Two experiments addressed contradictory claims about causal reasoning in elephants. In Experiment 1, 4 Asian elephants (Elephas maximus) were pretrained to remove a lid from the top of a bucket and retrieve a food reward. Subsequently, in the first 5 critical trials, when the lid was placed alongside the bucket and no longer obstructed access to the reward, each elephant continued to remove the lid before retrieving the reward. Experiment 2, which involved 11 additional elephants and variations of the original design, yielded similarly counterintuitive observations. Although the results are open to alternative interpretations, they appear more consistent with associative learning than with causal reasoning. Future applications of Fabrean methodologies (J. H. Fabre, 1915) to animal cognition are proposed.

How do animals react to covarations between events? The causal model theory posits that performance is, at times, based on the actor's belief and abstract knowledge about a causal connection (Dickinson, 1994; Dickinson & Shanks, 1995). In contrast, the associative model rejects the relevance of mental representation and belief about causality and ascribes the performance solely to the reinforcement of a stimulus–response (S-R) association (Cobos, López, Caño, Almaraz, & Shanks, 2002).

Although the existence of causal reasoning in animals has long received the attention of philosophers and experimentalists, a clear-cut answer still eludes us. To date, perhaps the least ambiguous evidence comes from Fabre's (1915) classical 19th century experiments with wasps. Fabre's straightforward interventionary procedures were later applied to bees (e.g., Gould & Gould, 1982), fruit flies (e.g., Nissani, 1975), and other insects, thus tending to support Fabre's views that insects are incapable of causal reasoning (but see Griffin, 2001).

Unfortunately, contemporary studies of causal reasoning often appear to lack the clarity and seeming finality of Fabre's (1915) and Thorndike's (1911) studies. For instance, whereas some observations of cotton-top tamarins, capuchins, and chimpanzees are consistent with causal reasoning, other observations offer ambiguous results and still others argue against it (Fujita, Kuroshima, & Asai, 2003; Hauser, 1997, 2003; Horner & Whiten, 2004; O'Connell & Dunbar, 2005; Premack & Premack, 1988). As just one example of this ambiguity, consider the trap-tube task. After presenting a similar task to the same species, one research team concluded that some chimpanzees seemed to understand the requirements of the task (reviewed in Visalberghi, 2002); another team concluded that the chimpanzees did not (reviewed in Povinelli, 2000).

In this article, I explore the utility of a consciously Fabrean approach to the perception of causality in elephants. In particular, this article raises the possibility of a cognitive resemblance between elephants and Thorndike's cats. This resemblance does not rule out—but still is not readily reconcilable with—the attribution to elephants of ideation (Rensch, 1957), thinking (Gordon, 1966), insight (Nissani, 2004; Williams, 1950), grief (Spinage, 1994), suicide (Gale, 1974), deception (Morris, 1986), and consciousness (Poole, 1997). The intent of this article is also to expand the meager database of elephant behavior and cognition (for recent claims and counterclaims about elephants' exceptional cognition, see Bradshaw, Schore, Brown, Poole, & Moss, 2005; Hart, Hart, McCoy, & Sarath, 2001; McComb, Moss, Sayialel, & Baker, 2000; Nissani, 2004; Nissani, Hoefler-Nissani, Lay, & Htun, 2005).

To determine whether elephants learn about the consequences of their actions in such a way that they can change their actions appropriately when circumstances change or, alternatively, whether instrumental conditioning simply strengthens their S-R habits, in Experiment 1 Burmese elephants were pretrained to remove food from a coverless bucket. They were next pretrained to remove a lid from the top of that bucket, insert their trunk into the bucket, and retrieve a food reward from the bottom of the bucket. Once this behavioral sequence was established, the lid was placed alongside the bucket so that it no longer obstructed access to the reward. If elephants can change their actions appropriately (Dickinson & Balleine, 2000), they might be expected to ignore the side lid in the first few trials and retrieve the reward directly, as they did in the early stages of pretraining. On the other hand, if instrumental conditioning merely strengthens an S-R habit, in the first few trials the elephants might be expected to continue removing the lid before retrieving the food. Experiment 2 tested the generalizability, reliability, and robustness of the results.

