Video analyses of naturalistic food acquisition and food
extraction/processing of primate species
Hill (2018) compiled a list of 113 primate species that forage on
cultivars and we have enlisted at least 16 primate species that are
known to subsist around urban landscapes. We mined videos of 3 primate
species from video databases/archives such that they spanned the
urban-forest gradient alongside the possibility of acquiring packaged
artificial food. Videos obtained covered foraging and feeding by
primates in as many microhabitats of each species as possible and
included their natural habitats (e.g., temperate evergreen forest,
savanna), altered habitats (e.g., botanical garden, nature-park, linear
structures through forests) and anthropogenic habitats (e.g., farmland,
tourist spots, residential areas). Based on availability of videos and
all relevant information, we included the following cercopithecid
species in the analyses: bonnet macaque (Macaca radiata ),
Japanese macaque (Macaca fuscata ) and vervet monkey
(Chlorocebus pygerythrus ). Several widespread primate species
(e.g., rhesus macaque, long-tailed macaque) could not be included in the
analyses due to inadequate video records or habitat representation(s).
We identified the year of video recording/uploading and the geographical
location of the primate group(s) based on the metadata of video. Among
videos containing foraging and feeding incidences, we shortlisted
instances in which food items met the conditions described in Table 1.
The conditions for food inclusion were based on delimiting the overlap
of mouth use and hand use for FA and the overlap of mouth use, hand use
and hand-mouth use for FP. Most conditions elucidated in Table 1 were
based on research literature on grasp, food acquisition, reach-for-food
and reach-for-grasp in primates (e.g. Peckre et al., 2019; Pouydebat &
Bardo, 2019; Sustaita et al., 2013). The remaining conditions were
determined based on the logical evaluation of an equal likelihood of
hand/mouth use in FA and FP with regard to species-food pairs. Several
covariates including variables for habitat (urbanization, packaged
food), individual characteristics (age), foraging style and
characteristics of food (attachment, embeddedness) were considered. The
description and inclusion/exclusion criteria of each covariate are
provided in Table 2. The covariates and their scopes described in Table
2 were also drawn from published materials (e.g. foraging style: Le Gros
Clark, 1959; Szalay et al., 1987; Van Horn, 1972; age: Colell et al.,
1995; food size/shape: Peckre et al., 2019; Reghem et al., 2011; food
consistency: Peckre et al., 2019; material texture: Yokoi et al., 2018;
prey mobility: Peckre, Lowie, et al., 2019; Toussaint et al., 2013;
foraging substrate: Fabre et al., 2017; Peckre, Lowie, et al., 2019;
Toussaint et al., 2013). Two researchers (TD and SD) naïve of the
habitat and ‘encounter with packaged food’ characterization of the
groups in the videos coded a random set of videos to (1) standardize the
coding framework mentioned in Table 2 and to (2) evaluate inter-coder
consistency in identifying hand/mouth use for FA and FP, age, foraging
style, food embeddedness and food attachment, i.e., attachment of the
fruit to the plant. To characterize the degree of exposure to packaged
anthropogenic food, we obtained information from either researchers
familiar with the groups, videographers, published/unpublished research,
residents of the region and/or from the narration in the videos.