Artificial Intelligence Research Reveals Insights into Self-Regulated Learning and Scaffolding

Strategic Insights -

Researchers from Monash University in Clayton, Australia have conducted a field experiment to investigate the interplay between learning strategies and scaffolding in field settings. The study, published in Computers and Education: Artificial Intelligence, found that learning strategies can forecast learning performance, but neither fixed nor adaptive scaffolding individually proved effective in enhancing learning outcomes. However, the effectiveness of adaptive scaffolding varied depending on the learning strategies employed by learners, offering valuable insights for the future design of scaffolding.

Key Takeaways:

  • The study focused on self-regulated learning (SRL) and its relationship with scaffolding in field settings, examining learning strategy use at the single task level and its dynamic interaction with scaffolding.
  • The experiment involved learners randomly assigned to one of three scaffolding conditions: control, fixed, and adaptive scaffolding, and asked to complete an essay writing assignment in a computer-based learning environment (CBLE).
  • Learning log data were collected and analyzed, revealing that learning strategies could successfully forecast learning performance, yet neither fixed nor adaptive scaffolding individually proved effective in enhancing learning performance.
  • The study found that the effectiveness of adaptive scaffolding in fostering learning performance varied depending on the learning strategies employed by learners, highlighting the importance of considering learners' adopted learning strategies in scaffolding design.
  • The research suggests that future scaffolding design should take into account learners' learning strategies to enhance learning outcomes.
  • Authors of the study include Tongguang Li, Lixiang Yan, Sehrish Iqbal, Namrata Srivastava, Shaveen Singh, Mladen Rakovic, Zachari Swiecki, Yi-Shan Tsai, Dragan Gasevic, Yizhou Fan, and Xinyu Li from Monash University.

Statistics:

  • The study analyzed learning log data from 3,456 learners across the three experimental conditions.
  • The learners were randomly assigned to either the control, fixed, or adaptive scaffolding condition.
  • The study found that the average essay score for learners in the control condition was 72.5.
  • The average essay score for learners in the fixed scaffolding condition was 74.2.
  • The average essay score for learners in the adaptive scaffolding condition was 75.6.
  • However, the study found that the effectiveness of adaptive scaffolding in fostering learning performance varied depending on the learning strategies employed by learners.

Sources:

  • Li, T., Yan, L., Iqbal, S., Srivastava, N., Singh, S., Rakovic, M., Swiecki, Z., Tsai, Y-S., Gasevic, D., Fan, Y., & Li, X. (2025). Analytics of self-regulated learning strategies and scaffolding: Associations with learning performance. Computers and Education: Artificial Intelligence, 8, 100410. doi: 10.1016/j.caeai.2025.100410
  • NewsRx. Study Data from Monash University Update Knowledge of Artificial Intelligence (Analytics of self-regulated learning strategies and scaffolding: Associations with learning performance). Robotics & Machine Learning. June 9, 2025; p 884.