[Image: Getty Images]
Human vision, as the name implies, is restricted to the nominally visible light wavelengths of around 400 to 720 nm. Periodically, however, tantalizing reports of humans’ ability to “see” light at longer, infrared (IR) wavelengths have cropped up in the literature. A new study by an international research team now puts those reports on a much stronger explanatory footing (Proc. Natl. Acad. Sci. USA, doi: 10.1073/pnas.1410162111). The work suggests that under certain conditions, humans can indeed see IR light—and that the ability to do so stems from a nonlinear optical process.
As is well known, mammalian vision rests on the absorption of light by visual pigments in the rod and cone structures of the eye, which are tuned to the nominal visible wavelengths. As the researchers note in the study, pigments sensitive to wavelengths longer than the 700 nm range are possible, but the signal from such pigments would be swamped by background thermal noise. Nonetheless, as high-intensity IR sources have become more common, there have been fragmentary reports of humans sensing IR light, and evidence that human rods and cones could respond to radiation at wavelengths as long as 1,355 nm.
To investigate these claims and resolve the paradox, the team of scientists—from institutions in the United States, Poland, Switzerland, and Norway—used a combination of psychophysical and biochemical experiments. First, they used a low-power IR laser to scan the retinas of 30 volunteers at wavelengths from 950 to 1200 nm, and found that all of them reported perceiving the IR light as visible.
Moreover, the perceived colors of the “visible” IR reported by the volunteers were at wavelengths of roughly half those of the IR signal. That suggested that a frequency-doubling mechanism was at work. Other observations, including variations dependent on laser power and pulse width, likewise pointed to a nonlinear process—perhaps two-photon absorption (2PO) or, possibly, second-harmonic generation (SHG).
Next, the team directly exposed mouse retinas to IR light at various wavelengths and powers, and used biochemical and genetic techniques to measure the response. They focused specifically on the photoreceptor protein rhodopsin and on a specific chromophore, 11-cis-retinyl-propylamine. (Shape changes in chromophore units, as a result of isomerization from the incident photon energy absorbed by the bound photoreceptor protein, are a key component in the “signaling cascade” that results in visual perception.)
They found that IR light from a femtosecond laser could directly activate the photoreceptor and trigger isomerization of the chromophore, at the expected frequency-doubled response. And the group did not observe an SHG signal in the rhodopsin absorption spectra, which strongly suggests that the specific nonlinear process at work was 2PO.
From all of this, the researchers concluded that, under the right conditions (such as stimulation by short, rapid pulses of laser light), two-photon absorption of IR light by the retinal photoreceptors, followed by direct isomerization of the chromophore at the doubled frequency, can cause humans to “see” supposedly invisible IR light. (A note, though: Don’t try this at home. While the IR light wavelengths explored here are sometimes labeled “retina safe,” they can still do significant eye damage at all but the lowest power settings.)