More recently, digital coordination made possible through apps like Slack, Zoom, and Google hangouts has sustained several industries during the COVID-19 Pandemic. However, their success grounds the potential for a virtual overhaul in the very near future.
In fact a new study conducted by researchers from two Moscow Universities, HSE University (Higher School of Economics) and Open University for the Humanities and Economics, concluded that modern facial recognition technology can determine a person’s personality type 58% of the time from selfies submitted by candidates.
Twelve thousand volunteers uploaded 31,000 selfies to aid in the research. These selfies were subsequently split into two groups. The first group was analyzed via a traditional AI system and the other was used to test a new, more intuitive neural network.
Ideally, employers could enlist the aid of advanced technological systems to remove hiring biases. The technology as is today, is imperfect but promising; suggesting the answer to an equal playing field might require some automated assistance.
“There is ample evidence that morphological and social cues in a human face provide signals of human personality and behavior. Previous studies have discovered associations between the features of artificial composite facial images and attributions of personality traits by human experts,” the authors wrote in the new paper. “The findings strongly support the possibility of predicting multidimensional personality profiles from static facial images using ANNs trained on large labeled datasets. Future research could investigate the relative contribution of morphological features of the face and other characteristics of facial images to predicting personality.”
Assessing the Big Five personality traits using real-life static facial images
The initial hypothesis isn’t unfounded.
Previously conducted research has suggested that genetic background contributes to both craniofacial structure and personality. Genetic predictors of facial characteristics have been discovered both in clinical data and non-clinical populations.
All of this to say a principled analysis could detect genetic correlates from a face alone that could be traced to behavior By employing the artificial neural network, the researchers behind the new study published in the journal Scientific Reports, assessed 128 different factors of a person’s face, including the width of the mouth, and height of the lips or eyes.
This process yielded five broad personality types: openness, conscientiousness, agreeableness, and neuroticism-OCEAN.
The pioneering technology accurately guessed the volunteers’ personality type 58% of the time when they compared the readings to the questionnaires filled in by volunteers before the test.
“The advantage of our methodology is it is relatively simple – it does not rely on 3D scanners or 3D facial landmark maps – and can be easily implemented using a desktop computer,” explained, Dr. Evgeny Osin, the study’s co-author. “This study presents new evidence confirming that human personality is related to individual facial appearance.”
According to Dr. Osin, the results of the study demonstrated that real-life photographs taken in uncontrolled conditions could be used to predict personality types with complex computer vision algorithms.
The AI was even able to estimate lifestyle factors like physical health, temperament, and relationship proclivities, with alarming accuracy.
“The highest correlations between observed and predicted personality scores were found for conscientiousness (0.360 for men and 0.335 for women) and the mean effect size was 0.243, exceeding the results obtained in prior studies using ‘selfies,” the author concluded.
CW Headley is a reporter for the Ladders and can be reached at firstname.lastname@example.org