Biological Anthropology — Article Review

Nocturnal Activity in Wild Chimpanzees (Pan troglodytes)
Evidence for Flexible Sleeping Patterns and Insights into Human Evolution

Paige Harris · paigeharris.io

Reviewed Article

Citation

Tagg, N., McCarthy, M., Dieguez, P., Bocksberger, G., Willie, J., Mundry, R., Stewart, F., Arandjelovic, M., Widness, J., Landsmann, A., Agbor, A., Angedakin, S., Ayimisin, A. E., Bessone, M., Brazzola, G., Corogenes, K., Deschner, T., Dilambaka, E., Eno-Nku, M., ... Boesch, C. (2018). Nocturnal activity in wild chimpanzees (Pan troglodytes): Evidence for flexible sleeping patterns and insights into human evolution. American Journal of Physical Anthropology, 166(3), 510–529.
https://doi.org/10.1002/ajpa.23478

Background

How I Got to This Topic

My path to this topic began at the intersection of storytelling, games, and conservation. Watching A Knight of the Seven Kingdoms prompted me to reflect on the figure of Sir Duncan the Tall and what it means to defend those whom the powerful find easy to ignore. That same instinct — protecting the vulnerable — shaped my Bannerlord mod Defend the Weak & Innocent, through which donations are directed to WCS.org.

Following WCS materials from there, including their work on migratory species, I began thinking about how animal movement and survival patterns can reveal something larger about adaptation and resilience. Once I reviewed the course assignment, it became clear that a primate-focused article would serve the material more directly than one centered on whales (cetaceans). Through a connection with my Canadian girlfriend, who has access to Canadian academic resources, I came across the Wiley synopsis that led me to this chimpanzee study.

Analysis

Journal Article Review

The article I reviewed asks whether wild chimpanzees are strictly diurnal animals or whether they sometimes engage in meaningful nighttime activity. Tagg et al. (2018) investigated terrestrial nocturnal activity across multiple African study sites and tested whether ecological variables — including habitat type, predator presence, temperature, rainfall, human activity, and moonlight — helped predict when and where that activity occurred. This question matters to biological anthropology because chimpanzees and bonobos are the closest living relatives of Homo sapiens, sharing a common ancestor and approximately 98–99% nucleotide similarity; understanding the boundaries and flexibility of their behavior can sharpen thinking about adaptation and the selective pressures that may have shaped early human evolution (Boyd et al., 2024, pp. 4–8, 115–120).

The methods are well suited to the research question. The researchers paired terrestrial camera-trap footage with ecological data drawn from 22 chimpanzee study sites participating in the Pan African Programme: The Cultured Chimpanzee, a large-scale initiative that collects standardized behavioral and ecological data across the species' range (Pan African Programme, n.d.). Sites spanned different habitats and chimpanzee subspecies, and cameras were positioned in locations already frequented by chimpanzees — such as established trails and feeding trees — to maximize the likelihood of capturing nighttime behavior. One important caveat is that this design is most informative about terrestrial activity in high-use locations; arboreal nocturnal behavior and movement away from camera placements likely go undetected, which means the study's estimates of nocturnal activity should be understood as conservative.

\[ P(X = k) = \binom{n}{k} p^k (1 - p)^{n-k} \]

For the analysis, the authors took care not to inflate the dataset by overcounting repeated detections of the same individuals. They constructed a conservative dataset based on whether identified chimpanzees were detected at night on specific nights at specific sites, then applied a generalized linear mixed model with a binomial error structure to assess which variables predicted the probability of nocturnal activity. A GLMM is useful here because the data are not normally distributed — each observation is binary, detected or not — and because observations are grouped within sites that each carry their own quirks. As Bono et al. (2021) explain, GLMMs are designed for exactly this kind of situation: they let researchers work with non-normal data while also accounting for the fact that observations within the same group are not truly independent of each other. That structure models the number of successes X out of n trials — in this case, the number of site-nights on which chimpanzees were detected after dark. This was a sound analytical choice because it directly mirrors the binary nature of the question and avoids treating every video clip as an independent event.

22
Study Sites
18
Sites with Nocturnal Activity
1.80%
Mean Nocturnal Activity

The results show that nocturnal activity was uncommon but considerably less rare than prior assumptions had implied. Chimpanzee nighttime activity was recorded at 18 of the 22 study sites, with an overall mean of 1.80% of total observed chimpanzee activity. It occurred across all hours of the night, though it was most concentrated near twilight rather than during the deepest nocturnal hours. Movement was the most frequently recorded behavior, suggesting that most of these events involved travel rather than extended feeding or social interaction.

The most consequential findings were that nocturnal activity was more likely at sites where human activity was lower, ambient temperatures were higher, and forest cover was greater. Predator abundance, large mammal presence, rainfall, and moonlight did not produce significant effects in the final model. This outcome challenges the intuitive assumption that chimpanzees move at night primarily in response to predation risk or lunar illumination. Instead, the pattern points toward behavioral plasticity driven by thermoregulatory pressure and ecological disturbance. As Boyd et al. (2024, pp. 130–145) explain, activity patterns in primates are shaped by ecology — diet, predation, ranging behavior, and resource distribution all influence when and how animals move through their environment, and natural selection favors individuals whose behavior fits the demands of their local conditions.

