Abstract: Although reinforcement often leads to repetitive, even stereotyped responding, that is not a necessary outcome. When it depends on variations, reinforcement results in responding that is diverse, novel, indeed unpredictable, with distributions sometimes approaching those of a random process. This article reviews evidence for the powerful and precise control by reinforcement over behavioral variability, evidence obtained from human and animal-model studies, and implications of such control. For example, reinforcement of variability facilitates learning of complex new responses, aids problem solving, and may contribute to creativity. Depression and autism are characterized by abnormally repetitive behaviors, but individuals afflicted with such psychopathologies can learn to vary their behaviors when reinforced for so doing. And reinforced variability may help to solve a basic puzzle concerning the nature of voluntary action.

Abstract: Six horses, highly trained for dressage competition, were used to study the stride kinematics of the walk, and to compare the kinematics of the collected, medium, and extended walks. Horses were filmed in a sagittal plane at a rate of 150 frames/s; temporal, linear, and angular data were extracted from the films. Results of ANOVA and Duncan's multiple range test indicated that the speed of the collected walk (1.37 m/s) was significantly (P < 0.01) slower than that of the medium (1.73 m/s) and extended (1.82 m/s) walks, values for which were not significantly different from each other. The increase in speed was associated with a significant increase in stride length, from 157 cm in the collected walk to 193 cm in the extended walk. This was a result of an increase in the over-tracking distance, whereas there was no significant difference in the distance between lateral placements of the limbs. Stride duration decreased (P < 0.01) from the collected walk (1,159 ms) to the extended walk (1,064 ms). Angles of the metacarpal and metatarsal segments, measured on the palmar/ plantar aspect, were higher at impact and lower at lift off in the collected than in the extended walk (P < 0.01). This indicated greater range of angular motion of this segment during the stance phase in the extended walk. Only 1 of the 6 horses had a regular 4-beat rhythm of the footfalls, with equal time elapsing between the lateral and diagonal footfalls.

Abstract: Body weight of 48 horses competing in a 3-day event was measured the day before the event (baseline), following the dressage phase of the event (day 1), after the endurance phases of the event (day 2), and 18 to 24 hours after the endurance phases (day 3). Plasma sodium and potassium concentrations were measured the evening before, immediately after, and 10 minutes after the endurance phases. Total body water, water loss, and net exchangeable cation loss were then calculated. Body weight and total body water were significantly decreased, compared with baseline values, at all times during the event, and significant water loss was detected. The largest changes were recorded after the endurance phases of the event. Water deficits were still detected 18 to 24 hours after the endurance phases of the event. Mean plasma sodium concentration was significantly increased immediately after the endurance phases of the event, compared with concentration measured the evening before, and remained increased after the 10-minute recovery period, presumably because of dehydration. Mean plasma potassium concentration was significantly increased immediately after the endurance phases of the event, compared with concentration measured the evening before, but was not increased after the 10-minute recovery period.

Abstract: REASONS FOR PERFORMING STUDy: MP3 recorders are readily available, small, lightweight and low cost, providing the potential for logging analogue hoof mounted accelerometer signals for the characterisation of equine locomotion. These, however, require testing in practice. OBJECTIVES: To test whether 1) multiple MP3 recorders can maintain synchronisation, giving the ability to synchronise independent recorders for the logging of multiple limbs simultaneously; and 2) features of a foot mounted accelerometer signal attributable to foot-on and foot-off can be accurately identified from horse foot mounted accelerometers logged directly into an MP3 recorder. METHODS: Three experiments were performed: 1) Maintenance of synchronisation was assessed by counting the number of samples recorded by each of 4 MP3 recorders while mounted on a trotting horse and over 2 consecutive 30 min periods in 8 recorders on a bench. 2) Foot-on and foot-off times obtained from manual transcription of MP3 logged data and directly logged accelerometer signal were compared. 3) MP3/accelerometer acquisition units were used to log accelerometer signals from racehorses during extended training sessions. RESULTS: Mean absolute error of synchronisation between MP3 recorders was 10 samples per million (compared to mean number of samples, range 1-32 samples per million). Error accumulation showed a linear correlation with time. Features attributable to foot on and foot off were equally identifiable from the MP3 recorded signal over a range of equine gaits. CONCLUSIONS: Multiple MP3 recorders can be synchronised and used as a relatively cheap, robust, reliable and accurate logging system when combined with an accelerometer and external battery for the specific application of the measurement of stride timing variables across the range of equine gaits during field locomotion. POTENTIAL RELEVANCE: Footfall timings can be used to identify intervals between the fore and hind contacts, the identification of diagonal advanced placement and to calculate stride timing variables (stance time, protraction time and stride time). These parameters are invaluable for the characterisation and assessment of equine locomotion.