It matters how we connect the dots.
Narrative fallacy is a term introduced by Nassim Taleb . It refers to our difficulty looking at a sequence of facts without connecting them in some way, either to each other or to something else we know. Here is an illustration. Can you memorize the following list: magnolia, 1964, firefighters, Chicago, jazz club, winter, 10. For most people, it is hard to retain all these details. Try this instead: “On a cold winter day in 1964, a jazz club ‘Magnolia’ in Chicago caught on fire. Firefighters were on the scene within 10 minutes.” Better?
While connecting things into a story is a useful strategy for memory exercises, in our daily lives, we routinely invent relationships between unconnected details without noticing that we do so. Our conscious minds resemble storytellers ; thinking statistically—leaving things unconnected with probabilities assigned to each one—is counterintuitive for us . “It will rain this afternoon with 65% probability,” says the weather forecast. Who remembers the 65%? We just grab the umbrella. Whether it was 65%, 58%, or 72%, it would make no difference—umbrella. This is what the forecast means to us.
We are all prone to committing the narrative fallacy when we do not apply focused effort to avoid its pitfalls. Our storytelling minds, which evolved to survive in a localized hunter-gatherer environment, are being exposed to a globalized and chaotic modern world [6,8]. Despite our brains and minds becoming more complex with evolution, our working memory remains limited. We can simultaneously keep in mind about seven, plus or minus two, pieces of information , which is enough to forage mushrooms, but certainly not enough to retain all the globally reported daily news.
Being overstimulated is tiring and we tend to avoid this regime if we can. Among various ways we do that, two common ones are chunking things together  and using what we already know to process the new data . Narrative fallacy is related to both—it is one of the shortcuts in our minds, that allows us to have a less effortful regime of processing, but often at the cost of accuracy.
Narrative fallacy directly applies to psychotherapy and psychiatry. A patient arrives to the consulting room. He isn’t sleeping well. He is constantly anxious. The clinician (because she’s human) would struggle not to wonder whether the two symptoms might be related. She may not say this out loud, but the thought would likely cross her mind. She probably studied years ago that insomnia could be a nonspecific response of the body and the mind and that it could be related to as many as five different factors at the same time. She also is aware that we may simply not knowexactly why the patient’s sleep is fragmented, despite three different examinations by various experts. However, the “not knowing” attitude is hard to maintain when lots of details are coming at her and the patient wants to hear an explanation.
The seasoned clinician may be able to tolerate a lack of connections in the data longer than novice practitioners, but no one is fully immune from the narrative fallacy. I think that instead of maintaining the “not knowing” position until enough data is accumulated many clinicians infer the cause of the problem, such as, in this case, that the anxiety caused the insomnia. Perhaps, they may feel justified to do so by telling themselves that this is only a preliminary hypothesis, which could be changed in the future if necessary.
However, testing such a hypothesis might not be possible , considering that the patient is overweight, barely goes outside, and drinks coffee to battle the afternoon fatigue. Yes, he is anxious, but this is just the tip of the iceberg. Can we isolate anxiety as the cause of insomnia? No, because anxiety, diet, physical activity, and sleep are all intertwined in a complex way.
This vignette simplifies things a great deal. What clinicians hear in the real first session in psychotherapy is not just two data points, but a vast amount of information, communicated explicitly and implicitly. A complex history is shared, traumatic experiences and complicated relationships are introduced, and symptoms are reported. The clinician listens carefully to the patient’s narrative, their speech patterns, and their nonverbal behaviors. This complex tapestry only begins to touch upon the whole experience of the patient.
Through the course of several initial sessions, the therapist comes up with a preliminary understanding of what kind of emotional problem the patient suffers from and why. She also establishes the relevant context of the patient’s challenges, such as his personality style , cultural factors, etc. These data are weaved into an organized “connected” tapestry, called a “case formulation .” The case formulation is a working theory that guides the treatment. It is, put another way, a story.
This story has many benefits and some limitations. One of these limitations is that well-intentioned, thoughtful clinicians often fill in the gaps in the patient’s data with the theories they have studied . The second is that once the gap is filled in, it is often not possible to verify the accuracy of the explanation . Finally, despite their best effort to stay open-minded, clinicians may hold on to their preliminary hypotheses . It is particularly hard to let go of a hypothesis, that was shared with the patient and resonated with him (which does not make this hypothesis accurate ).
How a clinician thinks about the mind can have an impact on the story they tell. A psychoanalytically oriented psychotherapist might infer that the patient’s reported panic attacks are related to the dysregulation of his “separation distress” system . A cognitive-behavioral (CBT) therapist might infer that the patient’s panic episodes are related to his “core beliefs” . A psychiatrist might infer that the patient’s reported depression is related to a low level of serotonin in the patient’s brain. Treatment choices will be guided by these preliminary ideas. The psychiatrist, for example, may prescribe a selective serotonin reuptake inhibitor (SSRI), which is supposed to increase the patient’s serotonin level in the brain. While the SSRI may in some cases bring modest relief in symptoms, let’s consider this intervention from the perspective of narrative fallacy.
The psychiatrist was once taught “a monoamine theory of depression .” She had read many peer-reviewed papers with various statistical results of the SSRI treatment, but none of them contained a convincing causal chain of influences from the molecular level of serotonin to the macro-level of depression. A psychiatrist cannot tell you why exactly the supposedly low levels of serotonin in your brain caused your depression, but that may be her story, her narrative fallacy. The relationship between A and B was made exclusively through the clinician’s theory and nothing else. This is how we tend to create fiction.
Should we abandon all clinical theories then and just treat the patient “intuitively”? No, this would take us back to astrology. What we can do instead is have an honest discussion about the narrative fallacy affecting our work. Simply being aware of its effects and adopting a non-defensive attitude in investigating it is a good start. We can also add checks and balances to our clinical work and introduce necessary updates in our graduate training programs that would help mitigate the risks of narrative fallacy .