New psychology research from the University of Otago, Warwick Business School, and University of California, San Diego, is helping explain why male faces with feminine features are considered attractive in some contexts but not others.
The study findings provide a new explanation for why the "Johnny Depp Effect" - which involves women tending to prefer men with more feminine faces - holds in some contexts, but not in others.
The international research team has found that when people are asked to rate the attractiveness of gender-blended face morphs they tend to judge them as less appealing if they are first asked to classify the face as male or female.
Across two separate experiments, gender blends were disliked when, and only when, the faces were first categorised by gender, despite an overall preference of the participants for more feminine features.
Study co-author Professor Jamin Halberstadt, of Otago's Department of Psychology, says that "processing fluency" - how easy it is to perceive, process, and categorise something - appears to account for this effect.
Piotr Winkielman, from the Warwick School of Business, and UCSD, adds that "mental effort can negatively color our initial impressions, even for things that are objectively pretty."
"The idea we tested is that the mental effort of having to assign a gender to an ambiguous face has a flow-on effect of negatively influencing how we feel about that face," says Professor Halberstadt.
In the second experiment, in which some participants were first asked to categorise gender-ambiguous faces by ethnicity, they did not subsequently judge the gender blends as less appealing.
This suggests that it is not merely a general aversion to facial ambiguity that is responsible for the effect, Professor Halberstadt says.
"It has previously been suggested that a woman's preference in male faces vary due to hormonal influences" that sometimes she is subconsciously looking for signs of a "nice dad" who will be a good provider, while other times it is the highly masculine "bad boy" with his 'better' genes.
"However, our research indicates that such changes in preferences can instead be explained by a simple cognitive process," Professor Halberstadt says.
"The more feminine faces are generally preferred, unless the context forces the viewer to put the face into rigid gender boxes," adds Professor Winkielman.
The findings are published in the international journal PLOS ONE. The study authors also include students Helen Owen from Otago and Evan Carr from the UCSD.