AI can decipher every possible facial expression and emotion, outperforming therapists

The human face is often described as a mirror, reflecting our inner emotional states. In the realm of psychotherapy, understanding these emotional cues is crucial. AI has proven to be without peer when analyzing facial expressions and emotions.

Dr. Martin Steppan, a psychologist from the University of Basel, highlights the importance of facial expression analysis in both research and therapeutic settings.

The journey of interpreting facial expressions in psychology dates back to the 1970s, when psychologist Paul Ekman pioneered a coding system to link basic emotions like happiness, disgust, and sadness to specific facial cues.

Dr. Steppan notes, “Ekman’s system is a cornerstone in psychological emotion research.”

AI and facial expression analysis

However, a significant challenge in this area is the time-intensive nature of analyzing facial expressions, especially in research and therapy.

Often, psychiatry specialists turn to less direct methods, like skin conductance measurements, to gauge emotional states.

Seeking a more efficient solution, Dr. Steppan, along with his colleagues including emeritus Professor Klaus Schmeck, Dr. Ronan Zimmermann, and Dr. Lukas Fürer, embarked on a new study.

Their objective? To test the effectiveness of AI in recognizing emotional states from video recordings.

“We wanted to find out whether AI systems can reliably determine the emotional states of patients in video recordings,” says Steppan.

Their research harnessed the power of artificial neural networks trained on over 30,000 facial photos to identify six basic emotions.

The team put this AI system to the test by analyzing videos from therapy sessions of 23 patients with borderline personality pathology at the University of Basel’s Center for Scientific Computing.

The AI had a daunting task ahead — processing over 950 hours of video recordings.

Remarkable emotional analysis results

The results were nothing short of remarkable. Statistical comparisons revealed a high level of agreement between the AI’s analysis and that of three trained therapists.

The AI matched the human capability in assessing facial expressions and excelled in detecting fleeting emotions, like a brief smile or a momentary expression of disgust, which are often missed or only subconsciously perceived by therapists.

An intriguing finding emerged from the AI analysis. Patients who showed emotional engagement and smiled at the beginning of a therapy session were less likely to cancel subsequent sessions compared to those who appeared emotionally detached.

This suggests that ‘social’ smiling might be a key predictor of therapy success in individuals with borderline personality pathology symptoms.

Dr. Steppan expressed his astonishment at the AI’s proficiency, “We were really surprised to find that relatively simple AI systems can allocate facial expressions to their emotional states so reliably.”

Future of AI in psychotherapy

This breakthrough indicates that AI could become an invaluable tool in therapy and research. It could aid in analyzing existing video recordings from studies, making it easier to identify emotionally significant moments during conversations. Such a capability could also enhance the supervision of psychotherapists.

In summary, while AI shows immense potential in the field of psychotherapy, Dr. Steppan reminds us, “Therapeutic work is still primarily about human relationships, and remains a human domain, at least for the time being.”

This study marks a significant step forward in the intersection of technology and psychology, opening new doors for understanding and treating emotional states more effectively.

The full study was published in the journal Psychopathology.

—–

Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.

—–