Hormones, ovulatory cycle phase and pathogen disgust: A longitudinal investigation of the Compensatory Prophylaxis Hypothesis. Julia Stern, Victor Shiramizu. Hormones and Behavior, Volume 138, February 2022, 105103. https://doi.org/10.1016/j.yhbeh.2021.105103
Highlights
• Testing associations between women’s cycle phase or hormone levels and pathogen disgust or contamination sensitivity
• No compelling evidence for upregulated pathogen disgust in the luteal phase or when progesterone levels are higher
• Results in contrast to the Compensatory Prophylaxis Hypothesis
• No differences in results for women with a current infection
Abstract: Multiple studies have argued that disgust, especially pathogen disgust and contamination sensitivity, change across women's ovulatory cycle, with higher levels in the luteal phase due to an increase in progesterone levels. According to the Compensatory Prophylaxis Hypothesis (CPH), women have a higher disgust sensitivity to pathogen cues when in the luteal phase (or when progesterone levels are higher), because progesterone is associated with suppressed immune responses. Evidence for this hypothesis is rather mixed and uncertain, as the largest study conducted so far reported no compelling evidence for an association between progesterone levels and pathogen disgust. Further, ovulatory cycle research has been criticized for methodological shortcomings, such as invalid cycle phase estimates, no direct hormone assessments, small sample sizes or between-subjects studies. To address these issues and to contribute to the literature, we employed a large, within-subjects design (N = 257 with four sessions each), assessments of salivary hormone levels and cycle phase estimates based on luteinizing hormone tests. A variety of multilevel models suggest no compelling evidence that self-reported pathogen disgust or contamination sensitivity is upregulated in the luteal phase or tracks changes in women's hormone levels. We further found no compelling evidence for between-subjects associations of pathogen disgust or contamination sensitivity and hormone levels. Results remain robust across different analytical decisions (e.g. in a subsample of women reporting feeling sick). We discuss explanations for our results, limitations of the current study and provide directions for future research.
Keywords: Pathogen disgustContamination sensitivityProgesteroneHormonesOvulatory cycleCompensatory Prophylaxis Hypothesis
4. Discussion
In the current study, we aimed to test the CPH by employing a longitudinal design, a large sample size, direct hormone assessments and LH-test validated cycle phase estimates. Across a variety of different analyses, we found no compelling evidence that pathogen disgust or contamination sensitivity is related to LH-validated cycle phase or different hormones levels (within-subjects and between-subjects). We further found no significant effects for women who reported feeling sick in our data.
Previous studies testing the CPH yield mixed findings. Our results are in line with the results by Jones et al. (2018) who also reported no compelling evidence that pathogen disgust tracks changes in women's salivary progesterone, estradiol, testosterone, or cortisol. They are further in line with studies not reporting compelling evidence for an association of different cycle phases and increased pathogen disgust (e.g. Fessler and Navarrete, 2003; Żelaźniewicz et al., 2016), but in contrast to previous studies reporting significant effects for either (within-subjects or between-subjects) hormone levels or different cycle phases.
Fleischman and Fessler (2018) published three possible explanations for the null results reported by Jones et al. (2018): a) that the CPH might be entirely wrong, b) measurement issues might explain differences in findings, c) progesterone might not be the driving factor. We argue that these three explanations might also pertain to our findings, as they were virtually identical to the findings published by Jones et al. (2018), although Jones et al. (2018) did not specifically assess women's cycle phases. Of course, it is possible that the CPH is wrong and that “changes in immune functioning are too small or not consistent enough to exert selective pressure on mechanisms governing behavior” (Fleischman and Fessler, 2018, p. 468). However, we refrain from such strong conclusions based on our findings, given that there are always limitations in single datasets, that absence of evidence does not equal evidence of absence and that our results might not be generalizable to other contexts or samples (e.g. pregnant women). Nevertheless, we think that our findings further challenge the CPH. We agree that differences in used measures might at least partly explain differences in findings between studies. For example, we did not investigate disgust responses to pictures depicting disease cues, for which some previous studies reported findings in line with the CPH (Fleischman and Fessler, 2011; Miłkowska et al., 2021). However, not using pictures in the current study does not explain differences in findings of self-reported pathogen disgust or contamination sensitivity via questionnaires also used by Fleischman and Fessler (2011) or Milkowska et al., 2019, Miłkowska et al., 2021. Other differences in methods (besides used stimuli) might be more likely to explain differences in findings. For example, the study by Jones et al. (2018) and the current study are the studies with the largest sample sizes so far, and also the only large-scale within-subject studies with direct hormone assessments. Further, our study used randomized sessions (e.g. not all participants started testing in the same cycle phase), whereas every participant had her first testing session in the fertile phase in the studies by Milkowska et al., 2019, Miłkowska et al., 2021. Interestingly, Fleischman and Fessler (2018) also stated that the higher test power in the study by Jones et al. (2018) suggests that the CPH might rather be wrong than differences in results could be explained by measurement issues. Regarding the potential explanation that progesterone is not the driving factor that regulates disgust sensitivity, we agree that other factors related to pregnancy might lead to upregulated disgust sensitivity. To investigate which factors are actually responsible for fluctuations in disgust sensitivity (and higher disgust sensitivity when pregnant), we suggest that future studies should collect data from pregnant women, as only testing pregnant women can answer this research question properly.
4.1. Limitations
We note several limitations regarding our study that might be addressed in subsequent research. First, whereas we investigated whether different self-report questionnaires or different analyses yield different findings regarding our research question, we did not investigate disgust responses to pictures or videos depicting disease cues. Second, we cannot draw strong conclusions on whether having an infection might moderate shifts in disgust sensitivity across the cycle. Although we did not find compelling evidence for this claim, only a very small number of our participants reported feeling sick (potentially as, if they were truly sick, they would not have attended the session). Third, due to ethical constraints, our study was observational, not experimental. Hence, it remains unclear whether progesterone administration might raise disgust, comparably to mice (Bressan and Kramer, 2021).