Abstract

The analysis of thousands of time series in different languages reveals that word usage presents oscillations with a prevalence of 16-year cycles, mounted on slowly varying trends. These components carry different information: while similar oscillatory patterns gather semantically related words, similar trends group together keywords representative of cultural and historical periods. We interpreted the regular oscillations as cycles of interest and saturation, whose behavior could be captured using a simple mathematical model. Driving the model with the empirical trends, we were able to explain word frequency traces across multiple languages throughout the last three centuries. Our results suggest that word frequency usage is poised at dynamical criticality, close to a Hopf bifurcation which signals the emergence of oscillatory dynamics. Crucially, our model explains the oscillatory synchronization observed within groups of words and provides an interpretation of this phenomenon in terms of the cultural context driving collective cognition. These findings contribute to unravel how our use of language is shaped by the interplay between human cognition and sociocultural forces. Significance: The frequency with which words are used presents regular oscillations of 16 years. We propose that these oscillations arise from a basic cognitive mechanism common to other cultural objects with life cycles, such as fashion. The words that belong to a topic of interest increase their frequency, which is then inhibited by saturation until interest is regained. Here we set up a simple mathematical model for the interaction of this cognitive mechanism and the sociocultural context, which explains the occurrence frequencies of thousands of words in different languages during the past three centuries. We show that oscillations are tuned to a critical point and are synchronized within word communities. (C) 2022 Elsevier Ltd. All rights reserved.