Art has no gender, only gender bias.

The current study investigated whether gender could be distinguished in a set of Abstract Expressionist paintings (N = 160) by an equal number of female (Krasner, Frankenthaler, Mitchell, Hartigan) and male (Pollock, Louis, Twombly, Kline) artists. The study consisted of three experiments and was performed by three separate groups of participants (total N = 800). In Experiment 1, participants judged whether the artwork was painted by a female or male artist and then reported whether their decision was based on colour, line, or composition of the artwork. In Experiment 2, participants assessed each work according to thirty-two characterizations drawn from art criticism, while in Experiment 3 they indicated their aesthetic appreciation for the artworks. Results showed that, overall, participants were significantly more likely to judge that the artworks were painted by males, regardless of the actual gender of the artist (2= 266.3, df = 1, p < 0.001). Participants relied significantly more on colour when judging artworks as painted by females, compared to line and composition (2= 697.5, df = 14, p < 0.001), in line with image feature analysis showing that women artists used more warm colours than men (F(1,152) = 7.78, p < 0.01). Contrary to the art critics’ view of female Abstract Expressionist art in the 1960s, artworks by women were rated higher on masculine attributes and lower on feminine attributes than the ones made by their male colleagues (beta = -0.31, t(2551) = -3.98, p < .001). Finally, liking, complexity, pleasure and order ratings of artworks made by women were not significantly different from those by men (all post-hoc comparisons: p > .05). Across the three experiments, we provide evidence that viewers have a male gender bias towards paintings’ authorship, although the artists’ gender is not a significant contributor to the appreciation of Abstract Expressionist paintings.