The article is persuasive on the whole. Its primary strength is scale: by analyzing 22 sites rather than a single field location, the authors can speak to patterns that transcend any one population or habitat. Camera traps also allowed nighttime observation without requiring sustained human presence, which would itself alter behavior. That said, the study carries real limitations. The most technically significant is the terrestrial placement of cameras, which likely results in a systematic undercount of arboreal nighttime behavior and introduces variation across sites in how much of the landscape was effectively monitored. The finding that nocturnal activity was higher where human pressure was lower does reduce the concern that detections were merely a function of camera placement, but the limitation remains.

A second limitation — and one I find more fundamental — concerns the use of physical dummies to conceal the recording equipment. The study employed artificial decoys placed on the ground to disguise the cameras, but chimpanzees are highly intelligent, socially aware animals, and this form of concealment is an unreliable method for capturing natural behavior. Research on chimpanzee theory of mind has demonstrated that chimpanzees can track what others have and have not seen, predict behavior based on others' knowledge states, and respond strategically to novel elements in their environment (Call & Tomasello, 2008). Placement on the ground versus in the canopy, or timing observations during the day versus at night, does not resolve the core issue: a sentient animal capable of reading its environment and tracking the reactions of its group is going to notice and respond to unfamiliar objects introduced into familiar habitat. Any behavioral data collected in the presence of such decoys carries genuine uncertainty about whether the observed individuals were acting naturally or reacting to the equipment. By 2018, camera technology had advanced sufficiently that small, low-profile trap cameras could be deployed without bulky housing or decoy structures. A less conspicuous setup would have produced cleaner data and reduced this source of bias considerably.

Even with those limitations, the methods and results support the authors' core argument. They do not claim chimpanzees are nocturnal. They argue, more carefully, that chimpanzees are predominantly diurnal but retain the capacity for flexible activity patterns under specific ecological conditions. This is consistent with earlier primatological work showing that sleep architecture and sleep-site behavior in primates are not rigidly fixed but vary with ecological context, social structure, and predation pressure (Anderson, 1998). That is a defensible conclusion, and it carries real weight for biological anthropology because it demonstrates how behavior can shift in a species sharing close ancestry with humans without requiring a fundamental change in species identity.

The article's broader contribution is to illuminate that chimpanzees engage in more nighttime activity than older frameworks assumed, and that this flexibility may help anthropologists think more carefully about how environmental pressure shapes behavior over time. Rather than being constrained by a fixed diurnal schedule, chimpanzees can adjust their sleep and movement in response to temperature, forest density, and human disturbance. This behavioral plasticity is significant: it suggests that even in species closely related to humans, daily rhythms can shift without major evolutionary restructuring. If similar flexibility existed in human ancestors or early Homo, it would offer a useful lens for interpreting how our lineage responded to shifting climates and environments across Africa. As Boyd et al. (2024, pp. 263–280) emphasize, adaptation in the hominin lineage was shaped not by a single stable habitat but by environmental variability itself — and a reminder that behavioral rigidity has rarely been a reliable survival strategy.

APA 7

References

Built With

Technologies

Markdown → HTML
The source file was a plain Markdown document. Every element was translated to semantic HTML # headings became <h1> / <h2> with anchor IDs wired to the nav links; *italic* became <em>; **bold** became <strong>; paragraphs became <p> tags; the APA citation block became a left-accented .citation-card div; the camera-dummy critique paragraph was wrapped in a .callout div to visually flag it; and the $$...$$ LaTeX formula was placed in a .formula-block div for MathJax. Abacus was used to assist in the markdown-to-JavaScript conversion workflow.
CSS Layout & .hero
All styling is a single embedded <style> block using CSS custom properties (--bg, --accent, --text, etc.) for the dark palette. The .hero header uses a linear-gradient base with a radial-gradient on its ::before pseudo-element to create the soft purple glow behind the title. The nav uses position: sticky and backdrop-filter: blur(12px) for the frosted-glass effect. Stat boxes use display: grid with repeat(auto-fit, minmax(160px, 1fr)) so they reflow on mobile automatically. The hero title scales with clamp(1.6rem, 3.5vw, 2.6rem). Sora and Source Sans 3 are loaded from Google Fonts with Helvetica Neue, Georgia, and Times New Roman as fallbacks.
MathJax 3
The LaTeX formula from the Markdown ($$P(X=k)=inom{n}{k}p^k(1-p)^{n-k}$$) was placed inside a \[...\] display block. MathJax 3 is loaded via jsDelivr CDN and scans the DOM on load, replacing the raw LaTeX with a rendered SVG equation — no build step or pre-compilation required.
Native Print CSS
Browser print styling via @page and @media print. The Print APA 7 button temporarily adds an apa-print-mode class to <body>, switching the font to Times New Roman 12pt double-spaced and indenting paragraphs to APA spec before calling window.print(). The class is removed via the afterprint event.

Technologies referenced here include MathJax 3 via jsDelivr; Google Fonts (`Sora` and `Source Sans 3`) with `Helvetica Neue`, `Georgia`, and `Times New Roman` fallbacks; HTML5, CSS3, and native browser print styling via `@page`, `@media print`, and `window.print()`; and Abacus as part of the markdown-to-JavaScript conversion workflow used to help produce the final static site