The example given of a complex mental task relying on complex data conversion is the exercise of writing one's name in such a way that it can be read in a mirror. The brain has all the visual information it needs and everyone knows how to write their name. However, this task is very difficult for most of us because the brain must perform an unfamiliar mental conversion: using what it sees in the mirror to guide the hand precisely enough to write upside down.
Performing mental transformations of information (or data conversion) induces variability : The researchers set out to explore this type of mental conversion in their study: participants were asked to perform 3 different tasks with different degrees of mental transformation required. The experiment shows that in the case of a task requiring difficult data conversion, participants optimize their performance by using the same strategies used to overcome noise in sensory perception. For example, in a line drawing task, in which participants have to draw lines of 7.5 to 15 centimeters, depending on the length of the original line, participants tend to draw lines of longer length. close to the average length of all lines. This allows them to have more precise tracings.
The brain interprets the data:Neuroscientists have known for many years that the brain does not accurately reproduce “what the eyes see or what the ears hear”. It has to deal with “background noise” linked to random fluctuations in electrical activity in the brain. This background noise can come from uncertainty or ambiguity about what we see or hear. This uncertainty also comes into play in social interactions, when we try to interpret other people's motivations or when we recall memories of past events. Previous research has revealed multiple strategies that help the brain compensate for this uncertainty. Using a framework called "Bayesian integration" or a probability-based model, the brain combines several potentially conflicting pieces of data and weights them according to their reliability. If this data comes from 2 different sources, it will rely more on the one that seems more credible. But that's not all: in this model, the brain also takes into account its past experiences. The example is given of finding a switch at night, which is based on past experience of locating the switch in question.
The brain "remembers" past experiences: a complex task that requires a more difficult mental transformation and therefore creates additional uncertainty and variability for the brain induces the brain to rely on its past experiences: "you show bias toward what you're good at, in order to compensate for that variability," comments lead author Mehrdad Jazayeri, professor of life sciences and fellow at MIT's McGovern Brain Research Institute. This strategy of recalling past experiences actually improves overall performance.
Reliability of the source and lessons from past experiences, these 2 strategies seem to work together to perfect the adaptation of the brain in favor of a particular result. This adaptation in the form of regression to the mean helps improve our overall performance by reducing variability and uncertainty. The researchers then hypothesize that, when one becomes very efficient in a task that requires complex calculations, the background noise is reduced and becomes less detrimental to overall performance.
The brain then trusts its own calculations more and stops getting closer to “averages”.
