Quick Decisions May Reveal Stronger Biases, Study Finds

Quick decisions and initial biases

Do you find it difficult to make decisions?

There is no need to rush your answer. A new study indicates that people who decide quickly are more likely to hold stronger biases than those who deliberate for longer.

The effect may be more far-reaching still. Fast decision-makers could be inclined to choose options that match their biases, even when the available evidence conflicts with those views.

A team headed by University of Utah mathematician Samantha Linn says the result indicates that slower decision-makers eventually make better choices, while offering insight into the mechanisms behind extremism.

"We show that in groups of agents who only differ in their initial biases, early decisions are made by agents with the most extreme predispositions," the researchers write in their paper. "On the other hand, late decisions do not depend on the initial bias and are more likely to be correct."

Taking a long time to reach a decision is often regarded as a sign of conditions such as anxiety, or simply as a lack of confidence. It may also be seen as a weakness.

However, although many people consider the evidence before making a choice, others begin deciding on the basis of instinctive responses, emotional reactions, incorrect assumptions, or the outcomes of earlier decisions.

Evidence accumulation models and decision-making

Linn and her colleagues set out to examine, from a strictly mathematical standpoint, how bias and the availability of information affect decision-making speed. They used numerical evidence accumulation models, which assess how agents gather and combine information to reach a decision.

The team developed models in which several agents needed to select between two choices, H+ and H-. Within the model, agents received random biases towards either option and were instructed to accumulate evidence.

"Here we ask how individuals' initial biases in a group affect the order and accuracy of their choices," they write. "When is a decision driven by an agent's initial bias as opposed to accumulated evidence?"

Why slower choices may be more accurate

The findings showed that, among groups of agents distinguished only by their starting biases, those with the strongest predispositions made the fastest decisions. This pattern appears even in small groups, though it is particularly marked in larger ones. These biased agents made choices consistent with their original bias, regardless of the information they could access.

Decision time was inversely related to the strength of a bias. Agents that took the longest to decide made choices least affected by their initial bias. Their decisions were also more likely to be correct than those made by snap decision-makers.

"Our decisions are often influenced by information we obtained previously and predilections we develop," the researchers write. "We have shown that these initial biases determine early decisions and have diminishing influence on later decisions."

The result is intuitive. When you have already formed an opinion on something, you do not need to spend much time considering it when faced with a choice. If Cajun food is your favourite cuisine, naturally you will choose it when the other option is hamburgers.

Since this work is entirely mathematical, psychology cannot be inferred from it. The study examines cause and effect without exploring the complex human behaviour involved. Nevertheless, it can provide a foundation for better understanding that behaviour in both small and large groups, and the ways our actions may be shaped by the beliefs we hold.

The findings will be published in Physical Review E.