Introduction

Individuals displaying cues of higher mate value are perceived as greater threats to pair bonds (Dijkstra and Buunk 1998, 2001, 2002) and threats to pair bonds typically trigger jealousy (Buss et al. 2000). Thus, individuals displaying cues of higher mate value would be predicted to trigger greater jealousy (O’Connor and Feinberg 2012). Since sexually dimorphic features are associated with markers of mate quality, such as attractiveness in women and dominance in men (Little et al. 2011), it may be linked with the perception of jealousy.

Consistent with this quality-jealousy hypothesis, men report greater jealousy when imagining scenarios in which their romantic partner interacts with men displaying more masculine face or body shapes (Dijkstra and Buunk 2001; Kruger 2006; Massar and Buunk 2009; O’Connor and Feinberg 2012) or speaking with a more masculine voice (O’Connor and Feinberg 2012). By contrast with this relatively large body of work on sexually dimorphic characteristics and men’s jealousy, only one study has tested for corresponding effects of sexually dimorphic face-shape characteristics on women’s jealousy. Consistent with the quality-jealousy hypothesis, O’Connor and Feinberg (2012) found that Canadian women reported greater jealousy to scenarios in which their romantic partner was imagined interacting with relatively feminine women. These results are consistent with other work reporting that women perceive other women with more feminine face and body shapes to present greater intrasexual competition (Fink et al. 2014) and that women wearing makeup that exaggerates feminine facial color patterns elicit greater jealousy (Mileva et al. 2016). Feminine physical characteristics are thought to reflect good physical condition in women, but not men (see O’Connor and Feinberg 2012), which would explain why feminine physical characteristics in men do not elicit jealousy.

The tendency for research on social judgments of faces to focus on Western participants’ perceptions of Western faces is a substantial barrier to our understanding of the generalizability of results across cultures (Sutherland et al. 2018; Han et al. 2018). Consequently, the current study investigated the effects of experimentally manipulated sexually dimorphic face-shape cues on Chinese and White UK women’s jealousy perceptions. Previous studies exploring the effects of physical cues on jealousy have examined only Western cultures.

Methods

Participants

Twenty-two heterosexual Chinese women (mean age = 23.98 years, SD = 3.07 years) and 15 heterosexual White UK women (mean age = 19.03 years, SD = 19.03 years) took part in the study. All participants were students at the University of Glasgow. Chinese participants were all born in China and reported having been in the UK for an average of 148 days (SD = 50 days).

Stimuli

Face stimuli were manufactured using the same computer graphic methods used by O’Connor and Feinberg (2012). Masculinized and feminized versions of 15 young adult Chinese women’s faces were manufactured by transforming the shape of each face image by ± 50% of the linear shape differences between Chinese male and Chinese female face prototypes. Similarly, masculinized and feminized versions of 15 young adult White UK women’s faces were manufactured by transforming the shape of each face image by ± 50% of the linear shape differences between White UK male and White UK female face prototypes. Male and female prototypes were manufactured by averaging the shape and color from 50 male and 50 female faces of the relevant ethnicity. Standard computer graphic methods developed for this specific task (DeBruine 2018; see also Tiddeman et al. 2001) were used to manufacture face stimuli. Example stimuli are shown in Fig. 1.

Fig. 1
figure 1

Feminized (left) and masculinized (right) versions of a Chinese face used in the study

Full size image

Procedure

Each participant was shown the 30 pairs of female faces (each pair consisting of a feminized and a masculinized version of the same face). Trial order and the side of the screen on which any given image was presented were fully randomized. Following O’Connor and Feinberg (2012), participants were instructed to “Click on the face of the person who would make you more jealous if they were flirting with your romantic partner (if you do not currently have a romantic partner, imagine that you do have one).”

Results

For each participant, we calculated the proportion of trials on which she chose the feminized face separately for Chinese faces and White UK faces. Data and analysis code are publicly available at https://osf.io/u24s9/.

