Carrying practices and infant object handling in two non-WEIRD communities

• In Event: 1-162 - New models of variation in everyday infancy: Insights from dense audiovisual corpora in six cultures

Integrative Statement

Dynamic systems theories of development assert that the acquisition of motor, cognitive, and language skills are mutually reinforcing (Smith & Thelen, 2003; see also Iverson, 2010; Walle & Campos, 2014). As children begin to sit, crawl, and walk, their ability to access and manipulate objects changes dramatically (e.g., Karasik et al., 2011; Sanchez et al., 2018), and their interest shifts from a primary focus on faces to a primary focus on hands (Fausey et al., 2016; Jayaraman et al. 2015; 2017). However, motor development differs across human societies (see Adolf et al., 2010 for a review). The frequency with which children hold and manipulate objects is bound to vary widely across cultures. What is the scope of cross-cultural variation in early object manipulation? How does it influence children’s motor and linguistic development?

To begin answering these questions, we analyzed the manual manipulation of objects in an annotated database of photos taken from daylong recordings of 100+ children 0;0–4;0 from two small-scale non-WEIRD societies: Tseltal Mayan and Papuan (Casillas, Brown, & Levinson, 2017). Caregivers hold and carry children differently in these communities: Tseltal Mayan infants are typically carried in a sling on their mothers’ backs, with limited ability to pick up objects spontaneously until they begin to walk. Rossel Island (Papuan) infants are typically carried in the arms or on the shoulders, and are more often set down onto the surface where the caregiver sits—they have more manual freedom earlier in development.

A wearable camera fitted with a fisheye lens took 2–4 photos per minute during a 8–11-hour at-home recording. The camera was attached to the child’s chest (ambulant children) or to the chest of a primary caregiver (non-ambulant children). Each photo was annotated for multiple characteristics, including whether the child or one of their interactants was manually manipulating an object/person of direct interest to the child (e.g., feeding or grooming the child). Annotation is ongoing, but at time of submission nearly 80k photos have been annotated (Mayan: 41K+ from 40 children and Papuan: 38k+ from 22 children). For each child’s recording, we computed the proportion of photos in which the child was handling something and in which the child’s interactants were handling something. We modeled these proportions using linear regression, including predictors of child age, cultural community, and their interaction.

Tseltal and Rossel infants show two distinct developmental trajectories for object handling (Figure 1). Before 10 months, Rossel children handle objects more often than Tseltal children do (10% vs. 3%). Afterwards, however, Tseltal children handle objects more often, averaging 26.5% between 0;10 and 4;0 (vs. a stable 10% of Rossel photos). These patterns bear out as effects of child age (B = 0.008; SE = 0.001; t = 8.62; p < .001) and an interaction of child age and cultural community (Tseltal-vs.-Rossel; B = -0.007; SE = 0.002; t = -4.33; p < .001). We discuss the implications of these trajectories for early motor and linguistic development and the next steps needed to further this investigation.

Authors

• Marisa Casillas, Max Planck Institute for Psycholinguistics
  Presenting Author

• Mary Elliott, University of Kaiserslautern
  Non-Presenting Author