Four female Asian elephants (Elephas maximus) served as subjects (identified by their government registration numbers): No. 6389, age 7 years at the time of the experiment; No. 5811, age 16; No. 5655, age 16; and No. 4800, age 31. Prior to the initiation of the critical side-displacement trials, 2 of these subjects took part in black+/white–visual discrimination and acuity tasks and 2 took part in black–/white+ visual discrimination and acuity tasks involving the same apparatus used in the present experiments (Nissani et al., 2005). In an average of 161 trials, the 3 younger elephants successfully met an overall learning criterion in that task; the older 31-year-old performed at the 50% level in 388 trials. Overall, just prior to the beginning of the critical side-placement trials, the 4 elephants took part in an average of 3.3 (range = 2–5) pretraining sessions, 16.4 (range = 7–22) discrimination sessions, and 870.5 (range = 342–1,180) discrimination trials.

Owing to lack of electricity and other conveniences, the present experiments relied on primitive equipment and extensive controls. All lid-removal tasks involved the use of a green plastic bucket. Each tapered bucket was 32 cm tall, had an internal top diameter of 30 cm, and had an internal bottom diameter of 25 cm. On one side of each bucket, about 10 cm from its bottom, there was a 10-cm hole—large enough for an experimenter to insert one hand containing a food reward but too large for an elephant to insert a trunk. The bucket was often secured to the ground with bamboo stakes around its perimeter.

The bucket had a green, black, or white, circular, 33-cm lid. A white, semicircular handle was attached to each lid. The lid and the handle were made of the same hard plastic material as that of the bucket.

The present experiments were carried out in the forests of central Burma (Myanmar), in an open clearing, in full daylight (for details, see Nissani et al., 2005). Burmese logging elephants feed on their own in the forest on rest days or when the day's work is done, and they are rarely given supplementary provisions. Thus, they responded eagerly to the food rewards used in our experiments: (a) a 3- to 8-cm piece of sugarcane, (b) a 1- to 2-cm tamarind ball, and (c) a 3- to 6-cm rice paddy wrapped in banana leaves.

The crucial phase involved the elephant's learning to remove a lid from the top of a bucket to obtain food from the bottom of the bucket. The elephant typically stood on one side of a heavy log, directly facing a single bucket on the other side of the log. The bucket was about 1 m from the elephant's feet. A single trainer squatted or sat behind the bucket, directly facing the subject, and gradually trained the elephant to retrieve a reinforcer from a coverless bucket. Once the elephant learned that task, a green, white, or black lid was introduced. At first, the lid barely covered the bucket's opening when the food was placed into the bucket; as the training proceeded, the lid covered an increasingly greater portion of the opening. Once the elephant learned to remove the lid, the trainer held the lid in place and allowed its removal only by the lid handle.

In the later pretraining stages, the trainer simultaneously inserted the food into the bucket through the bottom side hole with one hand and clicked the lid into place with the other hand. The trainer then removed his hand from the handle while still holding the food inside the bucket with the other hand. When the elephant grabbed the handle with its trunk and tossed the lid away, the trainer released the food and removed his hand from inside the bucket; the elephant then grabbed the food with its trunk and placed the food into its mouth. The performance criterion for this pretraining phase was defined as the point at which the elephant grabbed the lid by the handle within 3 s of lid placement, tossed the lid away, and removed the reward from the bottom of the bucket in 9 out of 10 consecutive trials.

Because the 4 elephants in this group had already taken part in black–white discrimination tasks prior to taking part in these side-placement trials, they were next trained to remove one of two lids and retrieve food from one of two adjacent buckets (Nissani et al., 2005). Reward placement in one of the two buckets followed a predetermined random sequence; the criterion was defined as 9 of 10 correct consecutive choices in two consecutive sessions.

