Fatal attraction phenomenon in humans: cat odour attractiveness increased for toxoplasma-infected men while decreased for infected women

Abstract

Background: Latent toxoplasmosis, a lifelong infection with the protozoan Toxoplasma gondii, has cumulative effects on the behaviour of hosts, including humans. The most impressive effect of toxoplasmosis is the "fatal attraction phenomenon," the conversion of innate fear of cat odour into attraction to cat odour in infected rodents. While most behavioural effects of toxoplasmosis were confirmed also in humans, neither the fatal attraction phenomenon nor any toxoplasmosis-associated changes in olfactory functions have been searched for in them.

Principal findings: Thirty-four Toxoplasma-infected and 134 noninfected students rated the odour of urine samples from cat, horse, tiger, brown hyena and dog for intensity and pleasantness. The raters were blind to their infection status and identity of the samples. No signs of changed sensitivity of olfaction were observed. However, we found a strong, gender dependent effect of toxoplasmosis on the pleasantness attributed to cat urine odour (p = 0.0025). Infected men rated this odour as more pleasant than did the noninfected men, while infected women rated the same odour as less pleasant than did noninfected women. Toxoplasmosis did not affect how subjects rated the pleasantness of any other animal species' urine odour; however, a non-significant trend in the same directions was observed for hyena urine.

Conclusions: The absence of the effects of toxoplasmosis on the odour pleasantness score attributed to large cats would suggest that the amino acid felinine could be responsible for the fatal attraction phenomenon. Our results also raise the possibility that the odour-specific threshold deficits observed in schizophrenia patients could be caused by increased prevalence of Toxoplasma-infected subjects in this population rather than by schizophrenia itself. The trend observed with the hyena urine sample suggests that this carnivore, and other representatives of the Feliformia suborder, should be studied for their possible role as definitive hosts in the life cycle of Toxoplasma.

Figure 1. Differences in the urine odour pleasantness scores attributed to various species of animals.

The empty circles and triangles denote mean Z-scores for women and men, respectively; the vertical bars denote 0.95 confidence intervals. The odour of samples with more positive Z-scores was scored as more pleasant (in comparison with other urine samples). The means were controlled for intensity of the odour attributed by particular rater, i.e. they were computed for mean intensity of the odour. A and B are the high and low concentration samples, respectively (see the Methods section).

Figure 2. The cat urine odour pleasantness.

The figure shows odour pleasantness scores attributed to the cat urine sample by Toxoplasma-infected and Toxoplasma-free male and female students. The circles and squares denote mean Z-scores for women and men, respectively; the vertical bars denote 0.95 confidence intervals. The odour of samples with more positive Z-scores was scored as more pleasant (in comparison with other urine samples). In contrast to results presented in Tab. 1, here the means were controlled for intensity of the odour attributed by particular rater, i.e. they were computed for mean intensity of the odour.