3 blind men approach an elephant. None of them has ever heard of an elephant before.
Curious, they each feel different parts of the elephant to figure out what the behemoth is. The first man happened to grasp the elephant’s trunk, concluding that the elephant must have been a type of thick snake. The second man conveniently felt the elephant’s ear, concluding that the elephant was a very large fan. The third man approached the elephant from the side, concluding that his accomplices must’ve been mistaken—it’s obviously a wall!
None of these men can figure out that the enormous object they’re grasping at is an elephant; they all reasoned rationally in this scenario, yet none of them could even come close to comprehending the nature of what they felt.
Now, imagine these blind men once again—this time with 8.29 billion people, all lined up shoulder-to-shoulder, with 1 foot of space between each person. They’re all standing around a perfectly spherical pond, a pond that measures 1.13 miles across and has a circumference of 3.54 miles; such a pond could have its edge wrap around the Earth 142 times. You are one out of the 8 billion ‘blind men’ standing at the edge of the astronomically large pond. You can grasp at branches and seaweed, while a person thousands—even millions of miles away from you grasps at fish and stones.
All 8 billion people, with you being a single one of them, constitute a single blind human, experiencing less than 0.0000001% of what’s inside the pond. Every day, you might exchange thoughts with those close to you, feeling a peculiar object such as a branch that catches your attention.
This is how large the scale of humanity is—you are a drop in a swimming pool full of others with completely different experiences—yet even while we are more connected than ever, many human beings never leave their tiny cognitive drop.
Most of us infer 80% of our beliefs from that 0.0000001%. We can make completely rational inferences from our personal experiences and still turn out completely wrong. From our own perspective, a sample size of 8 billion is impossible to conceive of, and, in turn, we are unable to sympathize with everyone’s perspectives. The human brain has only ever been trained on anecdotes, not data. This type of disproportionate emphasis on the infinitesimally small portion of humanity that surrounds us is what drives the majority of misguided beliefs.
(Data is just the aggregate representation of a large number of personal anecdotes)
Epistemic shortcutting based on availability.
In 1973, a paper by Tversky & Kahneman found that we tend to evaluate the probability and frequency of an event based on how easily we can cognitively retrieve it—if event K instantaneously comes to mind when thinking about immigration, we tend to emphasize the occurrence of event K when immigration comes to mind, even while event M could’ve occurred 10x more frequently.
Personal experiences and shallow but emotional anecdotes cause both weight skews and conspiratorial/associative skewing towards that anecdote or experience. This is known as the availability heuristic.
This built-in heuristic leads us to make conjunction and disjunction fallacies; the former of which is where we falsely assume that A and B happening is more likely than A happening, while the latter entails believing that A happening is more likely than A or B happening. This is due to the availability of events that are strongly associated with certain things.
Most importantly, the availability heuristic attaches disproportionate weight to dramatized, publicized, or personal events. It is this mental shortcut that causes us to give disproportionate credit to things that we’ve experienced before—numerous bad interactions with black people pre-empts you to accept racist beliefs, while a negative experience with a conservative family member or acquaintance, combined with a lack of self-awareness, can cause you to attack the character of conservatives.
We can see both the mechanisms derived from feeding anecdotes into the availability heuristic machine with most anti-vaxxers (self-proclaimed “vaccine skeptics”). The stereotypical anti-vaxxer usually gives 2 pieces of reasoning to justify their doubt in vaccines:
The “big pharma is really bad, Pfizer did a [insert concerning/overblown thing] with some vaccines, so it must be more probable that it’s a conspiracy surrounding vaccines!”
The “I have 4 children. They were perfectly young, healthy kids. Then my doctor forced me to vaccinate the youngest one. He got a stroke [insert time in minutes/hours/days] after the vaccination. This must be due to the vaccine!”
Reason a) is exactly the conspiratorial conjunction/disjunction fallacy caused by us associating specific events with pharmaceutical companies, or our own/our acquaintances’ negative experiences with such companies. Reason b) is caused by our emphasis on our own/our acquaintances’ negative experiences with vaccines. Reason a) can be thought of as caused by an implication of what causes reason b), such that they’re very closely linked back to our availability heuristic.
