Heuristics: Can We Trust Our Intuition?
Look at these pictures:
- Are these women working? If yes, what kind of job do they have?
- Where do they live?
- What languages do they speak?
- How educated are they?
- How old are they?
- How wealthy are they?
- Are they married?
- How many children do they have if any?
- Are they well organized?
- Do they have leadership competencies?
- What are their religions?
I can go on asking you tons of questions about these women and you will probably be able to make suppositions for each of them. In brief, you have been able to precisely judge these two women in less than two seconds. To do so, your brain has matched the few elements provided by these pictures with the tremendous quantity of related information present in your brain, from your past personal experiences. This super-fast process makes you have the feeling, for instance, that this woman is a young and rich housewife with basic education and limited individual freedom in a conservative Arabic country.
These are heuristics (i.e. experience-based shortcut techniques for problem solving), learning, and discovery. We usually use the term intuition in our daily life to refer to this way of thinking. Heuristics provide a fast and result-oriented solution to complex problems in our daily life, like quickly judging the person we will interact with. The issue, as for all shortcuts, is that the solution is not guaranteed to be optimal. To come back to our two pictures, the first woman is a young French demonstrator against the Syrian war in the streets of Paris. The second woman is Fatima Al-Jaber, the fifth most influential Arabic woman in the world, COO of the Abu Dhabi Al Jaber Group managing over 50,000 employees and $4.9bn in assets.
Heuristics are based on a Bayesian learning
How do we implicitely make these beliefs based on our own experience? Recent research in cognitive science has agreed on a Bayesian learning model. We simply have an a priori belief about, for instance, the link between wearing a veil and speaking Arabic that we update each time we interact with someone wearing a veil or speaking Arabic. These prior beliefs help us cope with uncertainty, like interacting with someone we see for the first time. To further understand this powerful learning system of the brain, you can look at this more theoretical talk from the University of Edinburgh:
Should we trust heuristics?
Heuristics or intuitions are an effective way to solve rapidly complex problems. Nevertheless, the provided answer is often an approximatin or even false, like our fast judgment of these two young women. It is thus important to learn the ”bad habits” of heuristics to use them adequately.
1. A machine to make convenient shortcuts…
2. A machine to produce (hyper)rationality: Six people are taken at random in the street, one after the other, for a marketing survey. Their gender is written on a paper. What is the most probable sequence? FFFFFF MMFFFF MFMFMF Even if we know that all sequences are equally probable, we will be more surprised by the first one and tend to find a sophisticated solution, while the only explanation is randomness!
3. A machine to provide intentionality
Look at this video and try to imagine what happens to the poor figures. What happened? From very simple moving geometrical figures, we have made a coherent and nice story! Our system 1 loves story-telling, even when there is no story and just randomness!
4. A machine to provide intentionality: the example of the Black Swan theory
As in the Black Swan Theory, our intuitive brain tends to rationalize a random event in the past with the benefits of the hindsight of the present.
References: Thinking fast and slow,D.Kahneman, 2001 Antigrale, Nassim Taleb, 2012 Knill, D. C., & Pouget, A. (2004). The Bayesian brain: the role of uncertainty in neural coding and computation. TRENDS in Neurosciences, 27(12)