Abstract: In modern intensive husbandry systems there is an increasing tendency for animals to interact with technical equipment. If the animal-technology interface is well-designed this may improve animal welfare by offering challenges for cognitive adaptation. Here a system and its application is presented that acoustically calls individual pigs out of a group (n = 8) to a feeding station. In three different learning phases, the computer-controlled “call-feeding-station” (CFS) trained the animals to recognize a specific acoustic signal as a summons for food, using a combination of classical and operant conditioning techniques. The experimental group's stall contained four CFSs, at each of which one animal at a time was able to feed. When an animal had learned to discriminate and recognize its individual acoustic signal it had to localize the particular CFS that was calling and to enter inside it. Then, it received a portion of feed, the amount of which was adapted to the respective age of the animals. Each animal was called at several, unpredictable times each day and the computer programme ensured that the total feed supply was sufficient for each animal. In the last phase of the experiment the animals, in addition, had to press a button with an increasing fixed ratio for the delivery of feed. It was demonstrated that the pigs were able to adapt quickly to the CFSs. Although they were challenged over 12 h daily by requirements of attention, sensory localization and motor efforts to gain comparatively low amounts of feed, they performed well and reached fairly constant success rates between 90 and 95% and short delays between 14 and 16 s between a summons and the food release in the last phase of the experiment. The weight gain during the experiment was the same as in a conventionally fed control group (n = 8). We therefore conclude that CFSs present a positive challenge to the animals with no negative effects on performance but with a potentially beneficial role for welfare and against boredom. The system is also a suitable experimental platform for research on the effects of successful adaptation by rewarded cognitive processes in pigs.

Abstract: A gray seal (Halichoerus grypus) was trained to touch a target on its left or right by responding to pointing signals. The authors then tested whether the seal would be able to generalize spontaneously to altered signals. It responded correctly to center pointing and head turning, center upper body turning, and off-center pointing but not to head turning and eye movements alone. The seal also responded correctly to brief ipsilateral and contralateral points from center and lateral positions. Pointing gestures did not cause the seal to select an object placed centrally behind it. Like many animals in similar studies, this gray seal probably did not understand the referential character of these gestures but rather used signal generalization and experience from initial operant conditioning to solve these tasks.

Abstract: The capacity to predict future events permits a creature to detect, model, and manipulate the causal structure of its interactions with its environment. Behavioral experiments suggest that learning is driven by changes in the expectations about future salient events such as rewards and punishments. Physiological work has recently complemented these studies by identifying dopaminergic neurons in the primate whose fluctuating output apparently signals changes or errors in the predictions of future salient and rewarding events. Taken together, these findings can be understood through quantitative theories of adaptive optimizing control.

Abstract: Processed feed for stabled horses is usually presented in buckets or mangers, and is easily and rapidly consumed. Foraging devices based on the Edinburgh foodball can be used to provide part of the ration. Current designs are all placed on the floor, raising concerns regarding ingestion of foreign materials along with the dispensed food. Alternative devices were evaluated, when presented within suitable, clean containers to prolong food-handling times but avoid such issues. In four Latin square designed replicated trials we investigated behaviour of 12 stabled horses with three foraging devices. These were separately presented for 5 min, varied in sensory complexity (round, square, polyhedral) and contained 500 g high fibre pellets. In Trials 1 and 2 six geldings were presented with devices in buckets then mangers. All individuals foraged successfully from at least one device and behaviour was compared. However, all individuals exhibited some frustration while using the devices (either pawing or biting them). Horses frequently removed the devices from the buckets in Trial 1 terminating these sessions. In Trial 2 mean device foraging duration was ranked polyhedral > round > square. Mean pawing rate in Trial 2 was calculated for horses (frequency of pawing per individual/summed duration manipulation and foraging) and was highest with square (0.11, npawers = 6). In Trial 3 six stabled mares were presented with the same foraging devices in mangers. Mean foraging duration with devices again ranked polyhedral > round > square. Mean pawing rate was highest with round device (0.08, npawers = 4). Trial 4 investigated behaviour of six horses when devices initially containing five high fibre pellets became empty. Mean foraging duration with devices ranked round > polyhedral > square. Mean pawing rate was highest with square device (0.11, npawers = 4). All horses foraged successfully from at least one foraging device in buckets and mangers. Devices met initial objectives but the unpredictability of reward suggests a source of frustration and warrants further investigation.