New Theory Of Consciousness Suggests The Brain Uses Analog Computing To Create Awareness And Experience

“These connections describe the detailed content of thought, the outside world, details of motor acts, and also store all the memories in your brain,” says study author Earl Miller from the Massachusetts Institute of Technology. 

“But something's got to organize that. It's like having a gazillion sheep and somehow you’ve gotta all herd them together,” he told IFLScience.

Traditionally, neuroscientists have worked on the assumption that the brain continually tweaks the activity of every individual in order to generate the constant ebb and flow of thoughts, memories, and shifting awareness that define our consciousness. 

Yet as Miller and his colleagues point out, such a model is impossibly complex and hugely inefficient.

“It’s like fans in a sports stadium doing the [Mexican] wave,” he says. “Imagine you tried to create that wave by influencing people one at a time - it'd be impossible. But the influence of the wave itself just naturally organizes the crowd, and that's what we're saying brainwaves do.”

The idea is therefore that as brainwaves sweep across the cortex, they create ripples of coordinated neuronal activity that then give rise to our conscious experience. And as these waves interact with each other, they generate new wave patterns and therefore enable the ever-changing constellation of thoughts that occupy our heads. 

“Those new wave patterns are new brain states,” explains Miller.

In this sense, the process mimics analog computing, which Miller describes as “math done by waves coming together.” For instance, as different brainwaves meet, they either double up if their peaks happen to coincide, or cancel each other out when a peak encounters a trough.

As this occurs, the activity of huge numbers of neurons is instantaneously modulated in a highly organized way. “That's something digital computing cannot do. Digital computing breaks everything down to one step at a time,” says Miller.

Summarizing this model, the study authors write that while the content and structure of our consciousness is determined by the electrical signals emitted by neurons, “moment to moment computation can be carried out by evolving wave patterns rather than by spikes alone.”

“Consciousness, then, emerges when these dynamic wave patterns bring the cortex into an organized, globally integrated state, one that naturally links and influences widespread activity,” they continue.

Moreover, the entire process is bi-directional and self-perpetuating, since the spiking of neurons due to external stimuli generates new electrical oscillations, which then spread across the brain as waves. As they do so, they organize the spiking of larger sets of neurons, creating new wave patterns that interact with others to generate entirely novel brain states.

“The efficiency of analog computation could explain how the brain performs vast amounts of computation while consuming only about 20 watts, roughly the power of a dim light bulb,” write the study authors. “By contrast, modern digital AI systems require enormous energy.”

The study is published in the Journal of Neuroscience and a preprint is available online here.