Young orangutans who explore little build fairly broad diets by watching others

The diet result comes from a smaller group than the 21 animals followed. According to New Scientist, the range of foods each youngster could eat at independence was estimated from about 4,700 hours of feeding observations on 13 of them ^[23]. That is about 62 percent of the cohort ^[21]. The learning records, nearly 7,000 instances, came from more than 1,700 hours of observation gathered over 12 years ^[5]. The Max Planck institute's own account describes 4,700 hours of observation covering peering, exploration and unaided feeding together ^[22]. The two accounts divide the hours differently.

Slow development is what made the question hard. "To really understand how these forms of learning translate into actual skills and knowledge, we needed to link learning events to concrete, measurable outcomes like diet. But because orangutans develop so slowly, there is a multiyear lag between learning and measurable behavioral change," said Caroline Schuppli, the senior author and a group leader at the institute ^{[19] [15]}. In the Science paper ^[3], Revathe Thillaikumar, a postdoctoral fellow at the institute ^[16], and a theoretical statistician at TU Dortmund University analysed just under 7,000 learning records with each animal's history kept separate ^[17]. "Analyzing each individual on its own was a linchpin for our study," Thillaikumar said ^[18]. The institute describes the work as the first to show several forms of learning interacting over many years to build survival-relevant skills in a wild great ape ^[24].

The mother is already inside the social side of the measure. Peering, the study's marker of social learning, counts close watching of the mother as well as of other orangutans nearby ^[14]. Thillaikumar told New Scientist that the spread in learning styles might partly reflect personality, and early development including how mothers care for their young ^[7]. "We were astonished by how different infants were from one another in their tendencies to learn," she said ^[6]. In my view the evidence supports this much: across a small group of wild apes, an infant's balance of watching and exploring tracks how broad a diet it reaches ^{[10] [11]}. It is an observed association from one field site.

Watching was often followed by doing. Thillaikumar's description of the time after peering was short. "Yeah, it's practice," she said ^[9]. Elizabeth Lonsdorf of Duke University said the study shows what researchers see in human children and other apes, "that they watch others perform a behaviour and then go try it out on their own" ^[12]. Michael Tomasello, also at Duke, said: "Predicting dietary breadth at maturity requires looking at all types of learning." ^[13]

The thing this doesn't tell you is whether a broad diet at independence keeps an orangutan alive afterward. An adult eats nearly 250 distinct foods chosen from thousands of options, some potentially toxic ^[2], and after about eight years with their mothers, young orangutans are on their own ^[1]. Diet breadth at that point is the study's measure of ecological competence ^[14]. Neither published account reports survival after independence or the size of the link between learning and diet.