Institute of Neuroscience and Psychology, University of Glasgow, 62 Hillhead Street, Glasgow, Scotland, G12 8QB, UK.
Institute of Neuroscience and Psychology, University of Glasgow, 62 Hillhead Street, Glasgow, Scotland, G12 8QB, UK; Department of Cognitive Science, Macquarie University, NSW 2109, Australia.
Trends Neurosci. 2020 Jun;43(6):373-384. doi: 10.1016/j.tins.2020.03.013. Epub 2020 Apr 30.
Artificial intelligence advances have led to robots endowed with increasingly sophisticated social abilities. These machines speak to our innate desire to perceive social cues in the environment, as well as the promise of robots enhancing our daily lives. However, a strong mismatch still exists between our expectations and the reality of social robots. We argue that careful delineation of the neurocognitive mechanisms supporting human-robot interaction will enable us to gather insights critical for optimising social encounters between humans and robots. To achieve this, the field must incorporate human neuroscience tools including mobile neuroimaging to explore long-term, embodied human-robot interaction in situ. New analytical neuroimaging approaches will enable characterisation of social cognition representations on a finer scale using sensitive and appropriate categorical comparisons (human, animal, tool, or object). The future of social robotics is undeniably exciting, and insights from human neuroscience research will bring us closer to interacting and collaborating with socially sophisticated robots.
人工智能的进步使得机器人具备了日益复杂的社交能力。这些机器满足了我们对感知环境中社交线索的内在需求,以及机器人提升我们日常生活的承诺。然而,我们的期望与社交机器人的现实之间仍然存在很大的不匹配。我们认为,仔细描绘支持人机交互的神经认知机制,将使我们能够收集到优化人类与机器人之间社交互动的关键见解。为了实现这一目标,该领域必须纳入人类神经科学工具,包括移动神经影像学,以探索长期的、身临其境的人机交互。新的分析神经影像学方法将能够使用敏感和适当的分类比较(人类、动物、工具或物体)更精细地描述社会认知表征。社交机器人的未来无疑令人兴奋,而人类神经科学研究的见解将使我们更接近与具有复杂社交能力的机器人进行交互和协作。