AIP-OS Dataset: Cognitive Performance in AI-Mediated University Learning
Description
This dataset provides the complete anonymized academic performance records of 182 undergraduate students from a quasi-experimental study conducted in Colombia. It supports the findings of the article “Redefining the classroom: posthuman cognition and algorithmic agency in higher education.” Students were divided into a control group and an experimental group exposed to the AIP-OS intelligent agent, developed under the Belief–Desire–Intention (BDI) architecture. The dataset includes individual-level scores on interpretive, argumentative, and propositional competencies across five curricular units, along with aggregate performance metrics and cognitive profiles. All data are de-identified and fully compliant with ethical research standards. Keywords Educational AI; BDI agent; Higher education; Cognitive performance; Posthuman pedagogy; Learning trajectories; Human–machine interaction; Distributed cognition