Skip to content High contrast mode

“Critical Thinking in Discovery Learning (for STEM education): A Cognitive-Computational Approach”

Tuition fee FREE
Application fee €80.00 one-time
More information 


According to Estonian Lifelong Learning Strategy, future school curricula need to integrate freer forms of learning to train students in evidence-based argumentation. Central to such innovation will be discovery learning scenarios for STEM education, where students can train two interrelated argumentation skills: the judgment of explanations for given data patterns and the self-directed search for new data to refine the judgment.
There is consensus among educational scientists, however, that younger students need to be assisted in discovery learning because their search and judgment skills are still in the process of being developed and consolidated. One way to realize such assistance is to make use of digital learning technology, so-called Assistive Learning Devices (ALDs), which attribute difficulties in students’ discovery learning (e.g., biased judgment or confirmatory search) to cognitive constraints (e.g., scope of attention) in order to provide guidance in search and judgment.
The development of effective ALDs requires modelling techniques of cognitive science, which formalize relationships between observable behaviour and cognitive states related to learning and attention. These cognitive-computational models have been developed under controlled conditions of laboratory experiments involving adult participants. Thus, for the development of an ALD, the question remains open whether these models generalize to the self-directed learning behaviour of younger students under natural conditions of everyday school life.
The goal of this PhD project, which is embedded into the project CEITER (, is to address this research question and to test as well as improve the ecological validity of existing cognitive-computational models of self-directed learning. Examining this question will take place in cooperation with a second already running PhD project at the School of Educational Sciences, which gathers empirical data on students’ search and judgment behavior in discovery learning scenarios at Tallinn high schools. Based on these data, a particular computational model will be tested and iteratively refined until accurate model-based predictions come to the fore.

Not available for applying at the moment

Application deadlines apply to citizens of: United States