Our overreliance on anecdotal evidence leads us to become less rational in more ways than simply ignoring valid statistics.
Prioritizing victims over numbers.
As a French diplomat once said, “The death of one man: that is a catastrophe. One hundred thousand deaths: that is a statistic!”
Such a quote, often falsely attributed to Stalin, is the embodiment of psychic numbing; our minds lose sight of large numbers and, therefore, do not scale our associations or attitudes to an event proportionally to the actual severity of the event. To us, seeing someone give away $100,000 or $1 million triggers a nearly identical reaction—we don’t feel 10 times more excited for the 1 million gift because it’s 10 times larger. They’re nearly identical in impact, and we are, therefore, unable to map out the magnitude of the event as opposed to smaller sums of money.
When it comes to real people, this effect gains another layer of anecdotal skew; we tend to lose the personal connection that comes with firsthand recounts and experiences.
This is known as the identifiable victim effect. We fail to give N times the help when N times the people are affected, we fail to feel N times the happiness when given N times the money, and the list goes on. People are more likely to donate to help a single starving girl introduced to them through a firsthand account than to alleviate an entire famine introduced to them with statistics; the latter is even tied to a decrease in charitable giving and compassion.
More surprisingly, Small, Loewenstein & Slovic found out, in a 2007 study, that people do tend to become conscious of the identifiable victim effect—leading them to arbitrarily decrease their aid and compassion to the identifiable victim while failing to increase aid to the much larger, statistically described impact. When applied to different contexts, the identifiable victim effect is much weaker than the 2007 study anticipated it to be, albeit still an occurring phenomenon.
Maier, Wong & Feldman do find that the specific results of Slovic’s study are irreproducible and carry weak empirical backing, though it doesn’t do much to downplay the effects of psychic numbing in general, which carries much better epistemics. During a later 2014 study, Slovic observed a decrease in empathy and giving when the sample size expanded from 1 → 2, showing a deterioration in empathy for any group larger than an individual. When the size continued to expand from 2 → 8, the amount of empathy decreased further.
“When we offered people the opportunity to help a child who’s starving, and we showed a picture of the child and information about her, we got a fairly strong response. But when we showed that same child and then, next to her picture, we gave the statistics of the larger problem of starvation, suddenly the donations dropped in half.” — Paul Slovic
The identifiable victim effect and psychic numbing are omnipresent anywhere there seems to be a victim. If you walk down the street and see a foreigner committing a crime, you’ll be able to draw implications on topics like immigration (e.g., ‘immigrants come from cultures of violence’, ‘mass immigration is destroying the country’) that are practically chained together with your strengthened overlap of “immigrants” and “crime”.
A propensity towards giving out excessive empathy towards victims with personal narratives is nearly unavoidable.
Forming beliefs against data.
If a scientist were to publish a study like this:
“Out of 1,000 participants sampled, 90% (900) of them had a preference for brownies with chocolate chips in them.”
This represents 900 people expressing a preference for chocolate chips in their brownies. If you were to try to counteract that study by bringing up a personal anecdote (e.g., ‘I actually have a friend who thinks putting chocolate chips in brownies is absolutely disgusting and ruins the texture of the brownie’), the personal anecdote would be orders of magnitude weaker—900 times weaker, to be exact!
A study telling you that 90% of 1,000 participants surveyed preferred chocolate chips in their brownies implies that you can bring in 900 people who can give a personal reason as to why they like chocolate chips in their brownies.
Such a study also implies that a very high percentage of people have a preference for chocolate chips in their brownies as opposed to those prefer those without chocolate chips, as they vast majority of people they sampled confirmed the claim that “people like chocolate chips in their brownies”, giving us ample reason to believe that a broader skew towards chocolate chips in brownies is present in larger groups (the state of New Jersey, the U.S, North America, Earth).
A preference for anecdotes of statistics shouldn’t even have an irrational emotional basis, since statistics are just better ways of representing thousands, if not millions of anecdotes—I could either recount the answers, preference level, and names of all 900 chocolate chip lovers to construct 900 individual anecdotes, or I can simply tell you that 900/1000 people sampled preferred brownies with chocolate chips.
