Historical narratives encode how people today perceive, understand, and reconstruct past realities. One basic form focuses on historical trajectories and encodes perceived changes. Here, we show that perceived changes in economic inequality are predicted by its objective changes over time, while both diverge from shifts in public attention to economic inequality. Using the Fill-Mask Association Test (FMAT), a new social computing paradigm, we quantified public perceptions of changes in economic inequality from 1800 to 2019, as represented semantically in large contemporary English-language texts. Perceived changes are consistent with how economic inequality has actually changed in the long history, which was replicated using various time series analysis methods (e.g., cross-correlation, vector autoregression, Granger causality test). In contrast, public attention to economic inequality, as measured by word frequency in Google Books, did not fluctuate in line with either objective or perceived changes. Our findings demarcate the objective–subjective relationships of economic inequality from a historical-psychological perspective. Objective changes are accurately perceived and encoded in natural language texts, forming a collective memory of economic inequality in history.
Bao et al. (Wed,) studied this question.