When Can We Trust an Expert’s Intuition?

What Malcolm Gladwell Missed in “Blink”

I recently read the book “Blink, the power of thinking without thinking”, by Malcolm Gladwell. A book about the magic of experts’ intuition.

The takeaway of the book is: the experts cannot be fooled easily because of their “thin-slicing” ability, that is, they can judge and find patterns in events based on a narrow window of experience, thanks to their intuition. On the other hand, experts can be easily (catastrophically) fooled, as in the case of “Warren Harding Error” (chapter 3).

The book left me with big questions on expertise and intuition, so I had to dig deeper. I am convinced that experts’ intuition is not always reliable, what I found is that it depends on the domain of expertise and the expert’s experience in the domain.

In this post I summarize 2 approaches to intuition:

• Naturalistic Decision-Making (NDM)
• Heuristics and Biases (HB)

The confrontation of these 2 approaches results in 3 conditions for reliable intuition, abbreviated as REF:

• Regularity
• Exposure
• Feedback

For starters, we define intuition as the ability to detect a pattern and solve problems rapidly, without relying on conscious reasoning.

Can you be an expert in a field ruled by random events?

Two Approaches to Intuitive Expertise: NDM vs. HB

The first approach to intuitive expertise is naturalistic decision-making (NDM). It is focused on the decisions made in real life by experienced people such as firefighting commanders, nurses, chess-masters, etc. NDM aims at demystifying the intuition by searching for the cues that led to the expert judgment.

The heuristics and biases (HB) approach argues that human intuition is often flawed. A statistical prediction algorithm, fed with the same information available, can perform better than human experts. The major examples of HB are clinical judgments. For instance, human judges can reach different conclusions for the same case information, when presented on different occasions. The same biases intervene when we decide to pick the best candidate for a job.

NDM focuses on the successes of the experts, and HB on their errors. But they don’t seem to address the same experts! The confrontation of these approaches gives the following conditions for reliable intuition.

The 3 conditions for Expertise Acquisition

To have reliable intuitive expertise you need to be a true expert, so know when you don’t know. Here are the 3 conditions to become a reliable expert.

Regularity/Predictability

You need to be in an environment that has rules in a way to allow for learning. You should be able to recognize regular patterns and learn from them. Some domains provide such an environment, such as medicine, computer science, firefighting, etc. Some domains are ruled by unpredictable events, which are extreme and rare but have a high impact (a.k.a. black swans). In such domains, it is impossible to develop intuition. You cannot be an expert in politics or the stock market for instance.

Exposure/Practice

Reading about something, or attending classes and conferences is not enough. You need to experiment yourself, and to have direct exposure to the subject-matter of expertise. Deliberate practice is a key element to develop reliable intuition. The longer your experience the better your judgment.

Feedback

Fast and unambiguous feedback is essential for expertise. You need to see the impact of your action as fast as possible, unequivocally, to improve your performance. If the feedback is delayed, the process of expertise acquisition (learning) becomes slower. In some domains, feedback is not so relevant to the action since other factors impact the output. Take for instance decision-making in big companies; it is often difficult to tell exactly what decision led to a consequence. Another technical domain where unambiguous feedback is challenging is Psychiatrie.

Finally, personal talents and skills play a significant role in accelerating the expertise acquisition process.

References

• Kahneman, Daniel, and Gary Klein. “Conditions for intuitive expertise: a failure to disagree.” American psychologist 64.6 (2009): 515.
• Taleb, Nassim Nicholas. The black swan: The impact of the highly improbable. Vol. 2. Random House, 2007.
• Gladwell, Malcolm. Blink: The power of thinking without thinking. (2006).
• Kahneman, Daniel. Thinking, fast and slow. Macmillan, 2011.