The thermal grill illusion and me
Like many people with Asperger Syndrome, I prefer a cool air temperature, and I will happily go around in shirt sleeves in weather that is so cold that other people find it uncomfortable to see me do it. I very rarely wear a jacket or coat. And I live in Scotland, which is a cold country.
By contrast, when I enter a swimming pool I find the water painfully cold, even if the pool is heated to a temperature at which most people find it comfortable.
How can that be? How is it that my atypical response to cool air and my atypical response to cool water apparently run in opposite directions? I think I may have stumbled on the answer.
Among the sensory channels from the skin, there are two that we will be concerned with. One of them, HPC, responds to noxious heat, pinch and cold, and produces pain. The other, COLD, responds to cool or cold alone and produces a cool feeling without pain.
There is a well‐known illusion called the thermal grill illusion. If a person’s skin is brought into contact with a grill consisting of alternating warm and cool bars, pain is felt. This is because the HPC channel responds to sharp temperature gradients on the skin, and the multiple sharp temperature gradients created by the alternating bars cause the HPC channel to be activated over a large area, causing pain. Some subjects describe the pain as burning heat, some as burning cold.
If the thermal grill experiment is done instead with only cool bars, most people feel no pain. And yet the HPC channel responds to the multiple sharp temperature gradients as before. The reason most people feel no pain is that the cool bars activate the COLD channel as well as the HPC channel, and in the brain the COLD channel masks the effect of the HPC channel. Adding the interleaved warm bars reduces the COLD activity without reducing the HPC activity, unmasking the HPC activity and causing pain.
So what is going on in my case? As I enter the water, there is a sharp temperature gradient between my skin just above the water (close to body temperature) and my skin just below the water, which has been rapidly cooled to something close to the water temperature (well below body temperature). This activates the HPC channel, as it does in any person. In a neurotypical person, that activity is masked by the simultaneous activation of the COLD channel. But in my case, I guess that there is some deficiency in the COLD channel, either in the cold‐specific nerve fibres in the skin or in the response of the brain to them, and it will be that same deficiency that prevents me from feeling cold in cold weather. So in my case the HPC channel is not masked by the COLD channel, and I therefore experience a painful thermal grill effect at the surface of the water. Paradoxically, I feel the water as burning cold because I can’t feel it as cool.
The matter is explained well in the Nature article How do you feel? Interoception: the sense of the physiological condition of the body (Craig, 2002). What the sentence quoted below calls “polymodal nociceptive activity” is what I have called HPC.
When the activity of cooling receptors is reduced, the polymodal nociceptive activity that is evoked by cooling is disinhibited centrally and causes a burning sensation at cool (< 24°C) temperatures that is normally felt only at noxious cold (< 15°C) temperatures.
(Craig, 2002, p 658)
How painful do I feel it? Surprisingly, it may be possible to quantify this. At 15°C most people find a pool painfully cold. But the thermal grill illusion produces a level of pain perceptually equivalent to the dangerously cold temperature of 10°C (Craig, 1996, p 30). It’s no wonder that I don’t like it.
Craig A D (1996) Pain, temperature, and the sense of the body. In: Franzen O, Johansson R, Terenius L (eds) Somesthesis and the neurobiology of the somatosensory cortex, Birkhäuser, Basel, pp 27–39.
Selected pages are free online at
Page 30, which is cited here, is at
Craig A D B (2002) How do you feel? Interoception: the sense of the physiological condition of the body. Nature Reviews Neuroscience 3, 655–666 (August 2002).
Abstract is at