Research‎ > ‎Papers‎ > ‎

Simoncelli - 2001 - Natural image statistics and neural representation


Simoncelli EP, Olshausen BA. Natural image statistics and neural representation. Annu Rev Neurosci. 2001;24:1193-216. PUBMED

10 Word Summary

Sensory signals are efficiently encoded using statistical representations.


It has long been assumed that sensory neurons are adapted, through both evolutionary and developmental processes, to the statistical properties of the signals to which they are exposed. Attneave (1954) and Barlow (1961) proposed that information theory could provide a link between environmental statistics and neural responses through the concept of coding efficiency. Recent developments in statistical modeling, along with powerful computational tools, have enabled researchers to study more sophisticated statistical models for visual images, to validate these models empirically against large sets of data, and to begin experimentally testing the efficient coding hypothesis for both individual neurons and populations of neurons.


  • Understand the function of neural systems by:
    • the tasks the organism performs
    • the computational capabilities and constraints of neurons
    • the environment in which the organism lives
  • Two approaches to understanding sensory processing:
    • examine statistical properties of neural processes to natural stimuli
    • derive a model of sensory processing
  • Efficient coding suggests what shape of distribution the individual neural responses should take and the statistical dependencies between neurons.
  • Visual information has the following statistics
    • Intensity
    • Color
    • Spatial correlation
    • Space-time correlations
  • It should be noted that although these techniques seek statistical independence, the resulting responses are never actually completely independent (p1206).
  • \tau