The critical phase of the experiment involved interspersing the base trials (lid on top of bucket) with the experimental trials. In these experimental trials, the lid was placed a short distance away from the edge of the bucket, onto the ground, randomly to the right or left of the bucket, so that it no longer obstructed access to the food inside the bucket—that is, during the last stages of pretraining and during base trials of the experiment itself, the elephant had to remove the lid to obtain the reward; during experimental trials, the elephant could ignore the lid and retrieve the reward directly from the bucket.

Two observers independently recorded each subject's behavior. Interobserver reliability for the first five trials, which entailed a judgment as to whether the elephant approached or ignored the lid, was 100%. This interobserver reliability was later confirmed by comparing the written records of one experimenter to the video record, again yielding 100% (20 of 20) concordance of this simple all-or-none behavior. When the elephant did not ignore the lid, the interaction was classified into three categories: full toss, touch, and approach (i.e., moving the trunk toward the lid but not touching it). Here, interobserver reliability was 95% (19 of 20), with the single discrepancy involving a disagreement between a touch and an approach in one of the three cases in which such an intermediate response was observed. That single discrepancy was resolved by consulting the video record.

The 4 elephants successfully achieved pretraining discrimination criterion within 2–5 sessions. Accurate records of the duration of each pretraining session were kept in just one case and comprised 125 min.

Figure 1 traces the performance of the 4 subjects in the critical side-placement configuration. In this figure, side-placement trials (top and bottom of y-axis) were interspersed with base trials in which the lid blocked access to the reward (center of y-axis). A trial that involved a full toss of the lid before food retrieval is shown at the bottom of the y-axis; a trial that involved the elephant merely approaching or touching the lid with its trunk is shown midway between the center and bottom axes. The top of the y-axis depicts experimental trials in which the elephant ignored the lid and directly inserted its trunk into the bucket.

Figure 1: Performance across trials of 4 elephants in lid–bucket side-placement task

 

For instance, in the first 6 trials of No. 5811's experimental session, a reward was placed inside the bucket while a lid was simultaneously placed on top of the bucket. In all 6 base trials, the elephant promptly tossed the lid, inserted its trunk into the bucket, and secured the reward. In the next 10 trials (7–16), the lid was placed onto the ground, to the right or left of the bucket, while the morsel was simultaneously placed into the bucket. In Trials 7–14, the elephant tossed the lid first and only then inserted its trunk into the bucket to retrieve the morsel. In Trial 15, it merely touched the lid and then retrieved the food. In Trial 16, it ignored the lid for the first time. Thereafter, before retrieving the food in the remaining 17 experimental trials, it ignored the lid in 7 of the trials, tossed the lid in 9 of the trials, and merely touched the lid in 1 trial.

The most relevant data involved the first few trials, before the elephants could revert to the early pretraining procedure (prior to the introduction of the lid) of retrieving the food directly from the bucket. In the first 5 side-placement trials, none of the 4 elephants ignored the lid, although their exact behavior varied somewhat. Before retrieving the food, No. 6389, No. 5811, and No. 4800 applied the same full toss to the lid on the ground as they had applied earlier to the lid on top of the bucket. No. 5655 performed a full toss before retrieving the food in its first two trials and merely touched the lid in its last three trials. Similarly, in experimental Trial 4 of the first 5 trials (and in subsequent Trials 7, 12, 13, and 17), the elephant searched for the food under the lid before retrieving it from the bucket (for visual details, see Hoefler-Nissani & Nissani, 2004).

