The Dunning-Krugger Effect in Learning Computer Science

From the beginning...

Every so often, a colleague asks me whether I have ever felt — or still feel — that sensation that the more we study a technology, the more we feel we know nothing about it. The feeling is very common among Computer Science professionals and students, especially while learning some new technology or discipline. Perhaps it comes from the nature of the field itself, which is in constant flux and therefore demands continuous study and renewal. Dunning-Krugger named this feeling the “Valley of Despair” [Image 1].

The effects of not knowing

In “Unskilled and Unaware of It: How Difficulties in Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments” [1], the researchers Justin Krugger and David Dunning studied how we come to recognize our own skills, and how that recognition plays a meaningful part in improvement and growth in any field of knowledge. The research gave rise to the concept of the “Dunning-Krugger Effect”:

The Dunning-Krugger effect is the phenomenon whereby individuals with little knowledge of a subject believe they know more than others who are better prepared, leading them to make poor decisions and reach unwarranted conclusions; it is their very incompetence that robs them of the ability to recognize their own mistakes.[1]

The study ran a series of experiments whose published results confirmed that “ignorance more frequently begets confidence than does knowledge.”

“If you’re incompetent, you can’t know you’re incompetent (…) [T]he skills you need to produce a right answer are exactly the skills you need to recognize what a right answer is. In logical reasoning, in parenting, in management, in problem solving, the skills you use to produce the right answer are exactly the same skills you use to evaluate the answer. And so we went on to see whether the same conclusion might hold in other areas. And to our surprise, it was very true.” [2]

In Computer Science, as in Medicine, specializing in a given discipline starts from an exceedingly broad generalist foundation. Even general practice, for instance, requires wide knowledge across a vast number of disciplines. The same holds true in Computer Science, and it is precisely this that so often produces the feeling of the “Valley of Despair.”

Dunning-Krugger describe this feeling — the one Socrates had already captured in the phrase “I know that I know nothing” — as a fundamental part of learning. Recognizing one’s own limitations in a given field of science or discipline naturally leads one to seek a deeper understanding of it. And yet, the further one advances in the study of a subject, the stronger the sense may become that the “Valley of Despair” keeps growing deeper.

But how do I know when I actually know something?

Impostor syndrome

Just as Dunning-Krugger found that ignorance breeds illusory superiority, mastery invariably breeds illusory inferiority: the student of a given subject, aware of how deep that discipline runs, may come to doubt their own abilities.

(…) These people suffer from illusory superiority. Conversely, real competence can weaken self-confidence, and some highly capable people may suffer from illusory inferiority, believing they are not as capable as they are, underestimating their own abilities, and even coming to believe that other, less capable individuals are just as capable as they are, or more so. This other phenomenon is called impostor syndrome [3].

How to learn anything

To help with this process of self-knowledge — of gauging how far our mastery of a subject really extends — one of the most interesting techniques I have had the chance to apply is the “Feynman Technique,” created by the physicist Richard Feynman. Feynman was a twentieth-century American theoretical physicist, one of the pioneers of quantum electrodynamics, known for his work on the path integral formulation of quantum mechanics. [4]

Beyond physics, Feynman was known for his ability to learn — and eventually master — different fields of science and of knowledge in general. Besides being a physicist, he was an accomplished visual artist and a musician — his favorite instrument was the bongo — among many other things. To learn a discipline, and to judge whether he had truly mastered it, Feynman devised a simple technique that proved extraordinarily effective: in four steps, it lets you learn about anything.

The Feynman Technique [5] boils down to this sequence:

1. Choose a concept
2. Write it out as if you were teaching it to a child
3. Go back to the subject and research it
4. Review and simplify even further

Simple as it is, the process described in the Feynman Technique — when combined with an understanding of what Dunning-Krugger describe regarding the ongoing need to question one’s own knowledge, and joined to the practical experience of applying it — will invariably lead to mastery and deep knowledge of a subject.

Final thoughts

So the feeling of the “Valley of Despair,” which I still experience at times even after 20 years in the field, as do my colleagues in industry and academia, is entirely natural — and, I would say, healthy. Knowing that one knows nothing, as Socrates so aptly put it, is the foundation of science and of human knowledge.

As I often say, what our times lack is a scientific spirit. The plain and simple method of scientific inquiry — observation, the pursuit of knowledge — is what we are missing. Had we been taught from an early age to embrace criticism and to change our minds in the face of new evidence, our situation as a society and as human beings would be very different.

References

[1] Justin Krugger; David Dunning (1999). «Unskilled and Unaware of It: How Difficulties in Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments». Journal of Personality and Social Psychology. 77 (6): 1121–34. PMID 10626367

[2] Errol Morris (2010). «The Anosognosic’s Dilemma: Something’s Wrong but You’ll Never Know What It Is (Part 1)». Available at: https://opinionator.blogs.nytimes.com/2010/06/20/the-anosognosics-dilemma-1/?_r=0.

[3] Fehlhaber, Kate; Fernanda Sucupira (translation) (June 2, 2017). «O que os sabe-tudo não sabem, ou a ilusão da competência». Nexo Jornal. Available at: https://www.nexojornal.com.br/externo/2017/06/02/O-que-os-sabe-tudo-n%C3%A3o-sabem-ou-a-ilus%C3%A3o-da-compet%C3%AAncia.

[4] Richard Feynman, biography. Available at: https://pt.wikipedia.org/wiki/Richard_Feynman.

[5] 4 passos para aprender tudo que você quiser, segundo um Nobel da Física. BBC News Brasil, July 12, 2016. Available at: https://www.bbc.com/portuguese/geral-36750825.