Lesson 2 of 8 · All lessons in this module
- Cognitive errors versus emotional biasesVideo lesson
- Belief perseverance I: conservatism and confirmation
- Belief perseverance II: representativeness, illusion of control and hindsight
- Information-processing biases: anchoring, mental accounting, framing and availabilityVideo lesson
- Emotional biases I: loss aversion and overconfidenceVideo lesson
- Emotional biases II: self-control, status quo, endowment and regret aversion
- Integrated case: a multi-bias file with 'least likely' and 'not shown' items
- Behavioural finance and market anomalies
Lesson 2 of 8 · 12 min
Belief perseverance I: conservatism and confirmation
Conservatism makes people update too slowly when new information arrives, and confirmation makes them go looking for information that supports what they already believe; both protect an existing view at the expense of an accurate one.
In short
- Conservatism bias: sticking with a prior view or forecast by giving too little weight to new, conflicting information. In Bayesian terms, the prior is overweighted and the new evidence underweighted, so beliefs underreact.
- Conservatism is worst with complex, technical or statistical information, because processing it takes more mental effort.
- Fix conservatism by weighing new information properly: ask 'how does this change my forecast?', update in the spirit of Bayes' rule, and get help interpreting hard data.
- Confirmation bias: noticing and seeking what confirms a belief while ignoring or discounting what contradicts it.
- Confirmation leads to one-sided research, biased screens, under-diversification and oversized holdings of the employer's stock.
- Fix confirmation by actively seeking contrary evidence and corroborating a decision with research from a different angle.
Unlock this lesson free for 7 days
Create a free account to get 7 days free: every lesson, video and flashcard, and the question bank. No card needed.