Empathy and compassion toward other species decrease with evolutionary divergence time

Abstract

Currently the planet is inhabited by several millions of extremely diversified species. Not all of them arouse emotions of the same nature or intensity in humans. Little is known about the extent of our affective responses toward them and the factors that may explain these differences. Our online survey involved 3500 raters who had to make choices depending on specific questions designed to either assess their empathic perceptions or their compassionate reactions toward an extended photographic sampling of organisms. Results show a strong negative correlation between empathy scores and the divergence time separating them from us. However, beyond a certain time of divergence, our empathic perceptions stabilize at a minimum level. Compassion scores, although based on less spontaneous choices, remain strongly correlated to empathy scores and time of divergence. The mosaic of features characterizing humans has been acquired gradually over the course of the evolution, and the phylogenetically closer a species is to us, the more it shares common traits with us. Our results could be explained by the fact that many of these traits may arouse sensory biases. These anthropomorphic signals could be able to mobilize cognitive circuitry and to trigger prosocial behaviors usually at work in human relationships.

Figure 1

Experimental procedure. Based on a focused question, each evaluator had 22 pairs of pictures to evaluate (randomly drawn from a total of 52 species). The question, also randomly drawn at the beginning of the test, was intended to assess either empathic or compassionate preferences. (photos by A. Miralles).

Figure 2

Empathy and compassion scores attributed to each organisms as a function of divergence time (Mya) between them and humans. The scores correspond to the probability that a given species is chosen from a pair of species that includes it and another randomly selected (n = 52 species). See SI Appendix, Results S1 for details. (Illustrations by A. Miralles).

Figure 3

Effect of confounding variables and response time. (A) Effect of rater’s traits on both questions. Odds ratio (for a qualitative variable: ratio of the odds of choosing the most phylogenetically related species in the depicted factor level to the odds of it occurring in the reference factor level; for age, centered variable: ratio of the odds of choosing the most phylogenetically related species in age 1 to the odds of it occurring in age 0) are represented by dots and 95% confidence interval by lines; blue or red dots indicate variables linked with an increased or decreased, respectively, choice probability for the most phylogenetically related species (n raters = 1134 for the empathy test and 1213 for the compassion test). (B) Predicted participants’ response time as a function of the absolute divergence time between the two species presented in each pair (area depicts the 95% confidence interval, n responses = 25001 for the empathy test and 26781 for the compassion test).

Figure 4

Relationships between empathy and compassion scores. While the oak benefits from an excessive compassion score when compared to its ability to arouse empathy, the tick suffers from a clear compassion deficit (n = 52 species).