Reducing the cognitive abstractness of climate change through an "engineering fiction" learning experience: A natural language processing study

The lackluster societal response to the climate crisis is partially attributed to the abstractness of people's mental construals of climate change given its vast spatial and temporal dimensions, which fail to evoke urgency to act. Prior efforts to measure mental construal levels of climate change are inconsistent, insufficient, and laborintensive. This study developed and implemented learning experiences for integrating engineering design and climate fiction writing to engage 48 high school students in concrete climate change thinking. A novel measure of cognitive abstractness overcomes previous methodological shortcomings by automatically quantifying the linguistic abstractness of participant-authored stories using natural language processing. Comparing participant stories written at the beginning and end of the intervention reveals a significant decrease in linguistic abstractness (Cohen's d = 1.01, p = 0.03). This study contributes to the nascent movement for greater use of narratives as data sources in environmental psychology research, which may uncover new insights into human behavior and decision making.