Measuring Dependencies between Biological Signals with Self-supervision, and its Limitations
This provides a tool for scientific discovery in fields like neuroscience and biology by enabling dependency measurement without a priori information, though it is incremental as it builds on self-supervision ideas.
The paper tackles the problem of measuring complex non-linear dependencies between biological signals without prior knowledge, introducing a self-supervised method called concurrence that distinguishes aligned vs. misaligned segments, and it is the first approach to expose relationships across a wide spectrum of signals like fMRI, physiological, and behavioral data without ad-hoc tuning.
Measuring the statistical dependence between observed signals is a primary tool for scientific discovery. However, biological systems often exhibit complex non-linear interactions that currently cannot be captured without a priori knowledge regarding the nature of dependence. We introduce a self-supervised approach, concurrence, which is inspired by the observation that if two signals are dependent, then one should be able to distinguish between temporally aligned vs. misaligned segments extracted from them. Experiments with fMRI, physiological and behavioral signals show that, to our knowledge, concurrence is the first approach that can expose relationships across such a wide spectrum of signals and extract scientifically relevant differences without ad-hoc parameter tuning or reliance on a priori information, providing a potent tool for scientific discoveries across fields. However, dependencies caused by extraneous factors remain an open problem, thus researchers should validate that exposed relationships truly pertain to the question(s) of interest.