We have thermoreceptors, thus we can sense temperature (both warm and cold). I'm interested in the sensitivity of our thermoreceptors -

What is the smallest temperature difference that we can sense?

I assume that different parts / organs may have different sensitivity (eg. lips vs fingers), thus I'd like to narrow my focus on the palms / fingers. But if someone has comparative data, that is welcomed too.

  • 5
    $\begingroup$ This varies enormously across the body, and also with age. One good starting place would be Stevens, J. C., & Choo, K. K. (1997). Temperature sensitivity of the body surface over the life span. Somatosensory & motor research, 15(1), 13-28. $\endgroup$
    – Corvus
    Mar 31 '15 at 23:44

Short answer
Temperature differences of 0.02 degrees Celcius can be distinguished, dependent on various factors including experimental conditions and bodily location.

The ability to discriminate temperature differences depends on whether it is a cooling or heating pulse, the skin temperature, the duration of the temperature stimulus, age, bodily location among other factors. Unfortunately I cannot access the primary literature other than a few isolated smaller studies. However, Scholarpedia has a very nice entry and associated references, and I quote:

The thermal sensory system is extremely sensitive to very small changes in temperature and on the hairless skin at the base of the thumb, people can perceive a difference of 0.02-0.07 °C in the amplitudes of two cooling pulses or 0.03-0.09 °C of two warming pulses delivered to the hand. The threshold for detecting a change in skin temperature is larger than the threshold for discriminating between two cooling or warming pulses delivered to the skin. When the skin at the base of the thumb is at 33 °C, the threshold for detecting an increase in temperature is 0.20 °C and is 0.11 °C for detecting a decrease in temperature.

The rate that skin temperature changes influences how readily people can detect the change in temperature. If the temperature changes very slowly, for example at a rate of less than 0.5 °C per minute, then a person can be unaware of a 4-5 °C change in temperature, provided that the temperature of the skin remains within the neutral thermal region of 30-36 °C. If the temperature changes more rapidly, such as at 0.1 °C/s, then small decreases and increases in skin temperature are detected. [...]

You also were interested in the palmar skin: here average difference limens are 0.02 - 0.06 °C, dependent on the temperature step used. These values were obtained using cooling pulses. In other words, differences in two cooling pulses of 0.02 - 0.06 °C were discernible, with higher limens needed for higher pulses. Baseline temperatures (29 - 39 °C) had little effect (Darian-Smith et al., 1977).

Darian-Smith et al., J Invest Dermat 1977;69:146-53

The Stevens, J. C., & Choo, K. K. (1998). Temperature sensitivity of the body surface over the life span. Somatosensory & motor research, 15(1), 13-28 article as commented on by Corvus is indeed a must-read, but my uni library does not have access to it unfortunately

  • $\begingroup$ This is an important point for our previous argument. As you now point out the perception of temperature is not only dependent on the relative temperature difference but also on the absolute temperature. $\endgroup$
    Apr 1 '15 at 4:57
  • $\begingroup$ @WYSIWYG - Yes.. and no. Surprisingly Darian-Smith et al., 1977 found no effect of base temperature (at least across 10 degrees C). However, according to Weber's law, the difference limen increases when the reference temperature is higher. This does apply to the temperature step, but base temperature may be another story? $\endgroup$
    – AliceD
    Apr 1 '15 at 13:49
  • $\begingroup$ Here's another question. Is the human body capable of sensing a continuous change in temperature or does it just sense when change occurs in bulk? I mean to say, is the "sense" continuous or discrete? $\endgroup$ Jan 20 '20 at 3:29
  • $\begingroup$ @PriyanshuDas good question - please ask new questions as a question - Make sure to add a little of your own research, for instance by linking to this question or answer and explaining it doesn't answer your current question. $\endgroup$
    – AliceD
    Jan 20 '20 at 7:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.