Different mechanisms are responsible for underconfidence

We make a myriad of decisions every day, and the confidence we experience in those decisions affects how we act on those choices. But what actually causes us to doubt our decisions?

A new study shows that self-doubt arises in different ways, depending on whether you are a woman or a man and whether you are prone to anxiety or not.

‘Underconfidence is not a phenomenon with a single cause. People may end up with the same pattern of self-deprecation, but via completely different mental processes,’ says Sucharit Katyal, postdoc at the Department of Psychology, University of Copenhagen.

Together with Stephen M. Fleming (University College London), he is author of the study, which has been published in the scientific journal Psychological Medicine.

Two paths to doubt

The researchers analysed data from 1,447 people and developed a dynamic model of how self-confidence changes over time after one makes a decision. They conclude that anxiety and gender in particular affect self-confidence in opposite ways:

• Anxiety: The more time a person with anxiety spends reflecting on their choices, the more their self-confidence declines. Negative thoughts accumulate and reinforce doubt.
• Gender: Women are typically less confident than men immediately after making a decision – but the difference diminishes when they have time to think things over. Extra reflection can therefore help to overcome gender-related bias.

"For women, it can be beneficial to take time to consider major decisions. For people with anxiety, on the other hand, it is better to trust your first instinct," explains Sucharit Katyal.

New knowledge can improve guidance

The study suggests that we should tailor interventions to help the groups most prone to self-doubt:

For people with anxiety: Trust your first gut feeling – extra time can give rise to pointless self-criticism. For women: Take time to reflect – it can boost your self-confidence and make you more certain of your good choices.

‘By uncovering the mechanisms behind these biases, we can design targeted interventions. It's basically about interrupting negative self-assessments in people with anxiety – and promoting reflection to reduce gender differences,’ concludes Sucharit Katyal.

The study is entitled “Gender and Anxiety Reveal Distinct Computational Sources of Underconfidence”. It can be read here.