Abstract: Like most large grazing herbivores, horses select their food based on visual cues, odour, taste, texture, availability and variety. There is relatively little published information about the role of flavour in diet selection by domestic horses in comparison with other domestic and companion animals. However, previous trials investigating effects of diet flavour in stabled horses indicated significant effects on foraging behaviour and selection. In this series of three trials we aimed to determine relative acceptance by presenting flavour preference tests to eight horses. Horses were stabled and fed hay ad lib on trial data collection days plus a standard unflavoured concentrate ration at 7:30 a.m. In Trial 1, 15 flavours were separately presented in standard 100 g cereal by-product meals and the trial was replicated. Quantity consumed, time of completion, partial rejection or refusal were recorded. Order of presentation was determined by a Latin Square design. Trial data were collected on five sampling days, separated by a minimum of 1 day. Horses were presented with six flavoured meals daily; minimum 1 h between the meals. Twelve flavours were universally accepted and of these the eight flavours with fastest mean consumption times (banana, carrot, cherry, cumin, fenugreek, oregano, peppermint and rosemary) were presented in paired preference tests in Trial 2. In Trial 2, all paired combinations of the eight flavours were presented, in two tests per day at noon and 4 p.m. Presentations of the same flavour were separated by at least 1 day. Paired presentations were in 300 g cereal by-product. Presentations were terminated when approximately half of the total amount presented had been consumed. Flavour preferences were expressed as a ratio from 0 (rejection) to 1 (exclusive consumption). Paired flavour preferences produced the following rank order: fenugreek, banana, cherry, rosemary, cumin, carrot, peppermint, oregano. In Trial 3, relative consumption times of mineral pellets flavoured with fenugreek and banana were significantly reduced in comparison with unflavoured pellets. In these short-term trials, flavour had significant effects on diet acceptance, selection and consumption times.

Abstract: By the age of 5.5 years, all of the young chimpanzees of Gombe National Park have acquired a skill known as 'termite fishing'. Termite fishing involves inserting a flexible tool made from vegetation into a termite mound and extracting the termites that attack and cling to the tool. Although tool use is a well-known phenomenon in chimpanzees, little is known about how such skills develop in the wild. Prior studies have found adult sex differences in frequency, duration and efficiency of tool-using tasks, with females scoring higher on all measures. To investigate whether these sex differences occurred in youngsters, I performed a 4-year longitudinal field study during which I observed and videotaped young chimpanzees' development of the termite-fishing behaviour. Critical elements of the skill included identifying a hole, making a tool, inserting a tool into a hole and extracting termites. These elements appeared in the same order during the development of all subjects, but females typically peaked at least a year earlier than males in their performance of the skills that precede termite fishing. In addition, young females successfully termite-fished an average of 27 months earlier than young males and were more proficient at the skill after acquisition had occurred. Furthermore, the techniques of female offspring closely resembled those of their mothers whereas the techniques of male offspring did not, suggesting that the process by which termite fishing is learned differs for male and female chimpanzees.

Abstract: We present a review of studies on elephant social and reproductive behaviour. While the social organization of the African savannah elephant (Loxodonta africana africana) has been intensively studied,that of the African forest elephant (Loxodonta africana cyclotis) and the Asian elephant (Elephas maximus) are poorly understood. Noninvasive molecular methods are useful in combination with behavioural data in understanding social organization and dispersal strategies. The ecological determinants of social organization, and the importance of matriarchal leadership to social groups, and relative importance of different forms of communication under various ecological conditions remain interesting topics that await investigation. Reproductive behaviour also has been examined in detail only in the African savannah elephant, although rigorous chemical analyses continue to be carried out using captive elephants of both species. Improved laboratory techniques may enable future work on reproductive signalling in free-ranging elephants, allowing for comprehensive studies of male-male interactions and mate choice by females.

Abstract: Dogs have an unusual ability for reading human communicative gestures (e.g., pointing) in comparison to either nonhuman primates (including chimpanzees) or wolves . Although this unusual communicative ability seems to have evolved during domestication , it is unclear whether this evolution occurred as a result of direct selection for this ability, as previously hypothesized , or as a correlated by-product of selection against fear and aggression toward humans--as is the case with a number of morphological and physiological changes associated with domestication . We show here that fox kits from an experimental population selectively bred over 45 years to approach humans fearlessly and nonaggressively (i.e., experimentally domesticated) are not only as skillful as dog puppies in using human gestures but are also more skilled than fox kits from a second, control population not bred for tame behavior (critically, neither population of foxes was ever bred or tested for their ability to use human gestures) . These results suggest that sociocognitive evolution has occurred in the experimental foxes, and possibly domestic dogs, as a correlated by-product of selection on systems mediating fear and aggression, and it is likely the observed social cognitive evolution did not require direct selection for improved social cognitive ability.

Abstract: In Pavlovian conditioning, animals acquire a response to a previously neutral stimulus (conditioned stimulus, CS), such as a light, if that stimulus predicts a biologically important event (unconditioned stimulus, US), such as delivery of food. Learning typically occurs when the CS precedes the US (forward conditioning), and not when the CS follows the US (backward conditioning). In social learning about predators, the predator stimulus is considered to be the CS to which observers acquire avoidance responses after the stimulus has been presented in contiguity with an alarmed demonstrator, the US. We tested the prediction that social learning of response to a predator would occur even if the social alarm cues (the US) appeared before the predatory stimulus (the CS). Carib grackles, Quiscalus lugubris, responded to a familiar predator presented at close range by suppressing alarm calls. Presentation of an unfamiliar avian model (black-and-yellow pigeon) also decreased calling, and this inhibition of calling was enhanced following a training session in which the model stimulus was presented in association with grackle alarm calls. Acquired inhibition of calling was independent of the order of presentation of the model and an alarm chorus. These are the first results to indicate that social acquisition of predator avoidance is not dependent upon a particular temporal relationship between predators and social alarm cues. Evolution may have modified some properties of Pavlovian conditioning to accommodate social learning about potentially dangerous stimuli.