Abstract: Previous observations of a New Caledonian crow (Corvus moneduloides) spontaneously bending wire and using it as a hook [Weir et al. (2002) Science 297:981] have prompted questions about the extent to which these animals 'understand' the physical causality involved in how hooks work and how to make them. To approach this issue we examine how the same subject (“Betty”) performed in three experiments with novel material, which needed to be either bent or unbent in order to function to retrieve food. These tasks exclude the possibility of success by repetition of patterns of movement similar to those employed before. Betty quickly developed novel techniques to bend the material, and appropriately modified it on four of five trials when unbending was required. She did not mechanically apply a previously learned set of movements to the new situations, and instead sought new solutions to each problem. However, the details of her behaviour preclude concluding definitely that she understood and planned her actions: in some cases she probed with the unmodified tools before modifying them, or attempted to use the unmodified (unsuitable) end of the tool after modification. Gauging New Caledonian crows' level of understanding is not yet possible, but the observed behaviour is consistent with a partial understanding of physical tasks at a level that exceeds that previously attained by any other non-human subject, including apes.

Abstract: Various experiments revealed that if an animal learns a stimulus-response-reinforcer relationship in one context and is then tested in another context there is usually a lessening of stimulus control, and the same discriminative stimuli that reliably controlled the behavior in the first context will have less effect in the new context. This reduction in performance is known as the “context shift effect.” The effect of changing context on the probability of detecting explosives was investigated in seven highly trained explosives detection dogs (EDDs). In experiment 1 the dogs were trained alternately on path A, which always had five hidden explosives, and on a very similar path B, which never had any explosives. Within a few sessions the dogs showed a significant decrease in search behavior on path B, but not on path A. In experiment 2 the same dogs were trained only on path B with a target density of one explosive hidden every 4th day. The probability of the dogs now detecting the explosive was found to be significantly lower than in experiment 1. In experiment 3 the effect of the low target density as used in experiment 2 was investigated on a new but very similar path C. Both the detection probability for the one explosive every 4th day on the new path and the motivation to search were significantly higher than found in experiment 2. Finally, in experiment 4, an attempt was made to recondition the dogs to search on path B. Although trained for 12 daily sessions with one explosive hidden every session, the dogs failed to regain the normal levels of motivation they had shown on both new paths and on the paths that they knew usually contained explosives. The findings reveal that even a very intensively trained EDD will rapidly learn that a specific stretch of path does not contain explosives. The dog will then be less motivated to search and will miss newly placed targets. This learning is specific to the formerly always-clean path and is to some extent irreversible. However, the dog will search and detect normally on new paths even if they are very similar to the always-clean path. The data are discussed in terms of variables affecting renewal. The results suggest that following training designed to make a behavior “context independent,” any extinction training will not generalize beyond that specific context used during the extinction training. In addition, if the behavior is extinguished in a specific context, it will be very difficult to restore that behavior in that context. These conclusions should be considered by anyone attempting to extinguish well-established trans-context behaviors.

Abstract: Non-human animals vary in their ability to make and use tools. The goal of the present study was to further explore what, if anything, differs between tool-users and non-tool-users, and whether these differences lie in the conceptual or motor domain. We tested two species that typically do not use tools-cotton top tamarins (Saguinus oedipus) and vervet monkeys (Cercopithecus aethiops)-on problems that mirrored those designed for prolific tool users such as chimpanzees. We trained subjects on a task in which they could choose one of two canes to obtain an out-of-reach food reward. After training, subjects received several variations on the original task, each designed to examine a specific conceptual aspect of the pulling problem previously studied in other tool-using species. Both species recognized that effective pulling tools must be made of rigid materials. Subsequent conditions revealed significant species differences, with vervets outperforming tamarins across many conditions. Vervets, but not tamarins, had some recognition of the relationship between a tool's orientation and the position of the food reward, the relationship between a tool's trajectory and the substance that it moves on, and that tools must be connected in order to work properly. These results provide further evidence that tool-use may derive from domain-general, rather than domain-specific cognitive capacities that evolved for tool use per se.

Abstract: The work of behavioural pharmacologists has concentrated on small animals, such as rodents and pigeons. The validity of extrapolation of their findings to humans depends upon the existence of parallels in both physiology and psychology between these animals and humans. This paper considers the question whether there are in fact substantial cognitive parallels between, first, different non-human groups of vertebrates and, second, non-humans and humans. Behavioural data from 'simple' tasks, such as habituation and conditioning, do not point to species differences among vertebrates. Using examples that concentrate on the performance of rodents and birds, it is argued that, similarly, data from more complex tasks (learning-set formation, transitive inference, and spatial memory serve as examples) reveal few if any cognitive differences amongst non-human vertebrates. This conclusion supports the notion that association formation may be the critical problem-solving process available to non-human animals; associative mechanisms are assumed to have evolved to detect causal links between events, and would therefore be relevant in all ecological niches. In agreement with this view, recent advances in comparative neurology show striking parallels in functional organisation of mammalian and avian telencephalon. Finally, it is argued that although the peculiarly human capacity for language marks a large cognitive contrast between humans and non-humans, there is good evidence-in particular, from work on implicit learning--that the learning mechanisms available to non--humans are present and do play an important role in human cognition.

Abstract: Rats were exposed to a radial maze containing six black smooth arms and six wire-grid-covered arms and a striped 'exit arm' in experiment 1. The probability of a black or grid arm being baited (5/6 vs 1/6) with sunflower seeds was associated with its proximal cue for some rats (the Relevant Arm Cue group) but not for others (the Irrelevant Arm Cue group). All rats could terminate a trial and receive a highly preferred morsel of apple by entering the exit arm only after having sampled all six seed-baited arms. Relevant Arm Cue rats usually chose some arms from the more densely baited set before choosing an arm from the less densely baited set and made fewer reentries than Irrelevant Arm Cue rats. Although such clustered, higher choice accuracy in the Relevant Arm Cue group corresponds to human clustered, better recall of verbal items from lists hierarchically organized by categories, these rats did not similarly exhaustively retrieve items (arm locations). That is, when required to terminate a trial by entering the 'exit' arm for an apple morsel after having sampled all seed-baited arms, both groups were equally unable to withhold making nonrewarded premature exits. This nonexhaustive foraging search pattern was maintained in the next two experiments in which the radial maze was reduced to three black and three grid arms along with the striped 'exit' arm and in which black and grid arm cues were paired with number of seeds (eight or one) in an arm for Relevant Arm Cue rats. Although Relevant Arm Cue rats displayed perfect clustering by entering all eight-seeded arms before a one-seeded arm, they made more premature exits and reentries into eight-seeded arms in experiment 2 or when forced to enter all eight-seeded arms in experiment 3 than did Irrelevant Arm Cue rats. These foraging tendencies prevent accurate estimations of the amount of information (i.e., arm locations) rats can 'chunk'.