One-sample t tests comparing the proportion of trials on which women chose the feminized faces with what would be expected by chance alone (i.e., 0.5) showed that Chinese women selected the feminized faces significantly more often than chance when assessing Chinese (t(21) = 6.481, p < 0.001, M = 0.739, SEM = 0.04, d = 1.382) and White UK (t(21) = 4.88, p < 0.001, M = 0.706, SEM = 0.04, d = 1.041) faces. One-sample t tests also showed that White UK women selected the feminized faces significantly more often than chance when assessing Chinese (t(14) = 6.17, p < 0.001, M = 0.769, SEM = 0.04, d = 1.593) and White UK (t(14) = 7.051, p < 0.001, M = 0.778, SEM = 0.04, d = 1.821) faces. Overall, 33 of the 37 women in the study chose feminized faces on more than half of the trials.

We also conducted a 2 × 2 mixed-design ANOVA, with face ethnicity as a within-subject factor, and participant ethnicity as a between-subject factor. This analysis showed no significant effect of the ethnicity of participants (F(1,35) = 0.95, p = 0.34, ηp2 = 0.03) or the ethnicity of faces (F(1,35) = 0.19, p = 0.67, ηp2 = 0.01), and no significant interaction between these two factors (F(1,35) = 0.56, p = 0.46, ηp2 = 0.02).

Finally, we conducted a binomial mixed effects model to simultaneously assess generalizability over stimuli and participants. The dependent variable was the binary preference for the feminized version of each stimulus (1 = chose the feminized version, 0 = chose the masculinized version). The model included face ethnicity as a within-subject fixed factor, participant ethnicity as a between-subject fixed factor, random intercepts for participant and stimulus, and maximally specified random slopes (i.e., face ethnicity within participants). This analysis showed a significant interaction (beta = 1.27, SE = 0.17, Z = 7.31, p < 0.001, 95% CI = [0.93, 1.61]) indicating that participants chose feminized faces more often than masculinized faces. There were no significant effects of the ethnicity of participants (beta = − 0.29, SE = 0.3, Z = − 0.97, p = 0.334, 95% CI = [− 0.89, 0.3]) or the ethnicity of faces (beta = 0.02, SE = 0.24, Z = 0.08, p = 0.94, 95% CI = [− 0.45, 0.49]), and no significant interaction between these two factors (beta = 0.25, SE = 0.33, Z = 0.76, p = 0.446, 95% CI = [− 0.4, 0.91]).

Discussion

We investigated the effects of sexually dimorphic facial characteristics on Chinese and White UK women’s jealousy. Both Chinese and White UK participants reported greater jealousy when imagining scenarios in which their romantic partner interacted with more feminine women. Moreover, both Chinese and White UK participants showed this effect when presented with Chinese and White UK faces. Our results for White UK women then replicate O’Connor and Feinberg’s (2012) findings for Canadian women’s jealousy. Our results for Chinese women demonstrate that the effect of sexually dimorphic facial characteristics on women’s jealousy is not unique to women who were raised in Western cultures and also occurs for women raised in an Eastern culture.

Although our analyses demonstrated that both Chinese and White UK women reported greater jealousy when imagining scenarios in which their romantic partner interacted with more feminine Chinese and White UK women, there may be subtle cultural differences in the magnitude of these effects that we did not detect in our study. Further research exploring this issue and also investigating how triggers of jealousy might change with experience of other cultures is a potentially important topic for future research. Similarly, exploring the developmental trajectory of jealousy perceptions to identify when sexually dimorphic physical characteristics in same-sex individuals trigger jealousy would provide further insight into the factors that shape jealousy perceptions.

There are limitations to the current study that should be acknowledged. First, the forced choice method we used shows that feminine shape characteristics influence jealousy when faces vary only on that dimension, but cannot show whether these effects are strong enough to influence jealousy when faces vary on other dimensions that also influence jealousy (e.g., makeup). Second, our sample size is relatively small. It would be useful for future work to address these issues.

In conclusion, we show strong effects of sexually dimorphic face shape on jealousy in women. That these strong effects were evident in both Western women (replicating O’Connor and Feinberg 2012) and Chinese women indicates that the effect of sexually dimorphic facial characteristics on women’s jealousy is not unique to women who were raised in Western cultures. Strong effects of sexually dimorphic face shape on jealousy in both Chinese and White UK women are also consistent with the strong effects sexually dimorphic characteristics have been shown to have on women’s facial attractiveness in both Western and Eastern cultures (Perrett et al. 1998).