Search for: All records

People are remarkably capable of generating their own goals, beginning with child’s play and continuing into adulthood. Despite considerable empirical and computational work on goals and goal-oriented behaviour, models are still far from capturing the richness of everyday human goals. Here we bridge this gap by collecting a dataset of human-generated playful goals (in the form of scorable, single-player games), modelling them as reward-producing programs and generating novel human-like goals through program synthesis. Reward-producing programs capture the rich semantics of goals through symbolic operations that compose, add temporal constraints and allow program execution on behavioural traces to evaluate progress. To build a generative model of goals, we learn a fitness function over the infinite set of possible goal programs and sample novel goals with a quality-diversity algorithm. Human evaluators found that model-generated goals, when sampled from partitions of program space occupied by human examples, were indistinguishable from human-created games. We also discovered that our model’s internal fitness scores predict games that are evaluated as more fun to play and more human-like. 

Question asking is a key tool for learning, especially in childhood. However, formulating good questions is challenging. In any given situation, many questions are possible but only few are informative. In the present work, we investigate two ways 5- to 10-year-olds and adults simplify the challenge of formulating questions: by reusing previous questions, and by remixing components of previous questions to form new questions. Our experimental results suggest that children and adults reuse and remix questions and adaptively modulate reuse depending on how informative a question will be in a particular situation. This work shows that task-relevant experience asking questions provides fodder for future questions, simplifying the challenge of inquiry and enabling effective learning. 

A fully occluded object cannot be perceived directly, but we can still infer its existence from the effect it has on the motion and behavior of other, visible objects. Here we report the results of a behavioral experiment designed to elicit these sorts of inferences and quantify their re- liability. Our experiment leverages videos of real-world objects interacting under real-world physics (specifically, interrupted pendulum motion). We propose a preliminary model for how the mind might efficiently infer the position and number of occluded objects simply from the effect they have on the visible physics of a scene. 

Humans have been faced with the challenges of pictorial production since at least the Paleolithic. Curiously, while the capacity to navigate layouts and recognize objects in everyday life comes almost effortlessly, inherited from our evolutionary past, the capacity to draw layouts and objects is more effortful, often needing time to improve over the course of an individual’s development and with the technological innovations acquired through culture. The present study examines whether young children might nevertheless rely on phylogenetically ancient spatial capacities for navigation and object recognition when creating uniquely human pictorial art. We apply a novel digital coding technique to a publicly available dataset of young children’s drawings of layouts and objects to explore children’s use of classic pictorial depth cues including size, position, and overlap. To convey pictorial depth, children appear to adopt several cues, without a preference among them, younger than had been suggested by previous studies that used other, less rich, analytic techniques. Moreover, children use more cues to pictorial depth in drawings of layouts versus objects. Children’s creation of uniquely human pictorial symbols may thus reflect their heightened use of depth for navigating layouts compared to recognizing objects, both cognitive capacities that humans share with other animals.