Hopper Lm, Holmes An, Williams LE, Brosnan Sf
Lester E. Fisher Center for the Study and Conservation of Apes , Lincoln Park Zoo , Chicago, IL , USA ; Language Research Center , Georgia State University , Atlanta, GA , USA.
PeerJ. 2013 Feb 12;1:e13. doi: 10.7717/peerj.13. Print 2013.
Although the social learning abilities of monkeys have been well documented, this research has only focused on a few species. Furthermore, of those that also incorporated dissections of social learning mechanisms, the majority studied either capuchins (Cebus apella) or marmosets (Callithrix jacchus). To gain a broader understanding of how monkeys gain new skills, we tested squirrel monkeys (Saimiri boliviensis) which have never been studied in tests of social learning mechanisms. To determine whether S. boliviensis can socially learn, we ran "open diffusion" tests with monkeys housed in two social groups (N = 23). Over the course of 10 20-min sessions, the monkeys in each group observed a trained group member retrieving a mealworm from a bidirectional task (the "Slide-box"). Two thirds (67%) of these monkeys both learned how to operate the Slide-box and they also moved the door significantly more times in the direction modeled by the trained demonstrator than the alternative direction. To tease apart the underlying social learning mechanisms we ran a series of three control conditions with 35 squirrel monkeys that had no previous experience with the Slide-box. The first replicated the experimental open diffusion sessions but without the inclusion of a trained model, the second was a no-information control with dyads of monkeys, and the third was a 'ghost' display shown to individual monkeys. The first two controls tested for the importance of social support (mere presence effect) and the ghost display showed the affordances of the task to the monkeys. The monkeys showed a certain level of success in the group control (54% of subjects solved the task on one or more occasions) and paired controls (28% were successful) but none were successful in the ghost control. We propose that the squirrel monkeys' learning, observed in the experimental open diffusion tests, can be best described by a combination of social learning mechanisms in concert; in this case, those mechanisms are most likely object movement reenactment and social facilitation. We discuss the interplay of these mechanisms and how they related to learning shown by other primate species.
虽然猴子的社会学习能力已有充分记录,但这项研究仅聚焦于少数几个物种。此外,在那些还纳入了对社会学习机制剖析的研究中,大多数研究的是卷尾猴(僧帽猴)或狨猴。为了更全面地了解猴子如何获得新技能,我们对松鼠猴(玻利维亚松鼠猴)进行了测试,此前从未在社会学习机制测试中对其进行过研究。为了确定玻利维亚松鼠猴是否能通过社交进行学习,我们对分住在两个社会群体中的猴子(N = 23)进行了“开放扩散”测试。在10次20分钟的实验过程中,每组猴子观察一名经过训练的群体成员从双向任务(“滑盒”)中取回黄粉虫。这些猴子中有三分之二(67%)既学会了如何操作滑盒,而且它们朝着经过训练的示范者所展示的方向移动门的次数明显多于另一个方向。为了厘清潜在的社会学习机制,我们对35只之前没有滑盒经验的松鼠猴进行了一系列三个对照实验。第一个实验重复了实验性的开放扩散实验环节,但没有加入经过训练的模型,第二个实验是猴子二元组的无信息对照实验,第三个实验是向单只猴子展示“幽灵”演示。前两个对照实验测试了社会支持(单纯在场效应)的重要性,并通过“幽灵”演示向猴子展示了该任务的可操作性。猴子在群体对照实验(54%的受试对象在一次或多次实验中完成了任务)和配对对照实验(28%成功)中表现出了一定程度的成功率,但在“幽灵”对照实验中无一成功。我们认为,在实验性开放扩散测试中观察到的松鼠猴的学习,最好用多种社会学习机制协同作用来描述;在这种情况下,这些机制很可能是物体移动重演和社会促进。我们讨论了这些机制之间的相互作用,以及它们与其他灵长类物种所表现出的学习之间的关系。