source: sciencedaily ai: scientists discovered the brain doesn't make decisions the way we thought
level: research
scientists at the university of illinois urbana champaign found that decision-making begins in early sensory brain areas, not just in higher regions. they recorded neural activity in mice navigating a virtual reality corridor and making perceptual choices. the primary somatosensory cortex, an early sensory processing area, showed decision-related activity influenced by feedback from higher brain regions. this challenges the traditional view that information flows one way through a hierarchy before decisions are made.
the research suggests the brain uses rapid feedback loops between regions, allowing continuous communication. this dynamic organization may explain how biological intelligence performs complex tasks with far less energy than current ai systems. the study, published in pnas, does not provide a direct blueprint for ai but offers architectural insights. the team plans to study the timing of these signals and develop new measurement technologies to understand how feedback loops coordinate processing.
understanding this bidirectional processing could inspire future ai architectures that are more capable and energy efficient. current ai, like convolutional neural networks, often relies on feedforward processing. incorporating feedback mechanisms similar to the brain's might improve decision-making in artificial systems. the researchers emphasize that learning from evolution's design could help address limitations in ai, particularly in complex decision tasks.
why it matters: this insight into brain feedback loops could guide the design of ai systems that make better decisions while using less power.
source: sciencedaily ai: scientists discovered the brain doesn't make decisions the way we thought