Congratulations to NC State Department of Computer Science Ph.D. students Yasitha Rajapaksha and John Thomas Bacher and Professor Tiffany Barnes on receiving an Honorable Mention at the Association for Computing Machinery (ACM) Conference on Human Factors in Computing Systems (CHI 2026). 

Their paper, “Investigating the Interaction of Game Features and Spatial Skills with Performance and Perceived Difficulty in a Block-Based 3D Programming Puzzle Game”, which studied how different game design features affect players’ performance and perceived difficulty in BOTs, a 3D educational programming puzzle game designed for beginners. 

“CHI is one of the most respected conferences in human-computer interaction, so having our paper accepted there already meant a lot to me,” said Rajapaksha. “Receiving an Honorable Mention made it even more special because it felt like recognition for the enormous amount of effort and care that went into getting the study right.”

Gaming to Learn

Block-based programming puzzle games are a popular tool for teaching novice learners the basics of programming. In BOTs, players control a character by building programs with drag-and-drop command blocks. Learners solve puzzles by navigating through 3D environments, moving boxes, activating switches and using programming concepts like loops and functions to complete levels. 

In the study, 60 participants played specially designed levels, where the NC State research team systematically changed features like layout size, dimensionality, playable character orientation, number of loops and number of functions. Performance was measured using metrics like completion time, number of attempts, interactions, deletions and how close players’ solutions were to the optimal solution. Players were asked how difficult each level felt through an in-game survey, and their spatial skills were determined using a standardized test.

“One of the most interesting findings was that spatial reasoning skills strongly predicted performance,” said Rajapaksha. “Players with stronger spatial skills were better at understanding the 3D layouts, planning the character’s movements and building efficient solutions.”

However, spatial skills did not predict how difficult players felt the levels were. 

“Our interpretation is that players judged difficulty less by how much mental effort they spent and more by whether they eventually solved the puzzle,” said Rajapaksha. “In other words, trying, failing, revising and trying again may have felt like a normal part of gameplay, rather than ‘difficulty’, as long as players were still making progress.”

Video games by their very nature exist in a virtual space, even 3D ones like BOTs. However, people exist in the real world, not a virtual one; this leads to basic human instincts — like the concept of ‘in front or behind’ — affecting their gameplay and ability to learn.

“We found that something as simple as the character’s starting orientation had a surprisingly large effect,” said Rajapaksha. “When the character started facing backward, players took much longer and made more attempts, deletions and interactions; this suggests that even small changes in perspective can make a 3D programming task much harder.”

Programming, as well as the broader field of computer science, are not easy concepts to learn, let alone master. That’s why games can be so useful in teaching these concepts.

“I’m especially excited about the broader impact of this work. We want educational programming games to better support students who may struggle with spatial reasoning or traditional approaches to computer science education. Research like this can help us design learning tools that are more adaptive, accessible and effective for a wider range of learners,” said Rajapaksha.

Read the paper here. To learn more about gaming research at NC State, check out the Game2Learn Lab.