How to reduce cognitive biases that hurt your problem-solving

February 5, 2019 9:04 am

Cognitive biases are more likely to have a detrimental impact on your problem-solving and decision-making when you are dealing with complexity or when the circumstances are changing rapidly.

Daniel Kahneman, the Noble prize-winning scientist who first described the concept of mental bias, is pessimistic about ‘training or conditioning’ to overcome these biases.

In keeping with Kahneman’s original prediction, the most recent research suggests that we require some type of external intervention to avoid falling into mental traps.  Some studies have shown that when people are cued that the challenge is likely to tap into a bias, and they have sufficient time and techniques to assist their thinking, they are better able to work around their natural biases. We also know that other people are better able to detect our biases than we are able to see them ourselves.  The natural differences in perception, emphasis and perspective create a window of ‘collective objectivity’, that is – IF we are open to being influenced by diverse views.

Research has also shown that implementing formal ‘thinking’ tools/techniques helps us ‘slow down’ our thinking, forcing us to think more deeply and completely – thus avoiding succumbing to these natural biases.

3 Tips for mitigating the risk of bias

  1. One technique we have found helpful for our clients is insisting that decision-makers systematically collect a variety of data including data that refutes their initial thinking.
  2. Carefully evaluating the robustness of each piece of data also helps – for example, statistical data is more robust than the one-off experience of the decision maker.
  3. Another technique involves seeking the perspective of people who are different (e.g. different values, needs, world-views) as this helps to broaden thinking and often results in better judgements and outcomes.

 

Photo credit Hans-Peter Gauster on Unsplash

Leave a Reply

Your email address will not be published.