Although the limited number of trials, the variable design, and the all-or-nothing nature of the results do not lend themselves to meaningful statistical analyses, the behavior of the elephants in the first 5 trials is consistent with earlier significant observations (Nissani et al., 2005) of a marked age effect in the learning ability of elephants. Thus, in its last 6 experimental trials, the 7-year-old elephant (No. 6389) ignored the lid and retrieved the food directly from the bucket even though, in this case, trials were often interspersed with base trials. The two 16-year-olds ignored the lid in an average of 1.5 trials out of 6, while the 31-year-old elephant (No. 4800) fully tossed the lid in 21 consecutive trials.

As mentioned previously, only No. 5655—in 5 of 17 experimental trials—searched for food under the lid before removing food from the bucket. In its remaining trials, and in all experimental trials of the other 3 elephants, interactions with the lid were immediately followed by food retrieval from the bucket. Because elephants can sense the presence of food at that distance only by sniffing and touching the ground (Nissani et al., 2005), this observation suggests that the 4 elephants did not form an alternative causal model, in which the lid is perceived as an obstacle to food underneath it (not to food in the bucket).

Eleven elephants from three geographically isolated logging camps served as subjects (see Table 1). As was the case with the 4 elephants of Experiment 1, the 8 older elephants of Experiment 2 previously took part in an extensive series of discrimination studies. Thus, just prior to commencement of the critical side-placement trials, these 8 elephants took part in an average of 3.9 (range = 1–6) pretraining sessions, 12.3 (range = 3–31) discrimination sessions, and 734 (range = 150–1,699) discrimination trials. In contrast, the 3 younger elephants in this group were tested immediately following the completion of pretraining with just a single bucket (2 subjects) or two buckets (1 subject).

 

 

 

With 8 elephants, the lid–bucket combination used in Experiment 1 was used here, as well. In addition to the elephants being consistently tested with either black or white lids, 1 elephant was consistently tested with a green lid.

To rule out the possibility that the observed behavior was traceable to some particular feature of the apparatus, 3 elephants were presented with a variation that involved a circular depression in the ground and black or white teak boxes. The depression was 10 cm deep and 19–23 cm wide. All boxes used were of similar height (7–8 cm), but their surface areas ranged from 245 cm² to 1,681 cm² (for details, see Nissani et al., 2005).

The basic design of Experiment 1 was used, with the following variations (see also Table 1). Experimenter bias was unlikely to account for the observations with the first elephant of Experiment 1 (No. 6389) because its behavior was surprising. However, by the time observations of the second elephant were conducted, it was clear that experimenter bias could influence the results. To rule out such a bias more methodically, more than five different experimenters carried out Experiment 2 in 3 different logging camps, with 11 new elephants. In particular, two experimenters—each working with the first elephant of each of the two additional camps—were deliberately not informed about the goals of the experiment and about the expectations of the senior experimenter, who recorded the observations from a distance and did not interact with the elephants during those critical trials.

As evidenced in the previous experiments, after mastering the lid-removal task, the 4 elephants of Experiment 1 and the 8 oldest elephants of Experiment 2 (see Table 1) took part in an average of 14.8 discrimination sessions and 840.7 discrimination trials before being faced with the lid-placement task. Thus, the possibility was raised that by then, the elephants' behavior had become ingrained, resembling functional fixedness (Luchins, 1959) and belief perseverance (Nisbett & Ross, 1980; Nissani & Hoefler-Nissani, 1992) in humans. Similarly, it could be argued that this overtraining preferentially strengthened the elephants' S-R habits over their action–outcome associations. To mitigate such potential problems, the experimenters implemented several additional controls with the 3 younger elephants of Experiment 2. First, these 3 elephants were presented with the side-displacement task as soon as they had achieved the pretraining criterion with a single bucket (see Table 1), thereby minimizing the probability that the behavior had become ingrained. Second, in each session involving 2 of these 3 youngest elephants, pretraining and critical trials were preceded by 20–64 control trials, in which a piece of sugarcane was noisily tossed into one of two open buckets. In these control trials, if the elephant inserted its trunk into the correct bucket, it was able to retrieve the sugarcane. Thus, throughout the training and just before the critical side-placement trials, these 2 elephants were engaged in a control variation that relied on their ability to retrieve food directly from an open bucket.

