Measuring Dependencies between Biological Signals with Self-supervision, and its Limitations

TL;DR

This paper introduces 'concurrence,' a self-supervised method for detecting complex dependencies between biological signals, demonstrating its broad applicability but also highlighting limitations due to extraneous factors.

Contribution

The paper presents a novel self-supervised approach called concurrence for measuring dependencies in diverse biological signals without prior assumptions.

Findings

Concurrence can detect relationships across fMRI, physiological, and behavioral signals.

It exposes scientifically relevant differences without ad-hoc parameter tuning.

Extraneous factors can cause spurious dependencies, requiring validation.

Abstract

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.