Neural Network in Russia Trained to Assess Human Creativity
AI tools are helping psychologists process creativity assessments based on multiple projective tests
Researchers at South Ural State University, working with colleagues from the Institute of Psychology of the Russian Academy of Sciences, have trained a neural network to analyze individual creativity levels, Russia’s Ministry of Science and Higher Education said.
The scientists patented two datasets for assessing creativity in children and adults. A neural network trained on these datasets can instantly generate results for the Torrance test. The team compiled and patented the data based on a sample of 3,007 participants in Russia who had previously completed the test developed by American psychologist Ellis Paul Torrance.
The researchers also built an AI model based on the patented datasets that helps process results from multiple creativity assessment tests.
Available to Everyone
According to Yulia Dmitrieva, an associate professor in the Department of Management Psychology and Professional Activities, the Torrance test was previously used mainly for children and adolescents, but demand for creative employees is now growing in the labor market.
Now, anyone can take a creativity test online using either the Urban method or the Torrance test via a dedicated web portal. The trained neural network generates results within minutes, eliminating the need to visit a psychologist.
The new test developed at South Ural State University evaluates not only the six universal Torrance creativity factors but also additional indicators, including cultural code and personality traits.
