Rats and other animals need to be highly attuned to social signals from others so they can identify friends to cooperate with and enemies to avoid. To find out if this extends to non-living beings, Laleh Quinn at the University of California, San Diego, and her colleagues tested whether rats can detect social signals from robotic rats.

    They housed eight adult rats with two types of robotic rat—one social and one asocial—for four days. The robot rats were quite minimalist, resembling a chunkier version of a computer mouse with wheels to move around and colorful markings.

    During the experiment, the social robot rat followed the living rats around, played with the same toys, and opened cage doors to let trapped rats escape. Meanwhile, the asocial robot simply moved forwards and backwards and side to side.

    Next, the researchers trapped the robots in cages and gave the rats the opportunity to release them by pressing a lever.

    Across 18 trials each, the living rats were 52 percent more likely on average to set the social robot free than the asocial one. This suggests that the rats perceived the social robot as a genuine social being. They may have bonded more with the social robot because it displayed behaviours like communal exploring and playing. This could lead to the rats better remembering having freed it earlier, and wanting the robot to return the favour when they get trapped, she says.

    The readiness of the rats to befriend the social robot was surprising given its minimal design. The robot was the same size as a regular rat but resembled a simple plastic box on wheels. “We’d assumed we’d have to give it a moving head and tail, facial features, and put a scent on it to make it smell like a real rat, but that wasn’t necessary,” says Janet Wiles at the University of Queensland in Australia, who helped with the research.

    The finding shows how sensitive rats are to social cues, even when they come from basic robots. Similarly, children tend to treat robots as if they are fellow beings, even when they display only simple social signals. “We humans seem to be fascinated by robots, and it turns out other animals are, too,” says Wiles.

答案精析：本题为细节题。根据题干中的Janet Wiles可定位至原文第六段。该段第三句引号里的内容为Janet Wiles所说的话：“我们原本以为必须要给它（机械鼠）安装一个可以活动的头和尾巴，加上一些面部特征，并在它身上添加气味，让它闻起来像一只真正的老鼠，但这些都没有必要。”由此可知，机械鼠的外表和气味并不会对实验有太大影响，再结合前文第五段第二、三句可知，因为有社交行为，真老鼠才会把机械鼠看作真正的社交个体。由于机械鼠表现出了共同探索和玩耍的行为，真老鼠可能会和社交机械鼠更加亲近。因此可推断出，真老鼠会更多地对行为作出反应，而非外表，故正确答案为C。

错项排除：A项利用原文中的facial features进行干扰，但原文说的是面部特征并不会影响老鼠觉察社交信号，也没有提到老鼠会记住其他老鼠的面部特征，故A项错误。B项利用原文中的scent进行干扰，但文章并没有把气味和大小（sizes）进行比较，故B项错误。D项利用原文中的plastic box on wheels进行干扰，但选项中的关键词be scared（被吓到）在原文中并无依据，故排除D项。