From a tender age we are made to learn about our "five senses"-just five, namely sight, hearing, smell, taste and touch. We are told that beyond the bounds of this tidy sensorium lies only the murky realm of the paranormal, where clairvoyants supposedly enjoy the use of an extra, "sixth sense."
But this five-ordinary-sense dogma seems badly in need of revision. Studies in sensory neuroscience have been revealing such a complexity that it is now hard to say how many distinct "senses" we have.
A multitude of touches
Consider the "sense of touch." We should know from experience that the term is really a grab-bag for at least several basic kinds of sensation. Texture and pressure, heat and cold, thermal conductance, vibration, various pains and pleasures, and itch-all these (and more) our skin can somehow detect and distinguish when it touches things. Neuroscientists once wondered if one or a few types of "touch" neuron and their associated nerve ends in the skin were somehow responsible for conveying these different sensory messages to the brain. But it's now clear that we have an abundance of primary sensory pathways, each specialized for a distinct element of sensation.
The sensation of itch, for example, was long suspected to originate from nerve ends that mediate pain and touch, as if it were a blend of the two and not a sense in its own right. But just this year neuroscientists have shown that "a separate set of primary sensory neurons works to detect the small class of itch-producing compounds," says Mark Hoon at the National Institutes of Health.
Like other primary sensory neurons for touch-related sensations, those for itch are located in the spine. Each extends one long nerve fiber to a particular patch of skin, where the fiber branches out in a rootlike mesh of nerve-ends. Experiments with transgenic mice have revealed that these nerve ends are studded with special receptors-with names such as MrgprC11 and MrgprA3-that can be activated specifically by contact with histamine or other itch-making compounds. When these itch receptors, or the sensory neurons to which they belong, are deleted from mice, the mice no longer respond to itch-inducing stimuli, yet they continue to respond normally to other skin-delivered stimuli.
With its own, dedicated set of primary sensory neurons and nerve-end receptors, itch arguably should be considered separate from other touch senses. "Itch is definitely a unique sensation," says Xinzhong Dong from Johns Hopkins University. On the other hand, itch is hard to separate completely from its sensory cousin, pain. As Dong explains, an itch begs to be scratched, and scratching causes pain-and that pain from scratching somehow cancels out the itch signals, apparently even before they reach the brain. "It has been speculated that there are inhibitory interneurons in the spinal cord that can cross-talk from the pain pathway to the itch pathway, so that when you scratch, you activate those interneurons and block itch sensations," Dong says.
In the grand scheme of the senses, itch's relationship with pain isn't the most complicated one. "Itch is a relatively basic sensory input, in that there are very few areas in which it is modulated [on the way to the brain]," says Hoon. Other types of touch undergo much greater processing before we experience them. Veronica Abraira and David Ginty, also of Johns Hopkins, described in a recent paper how the most ordinary sense of touch-so-called innocuous touch-is mediated by groupings of different "low-threshold mechanoreceptors" (LTMRs) on skin-embedded nerve ends. LTMRs come in a variety of subtypes, each designed to detect a different sort of mechanical, skin-moving stimulus, and they are arranged in different patterns in different places in the skin. For example, according to Abraira and Ginty, "each of the three major hair follicle types of trunk hairy skin … is innervated by a unique and invariant combination of LTMRs." The nerve ends that bear these LTMRs send their signals to touch neurons in the spine, which in turn are arranged in map-like patterns corresponding to their zones of innervation on the body.
What remains to be determined precisely is how these touch neurons in the spine work together to, as Hoon puts it, "produce our amazing ability to detect an almost infinite variety of mechanical pressures on our skin."
The sensory blender
This basic pattern-in which multiple types of primary input are somehow woven together to make a holistic sensory experience-holds for most other sense categories too. We perceive blends, in other words, and hardly ever the individual, raw inputs. How do we know that our bladder is full? We just seem to know-the processing of various stretch- and pressure- and pain-related sensors in and around the bladder gives us a simple, integrated "bladder-full" feeling, under the sense category known as interoception.
Why do the police sometimes ask an automobile driver to touch the tip of his nose while keeping his eyes closed? Because a driver's brain should be able to accomplish this feat easily, if its proprioceptive sense-based on the integration of complex signals from stretch and pressure-related sensors in skin, muscles and joints-has not been impaired by alcohol.
Unsurprisingly, given our complex cuisines, gustatory sensations also come to us as integrations of distinct signal types. What we commonly regard as "taste" is mediated by distinct receptors for the sweet, bitter, sour, salt and umami chemicals in our food. What about the "taste" of peppery heat in spicy food? It turns out that that doesn't come via taste neurons at all-but from painful-heat-sensing neurons, which are essentially short-circuited by the classic hot-pepper ingredient capsaicin. Szechuan peppers, which have a buzzy feel on the tongue, have their own non-taste ingredient, which apparently activates the same touch-related nerve fibers that detect mechanical vibrations. To volunteers in a recent study, the "taste" of the peppers was like touching an object that vibrated at 50 Hz.
In general, the boundaries of our sense organs and sense categories are hazier than we have been led to believe. Our ears contain not only a mix of cochlear hair cell types that give us hearing, but also vestibular hair cells that help give us our sense of balance, head position and even gravity. When we see visual images, we do so via distinct rod and cone photoreceptor cells-although our eyes also contain a recently discovered third type of light-sensing cell, the melanopsin-expressing retinal ganglion cell, which gathers light not to form images but to help regulate pupil diameter as well as sleep and related circadian (24-hour-cycle) processes.
Some of our sensory experiences are so "meta" that they don't necessarily depend predominantly on a single sensory mode. The V5 region of the visual cortex, for example, appears to be specialized for the sensing of visual motion. Yet studies have shown that in people who are blinded at a young age, V5 starts to respond prominently to auditory inputs, giving such people superior auditory-motion sensing abilities. The conservation of the motion-sensing function in such cases suggests that V5 develops auditory inputs naturally, even in the sighted, and can strengthen them when necessary. Thus, especially at young ages, we might possess a "sense of motion" that doesn't belong to just sight or hearing, but is informed by either or both.
Higher in the cortical processing hierarchy lie other, more obviously multimodal or meta-sensory types of perception. Arguably one is the sense of the passing of time. Martin Paulus's laboratory at the University of California, San Diego, for example, has found evidence from brain-imaging experiments that this time-sense is a feeling of accumulating moments of sensory experience-mediated chiefly by the insula, a deep-seated brain region with extensive interoceptive and other sensory inputs. "The sense of time seems to be an integrated sense of what's happening to us," Paulus says. "If we had no experience of the inside or the outside of our bodies, we would not have this sense of time."
Higher-order, highly integrated sensory processes such as these beg the question: what is a sense? But sensory neuroscientists don't seem yet to have found a good, "objective" definition-in terms of sense-specific receptors, sense-specific neurons, or sense-evoked behaviors.
Meanwhile our list of senses-candidate senses anyway-keeps getting longer, as we reflect upon our ordinary sensory experiences and design new experiments. "I was at a meeting on itch recently," says Dong, "and it was noted that some compounds that induce an itch sensation can also induce a prickling, tingling sensation. What mediates that?"