Context dependent adaptation in a neural computation

The preprint operates within the paradigm of efficient coding, specifically focusing on how neural systems adapt to the statistics of sensory inputs to maximize information transmission. It uses the established framework of spike-triggered analysis and information theory to investigate the adaptive behavior of the fly visual neuron H1.

This preprint presents a detailed investigation of context-dependent adaptation in the fly visual neuron H1. It employs well-established experimental and analytical techniques within the efficient coding paradigm. The study focuses on how H1 adapts to different contrast distributions, a phenomenon within the scope of normal science. While the findings regarding two-dimensional adaptation and the link between integration time and spike rate are interesting, they represent refinements of existing knowledge rather than a challenge to the prevailing paradigm. There's no indication of a paradigm shift, model drift, or model crisis. The research deepens the understanding of adaptation within the efficient coding framework, solidifying its explanatory power. The work could also be considered as Model Drift as the two-dimensional adaptation and scaling behavior of the neuron's response are unexpected and deviate slightly from the standard model of contrast adaptation. However, these deviations are not radical enough to constitute a model crisis or revolution, and the core principles of efficient coding remain unchallenged. Thus, the classification of Normal Science is favored, with Model Drift as a secondary, less likely possibility.