UA Little Rock Ph.D. Student Uses AI to Study Emotion

What role do emotion and social connection play in scientific discovery?

Praveshika Bhandari, a researcher from Nepal pursuing a doctorate in computer science at the University of Arkansas at Little Rock, is using artificial intelligence and natural language processing — a field that enables computers to analyze and interpret human language — to explore that question. By analyzing Albert Einstein’s personal writings alongside his scientific work, her research traces the often-hidden emotional roots of innovation.

Her research sits at the intersection of two long-standing interests. Along with her bachelor’s degree in computer science, she also holds a degree in psychology. She said she has long been fascinated by how people think, feel and make decisions.

Bhandari’s work uses AI to examine how emotions and social relationships influence intellectual work. Her goal is to better understand how emotions connect to creativity and intellect and how social influences shape research outcomes. Scientific publications tend to present ideas in polished, emotionless terms, but private correspondence and personal writings often tell a different story. Her central question is how emotional states and social context influence the direction and outcomes of scientific research.

“Say there’s a scientist and he has publications, but also has personal letters, emails and other writings,” Bhandari said. “The publications are edited down to only the facts, but when the scientist talks about the same research topics in their letters, we are able to see what the scientist is feeling and where they are aiming to go with the research next.”

Bhandari uses Einstein as her case study to examine how personal experiences and emotion intersect with scientific innovation. He was chosen not only for his scientific legacy but also for the extensive personal writings he left behind and the complexity of his life beyond his work. While he is often viewed only through the lens of his genius, Einstein’s writings reflect a life shaped by migration, war and family challenges, including divorce. These experiences unfolded alongside his major intellectual contributions, making his letters a rich source for studying how creativity persists through emotional highs and lows.

“It was really interesting to see the ways he was thinking about different things,” Bhandari said. “He was still able to make all these genius contributions while having all the ups and downs of normal human life.”

The project began as an idea developed by Dr. Arya Basu during his earlier work at Emory University, where he explored how AI could reveal the emotional and social context behind scientific discovery. After bringing that line of inquiry to UA Little Rock, Basu became Bhandari’s primary doctoral adviser, working with her to shape the concept into a feasible research design. Dr. Jan Springer served as a secondary adviser, offering feedback as the study took form.

In the early stages of the research, Bhandari examined whether similar methods could identify signals of suicidal ideation in social media posts. However, privacy concerns and limited access to data made that approach impractical. Basu then suggested applying the techniques to Einstein’s publicly available writings, allowing the project to move forward using a rich archival record without the same ethical constraints.

Using a technique known as context-aware emotion modeling, Bhandari feeds AI models both Einstein’s personal letters and his scientific writings to analyze emotional patterns over time. The approach allows the system to account for what was happening in Einstein’s personal life as well as his professional work. By grouping his writings around subjects such as relativity or gravity, the model can identify whether Einstein expressed frustration, excitement or optimism as his ideas evolved. One pattern shows rising frustration before major publications, followed by an increase in positive emotions after his work was published.

Bhandari’s findings highlight the role emotions play in intellectual creativity.

“Emotion shapes everything we do every day in our lives,” she said. “Even in science, the way you feel about a problem can shape how you approach it. Emotions matter and impact your work. By studying how great people achieve great things, we can learn lessons that may help us in our own work and creativity.”

Bhandari is working to automate the process through machine learning so it can be applied beyond Einstein. Her goal is to better understand how people think, whether they are renowned figures or everyday individuals.

“We want to see the lessons hidden behind how anyone thinks or creates,” she said.

Her work is not limited to scientists. The same techniques could be applied to artists, athletes or anyone engaged in creative work. She also plans to expand her analysis beyond text to include images, paintings and handwriting in an effort to better understand the hidden processes of the human mind.

Bhandari said the significance of her research lies in addressing an underdeveloped area of study: the relationship between human emotion and professional life. While therapists will always play a central role, she said AI tools could one day complement human care by helping identify emotional patterns earlier or by offering additional analytical insight.

The project extends beyond her dissertation, and she plans to continue the work after graduating.

“I’m really fascinated by machine learning and natural language models, but I’m also very interested in human emotions and what goes on within the mind,” Bhandari said. “So I hope to keep doing research in this interdisciplinary area where humans and computers can come together — not just meet, but complement one another.”

As Bhandari prepares to graduate with her Ph.D. next semester, her work reflects the kind of interdisciplinary research underway at UA Little Rock. By examining the emotional lives behind scientific breakthroughs, her research challenges the idea of science as purely rational and detached, suggesting instead that creativity, persistence and discovery are deeply human processes shaped by emotion as much as intellect.