Exploring the origins of human cognition: Integrating physical knowledge and learning from others

人类思维的独特之处是什么？它为什么会进化？从找出哪个按钮可以从自动售货机中取出糖果，到决定沿着两根不稳定的树枝爬上哪根去摘美味的水果，人类和非人类动物每天都面临着具有挑战性的因果关系和物理推理问题。先前在现实世界的经验可能会影响我们的决策，但社会信息也很有价值，特别是在我们自己收集证据的能力有限的情况下。社会学习能力和身体推理能力都被认为是人类独特认知的来源。不同物种是如何整合这些线索的？

What is unique about the human mind and why did it evolve? From figuring out which buttons will get you candy from a vending machine to deciding which of two precarious branches to climb along to reach a tasty fruit, human and non-human animals are confronted with challenging causal and physical reasoning problems on a daily basis. Prior experience with the physical world is likely to influence our decision-making, but social information is also valuable, particularly in cases where we have limited ability to collect our own evidence. Social learning and physical reasoning abilities have both been proposed as the source of uniquely human cognition. How are these cues integrated by different species? In the next five years I will extend my ongoing research program with children and non-human primates on understanding of cause and effect to include dogs a species that is extremely interesting from an evolutionary perspective given their “human-like” social skills that are not even shared by our closest primate relatives. Of interest is how these different species interpret the intentions of a human demonstrator, and whether they are influenced by how mechanically necessary actions seem in relation to a desirable outcome. I will also expand this research theme to investigate how dogs, and children of different ages, integrate their physical beliefs (e.g., that objects fall downwards when dropped) with information communicated to them by someone particularly when these sources of information are in conflict. In addition to collecting empirical data from dogs and children, I will use computational modelling to disentangle the complex interactions between different sources of information that individuals use to solve problems. Bayesian inference models give us a way of precisely characterizing hypotheses about what both children and non-human animals think and know and testing them against experimental performance. This interdisciplinary approach has been previously successful in capturing children's performance on a variety of problem-solving tasks, including in my own work, but has rarely been applied to animal cognition. My research program combines mathematical tools from computer science and statistics with the empirical methods of developmental psychology and comparative cognition. Training in my lab will provide HQP with a combination of knowledge, experience and technical skills unique within Canada, and rare internationally. Our findings are likely to have broad interdisciplinary appeal and will be relevant to researchers in developmental and comparative psychology, animal behaviour, and computer science. From a practical standpoint, results are likely to be relevant to early childhood education, as well as enhancing pet and service-dog training.