A new preprint on arXiv applies Benoît Mandelbrot's coastline paradox to neuroscience, asking how the resolution of our measurements shapes what we see in neural computation. Just as a coastline's length depends on the ruler used, the paper suggests that a neuron's apparent computational structure may depend on the observational scale.

The authors introduce a scale-indexed framework to examine how interaction order and computational capacity change with resolution. Their central claim, based on the abstract, is that measurement is not a neutral window into neural function but a joint property of the object and the tool used to observe it.

The work is conceptual, drawing a parallel between a classic geometric paradox and the study of neural systems. It does not present experimental results in the abstract, but rather offers a new way to think about how we characterize neural computation across scales.