Tool Use

Why is tool use in animals significant?

Tool use in non-human animals has come to be a marker in comparative research for intelligence and comprehension. While the frequency and ingenuity of tool use in animals can be a marker for brain size (Emery & Clayton, 2009), tool use as a proxy for intelligence has a controversial history that requires further objective, non-anthropomorphic analysis.  Before getting too deep into the politics of tool use, we first need to define tool use the way scientists define it.


What exactly do we mean by tool use?

A common, succinct way of defining tool use is by describing the use of “an external object used to assist in gaining some desired end” (Wynne & Udell, 2013, p. 118). By this definition, objects that are used to act on the environment or the organism itself would fall into this category. Though this definition might lead one to think that animals that can use tools can only use one tool at a time, there are instances in which specific animals in specific contexts demonstrate the capacity to use more than one tool at a time, or use tools in succession (Emery & Clayton, 2009).


What are the variants of tool use?

When animals use tools in succession to achieve a goal, this can be defined as either using a tool kit or engaging in meta-tool use. A tool kit is a behavior in which an animal “[relates] each tool to a particular task and [manufactures] the tool to meet the corresponding demand” (Fox, Sitompul, & Van Schaik, 1999, p.103). An example of this is when a chimpanzee combines the use of a probing stick and a fishing stick when foraging for termites in a mound (Suzuki, Kuroda, & Nishihara, 1995).

Meta-tool use is when animals use tools that cannot be used to attain the current goal, but they can be used to acquire other tools that can be used to in order to attain the current goal (Emery & Clayton, 2009). An example of this is when an animal uses a short rod within reach to access a long rod out of reach, and this long rod can be used to forage for food that is out of reach of the animal (Emery & Clayton, 2009).

This video is an example of a crow using a tool kit by using tools in succession to solve a foraging puzzle. (Video from BBC Earth on YouTube.)


Does tool use require flexibile behavior across different environments?

Some researchers further operationalize tool use by discussing its mechanical properties, such that tools are “freely manipulable object to modify the physical properties of a target object through a complex mechanical interaction” (Deecke, 2012, p. 725). This definition of tool use indicates that it is necessary for these tools to be flexible across environments, such that the animals are displaying a sensitivity to the context in which a tool will achieve a desired goal, a highly adaptive skill.

There are instances where animals have been found to use tools to achieve a desired goal, yet these animals are not susceptible to flexible tool use. For instance, capuchin monkeys have been tested using tool tasks where reinforcers can become “trapped” if the monkey does not use the stick tool correctly, and it has been found that while these monkeys learn strategies to obtain the reinforcer without trapping it, they are not flexible to manipulations in the task and as such do not demonstrate either behavioral flexibility or causal reasoning with regards to how their use of the tool affects the environment (Visalberghi & Limongelli, 1994).

This type of reasoning that does not require causal inference is called folk physics, in which an animal’s performance may appear to be based on intuitive reasoning that relies on fixed strategies and not based on flexible trial-and-error reasoning (Emery & Clayton, 2009).

Though the capuchin monkeys were using a stick in a tool-like manner, can we say they were engaging in tool use if there was little evidence of causal reasoning or flexibility in behavior? Some researchers further define tool-use behavior as “only [occurring] in response to very specific stimuli” (Finn, Tregenza, & Norman, 2009, p. R1069), as when octopi only use coconut shells to protect themselves only when necessary or when the shells will be needed in the foreseeable future (Finn et al., 2009).


Does tool use require intelligence? What about direct manipulation of a tool-like object?

While the intention of this research was not to devalue capuchin tool use, but rather to discuss how “simple behaviours, such as the use of an object (or objects) as shelter, are not generally regarded as tool use, because the shelter is effectively in use all the time, whereas a tool provides no benefit until it is used for a specific purpose” (Finn et al., 2009, p. R1069), some researchers defend the argument that tool use does not require intelligence, and intelligence does not require tool use (Emery & Clayton, 2009).

Specifically, a tool can be used without achieving a desired outcome, so long as the tool is an object that is used to manipulate the environment (Visalberghi & Limongelli, 1994). Further, tool use does not require an overarching label of “intelligence” to be credible tool use, as intelligence may be broken down more objectively into simpler cognitive categories, such as prospective memory (Emery & Clayton, 2009; Finn et al., 2009), motivation (Auersperg, Von Bayern, Gajdon, Huber, & Kacelnik, 2011; Emery & Clayton, 2009; Hart, Hart, McCoy, & Sarath, 2001; Visalberghi & Limongelli, 1994), and attention (Auersperg et al., 2011; Visalberghi & Limongelli, 1994).

Capuchins, while not the best models for studying causal reasoning and flexible behaviors, are capable of using tools to crack open food items. (Image from the New York Times.)

Some current research is even now defending the argument that foraging behavior that does not require the direct use of tools still constitutes tool-using behavior, as compared to “pure tool use” (Allen, Bejder, & Krützen, 2011). This also brings into question the role of using found objects or natural tools as compared to manufacturing objects for specific tasks (Fox et al., 1999; Hart et al., 2001).


Why does tool use exist?

With all of these definitions in mind, why, in the grand scheme of things, would tool use ever be a behavior that could emerge and be evolutionarily selected for among specific species? Morphological adaptations are usually “cheaper” in these sense that they do not require specific learned behaviors in order to ensure survival, but sometimes it is cheaper for an animal to expend cognitive resources on tool use in order to increase the effectiveness of foraging behaviors, as in New Caledonian crows with long juvenile periods that ultimately gain more nutrients from tool-extracted prey (Rutz & St Claire, 2012).


How can we comprehensively discuss tool use and variable tool-using behavior among species?

As is expected from some animals using tools regularly, some using tools infrequently, and some not using tools at all, these animals across species have varying levels of sophistication when it comes to how the animals can manipulate tools and if they can modify them, when the animals can use tools, why the animals use tools, and what ties these animals together evolutionarily with regards to tool use.

For example, variability among tool-using animal species in tools comes into play when considering that tool use is the rule while tool manufacture, in which only corvids and primates are recorded as manufacturing tools beyond simple modification, is the exception (Emery & Clayton, 2009).

The format of discussing this variability among tool-using animal species corresponds to Tinbergen’s foundational four questions when studying animal behavior (Tinbergen,1963). When discussing the “how” of tool use, this relates to Tinbergen’s question of “mechanism,” or the internal or external causes of behavior. When discussing the “when” of tool use, this relates to Tinbergen’s questions of “ontogeny,” or the development of the behavior. When discussing the “why” of tool use, this relates to Tinbergen’s questions of “adaptive value,” or the function of the behavior. And finally, when discussing the “what” of tool use across time, this relates to Tinbergen’s questions of “phylogeny,” or the evolution of the behavior or capacity for the behavior.

The format of this website follows Tinbergen’s four questions in order to provide a comprehensive view of tool use across a variety of species, all using different and similar mechanisms, involving different and similar developmental practices, arising for different and similar functional reasons, and relating to different and similar evolutionary histories.