whittle out a single aberrant phrase concealed among hundreds of millions of books and papers, is how the architects of AI rate their tools’ latest iterations. Burping their new babies. But for us, as humans, it is precisely that aberrant phrase or sentence that confers the appearance of truth. At least in works of fiction. Because, as humans, that most suspicious, and justifiably suspicious, of species, we inherently mistrust any story that conforms too closely to our expectations. And anything that jars with those expectations, however minimally, can imbue a fictional world with a veneer of reality. Even if it’s only a porcine sticking plaster.
Copyright David Thomas 2024
ALL RIGHTS RESERVED