As in Experiment 1 and in the bucket–lid variation of Experiment 2, the box–hole variation involved the elephants learning to displace the box to retrieve a reward from a circular depression below the box. Again, in critical trials, the box was placed alongside the circular depression while the reward was simultaneously placed inside it. Additional minor procedural variations involving only 1 or 2 elephants are briefly described in the Results section.

Interobserver reliability for the first five trials, which entailed a judgment of whether the elephant interacted with or ignored the lid or box, was 98% (54 of 55), with the single discrepancy being resolved by the video record. When the elephant interacted with the lid or box, the behavior was classified into three categories: full toss (lid) or displacement (box), touch, and approach (moving the trunk toward the lid or box but not touching it). Two discrepancies occurred. One was resolved by consulting the video record; the other was arbitrarily classified as a touch (as opposed to a toss).

Table 1 summarizes the first 5 side-placement trials. In 3 of 55 trials, 3 different elephants (ages 12–17) ignored the lid and proceeded directly to the bucket or hole. In the remaining 52 trials, 2 food retrievals were preceded by trunk movement toward the box without coming into contact with it, 9 retrievals were preceded by touching (but not moving) the lid or box, and 41 retrievals were preceded by a full toss (of the lid) or displacement (of the box). None of the procedural variations seemed to affect the results, thus rendering unlikely the competing interpretations of experimenter bias and peculiarities of the lid–bucket apparatus.

Observations of the 2 youngest elephants in this group are of particular interest here, for they seem to argue against the functional fixedness (or rigidity-of-behavior) interpretation. These 2 elephants were confronted with the side-placement task as soon as they had mastered the removal of a single lid. Moreover, every pretraining and experimental session with these 2 elephants was preceded by at least 20 control trials in which a piece of sugarcane was tossed into one or another bucket, followed in all cases by trunk insertion into one of the buckets. Despite these two additional controls, and despite the 2 elephants' comparative youth, they tossed or touched the side lid first in 9 of 10 trials.

Only 3 elephants were tested beyond the first five trials. In 2 elephants, with a mean of 6.4 additional experimental trials (beyond the first five), no change took place—they continued to remove the lid or box as they did in the first trial. The performance of the other elephant, however, improved over time (the lid was ignored in three of four additional trials).

Only 1 elephant consistently searched for food under the displaced box before retrieving food from the hole; the other 10 elephants ignored the ground under the box or lid and moved directly to the hole or bucket. This observation again seems to rule out the notion that the elephants entertained a causal theory, but one of a different kind (that the lid or box hindered access to food underneath).

These results are perhaps more readily reconcilable with the associationist model of elephant learning than with a causal learning model. The following qualitative observations likewise do not prove, but appear more consistent with, the associationist model.

The results of this study could be ascribed to a variety of causes besides the failure to form a causal connection between an obstacle and a reward. One cause could be that captive elephants in Myanmar, at times, undergo harsh training (Himmelsbach, 2002) starting at about age 5—a biographical detail that could impair their emotional and intellectual development (Harlow & Harlow, 1965). Another cause could be captivity itself. Still another cause involves the visual nature of the task; it is conceivable that elephants would have grasped the concept of obstacle removal in an equivalent task that relied on olfaction or hearing. Likewise, although this experiment devised two control variations to minimize the effects of functional fixedness and behavioral rigidity, we cannot altogether rule out the possibility that the results speak more to the rigidity of elephants' learned behavior than to their lack of causal reasoning. Finally, although elephants in the wild are known to spontaneously engage in obstacle-removal operations—for example, both African (Gordon, 1966) and Asian (U Tin Lay, personal communication, December 24, 2002) elephants remove sand to gain access to water in a dry riverbed—tasks with human-made artifacts may lack ecological validity.

