home
Search
Man operating a computer in a factory.

Scientific publications

Industry 5.0: Robots reducing workers’ cognitive load

Resource Details

  • Written by

    Ieva Miseviciute

  • Read time

    3 min

Industry 5.0 places a worker’s wellbeing at the center of the manufacturing process. Rather than replacing humans with machines, the new industrial era envisions seamless human-robot collaboration (HRC), where collaboration robots – cobots - bring repeatability and precision, complementing humans’ dexterity and flexibility. While HRC increases productivity in the manufacturing process, the user experience and human wellbeing during this interaction are less defined. Recognizing this gap, a research team from Politecnico di Torino, Italy, embarked on a mission to quantify human cognitive load during HRC in a product assembly task.  

Cognitive load, the mental effort required to perform a task, is essential in human work performance and wellbeing. Excessive load can hamper productivity, while optimal load can enhance it. By measuring and understanding cognitive load, the team aims to pave the way for a future where HRC not only boosts efficiency but also improves human wellbeing at its core. 

The study participants were tasked to assemble products (e.g., a diaphragm water pump) of varying complexity, using two methods: manual and with cobot assistance. The scientists hypothesized that tasks with greater complexity would lead to increased cognitive load, while the support of a cobot could alleviate it. To gauge the influence of product assembly complexity and HRC on cognitive load, the researchers measured heart rate variability, galvanic skin response, and eye movements with Tobii Pro Glasses 3 mobile eye tracker. 

The study findings underscore the benefits of cobot assistance in mitigating cognitive load during complex assembly tasks. The impact of cobot on cognitive load varied across all physiological measures – galvanic skin response showed an upward trend as assembly complexity increased, while no particular trend was observed for the heart rate variability measures. Notably, eye tracking metrics yielded the most significant results indicating changes in cognitive load: 

  • The average amplitude of saccades was greater in collaboration with cobots than in manual tasks, indicating reduced cognitive load levels when cobots are involved in the assembly process. 
  • The average peak velocity of saccades was higher when collaborating with cobots, compared to manual modality, suggesting reduced cognitive load when working with cobots. 
  • The average duration of fixation was longer in more complex tasks, supposedly requiring more cognitive effort. However, the study authors suggest that longer fixations during complex tasks might as well reflect a flow state – deep engagement in an activity. 
  • The average pupil diameter was larger during more complex tasks, suggesting greater cognitive load than tasks with lower complexity. 

Study authors emphasize that no single physiological metric can individually describe cognitive load, and therefore, a comprehensive, multimodal approach remains essential. In practical scenarios, physiological data from sources like eye tracking, heart rate variability, and skin conductance can be fed to cobots to customize their interactions based on the individual cognitive loads of human workers.  

Cited publication

Capponi, M., Gervasi, R., Mastrogiacomo, L., & Franceschini, F. (2024).
Assembly complexity and physiological response in human-robot collaboration: Insights from a preliminary experimental analysis. Robotics and Computer-Integrated Manufacturing, 89, 102789.

Read more

Interested in similar articles? Visit our scientific publication hub to see all our scientific publication highlights.

Resource Details

  • Written by

    Ieva Miseviciute

  • Read time

    3 min

Author

  • Tobii employee

    Ieva Miseviciute, Ph.D.

    SCIENCE WRITER, TOBII

    As a science writer, I get to read peer-reviewed publications and write about the use of eye tracking in scientific research. I love discovering the new ways in which eye tracking advances our understanding of human cognition.

Discover more about human-machine interaction