Thinking (Systemized)

Decision science is helping us understand how the speed at which we make choices can affect how good they are. A new study uses math to show that our first biases are more likely to show up in quick, impulsive decisions. On the other hand, decisions that take longer and are more thought out are more likely to be right and less biased. The study found that "fast deciders," or people who rushed to make decisions, often made choices based on their biases and were wrong about half the time, even when they looked at the facts. The "slow deciders," or people who waited and got more information, were more likely to get over their biases and make the right choice. This conclusion quantitatively supports an intuitive notion: our initial instincts, though occasionally beneficial, may mislead us, particularly in group contexts where the most assertive initial assessments can exert excessive influence. Taking a little more time to think about the evidence can help us make better choices. The 2024 study used a drift-diffusion model of decision-making and proposed ways to make better decisions in real-life situations. The authors assert that it may aid in recognizing instances where group members are excessively influenced by impulsive, biased choices or "groupthink," and when encouraging a moment of contemplation could produce more advantageous results. These ideas are similar to what people used to think about System 1 (quick, intuitive) and System 2 (slow, reflective) thinking, but now they are based on strong math. In short, taking our time with important decisions helps us get past our biases, which leads to better, more reliable choices.

   Sources:

   Florida State University, August 12, 2024. Making decisions quickly or not: The math behind it

   daily science.com

   science daily.com. ScienceDaily, Physical Review E, Vol. 110, Article 024305.

   Linn, S., Lawley, S. D., Karamched, B. R., Kilpatrick, Z. P., & Josić, K. (2024). Quick decisions show biases, but slow ones don't. Physical Review E, 110(2), 024305. ScienceDaily talked about the work.