How strongly does our metabolism influence thinking? This question has been examined by Dr. Philipp Haueis, a philosopher of science at Bielefeld University, in a recent publication in the journal Behavioral and Brain Sciences. Together with his co-author David J. Colaço, PhD, from Ludwig-Maximilians-Universität Munich, Haueis explores how the brain’s metabolism shapes our cognitive abilities—processes such as memory, perception, and attention.

“We wanted to find out what happens when you take the brain’s energy demands seriously,” says Haueis. “The brain is not an abstract computer but an organ that needs energy and has limits. These limits influence how we think, remember, and make decisions.”

The human brain accounts for only about two percent of body mass but consumes around 20 percent of the body’s energy. Despite this high consumption, it operates far more efficiently than modern computers. Yet many scientific models designed to explain perception, memory, or attention hardly take these energetic constraints into account. This is precisely where the study comes in.

The researchers show that metabolism—the chemical processes that supply and consume energy in the body—plays two central roles in the study of thinking. First, it can help determine whether existing cognitive models (theoretical descriptions of mental processes) are biologically plausible at all. Any model that requires more energy than the brain can provide is unrealistic. Second, knowledge about metabolism can be used to develop new models. These models may offer new insights into previously overlooked relationships between brain structure and information processing—for example, how neural networks use energy to learn efficiently.

The publication is the first to systematically summarize how metabolic insights can be used for modeling mental processes and why this is philosophically necessary. Through the journal’s “Open Peer Commentary” format, in which researchers from various disciplines publicly comment on the work, a broad debate is expected to emerge.

This debate reaches far beyond specialist communities. The study has far-reaching implications for understanding mental effort, for theories of computation in the brain, and for questions about how artificial intelligence compares to biological intelligence.

“If we understand that thinking costs energy, we also better understand why attention is limited and why machine learning—unconstrained by biology—follows different paths,” says Haueis. Thus, the study not only contributes to basic research but also provides impulses for societal discussions about AI, energy efficiency, and the nature of intelligence itself.

The study was conducted within the Institute for Studies of Science (ISoS) at Bielefeld University, which received the status of a central academic unit in May 2025. ISoS brings together interdisciplinary research on science, medicine, and technology and examines how scientific practices are embedded in societal contexts.