Beyond the Lombard concepts that have already been stored in the trunk of pseudosciences, artificial intelligence does seem to have found correlations between our face and our personality .
This is what a new study published in Nature suggests, the results of which indicate that real-life photographs taken under uncontrolled conditions can be used to predict personality traits using computer vision algorithms.
In psychology, the Big Five Model (or simply Big Five) is a taxonomy or classification of personality traits that analyzes the composition of five personality dimensions in its broadest sense, and can be correlated with certain faces , according to the aforementioned study . The big five would be: Openness to experience, Awareness, Extraversion, Kindness and Neuroticism.
According to the study, these types of faces would be correlated with such dimensions of personality:
The initial sample included 25,202 participants who completed the questionnaire and uploaded a total of 77,346 photographs. Finally, artificial neural networks (ANNs) were trained on a large set of labeled data to predict the scores.
It is not very well known if and why these correlations have a true causal link. But if they did, there are several theoretical reasons to expect associations between facial images and personality .
First, the genetic background contributes to both the face and the personality. In addition to shaping the face, genes also play a role in the development of various personality traits, such as risk behavior, and the contribution of genes in some traits outweighs the contribution of environmental factors.
Our brain, intuitively, would also find a face more beautiful or reliable based on face patterns , just as we unconsciously detect the health of the other person, as suggested by a study a few years ago , led by researchers at the University de Macquarie, who developed a new computer model that can predict the state of certain health characteristics simply by analyzing a person’s face.
The study, which was published in the journal Frontiers in Psychology , shows that the model is capable of detecting information about BMI, body fat, and blood pressure levels .
This allows us to identify healthy people to mate or to form cooperative relationships, it is a mere evolutionary mechanism, which algorithms have now learned to identify with much more certainty. You can delve into all this in the following video :