Summary: Researchers have discovered how human brains inherently compute like high-powered computers through Bayesian inference, enabling accurate, rapid environmental interpretation. This statistical method combines prior knowledge and new evidence, allowing us to quickly and accurately perceive our surroundings.
This study shows how our brain’s visual system is innately designed to perform Bayesian inference on the sensory data it collects. Such revelations promise advances in areas ranging from machine learning of AI to new therapeutic strategies in clinical neurology.
Key facts:
- Human brains use Bayesian inference, a statistical method, to effectively combine prior knowledge with new data to effectively interpret the environment.
- Research suggests that our brain’s visual system has the basic structure and connections for Bayesian computation.
- The discovery is important in fields ranging from artificial intelligence, where simulating these brain functions through machine learning, to innovative therapeutic approaches in neurology.
Source: University of Sydney
Scientists have confirmed that human brains are naturally wired to perform advanced calculations, much like high-powered computers, to interpret the world through a process known as Bayesian inference.
In a study published in the journal Nature communication, researchers from the University of Sydney, the University of Queensland and the University of Cambridge have developed a unique mathematical model that closely matches how the human brain works when it comes to sight reading. The model had everything needed for Bayesian inference.
Bayesian inference is a statistical method that combines prior knowledge with new evidence to make intelligent predictions. For example, if you know what a dog looks like and you see a furry animal with four legs, you can use your prior knowledge to guess that it is a dog.
This inherent ability enables people to interpret environments with extraordinary accuracy and speed, best done by simple captcha security measures when asked to identify fire hydrants in a panel of images.
The senior investigator of the study Dr. Reuben Redoux, of the University of Sydney’s School of Psychology, said: “Despite the conceptual appeal and explanatory power of the Bayesian approach, how the brain computes probabilities remains largely mysterious.”
“Our new study sheds light on this mystery. We found that the basic structures and connections in our brain’s visual system are set up in such a way that it can make Bayesian inferences on the sensory data it receives.
“What makes this discovery important is that there is an underlying structure in our brains that allows this advanced form of processing, allowing us to interpret our surroundings more effectively.”
The study’s findings not only confirm existing theories about the brain’s use of Bayesian-like inference, but also open the door to new research and innovation, where the brain’s natural capacity for Bayesian inference can be harnessed for practical applications that benefit society.
“Our research, while largely focused on visual perception, holds broad implications across the spectrum of neuroscience and psychology,” said Dr Ridoux.
“By understanding the fundamental mechanisms the brain uses to process and interpret sensory data, we can pave the way for advances in areas from artificial intelligence, where mimicking such brain functions to machine learning, can revolutionize clinical neurology, potentially offering new strategies. Therapeutic Interventions in the Future.”
The research team, led by Dr William Harrison, made the discovery by recording brain activity while passively viewing a display engineered to elicit specific neural signals associated with visual processing. They then created mathematical models to compare a spectrum of competing hypotheses about how the human brain perceives vision.
About this neuroscience research news
Author: Philip Ritchie
Source: University of Sydney
Contact: Philip Ritchie – University of Sydney
Image: Image is credited to Neuroscience News
Original Research: Free access.
“Neural Tuning in the Human Visual System Instantiates Prior Expectations” by Ruben Redoux et al. Nature communication
Abstract
Neural tuning instantiates prior expectations in the human visual system
Perception is often constructed as a process of active inference, whereby prior expectations are combined with noisy sensory measurements to infer the structure of the world. This mathematical framework has proven critical to understanding perception, cognition, motor control, and social interaction.
While theoretical work has shown how priors can be computed from environmental statistics, their neural instantiation can be realized through multiple competing encoding schemes.
Using a data-driven approach, we here extract the brain’s representation of visual orientation and compare it with simulations from different sensory coding schemes. We find that the tuning of the human visual system is highly conditional on stimulus-specific variation that is not predicted by previous proposals.
We further show that the adopted encoding scheme effectively embeds environmental priors for natural image statistics in sensory measurement, providing the functional architecture necessary for optimal inference at early stages of cortical processing.