Thin-Slicing: Decisions And Intuition

I guess some people have started looking at thin-slicing as a means to achieving optimal quality in decision making. Let’s take a look at where the social research has brought us. Malcolm Gladwell talks about it in his book, Blink. Of course, when the lay population looks at the implications of such an ability they go, “wow, that’s cool… since it’s scientific, let’s use it on everything”.

Scientific basis, yes. Scientific decision-making? Hmm… not quite.

I did a little bit of digging up on the original authors of the paper “Can intuition improve deception detection performance?” by Justin S. Albrechtsen, Christian A. Meissner and Kyle J. Susa, just to see what they had to say. Of course, we need to understand that we are looking at various snapshots, and although the hypothesis is supported, it’s going to be a little more difficult to extrapolate this idea beyond the current form. They themselves acknowledge that “current studies used methodologies that may not be easily implemented outside of the laboratory“, signalling the birth of more research to come in order to support intuition.

Let’s not get ahead of ourselves. In my recent studies on modeling from the domain of organizational psychology, a few researchers are also looking into expertise and the way they process knowledge and information (see the meta-analysis of the literature, “Expertise-Based Intuition and Decision Making in Organization” by Eduardo Salas, Michael A. Rosen and Deborah DiazGranados, 2010, Journal of Management); it is clear that you need expertise in order to rely on intuitions.

~ Salas, et. al.

They go on to outline some of the key mechanisms of expert performance which include the following (refer to the paper for details on the empirical support for each item):

• Large and well-developed knowledge base
• Pattern recognition
• Sensemaking
• Situation assessment and problem representation
• Automaticity
• Mental stimulation

They also outline some of the factors that determine expertise building:

• Deliberate and guided practice
• Self-regulation
• Feedback seeking
• Motivation
• Goal setting

At times, such expertise will break down simply because of a non-conformance of effective team working skills (e.g. collaboration, knowledge sharing, task conflict, competence based trust, etc), which points to additional factors that enables effective team intuitions:

• Learning and adapting
• Clarity of roles and responsibilities
• Prebrief and debrief cycle
• Coordination and cooperation

Salas, et. al. further suggest that “more rigorous studies in the field are needed… [and] brings to bear the methods that are used to investigate expertise-based intuition in the field. Methodology such as think-aloud protocols, narratives, or shadowing may deem insightful in unpacking the black box of intuition. In addition, systematic longitudinal evaluations of interventions designed to develop expertise-based intuition need to be conducted.”

From this, we know that the literature and support for expertise and intuition is weak. It is therefore important that modeling and similar cognitive analysis protocols be used to start the ball rolling. After all, it is important to note that experts are generally as overconfident as non-experts (see Craig R.M. McKenzie, Michael J. Liersch and Ilan Yaniv in their 2008 paper “”Overconfidence in interval estimates: What does expertise buy you?”), and people often expect experts to be less overconfident!

So… am I an expert in experts? Not as much as I want to be, even though many people might deem me to be!

Am I going to suggest that thin-slicing is an approach that everyone jumps on? Well, a useful representation in application could be useful, but be wary of the individual who says you can master it simply because the support for this is limited, and also because, well, you yourself need to become an expert.