New Software Revolutionizes Brain Simulations for Cognitive Tasks

Researchers at the University of Tübingen have developed a groundbreaking software, named Jaxley, that enhances brain simulations to successfully perform complex cognitive tasks. This development represents a significant advancement in the field of neuroscience, allowing for deeper insights into the brain’s functioning and capabilities. The findings were published in the journal Nature Methods in early 2025.

For decades, scientists have aimed to create accurate computer models of the brain to better understand its intricate processes. Traditional simulations often fell short due to their reliance on either oversimplified neuron models or detailed biophysical processes that failed to replicate brain-like tasks. Michael Deistler, the first author of the study, noted, “Either the path is similar to that in the brain, but the result is not, or the result is correct but the process that leads there does not compare with the processes in the brain.”

Jaxley addresses these limitations by enabling brain models to be trained using a technique known as backpropagation of error, commonly used in artificial neural networks. This method allows the software to adjust its parameters repeatedly until the desired output is achieved. By training brain models with this approach, the software can now simulate complex tasks more accurately.

Unlocking the Complexity of the Brain

The human brain involves numerous parameters during task execution, such as neuron size, connection strength, and ion channel quantity. Many of these important factors are difficult to measure, which has hindered progress in creating precise simulations. Deistler explained, “Jaxley can train these non-measurable parameters in brain models. The software repeatedly changes their values, repeatedly readjusts, until the simulation reaches the desired result.”

After the training process, the brain models created with Jaxley demonstrated capabilities such as classifying images and storing memories. Jakob Macke, Professor of Machine Learning in Science at the University of Tübingen and the study’s last author, emphasized the implications of this software: “Thanks to Jaxley, we can now study how neuronal mechanisms contribute to solving tasks.”

The potential applications of Jaxley extend beyond research. In the long term, these advanced simulations could play a crucial role in medicine, offering new insights into neurological diseases and allowing for virtual testing of treatments before they are administered to patients.

With the launch of Jaxley, the University of Tübingen is poised to make significant contributions to the understanding of brain dynamics, opening new avenues for both scientific inquiry and medical advancements. As researchers continue to explore the capabilities of this software, the implications for cognitive science and neurology are vast and promising.

For further details, refer to the published work: Michael Deistler et al, “Jaxley: differentiable simulation enables large-scale training of detailed biophysical models of neural dynamics,” Nature Methods, 2025, DOI: 10.1038/s41592-025-02895-w.