AI improves personality tests for faster, more accurate results

Personality tests are widely used in the workplace for recruiting, leadership training, and team building. But what if artificial intelligence could make them faster, smarter and more accurate? New research from the University of East London (UEL) suggests that machine learning could significantly improve the way organizational psychologists and managers use one of the most widely used personality tools, the DISC assessment.

The DISC assessment categorizes individuals into four behavioral styles (Dominance, Influence, Firmness, and Conscientiousness) and is commonly used in organizations to understand how people communicate, lead, and work within teams. The model’s appeal lies in its simplicity, allowing organizational psychologists and managers to quickly gain insight into behavioral trends.

However, traditional DISC ratings rely on simple scoring rules that assign users to one category based on their highest score. Although efficient, this approach can miss individuals with traits that span multiple behavioral styles and oversimplify personality.

New research investigates whether machine learning offers a more flexible, data-driven way to analyze DISC responses, potentially providing more accurate and nuanced personality insights. Rather than assigning people to a single category, this approach can also identify mixed behavioral patterns when individuals exhibit characteristics of multiple DISC styles.

Researchers tested several machine learning models that predict DISC personality types based on a standard 40-question assessment using responses from more than 1,000 participants. The most successful model achieved an accuracy of over 93%, demonstrating that artificial intelligence can reliably reproduce traditional DISC classification.

The study also considered whether the questionnaire itself could be made more efficient. The team showed that by identifying the most informative questions within an assessment, a much shorter version can still produce reliable results.

The model using only 10 carefully selected questions maintained an accuracy of over 91%. This suggests that DISC assessments can be delivered much more quickly with little loss of predictive strength.

Beyond prediction, the researchers applied clustering techniques to investigate how people naturally group together based on behavioral characteristics. This analysis reveals four distinct personality clusters that closely align with established DISC categories, while also highlighting subtle overlaps between behavioral styles.

Dr. Mohammad Hossein Amirhosseini, associate professor of computer science and digital technology at UEL and lead researcher, said the findings demonstrate how modern data science can enhance established psychological tools without losing their practical value.

DISC has long been a favorite in the workplace because it is simple and easy to apply. Our research shows that machine learning maintains its simplicity while adding deeper insights, allowing organizations to understand behavioral patterns with greater accuracy and flexibility. ”

Dr. Mohammad Hossein Amirhosseini, UEL Associate Professor of Computer Science and Digital Technology

Shorter assessments could make personality profiling easier to use in fast-paced professional environments where time is limited.

“A 10-question assessment tool that captures the underlying personality structure would make these assessments much more practical in situations such as recruiting, leadership development, and team building,” said Dr. Amirhosseini.

The study also suggests that machine learning may help advance personality assessment beyond rigid categories by identifying hybrid or blended behavioral profiles that may be missed by traditional scoring methods.

As organizations increasingly turn to data and artificial intelligence to support decision-making, such an approach could help usher personality assessment into a more flexible and evidence-based era.

“Human personality rarely fits neatly into a single box,” Dr. Amirhosseini added. “By using machine learning, we can better reflect the complexity of behavior while maintaining the clear, actionable insights that have made DISC so widely used.”