It may be instructive to suitably modify and apply the side-placement design to great apes and corvids—vertebrates whose visual acuity is sharper than that of macrosmatic elephants but whose cognitive exceptionality is similarly controversial.

Bradshaw, G. A., Schore, A. N., Brown, J. L., Poole, J. H., & Moss, C. J. (2005, February 24). Elephant breakdown. Nature, 433, 807. [Context Link]

Cobos, P. L., López, F. J., Caño, A., Almaraz, J., & Shanks, D. R. (2002). Mechanisms of predictive and diagnostic causal induction. Journal of Experimental Psychology: Animal Behavior Processes, 28, 331–346. Ovid Full Text ExternalResolverBasic Bibliographic Links Library Holdings [Context Link]

Dickinson, A. (1994). Instrumental conditioning. In N. J. Mackintosh (Ed.), Animal learning and cognition (pp. 45–79). San Diego, CA: Academic Press. [Context Link]

Dickinson, A., & Balleine, B. W. (2000). Causal cognition and goal-directed action. In C. Heyes & L. Huber (Eds.), The evolution of cognition (pp. 185–204). Cambridge, MA: MIT Press. [Context Link]

Dickinson, A., & Shanks, D. (1995). Instrumental action and causal representation. In D. Sperber, D. Premack, & A. J. Premack (Eds.), Causal cognition (pp. 5–25). Oxford, England: Clarendon Press. [Context Link]

Fabre, J. H. (1915). The hunting wasps (A. Teixiera de Mattos, Trans.). New York: Dodd, Meade. [Context Link]

Fujita, K., Kuroshima, H., & Asai, S. (2003). How do tufted capuchin monkeys (Cebus apella) understand causality involved in tool use? Journal of Experimental Psychology: Animal Behavior Processes, 29, 233–242. Ovid Full Text ExternalResolverBasic Bibliographic Links Library Holdings [Context Link]

Gale, U. T. (1974). Burmese timber elephant. Rangoon, Myanmar: Trade Corporation. [Context Link]

Gordon, J. A. (1966). Elephants do think. African Wild Life, 20, 75–79. [Context Link]

Gould, J. L., & Gould, C. G. (1982). The insect mind: Physics or metaphysics? In D. R. Griffin (Ed.), Animal mind—Human mind (pp. 269–298). Berlin, Germany: Springer-Verlag. [Context Link]

Griffin, D. R. (2001). Animal minds. Chicago: University of Chicago Press. [Context Link]

Harlow, H. F., & Harlow, M. K. (1965). The affectional systems. In A. M. Schrier, H. F. Harlow, & F. Stollnitz (Eds.), The behavior of non-human primates (pp. 287–334). New York: Academic Press. [Context Link]

Hart, B. S., Hart, L. A., McCoy, M., & Sarath, C. R. (2001). Cognitive behaviour in Asian elephants: Use and modification of branches for fly switching. Animal Behaviour, 62, 839–847. ExternalResolverBasic Bibliographic Links Library Holdings [Context Link]

Hauser, M. D. (1997). Artifactual kinds and functional design features: What a primate understands without language. Cognition, 64, 285–308. ExternalResolverBasic Full Text Bibliographic Links Library Holdings [Context Link]

Hauser, M. D. (2003). To innovate or not to innovate? That is the question. In S. M. Reader & K. N. Laland (Eds.), Animal innovation (pp. 329–338). Oxford, England: Oxford University Press. [Context Link]

Himmelsbach, W. (2002). Ausbildung und Einsatz von Arbeitselefanten in der Forstwirtschaft Myanmars: Beziehungen und Konflikte zwischen Wald, Tieren und Menschen [The training and use of work elephants in Myanmar forestry: Relations and conflicts among forest, animals, and human beings]. Unpublished master's thesis, Georg-August-Universitat, Gottingen, Germany. [Context Link]

