Making an accurate decision requires one to estimate the certainty of an anticipated outcome, given existing evidence. Currently, it is believed that activity in prefrontal cortical regions like the orbitofrontal cortex reflects the accumulated evidence of all possible outcomes, which then prompts the choice of the option with the most compelling evidence. It is assumed that the choice with the most evidence is therefore the most certain option.
Two prominent measures of certainty are the speed and energy put into moving after a decision is made, i.e., the less time and the more forceful a movement is, the more certain (and more accurate) the decision tends to be. Conversely, uncertain decisions take longer to make, which can be a reflection of ambiguous evidence, and choice movements are generally slower and less energetic. Typically, the accumulation of evidence and time to make a decision are confounded. In their study, Kiani, Corthell and Shadlen tease apart these variables to show how evidence and decision time affect the certainty of a decision independently.
In this experiment participants were asked to fixate on a central point. Then, two horizontal bars were displayed at the top and bottom of the screen. Finally, a group of small white dots were presented on the screen with dynamic random motion, each dot moving independent of other dots. By purposely moving their gaze (saccade), participants decided whether the random dots were moving upward or downward (toward the corresponding horizontal bar) and indicated how certain they were of their decision. Certainty was determined because a saccade to the left-most part of the bar identified completely uncertain, with ‘certainty’ increasing from left to right until the right-most end of the bar reflected a completely certain decision.
Surprisingly, subjects used the length of time to make a decision as a measurement of certainty of that decision. Kiani and colleagues discovered that while longer reaction times are associated with noisier data and greater uncertainty, people also use the length of time they’ve taken before deciding as a measure of how certain they are of a particular choice.
This study verifies that choice certainty is related to task accuracy and shows that deciders use the length of time to make a decision as a decision variable. Further research is needed to test how these principles of decision certainty relate to more complex and abstract choices rather than simpler perceptual binary-type decisions.
- Kiani R, Corthell L, Shadlen MN (2014) Choice Certainty Is Informed by Both Evidence and Decision Time. Neuron 84(6):1329-1342. doi: 10.1016/j.neuron.2014.12.015