The shifting shelf task: a new, non-verbal measure for attentional set shifting

Attentional set shifting is a core ingredient of cognition, allowing for fast adaptation to changes in the environment. How this skill compares between humans and other primates is not well known. We examined performance of 3- to 5-year-old children and chimpanzees on a new attentional set shifting task. We presented participants with two shelves holding the same set of four boxes. To choose the correct box on each shelf, one has to switch attention depending on which shelf one is currently presented with. Experiment 1 (forty-six 3- to 5-year olds, predominantly European White) established content validity, showing that the majority of errors were specific switching mistakes indicating failure to shift attention. Experiment 2 (one hundred and seventy-eight 3- to 6-year olds, predominantly European White) showed that older children made fewer mistakes, but if mistakes were made, a larger proportion were switching mistakes rather than 'random' errors. Experiment 3 (52 chimpanzees) established suitability of the task for non-human great apes and showed that chimpanzees' performance was comparable to the performance of 3- and 4-year olds, but worse than 5-year olds. These results suggest that chimpanzees and young children share attentional set shifting capacities, but that there are unique changes in the human lineage from 5 years of age.

Young children spontaneously invent three different types of associative tool use behaviour

Associative Tool Use (ATU) describes the use of two or more tools in combination, with the literature further differentiating between Tool set use, Tool composite use, Sequential tool use and Secondary tool use. Research investigating the cognitive processes underlying ATU has shown that some primate and bird species spontaneously invent Tool set and Sequential tool use. Yet studies with humans are sparse. Whether children are also able to spontaneously invent ATU behaviours and at what age this ability emerges is poorly understood. We addressed this gap in the literature with two experiments involving preschoolers (E1, N = 66, 3 years 6 months to 4 years 9 months; E2, N = 119, 3 years 0 months to 6 years 10 months) who were administered novel tasks measuring Tool set, Metatool and Sequential tool use. Participants needed to solve the tasks individually, without the opportunity for social learning (except for enhancement effects). Children from 3 years of age spontaneously invented all of the types of investigated ATU behaviours. Success rates were low, suggesting that individual invention of ATU in novel tasks is still challenging for preschoolers. We discuss how future studies can use and expand our tasks to deepen our understanding of tool use and problem-solving in humans and non-human animals.