Export Reviews, Discussions, Author Feedback and Meta-Reviews

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. One of the chief concerns of systems neuroscientists is characterising how individual neurons respond to sensory stimuli. Since the stimulus space is often huge, data is always limited, and neurons are fundamentally noisy, the statistical challenges involved with this characterisation have spurned a vibrant field of computational neuroscience. This paper considers a particular form of this task, where recordings are made of local populations of neurons, at some intermediate point in the processing hierarchy in the brain (e.g. Such recordings are very common already, and also are growing in number and fidelity.