Promoting reflexive learning in cartography in an international classroom

Juliane Cron, Melissa del Carmen Ernstberger, Liqiu MengAbstracts 2025 · DOI

Abstract

As generative AI begins to automate the technical aspects of map-making, cartographic education must pivot toward fostering "soft skills" such as critical thinking, social responsibility, and intercultural communication. This paper presents a case study of the "Mapping Project" course at the Technical University of Munich (TUM), which utilizes a reflexive learning framework. By strategically pairing international students with complementary skills in art, science, and technology, the course encourages students to move beyond technical execution to reflect on the biases and social implications of their work. Results demonstrate high student engagement and the creation of innovative cartographic prototypes that may eventually serve as high-quality training data for specialized AI models like MapGPT.

Introduction

In the contemporary landscape of geospatial science, technical proficiency is no longer the sole benchmark of a successful cartographer. The rise of automated tools and generative AI necessitates a shift in higher education toward "reflexive learning"—the process of questioning one’s own assumptions, biases, and methods through experience and dialogue. Cartography, situated at the intersection of art, science, and technology, provides a unique fertile ground for this pedagogical evolution.

Traditional cartography curricula often focus on the how of map-making. However, the "Mapping Project" at TUM seeks to address the why, leveraging the diverse backgrounds of an international student body to foster a more socially-conscious and critical approach to the discipline. The challenge lies in moving beyond teaching map production to teaching students how to reflect on their choices and the broader social impact of their work.

Methodology: The Reflexive Framework

The "Mapping Project" is a 5-credit elective within the International MSc Cartography program. The pedagogical approach is built upon three core pillars:

Skill Categorization and Strategic Pairing

At the start of the course, students self-identify their strengths and weaknesses across three domains: Art, Science, and Technology. This categorization acknowledges that modern cartography is rarely a solitary endeavor. Students are then strategically paired with partners who possess complementary strengths. For instance, a student with a background in data processing (Technology) might be paired with one strong in design (Art). This collaboration necessitates a "co-learning" environment where students must translate their specialized knowledge for their partners, thereby identifying their own knowledge gaps.

Dialogue-Based Learning

The course employs several dialogue-based methods to encourage continuous reflection:

  • Lightning Talks: Two-minute presentations where students share progress and receive immediate peer feedback.
  • Continuous Peer Review: Structured sessions where pairs critique each other's methodologies and creative choices.
  • Lecturer as Facilitator: Moving from a top-down instruction model to a mentorship role, where lecturers facilitate unconventional learning paths and identify new topics for their own teaching portfolios.

Process-Oriented Assessment

To ensure students value the journey as much as the destination, the assessment is split to reward documentation and reflection:

  • 50% Final Map: The technical and aesthetic quality of the product.
  • 20% Descriptive Poster/Methodology: A detailed account of the technical choices and creative decisions made.
  • 30% Final Presentation/Exhibition: The ability to defend the work and reflect on the collaborative process.

The Student Cohorts

As a pedagogical study, the data consists of student outputs and feedback from two consecutive years:

  • 2022 Cohort: 19 students from 16 countries (9 projects).
  • 2023 Cohort: 27 students from 17 countries (14 projects). Students entered the program with diverse backgrounds in geography, urban planning, computer science, design, architecture, and physics, providing a rich cross-disciplinary environment.

Insights and Results

The study found that the reflexive approach led to significant improvements in both student output and institutional engagement.

The "Co-Learning" Effect

Reflexive learning was found to be a bidirectional process. Lecturers reported becoming more open to unconventional learning paths by observing student-led projects. This "co-learning" effect helped bridge the gap between the disparate skill sets of art, science, and technology, proving that cartography requires a synthesis of these fields rather than a mastery of just one.

Innovative Cartographic Outputs

The reflexive approach led to projects that transcended standard navigation. Notable outputs included emotion maps for urban wellbeing, social advocacy tools for "period poverty" awareness, museum accessibility tools, and AR-enhanced 3D models for the visually impaired.

Student and Institutional Impact

Course evaluations were exceptionally high, with students rating "Learning Progress" at 4.8/5.0 and "Teaching Approach" at 4.9/5.0. Despite being an elective, nearly the entire cohort attended, and students have advocated for making the course a compulsory part of the curriculum.

Discussion: The Human-in-the-Loop for AI

A significant takeaway of this study is the link between reflexive learning and the development of generative AI. The authors suggest that the high-quality, "reflexive" map prototypes created by students—which are grounded in social ethics and design principles—represent superior training data compared to raw, uncurated geospatial datasets. These high-quality outputs are intended to train MapGPT, a future generative model designed to understand the nuances of cartography as a human-centric discipline rather than just a data-processing task.

Limitations and Future Work

While successful, the authors acknowledge that assessing "reflexive learning" and teamwork is inherently more subjective than grading technical exams. Furthermore, scaling this high-touch, dialogue-heavy model to larger undergraduate classes may require significant resource adjustments. Future work will focus on the development of MapGPT and refining the assessment of collaborative soft skills.

Conclusion

By prioritizing reflection over mere production, cartographic education can prepare students to lead in an AI-driven world. The "Mapping Project" demonstrates that an international, cross-disciplinary classroom is the ideal environment for fostering the critical thinking necessary to ensure that the maps of the future are not just accurate, but socially responsible and well-designed.

Summary updated

BibTeX
@inproceedings{cron2025promoting,
  author = {Juliane Cron and Melissa del Carmen Ernstberger and Liqiu Meng},
  title = {{Promoting reflexive learning in cartography in an international classroom}},
  booktitle = {Abstracts},
  year = {2025},
  doi = {10.5194/ica-abs-10-49-2025},
}
Figure