In the early morning hours of 22 October 1850, a German doctor called Gustav Fechner erupted from sleep in a paroxysm of insight. Fechner, a philosopher and experimental psychologist, had dozed off while contemplating the fundamental relationship between the physical world and human perception. Thirty years earlier, Fechner's former professor, Ernst Weber, had performed an experiment in which he had asked subjects to tell him which of two weights was heavier. Weber was surprised to discover that people found it much harder to discriminate small differences if the weights were large than if they were small, a result that became known as Weber's law.

Fechner's convulsive realization was that Weber's law could be described by a simple mathematical relationship: P=log(I), where P is perception and I is the stimulus intensity. This equation predicts that the smallest noticeable difference between two stimuli should increase linearly as the stimuli get larger. To test this, Fechner performed a classic series of psychophysics experiments in which he demonstrated that visual thresholds increase linearly with increases in background illumination.

Despite the elegance of the Weber–Fechner law, the underlying reason why it describes human visual perception has remained elusive. In addition, experiments by Horace Barlow and others in the 1950s showed that the Weber–Fechner law does not apply to human vision at low light levels. Now, a study of the retina from Juan Angueyra and Fred Rieke suggests a possible explanation for why human vision obeys the Weber–Fechner law, including why it only applies at high light intensities.

The mammalian retina consists of two types of primary photoreceptors: rods, which handle vision in low light, and cones, which operate at higher light intensities and contribute to color vision. Recording from primate cone photoreceptors, Angueyra and Rieke investigated how adaptation of cone signals and noise set limits on perceptible intensity differences. The authors varied the background luminance and measured both the noise in the transduction currents during constant light and the signal evoked by a brief light flash. They found that the light-evoked signals were attenuated as background light intensity increased, while noise levels remained relatively constant. These properties are consistent with the Weber–Fechner law and are in stark contrast to findings in rods, in which signal and noise both grow as light levels increase. This difference between rods and cones may explain why cone vision follows the Weber–Fechner law, while rod vision does not.

In addition to quantifying signal and noise in cones, the authors investigated the source of cone noise. By recording from single cones while pharmacologically manipulating the phototransduction pathway, they identified two primary sources of cone noise: gating of the cyclic nucleotide-gated channels that produce phototransduction currents, and fluctuations in the concentration of the second messenger molecules that open these channels. These data provide additional support that, in contrast to rods, spontaneous activation of cone pigments is not likely to contribute to perceptual thresholds.

One-hundred and fifty years after Fechner's pioneering pyschophysics experiments, Angueyra and Rieke have proposed a mechanistic basis for the Weber–Fechner law that relies on detailed measurements of photoreceptor currents and pharmacological manipulation of phototransduction chemistry. An interesting question is whether Fechner, if he were alive today, would accept this reductionist explanation for what he originally considered a phenomenon of the mind. My guess is that Fechner, who was among the first to consider how physiology constrains subjective human experience, would be satisfied.

J. M.
Origin and effect of phototransduction noise in primate cone photoreceptors
Nat. Neurosci.