Challenges of Brain Research

There has been and continues to be a prevalent opinion in neuroscience that fundamental insights into how the brain works can be gained by generating large datasets and advanced algorithms. However, this view has been shaken in recent years by three groundbreaking articles that vividly describe the conceptual challenges of brain research. Yuri Lazebnik criticizes the lack of formal language in the life sciences and the inability to translate complex biological processes into mathematical models. Joshua Brown emphasizes the need for a unified mechanistic framework for neuroscience. Eric Jonas and Konrad Kording conducted an experiment in which they emulated the MOS 6502 microprocessor and applied neuroscientific methods to analyze it. Despite interesting results, these methods could not provide a mechanistic understanding of how the microprocessor works. A system is considered understood when hypotheses about its structure and function are formulated and tested, with these hypotheses being derived from a more general theory. According to David Marr, a system should be understood at three levels—computational, algorithmic, and implementational—to fully capture its function and behavior. To prove that the brain is understood, an artificial system would have to be developed that performs the same inputs, transformations, and outputs of information as the biological model.