How being perceived to be an artist boosts feelings of attraction in others

Music production is a universal phenomenon reaching far back into our past. Given its ubiquity, evolution theorists have postulated adaptive functions for music, such as strengthening in-group cohesion, intimidating enemies, or promoting child bonding. Here, we focus on a longstanding Darwinian hypothesis, suggesting that music production evolved as a vehicle to display an individual’s biological fitness in courtship competition, thus rendering musicality a sexually selected trait. We also extend this idea to visual artists. In our design, we employed different versions of naturalistic portraits that manipulated the presence or absence of visual cues suggesting that the person was an artist or a non-artist (e.g., farmer, teacher, physician). Participants rated each portrayed person’s appeal on multiple scales, including attractiveness, interestingness, sympathy, and trustworthiness. Difference scores between portrait versions revealed the impact of the artistic/non-artistic visual cues. We thus tested Darwin’s hypothesis on both a within-subject and within-stimulus level. In addition to this implicit approach, we collected explicit ratings on the appeal of artists versus non-artists. The results demonstrate divergent findings for both types of data, with only the explicit statements corroborating Darwin’s hypothesis. We discuss this divergence in detail, along with the particular role of interestingness revealed by the implicit data.

as compared to non-artists.Among each other, the visual artists' boost is steeper than the boost of musicians (***p < 0.001).Right: The boost in Interestingness from torsos to facials is significantly higher for visuals artists as compared to both musicians and non-artists (***p < 0.001).Due to the stricter way of testing in this approach, the difference between musicians and non-artists is no longer significant (p = 0.263), but only visible by trend.
Table S7.LMM results predicting Sympathy through the portrait's Layout (face, torso) and Status of the depicted person (musician, visual artist, non-artist) as well as the Sympathy-Boost just through Status.To test H2, the following models have been fitted:  S7).Left: EMMs including 95 % CI boundaries for facial portraits (F) and half-length torsos (H) across the three Status groups.The boosting effect is significantly steeper for non-artists as compared to musicians (*p < 0.05).Right: The boost in Sympathy from torsos to facials is not significantly different between the three status groups.Due to the stricter way of testing in this approach, the difference between musicians and non-artists is no longer significant (p = 0.146), but only visible by trend.

Figure S6
. LMM results for Wish-to-Meet ratings (cf.Table S9).Left: EMMs including 95 % CI boundaries for facial portraits (F) and half-length torsos (H) across the three Status groups.The boosting effect of visual artists is significantly steeper than that for musicians and non-artists (***p < 0.001).Right: The same result pattern is reflected in the Model-M-Boost: The boost in Wish-to-Meet from torsos to facials is significantly higher for visual artists than for both non-artists and musicians (**p < 0.01).
boundaries for facial portraits (F) and half-length torsos (H) across the three Status groups.The boosting effect is significantly steeper for both musicians (*p < 0.05) and visual artists (***p < 0.001)

Figure S5 .
Figure S5.LMM results for Trustworthiness ratings (cf.TableS8).Left: EMMs including 95 % CI boundaries for facial portraits (F) and half-length torsos (H) across the three Status groups.The boosting effect of non-artists is significantly steeper than that of musicians (***p < 0.001) and visual artists (*p < 0.05).Right: The boost in Trustworthiness from torsos to facials is significantly higher for non-artists as compared to musicians (***p < 0.001).Due to the stricter way of testing in this approach, the difference between non-artists and visuals artists is no longer significant (p = 0.15), but only visible by trend.

Figure S7 .
Figure S7.LMM results for ViewingTime (cf.TableS10).Left: EMMs including 95 % CI boundaries for facial portraits (F) and half-length torsos (H) across the three Status groups.The boosting effect is significantly flatter for musicians as compared to both non-artists and visual artists (***p < 0.001).Right: The same result pattern is reflected in the Model-V-Boost: The boost in ViewingTime from torsos to facials is significantly lower for musicians as compared to the other two Status groups (**p < 0.01).

Figure S8 .
Figure S8.Effect sizes of boosts per subgroup for all remaining variables beside Attractiveness (which is shown in Fig.6).The forest plots depict EMMs for the boosts expressed as Cohen's d for each stimulus and predicted in a linear model by subgroup membership.The error bars represent 95 % confidence limits.For better readability, the superordinate status membership of the subgroups is coded in color: non-artists are shown in blue, musicians in orange and visual artists in green.

Table S2 .
General Observations II: LMM results predicting Ratings of each scale through the portrait's Layout format.The following model has been fitted:Level 1: Ratingij = 0j + i0 + 1j Layout + ij

Table S2 [
continued].LMM results predicting Ratings through the portrait's Layout.

Table S3 .
LMM results predicting Attractiveness through the portrait's Layout (face, torso) and Status of the depicted person (musician, visual artist, non-artist), as well as the Attractiveness-Boost just through Status.To test H2, the following models have been fitted:

Table S4b .
LMM results for a reduced data set of female raters and male portraits, predicting Attractiveness-Boost through Status of the depicted person(musician, visual artist, non-artist).This is thus a modified version of Model-A-Boost for a reduced data set:

Table S4c .
LMM results for a reduced data set of male raters and female portraits, predicting Attractiveness-Boost through Status of the depicted person(musician, visual artist, non-artist).This is thus a modified version of Model-A-Boost for a reduced data set:

Table S6 .
LMM results predicting Interestingness through the portrait's Layout (face, torso) and Status of the depicted person (musician, visual artist, non-artist) as well as the Interestingness-Boost just through Status.To test H2, the following models have been fitted: Figure S3.LMM results for Interestingness ratings (cf.TableS6).Left: EMMs including 95 % CI

Table S8 .
LMM results predicting Trustworthiness through the portrait's Layout (face, torso) and Status of the depicted person (musician, visual artist, non-artist) as well as the Trustworthiness-Boost just through Status.To test H2, the following models have been fitted:

Table S9 .
LMM results predicting Wish-to-Meet through the portrait's Layout (face, torso) and Status of the depicted person (musician, visual artist, non-artist) as well as the Wish-to-Meet-Boost just through Status.To test H2, the following models have been fitted:

Table S10 .
LMM results predicting ViewingTime through the portrait's Layout (face, torso) and Status of the depicted person (musician, visual artist, non-artist) as well as the ViewingTime-Boost just through Status.To test H2, the following models have been fitted:

Table S21 .
LMM results for the comprehensive covariates model (model.9 in R script) predicting Attractiveness through Layout and Status, while controlling for stimulus order, character's smile intensity, character's gaze direction, character's estimated income, character's estimated age, character's gender, rater's gender, rater's preference for professions and their own music production (today and overall, including their past).

Table S22 .
Model comparison.The initial model (model.0)predicts the Attractiveness by the factors Layout and Status.It is Model-A as described in the manuscript.The consecutive models (model.1 to 9) append cumulatively one of the following covariates: stimulus order, estimated income of the depicted character, rater's preference for professions (Prof_Pref), character's smile intensity, character's gaze direction, character's estimated age, character's gender, rater's gender, rater's own musicality.The statistical model fit to the data is tested via the anova function in R.