Hoefler-Nissani, D. M., Producer & Nissani, M., Writer/Director (2004). The inner life of elephants [University film]. Detroit, MI: Wayne State University. [Context Link]

Horner, V., & Whiten, A. (2004). Causal knowledge and imitation/emulation switching in chimpanzees (Pan troglodytes) and children (Homo sapiens). Animal Cognition, 8, 164–181. ExternalResolverBasic Bibliographic Links Library Holdings [Context Link]

Köhler, W. (1925). The mentality of apes. New York: Harcourt and Brace. [Context Link]

Luchins, A. S. (1959). Rigidity of behavior. Eugene: University of Oregon Press. [Context Link]

McComb, K., Moss, C., Sayialel, S., & Baker, L. (2000). Unusually extensive networks of vocal recognition in African elephants. Animal Behaviour, 59, 1103–1109. ExternalResolverBasic Bibliographic Links Library Holdings [Context Link]

Morris, M. D. (1986). Large scale deceit: Deception by captive elephants? In R. W. Mitchell & N. S. Thompson (Eds.), Deception: Perspectives on human and nonhuman deceit (pp. 183–191). Albany: State University of New York. [Context Link]

Nisbett, R., & Ross, L. (1980). Human inference. Englewood Cliffs, NJ: Prentice Hall. [Context Link]

Nissani, M. (1975). A new behavioral bioassay for an analysis of sexual attraction and pheromones in insects. Journal of Experimental Zoology, 192, 271–275. ExternalResolverBasic Bibliographic Links Library Holdings [Context Link]

Nissani, M. (2004). Theory of mind and insight in chimpanzees, elephants, and other animals? In R. H. Tuttle (Series Ed.), & R. J. Lesley & G. Kaplan (Vol. Eds.), Developments in primatology: Progress and prospects: Vol. 4. Comparative vertebrate cognition pp. 227–261). New York: Kluwer Academic/Plenum Publishers. [Context Link]

Nissani, M., & Hoefler-Nissani, D. M. (1992). Experimental studies of belief-dependence of observations and of resistance to conceptual change. Cognition and Instruction, 9, 97–111. [Context Link]

Nissani, M., Hoefler-Nissani, D., Lay, U. T., & Htun, U. W. (2005). Simultaneous visual discrimination in Asian elephants. Journal of the Experimental Analysis of Behavior, 83, 15–29. (Also featured at: http://seab.envmed.rochester.edu/jeab/articles/2005/jeab-83-01–0015.pdf ). [Context Link]

O'Connell, S., & Dunbar, R. I. M. (2005). The perception of causality in chimpanzees (Pan spp.). Animal Cognition, 8, 60–66. ExternalResolverBasic Bibliographic Links Library Holdings [Context Link]

Poole, J. (1997). Elephants. Stillwater, MN: Voyageur Press. [Context Link]

Povinelli, D. J. (2000). Folk physics for apes: The chimpanzee's theory of how the world works. New York: Oxford University Press. [Context Link]

Premack, D., & Premack, A. J. (1988). The mind of an ape. New York: Norton. [Context Link]

Rensch, B. (1957). The intelligence of elephants. Scientific American, 196, 44–49. ExternalResolverBasic Library Holdings [Context Link]

Spinage, C. A. (1994). Elephants. London: Poyser. [Context Link]

Thorndike, E. L. (1911). Animal intelligence: New York: Macmillan. [Context Link]

Visalberghi, E. (2002). Insight from capuchin monkey studies: Ingredients of, recipes for, and flaws in capuchins' success. In M. Bekoff, C. Allen, & M. Burghardt (Eds.), The cognitive animal (pp. 405–411). Cambridge, MA: MIT Press. [Context Link]

Williams, J. H. (1950). Elephant Bill. Garden City, NY: Doubleday. [Context Link]

Keywords: animal intelligence; causal model theory; stimulus–response association; reasoning; Asian elephants

Back to Elephant Corner      Moti Nissani's Homepage