Cities across Europe are grappling with complex challenges like climate adaptation, social inequality, and resource pressures. These issues require transformative changes in decision-making, collaboration, and community engagement rather than just technical fixes.
The conditions that enable deep learning
Knowledge sharing has emerged as a crucial tool for transformation, with city networks and European programmes facilitating the exchange of experiences. Research from the Dutch Societal Innovation Hub (DSIH) indicates that simple sharing does not guarantee deep learning. Authentic transformative learning requires questioning assumptions and redefining problems and solutions. Without the right conditions, knowledge sharing can be superficial and fail to effect genuine change.
Over two years, DSIH brought together stakeholders from Rotterdam, Utrecht, and the Helmond-Eindhoven region to explore digital innovation for societal challenges. Through interviews, observations, hackathons and several thematic knowledge groups, researchers found significant variation in how learning unfolded. Some groups focused on producing concrete, codifiable outputs such as while others engaged in more open-minded reflection, grappling with uncertainty, values, and underlying assumptions.
This variation revealed a central insight: deep learning happens mostly when specific enabling factors are present. These include developing a transformational language, using reflexive facilitation styles, embracing uncertainty, engaging with both conceptual and tacit knowledge, maintaining a strong connection to the broader mission, and allocating sufficient time for reflection.
The tension between project efficiency and transformative learning
A key finding of the study is the persistent tension between the output-oriented logic of EU-funded projects and the open-ended nature of deep learning. While funders often require tangible deliverables within tight timeframes, transformative learning is non-linear, slow, and sometimes uncomfortable. It demands space for ambiguity, experimentation, and dialogue about values; conditions that traditional project structures rarely prioritise. As a result, some knowledge groups in DSIH gravitated towards surface-level results. In contrast, others struggled to balance expectations for concrete outputs with the need to reflect and challenge underlying norms. Recognising this tension is essential; without addressing it, knowledge-sharing initiatives risk falling short of their transformative ambitions.
"Learning that critically reflects on existing systems and questions societal behavior can be a vital first step in achieving actual change toward a better world."
Peter Scholten, Senior Specialist in Governance of Sustainable Urban Resources
Designing knowledge-sharing initiatives for more profound impact
The DSIH study offers several practical lessons for cities, facilitators, and programme designers seeking to foster deep learning. First, facilitators should not avoid discomfort; pushing participants gently out of their habitual ways of thinking can spark deeper reflection. Incorporating artistic or unconventional methods can also help participants access implied knowledge and explore emerging ideas beyond technical problem-solving. Second, transformation-oriented projects should explicitly state that deep learning is a core objective, not merely the production of outputs. Finally, deep learning requires dedicated time and space for reflection, meaning project timelines and structures must accommodate slower, repetitive processes.
By embracing these principles, cities can move beyond exchanging best practices toward cultivating the deeper learning necessary for meaningful urban transformation.
You might also like
- More information
The insights presented in this article draw on research conducted by IHS academic staff Peter Scholten, Jan Fransen and Elena Marie Enseñado, in collaboration with Esmeralde Marsman from the Municipality of Rotterdam.
