Our brains can make fast decisions, and accurate decisions — but not at the same time

When it comes to making decisions on the fly, we sacrifice accuracy for speed. It's true for humans, and it's true for most other species — rapid fire answers are less likely to be correct. Called the speed-accuracy tradeoff (SAT), and while we know it's a thing, the scientific basis for it has been poorly understood. However, a new paper is claiming to have unlocked how the brain handles this — and it's turning the traditional model on its head.

The problem is, while it's easy to test speed and accuracy with humans, it's not so straightforward to analyze what's going on on a neuronal level. Conversely, with monkey models, we can accurately monitor signals from individual neurons, but can't get them to switch back and forth from high-speed to high-accuracy decision making.


A new paper published in the journal Neuron has solved this conundrum. Richard Heitz and Jeffrey Schall of Vanderbilt University trained macaque monkeys in two different games — one which rewarded high speed choices, and allowed for some incorrect guesses; the other which only rewarded the right answer with no regard for time.

In a press release, Heitz described the two games:

Our tests are like two different game shows. One – call it Fast Fury – is like Jeopardy. In order to answer a question you must be the first to hit the buzzer. Buzzing in and answering incorrectly is bad, but being slower than the other contestant means you will never earn a reward. That is much different than the second game show – call it High Stakes Showdown – where buzzing in at any time gives you the opportunity to answer a question, but being wrong results in a serious penalty.

The first thing you see is that neural activity of the player of Fast Fury jumps up even before the question is read. The subjective experience of getting ready that we all experience appears to be reflected in the background activity of neurons in prefrontal cortex.

Conversely, that neural activity drops to a very low level during the mode that requires accuracy over speed. The data suggests that when faced with a problem, activity of the prefrontal cortex neurons are amplified when speed is required, and suppressed when accuracy is paramount.

This is contrary to the current theory, which thinks that our brain uses the same processes for all types of decision. The data from this study seems to suggest that our minds deal with information in a different way, depending on what we want of it. Schall said:

What this means is that identical information presented to the brain is analyzed differently under speed stress than under accuracy stress.

If this model holds up under further scrutiny, it seems to show that our brain has a separate mechanism for when running at high speed — which could mean an awful lot for how we understand decision making in the human brain.



Reminds me a little of our old phrase in designing networks in IT . . . Fast, Cheap, Reliable . . . you get to pick two . . .