What makes a good tool user? A species-wide experimental study of the Hawaiian crow

在动物王国里，使用工具是极其罕见的。在所有物种中，只有不到0.01%的物种有这种行为，而且这些物种中只有少数每天使用工具或能够用原材料制造工具。这是一个重大的科学难题，因为使用工具可以带来巨大的好处，例如，可以方便地获得营养丰富的无法获得的食物。在这个项目中，我们想通过提出一个看似简单的问题来解决这个难题：“是什么造就了一个好的工具用户？”解决这个问题需要两个关键因素：合适的研究物种和合适的研究策略。我们建议研究夏威夷乌鸦（通常被称为“Alala”），而不是像黑猩猩或新喀里多尼亚乌鸦那样众所周知的工具使用者，这个物种的工具使用能力在几年前才被发现并系统地记录下来。可悲的是，这个物种在2002年在野外灭绝了，但在夏威夷的两个圈养设施中幸存下来，我们的合作者正在那里为一个正在进行的重新引入计划繁殖鸟类。有机会接触到一个物种的所有成年成员，已知的遗传和社会背景，提供了一个杰出的研究机会。关于研究策略，我们首先需要记录个人在使用、制造和发明工具（“技术能力”）方面的能力差异。我们从之前的观察中得知，一些“Alala”是熟练的工具使用者，而另一些则相当笨拙，或者根本不使用工具。为了系统地研究这些差异，我们设计了一些任务，来衡量鸟类用工具获取食物的速度有多快，或者它们的灵活性和创造性有多强——这是成为一个好的工具使用者的两个互补方面。例如，我们的一些任务包括使用一根棍子从越来越困难的洞中取出小块食物，修改原材料以创造一种功能工具，或者发明一种全新的工具类型。下一步包括寻找可能解释鸟类技术能力差异的因素。为此，我们将收集两个额外的数据集。首先，我们将测试我们的受试者在不需要使用工具的情况下评估各种认知和行为能力的任务。例如，我们包括了测量记忆能力、学习速度、创新倾向、解决问题的持久性和对新事物的吸引力的任务。这将帮助我们确定那些总体上“更聪明”的鸟类，或者具有特定行为倾向的鸟类，是否更善于使用工具。其次，我们将收集鸟类的广泛“个人”信息，包括它们的年龄、饲养历史、与其他个体的关系，甚至它们的形态。这是因为Alala可能会随着年龄的增长而磨练他们使用工具的技能，通过观察精通工具的成年人来学习，从父母那里继承技术能力，或者受益于特别直的喙，可以灵巧地处理棍棒和原材料。一旦收集了这两个大数据集，我们将结合所有信息并使用先进的统计技术来检查是什么造就了一个好的工具用户。总之，我们将使用迄今为止开发的最全面的实验任务集来调查动物工具的使用，收集整个物种的数据。这种规模的研究是前所未有的，有可能使我们对制造和使用工具的特殊之处的理解发生重大变化。重要的是，我们对个体行为差异的详细调查将有助于揭示为什么工具使用在动物王国中如此罕见，甚至可能为人类如何进化出惊人的技术能力提供新的线索。

Tool use is extremely rare across the animal kingdom. The behaviour has been observed in less than 0.01% of all species, and only a handful of these use tools on a daily basis or are capable of manufacturing tools from raw materials. This presents a major scientific puzzle, given that tool use can afford significant benefits, for example, by providing easy access to nutritious out-of-reach food. In this project, we want to address this conundrum by posing a seemingly simple question: "What makes a good tool user?"Two key ingredients are needed to address this question: a suitable study species and an appropriate research strategy. Rather than working with one of the well-known tool users, like chimpanzees or New Caledonian crows, we propose to study the Hawaiian crow (usually called 'Alala) -- a species whose tool-using abilities were discovered and systematically documented only a couple of years ago. The species sadly became extinct in the wild in 2002, but survives in two captive facilities in Hawaii where our collaborators are breeding birds for an ongoing reintroduction programme. Having access to all adult members of a species, of known genetic and social background, offers an outstanding research opportunity.With regard to the research strategy, we first need to document how individuals vary in their ability to use, make and invent tools ('technical competency'). We know from our earlier observations that some 'Alala are highly skilled tool users, while others are quite clumsy, or do not use tools at all. To study these differences systematically, we have designed tasks that measure either how fast birds can extract food with tools, or how flexible and inventive they are -- two complementary facets of being a good tool user. For instance, some of our tasks involve using a stick to extract small pieces of food from holes of increasing difficulty, modifying raw materials to create a functional tool, or inventing an entirely new tool type.The next step involves searching for factors that may explain variation in birds' technical competency. We will collect two additional datasets for this purpose. First, we will test our subjects on tasks that assess -- without requiring the use of tools -- various cognitive and behavioural capacities. For example, we have included tasks measuring memory ability, learning speed, innovation tendency, problem-solving persistence, and attraction to novel objects. This will help us determine whether birds that are 'smarter' overall, or have particular behavioural dispositions, are better tool users. Second, we will collect extensive 'personal' information on the birds, including their age, rearing history, relatedness to other individuals, and even their morphology. This is because 'Alala may hone their tool-using skills as they get older, learn by observing tool-proficient adults, inherit technical competency from their parents, or benefit from having particularly straight bills that allow dexterous handling of sticks and raw materials. Once both of these large datasets have been collected, we will combine all information and use advanced statistical techniques to examine what makes a good tool user.In summary, we will use the most comprehensive set of experimental tasks ever developed for investigating animal tool use, collecting data for an entire species. A study of this scale is unprecedented and has the potential to produce a step change in our understanding of what is special about making and using tools. Importantly, our detailed investigation of behavioural variation across individual 'Alala will contribute to uncovering why tool use is so rare across the animal kingdom, and it may even shed new light on how humans have evolved their astonishing technological prowess.