Educational digital assistants designed to understand and support students emotionally as well as academically are increasingly becoming a vital tool in learning environments. These assistants are similar to devices like Siri or Alexa but are specifically made for educational settings. They are designed to recognise and respond to students’ feelings, offering both emotional support and academic guidance. When used in digital tutoring platforms – such as Duolingo for language learning or ALEKS for maths – these assistants can help students manage feelings such as anxiety and to build confidence, making learning more engaging.

This blog post looks into the current research on these empathetic digital teaching assistants by summarising a thorough review of studies led by Ortega-Ochoa et al. (2024). The review examined 1,162 studies from databases such as Scopus and Web of Science, covering research from 2018 to 2022. Out of these, 13 studies were selected for a detailed analysis based on criteria including relevance and quality. The summary below explains how these assistants are developed and how their success is measured.

Design principles of empathetic digital teaching assistants

The main design features include a strong emphasis on the ability to understand and share the feelings of students, encouraging conversations that aid learning, expertise in the subject matter, and customised feedback that matches the student’s learning level. For example, Kumar (2021) described an assistant that can interact in a very human-like way, using personal greetings which helps build a connection. Another important feature is how these assistants can use information about what a student already knows to make the learning experience more personalised and effective. For instance, Wu et al. (2020) used initial tests to understand each student’s knowledge level, allowing the assistant to adjust its interactions for more personalised support.

‘Another important feature is how these assistants can use information about what a student already knows to make the learning experience more personalised and effective.’

Measuring success

The effectiveness of these digital teaching assistants is measured by various factors identified in the review. These include the students’ learning performance, evaluated through tests on different types of knowledge and skills, and how students feel about the assistant and their learning experience, assessed through surveys and interviews. These measurements look at how connected students feel to the assistant, how much they enjoy the interactions, and how confident they feel about what they’re learning. Some studies, like those by Ayedoun et al. (2020), mostly use numbers to measure impact, while others use a combination of methods to get a fuller picture of the assistant’s effect on learning. The review found that the kind of feedback these assistants give can greatly affect learning outcomes (Wambsganss et al., 2021). For example, Oker et al. (2020) observed that students who received feedback through both verbal and facial expressions from a virtual tutor were more engaged and performed better in maths tests than those who received feedback through just one mode. Based on the evaluation indicators, the review introduces a framework for evaluating the learning outcomes facilitated by empathetic digital teaching assistants.

The review provides valuable information for researchers, educators and policymakers on how to develop these empathetic digital teaching assistants and assess their effectiveness. This blog post gives an overview of the latest findings in this field. For more in-depth information, please refer to the full review by Ortega-Ochoa et al. (2024).

References

Ayedoun, E., Hayashi, Y., & Seta, K. (2020). Toward personalized scaffolding and fading of motivational support in L2 learner-dialogue agent interactions: An exploratory study. IEEE Transactions on Learning Technologies, 13(3), 604–616. https://doi.org/10.1109/TLT.2020.2989776

Kumar, J. A. (2021). Educational chatbots for project-based learning: Investigating learning outcomes for a team-based design course. International Journal of Educational Technology in Higher Education, 18(65), 1–28. https://doi.org/10.1186/s41239-021-00302-w

Oker, A., Pecune, F., & Declercq, C. (2020). Virtual tutor and pupil interaction: A study of empathic feedback as extrinsic motivation for learning. Education and Information Technologies, 25, 3643–3658. https://doi.org/10.1007/s10639-020-10123-5

Ortega-Ochoa, E., Arguedas, M., & Daradoumis, T. (2024). Empathic pedagogical conversational agents: A systematic literature review. British Journal of Educational Technology, 55(3), 886–909. https://doi.org/10.1111/bjet.13413

Wambsganss, T., Weber F., & Söllner, M. (2021). Designing an adaptive empathy learning tool. In T. X. Bui (Chair), Proceedings of the 54th Hawaii International Conference on System Sciences (pp. 54–63). University of Hawaii at Manoa. https://doi.org/10.24251/HICSS.2021.007