Category: Neurotech

How the New Science of Brainwaves Reads Minds, Tells Us How We Learn, and Helps Us Change for the Better

In Electric Brain, neuroscientist R. Douglas Fields explores the hidden electrical dimension of the human brain. The book explains how rhythmic patterns of neural activity—known as brainwaves—shape perception, thought, emotion, and consciousness. Rather than being simple by-products of neural firing, these oscillations help organize communication across the brain, synchronizing networks of neurons so that information can move efficiently between regions.

Fields traces the scientific discovery of brainwaves back to the early twentieth century, when German psychiatrist Hans Berger performed the first human electroencephalogram (EEG). Berger demonstrated that the brain produces detectable electrical signals that change with mental activity. Opening the eyes, concentrating on a task, or experiencing emotion all alter the rhythm of these electrical oscillations. Berger’s work revealed for the first time that cognitive and emotional states could be directly monitored through measurements of brain activity.

The book explains how different frequency bands—delta, theta, alpha, beta, and gamma—are associated with different mental states. Delta waves dominate deep sleep, alpha waves appear during relaxed wakefulness, beta waves accompany active thinking, and gamma oscillations are linked with complex cognitive integration. These rhythms coordinate neural circuits much like sections of an orchestra playing together and separating again as needed. Through this synchronization, the brain binds information from different regions into coherent perception and thought.

Fields also examines the growing technological ability to measure and influence brain activity. EEG recordings, quantitative EEG analysis (qEEG), and neurofeedback techniques allow researchers to detect abnormal electrical patterns associated with neurological or psychological conditions. In neurofeedback therapy, patients receive real-time feedback on their brainwave activity, allowing the brain to gradually adjust its own patterns without drugs or surgery.

The electrical nature of the brain has also opened new possibilities in medicine and technology. Techniques such as deep brain stimulation and transcranial direct current stimulation (tDCS) can modulate neural activity by applying small electrical currents to specific brain regions. Research suggests that abnormal oscillations—particularly excessive beta activity—play a role in disorders such as Parkinson’s disease, and modifying these rhythms may reduce symptoms. Brain-computer interfaces are also emerging that allow people to control prosthetic devices or computers directly through neural signals.

At the same time, Fields emphasizes that the science of brainwaves remains incomplete. Many observed relationships between oscillations and mental processes are correlations rather than proven causal mechanisms. Scientists continue to debate whether brainwaves actively drive neural computation or simply reflect the underlying activity of neurons.

Ultimately, Electric Brain portrays the mind as a dynamic electrical system—a constantly shifting landscape of synchronized rhythms and interacting signals. By revealing the brain’s electrical language, Fields argues, neuroscience may gain deeper insight into consciousness, neurological disease, and the future integration of human brains with emerging technologies.

Comparing the energy of the human brain and artificial intelligence

It is often remarked, sometimes with unease, that artificial intelligence consumes enormous amounts of energy while the human brain runs on little more than the power of a small light bulb. The contrast is striking. But before we marvel too quickly at the efficiency of the brain, we should pause and ask whether we are making a fair comparison.

What exactly is the human equivalent of an AI system answering a question?

The simplest comparison would be a single answer. You ask a question and either a person or an AI replies. But that framing quietly hides a large part of the machine’s cost. An AI answer depends on a vast training process that took place earlier in large data centres. The electricity used to train the model does not appear in the moment of answering. If we compare a human reply with a single AI response, we are ignoring the energy required to build the machine’s knowledge in the first place.

Another possibility is to compare a trained AI model with a trained human expert. A PhD, for example, represents decades of learning. The AI equivalent is its training phase, during which the model absorbs enormous amounts of text and data. Both systems require a long investment before they are able to produce sophisticated answers.

We could widen the frame further. AI models are trained on the accumulated output of millions of people: books, research papers, code, and conversations. In that sense an AI model resembles a compressed form of collective knowledge. The human comparison might not be a single expert at all, but something closer to a research community.

There is an even deeper perspective. Human intelligence itself is the product of hundreds of millions of years of evolution. If we tried to account for the energy required to evolve brains capable of language and reasoning, biological intelligence would hardly look inexpensive.

For practical purposes, however, the clearest comparison is this: a trained AI model and a trained human expert.

Once we make that comparison, the numbers become interesting. Training a frontier AI model today can require several million kilowatt-hours of electricity. The cost is paid up front during training, after which the model can generate answers at relatively low additional cost.

The human brain, by contrast, runs on about twenty watts of power. Over a full day that amounts to roughly half a kilowatt-hour. Within that modest energy budget the brain performs perception, memory, learning, language, and reasoning.

The real puzzle is not that AI systems use a great deal of energy. Modern computers were built for speed and scale, not for metabolic thrift. The deeper puzzle is why the brain is so efficient.

Evolution had a strict energy budget. Brains that wasted energy did not survive. Neurons fire sparsely, meaning most of the brain is quiet most of the time. Memory and computation happen in the same place, reducing the need to move information around. And the brain relies heavily on prediction, focusing effort on what changes rather than recalculating everything continuously.

The result is a form of intelligence that runs steadily on the power of a small light bulb.

Perhaps the real surprise is not that artificial intelligence consumes so much energy, but that human intelligence runs on just twenty watts.

Neurotechnology is the use of light (such as infrared or photobiomodulation), sound (as in neurofeedback or binaural beats), vibration or ultrasound, electricity or magnetism (as in EEG, tDCS, or TMS), and plant compounds (including psychedelics), sometimes combined with feedback or digital control systems, to measure or modulate activity in the brain, nervous system, and body. It is used in meditation and medicine to treat conditions such as depression or Parkinson’s disease, and to enhance learning, attention, and states of consciousness.