Imperfect execution of the brain of mathematically optimal perception

Human perception is based on mathematically optimal principles, but the brain applies these principles imperfectly, suggests a new research by Elina Stengård and Ronald van den Berg of the University of New York. Uppsala in Sweden. They present their discoveries in PLOS Computational Biology.

The human brain uses imprecise sensory inputs to determine truths about the environment. Previous research has suggested that human perception is "Bayesian", meaning that the brain reports the uncertainty of sensory observations in a mathematically optimal way. However, some of these studies have been criticized mathematically and further research suggests that the brain is inherently inaccurate at the neural level.

To remedy these problems, Van den Berg and Stengård presented 30 volunteers with a series of perception tests. These tests consisted of determining whether the ellipse shapes appearing on a screen were inclined clockwise or counterclockwise with respect to the vertical. Different tests have incorporated sensory uncertainty in different ways, such as varying degrees of lengthening the shape of the ellipse, distractions in the form of near ellipses and a short time of ## EQU1 ## 39; display the ellipse on the screen.

The researchers then analyzed their results against a series of different mathematical models. They found that the data is better explained by a Bayesian model at the base, but also subject to imperfections. This model outperformed both an optimal Bayes model and all non Bayesian models tested.

"Our findings suggest that human perception is designed on optimal strategies, even though the execution of these strategies by the brain seems to be imperfect," said Van den Berg. "This new concept provides a theoretical ground for understanding the seemingly opposite literatures of optimal models and heuristic models."

Further research is needed to identify the causes of apparent imperfections in the decision-making process during ellipse perception tests. Future research could also test whether the imperfect Bayesian model can account for human behavior in other types of perception tests and in higher-level cognitive decision-making tasks.