Kubin et al. (2021) found that both liberals and conservatives believe that using facts helps foster mutual respect in discussions, while they actively engage in the opposite. Across 15 studies, people respected those they disagreed with more when they cited personal experiences or anecdotes over objective studies and facts. They were even viewed as more rational when they rationalized moral and political views by utilizing subjective experiences or anecdotes.
Due to statistics being methodologically rigorous, subject to revision/error, and often misrepresented, personal stories and experiences gain a sense of innate credibility that’s verifiable—if someone had a personalized, emotional experience that led them to a conclusion, it sounds a lot more grounded in “reality” than an abstract study would; we tend to view emotional, irrational ways of justifying beliefs as more convincing, genuine, and rational since objective statistics feel disconnected from our own experiences.
Learning to reason with personal experience makes reasoning built upon the impersonal seem further from reason.
Rejecting irrationality.
The 3 blind men would’ve never figured out that the object was an elephant without mutual communication from tons of further observations—you and I will also never see into reality when our sample size is 1.
In our world, we cannot hope to do that. There are 8.29 billion of us standing on the edge of the metaphorical pond, millions of miles across. An attempt to understand the world we live in using purely personal experience would be doomed to fail; our biological adaptations, however, predispose us to take that path. We no longer live in tribal, ape-like societies.
Our development in the past 300 years has been far too fast by evolutionary standards, leaving our evolutionary clock millions of years behind. Every single cognitive shortcut that leads you to embrace conspiracy, to embrace stories and narratives over reality, has served humanity well up until this point.
How do we escape this madness that is irrationality?
Before you attach any weight to an experience, ask:
“What is this experience objectively proving?”
“Would I have changed my mind if my experience here turned out to represent the opposite conclusion?”
And, most importantly, “how is this experience reflective of overall objective trends or factors?”
And before you attach any weight to a statistic, consider:
How large the sample was
The methodology used in the study
If it’s been replicated rigorously
If funding or interests affected the results
A valid statistic with a sample size of 10,000 is worth 10,000 anecdotes! It’s not that personal experience is irrelevant, but that a single experience, no matter how emotionally inducing, is only worth a single data point. Don’t choose between basing your beliefs on stats or events; choose between basing beliefs on a single source of data or 10,000.
Some citations:
Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5(2), 207–232. https://people.umass.edu/biep540w/pdf/Tversky%20availability.pdf
Small, D. A., Loewenstein, G., & Slovic, P. (2007). Sympathy and callousness: The impact of deliberative thought on donations to identifiable and statistical victims. Organizational Behavior and Human Decision Processes, 102(2), 143–153. https://www.sas.upenn.edu/~baron/journal/7303a/jdm7303a.htm
Slovic, P. (2007). “If I look at the mass I will never act”: Psychic numbing and genocide. Judgment and Decision Making, 2(2), 79–95. https://www.cambridge.org/core/journals/judgment-and-decision-making/article/if-i-look-at-the-mass-i-will-never-act-psychic-numbing-and-genocide/0E55D099E133068F9ACD5A0DBBE1E4E2
Västfjäll, D., Slovic, P., Mayorga, M., & Peters, E. (2014). Compassion fade: Affect and charity are greatest for a single child in need. PLOS ONE, 9(6), e100115.
https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0100115
Maier, M., Wong, Y. C., & Feldman, G. (2023). Revisiting and rethinking the identifiable victim effect: Replication and extension of Small, Loewenstein, and Slovic (2007). Collabra: Psychology, 9(1), 90203. https://online.ucpress.edu/collabra/article/9/1/90203/199223
Kubin, E., Puryear, C., Schein, C., & Gray, K. (2021). Personal experiences bridge moral and political divides better than facts. Proceedings of the National Academy of Sciences, 118(6), e2008389118. https://www.pnas.org/doi/10.1073/pnas.2008389118
Just used your famously misattributed quote 😎
Is there a belief shop one can purchase from? Asking for a friend.