Orthogonal representations for robust context-dependent task performance in brains and neural networks

By Timo Flesch, Keno Juechems, Tsvetomira Dumbalska, Andrew M. Saxe, Christopher Summerfield. 2022

194 citations recorded by the source corpus.

Summary How do neural populations code for multiple, potentially conflicting tasks? Here we used computational simulations involving neural networks to define “lazy” and “rich” coding solutions to this context-dependent decision-making problem, which trade off learning speed for robustness. During lazy learning the input dimensionality is expanded by random projections to the network hidden layer, whereas in rich learning hidden units acquire structured representations that privilege relevant over irrelevant features. For context-dependent decision-making, one rich solution is to project task representations onto low-dimensional and orthogonal manifolds. Using behavioral testing and neuroimaging in humans and analysis of neural signals from macaque prefrontal cortex, we report evidence for neural coding patterns in biological brains whose dimensionality and neural geometry are consistent with the rich learning regime.

Subfields: Cognitive Neuroscience, Representation & Self-Supervised Learning, Neuroscience

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

1. Carlos González-García
2. M. Ruz
3. Paula Pena
4. Ana F. Palenciano

Citations in the corpus

• Novel Verbal Instructions Recruit Abstract Neural Patterns of Time-Variable Information Dimensionality (2025, methodology): “Additionally, it has been shown that task settings with a similar structure where information can be reused, or with high task complexity, favor generalization (Aoi et al., 2020, Flesch et al., 2022; Badre et al., 2021; Woolgar et al., 2015).”