Sequential Tool Use
KeywordsNonhuman Primate Cognitive Mechanism Termite Mound Single Tool Sequential Tool
The use of two or more tools to obtain a goal.
Tool use is a widespread behavior among many nonhuman animal species (e.g., Shumaker et al. 2011). However, nonhuman primates and corvids have produced the most extraordinary examples of tool-use behaviors. In the wild, for example, chimpanzees at Gombe (Tanzania) have been reported to plunge little sticks into termite mounds in order to access the termites. Similar behaviors have been described in New Caledonian crows – they use small twigs to extract insects from holes in logs. These two examples represent instances of using a single tool to obtain food (Shumaker et al. 2011).
Although more uncommon, instances in which the use of two or more tools is required to obtain a particular goal – so-called associative tool behavior (Shumaker et al. 2011) – have also been demonstrated in animals. Associative tool behaviors encompass cases in which animals use a tool to support another tool, so-called metatool use, to modify or manufacture another tool, secondary tool use, as well as instances in which two or more tools are used in sequence and each tool is used in a different mode, so-called serial tool use or tool set. An example that falls in the last category has been described in the chimpanzees of the Goualougo Triangle (Republic of Congo). When trying to access the termites from subterranean nests, the chimpanzees first use a puncturing stick on the ground and second a fishing probe that allows them to obtain the termites (Sanz et al. 2004).
Importantly, a tool can also be used to obtain a second out-of-reach tool, which subsequently will serve to obtain an out-of-reach goal – so-called sequential tool use. According to Shumaker et al. (2011), in the animal tool-use research, the categories metatool use and sequential tool use have been mistaken and considered to be equivalent categories. However, as pointed out by Shumaker et al. (2011), “a metatool is not used to acquire another tool but rather to increase the efficiency or effectiveness of the second tool” (p. 20). Thus, an orangutan using a stick to push a paper towel into a puddle of edible liquid is not an example of sequential tool use but an example of metatool use (Shumaker et al. 2011). In addition, whereas sequential tool use has been reported in primates and corvids, metatool use has only been observed in primates.
Sequential tool use has been commonly tested in lab settings by presenting subjects with an out-of-reach reward, an immediately available tool, and an out-of-reach second tool. The first tool is not useful to obtain the reward but useful to reach the second tool, which can then be used to obtain the reward. For example, Bird and Emery (2009) showed that rooks could spontaneously drop a large stone into a container to release a small stone, which was then used to acquire food. New Caledonian crows could use a directly available short stick to reach for an out-of-reach long stick, and then use the long tool to reach for a reward placed in a vertical tube (e.g., Taylor et al. 2007). Likewise, Bluff et al. (2007) showed that New Caledonian crows could first obtain a tool by pulling up a string; next, use the tool to extract a longer second tool from a toolbox; and finally, use the long tool to obtain food from a hole. Using a very different setup, Wimpenny et al. (2009) presented New Caledonian crows with the following situation: an out-of-reach reward, two tools that were available but too short to reach the food, and four out-of-reach tools differing in functionality. The distance of the food and/or which tools (and how many) were needed to obtain the food defined the sequence of actions required for successful performance. Wimpenny et al. (2009) found that crows could use up to three tools in sequence in order to obtain the reward.
As mentioned above, nonhuman primates have also been reported to use tools to access other tools. For example, Martin-Ordas et al. (2012) showed that apes could spontaneously use up to five tools in sequence. Macaques (Hihara et al. 2003) and cotton-top tamarins (Santos et al. 2005) can also do so – although only after receiving some training.
Cognitive Mechanisms Involved in Sequential Tool Use
Wimpenny et al. 2009 (see also Shumaker et al. 2011) argued that the use of two or more tools is more cognitively demanding than the use of a single tool – the number of objects required to succeed will certainly increase the level of difficulty of the task (see also Fragaszy and Cummins-Sebree 2005). Taylor et al. (2007; see also Bird and Emery 2009) have also suggested that using tools in sequence is a demanding task because the individual has to (1) understand that the tool can be used to access food but also to access nonfood items, (2) inhibit the drive to use the tool to access the food, and (3) plan the sequence of actions, that is, anticipate the series of sub-actions needed to achieve a final goal. Importantly, Wimpenny et al. (2009) stressed that most of the research done on sequential tool use did not help to unravel the cognitive underpinnings of animal performance. In particular, they argued that learning history or trial error – as opposed to planning or causal understanding – could explain animals’ performance in those particular tasks. Wimpenny et al.’s (2009) task controlled for some of these issues (see also Martin-Ordas et al. 2012). Consequently, the authors concluded that, at least, the performance of one subject “in particular leaves open the possibility that crows may solve sequential tool problems by planning their actions, rather than having to build up associations by repeated experience.” (p. e6471).
Sequential tool use is defined as the ability to use a tool in order to obtain a second out-of-reach tool, which subsequently will serve to obtain an out-of-reach goal. Whether planning, reasoning, or associative learning are cognitive mechanisms involved in sequential tool use is still an open question. As such, future research should develop experimental paradigms to further investigate the cognitive mechanisms underlying this particular tool-use behavior.
- Bluff, L. A., Weir, A. A. S., Rutz, C., Wimpenny, J. H., & Kacelnik, A. (2007). Tool-related cognition in New Caledonian crows. Comparative Cognition and Behavior Reviews, 2, 1–25.Google Scholar
- Shumaker, R. W., Walkup, K. R., & Beck, B. B. (2011). Animal tool behavior: The use and manufacture of tools by animals. Baltimore: The Johns Hopkins University Press.Google Scholar