Back

Modern Times: Longitudinal Study of Toba/Qom Communities Reveals Delay and Shortening of Sleep in Real-Time Across Electrification

Casiraghi, L. P.; Spiousas, I.; Trebucq, L. L.; Coldeira, M. F.; Godoy Peirone, C. R.; Zhang, V. Y.; Moyano, M. D.; Golombek, D. A.; de la iglesia, H.

2025-08-01 physiology
10.1101/2025.07.28.667301 bioRxiv
Show abstract

While artificial light and digital technologies are widely assumed to delay and curtail sleep, direct, longitudinal evidence documenting these changes is remarkably scarce. From 2012 to 2024, we conducted a longitudinal study of native Toba/Qom communities in northern Argentina, including rural groups newly introduced to electricity and semi-urban groups with earlier access to electricity. Using linear mixed-effects models on a dataset of over 12,000 actigraphically recorded sleep events from 156 participants at different times, we observed striking shifts in sleep dynamics. Remarkably, both groups showed delays of up to 1.4 hours in sleep timing, and participants from rural communities lost a full hour of sleep within just a decade. These changes occurred across a period that included, first, the introduction of electricity to rural communities after 2016, and, in more recent years, the spread of the internet and smartphone use among the Toba/Qom. Our models reveal that these events alone cannot fully explain the observed shifts in sleep timing, suggesting that broader processes of modernization and acculturation, beyond electric light and digital devices alone, may be driving sleep changes. The rapid transformations observed in these communities mirror, within a single decade, the likely trajectory of human sleep across the 20th and early 21st centuries, underscoring the complex ways modern environments reshape sleep behavior.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.