Measuring animal emotions with infrared thermography: How to realise the potential and avoid the pitfalls

Infrared thermography, also known as thermal imaging, is expected to yield some of the next major steps in improved monitoring of animal welfare. As a non-contact technology enabling rapid measurements on large numbers of individuals, it holds considerable future potential for automated monitoring and handheld assessment tools (eg for farms, zoos and laboratories). In this review, we give an overview of the current body of knowledge; highlight areas requiring further research for practical applications; and offer technical and methodological hints on how to avoid pitfalls and to succeed in thermographic research on animal welfare.
Infrared thermography involves the use of a thermal camera to measure subtle differences and changes in animals’ skin or eye temperatures. These reflect variations in blood flow and muscle contraction, some which are linked to emotions. Thermographic studies on animal emotions have so far mainly focussed on the sympathetic nervous system (SNS). Activated during emotional arousal, the SNS causes vasoconstriction in peripheral blood vessels (eg in the skin and behind the eyes). A pattern of change in eye temperature over time, consisting of a reduction and a subsequent increase, have been physiologically validated in cattle to be triggered by acute sympathetic activity. Another consequence of SNS activity is a temperature reduction in peripheral body parts, and this has been investigated, using thermography, in a number of species, including measurements of the ears of rabbits and dogs, the combs of hens, and the noses of cattle and several species of primates including humans. We review these studies and highlight the emotional and physiological processes that have yet to receive extensive
investigation, including measurement of facial expressions, long-term emotional/mood states and interactions between valence and sympathetic arousal. When investigating emotional effects on thermographic recordings, other factors also have substantial effects on animals’ surface temperature distributions. Recent exercise, metabolic activity, diurnal rhythm, ambient temperature, humidity and nearby heat-reflecting surfaces are some of the most important. Systematic measures of these, and development of new methods to standardise and correct their effects on thermal images, will be a crucial component of developing thermographic techniques for measuring
animal welfare, and improving the quality and reliability of future research. A key component in future success in developing thermographic measures for animal welfare will be interdisciplinary collaboration involving expertise in animal welfare, cognition, physiology and physics.