In urban environments, humans often encounter other people that may engage one in interaction. How do humans perceive such invitations to interact at a glance? We briefly presented participants with pictures of actors carrying out one of 11 behaviors (e.g., waving or looking at a phone) at four camera-actor distances. Participants were asked to describe what they might do in such a situation, how they decided, and what stood out most in the photograph. In addition, participants rated how likely they deemed interaction to take place. Participants formulated clear responses about how they might act. We show convincingly that what participants would do depended on the depicted behavior, but not the camera-actor distance. The likeliness to interact ratings depended both on the depicted behavior and the camera-actor distance. We conclude that humans perceive the “gist” of photographs and that various aspects of the actor, action, and context depicted in photographs are subjectively available at a glance. Our conclusions are discussed in the context of scene perception, social robotics, and intercultural differences.

在城市环境中，人类经常会遇到其他人，并可能进行互动。人们如何一眼就能看出这种互动邀请？我们向参与者简要展示了演员在 4 个摄像机与演员距离下执行 11 种行为之一（例如挥手或看手机）的照片。参与者被要求描述在这种情况下他们可能会做什么、他们如何决定以及照片中最引人注目的是什么。此外，参与者还评估了他们认为互动发生的可能性。参与者对他们可能如何行动做出了明确的回应。我们令人信服地证明，参与者会做什么取决于所描绘的行为，而不是摄像机与演员的距离。互动评分的可能性取决于所描述的行为和摄像机与演员的距离。我们得出的结论是，人类感知照片的“主旨”，并且照片中描绘的演员、动作和背景的各个方面都可以主观地一目了然。我们的结论是在场景感知、社交机器人和跨文化差异的背景下讨论的。