Technical note: Instantaneous sampling intervals validated from continuous video observation for behavioral recording of feedlot lambs

Validation of a methodology for characterization of rumination, lying, standing, and performing non-nutritive oral behaviors and behavioral patterns in Holstein dairy calves

Calf behavior is closely related to its early growth, production performance, and health performance. Continuous behavior recording is the most accurate but also time-consuming method used for monitoring animal behaviors, so the instantaneous sampling method is often adopted to minimize the time required to quantify behavioral observations in animal studies. Moreover, the optimal sampling intervals required to yield accurate information for estimating Holstein dairy calves' behaviors are still unknown. Our primary objective was to determine the most optimal sampling intervals for monitoring behaviors of Holstein dairy calves during preweaning and weaning periods to improve efficiency while maintaining reliability. The secondary objective was to describe their behavioral patterns. Rumination, lying, standing, and non-nutritive oral behavior (NNOB) data of 18 calves (observation time: 360 h/calf, 6,480 h in total) were continuously recorded for 15 d (3 d at 1, 3, 6, 9, and 12 wk of age). The continuous behavioral data were compared with instantaneous sampling at 5 s, 10 s, 15 s, 30 s,1 min, 3 min, 5 min, 10 min, 15 min, 30 min, and 60 min intervals. Sampling intervals were considered accurate if they met 4 criteria: coefficient of determination ≥0.90 (i.e., strongly related to true values), slope = 1, intercept = 0 (i.e., they did not over- or underestimate true values), and relative error <10%. The most optimal sampling interval was considered the highest sampling interval among the 11 sampling intervals that meet the criteria for accurate monitoring. As expected, the strength of the linear relationship between the continuous recording and instantaneous sampling decreased as the sampling intervals increased. The results varied across the different behaviors, with rumination, lying, standing, and NNOB being reliable at instantaneous recordings of 3 min, 10 min, 10 min, and 1 min for the preweaning period (1, 3, and 6 wk of age) and 10 min, 10 min, 15 min, and 3 min for the postweaning period (9 and 12 wk of age). In terms of behavioral patterns, lying time decreased, whereas rumination, standing, and NNOB time increased with age. After weaning, no significant changes in time spent performing these behaviors. Additionally, the rumination behavioral pattern becomes stable after wk 6 with decreasing after the morning feeding and occurring mainly in the morning. In conclusion, instantaneous sampling is a reliable method for monitoring the behaviors of dairy calves, but the optimal sampling intervals should be selected based on different ages and management conditions.

A simulated comparison of behavioural observation sampling methods

Behavioural research requires the use of sampling methods to document the occurrence of responses observed. Sampling/recording methods include ad libitum, continuous, pinpoint (instantaneous), and one-zero (interval) sampling. Researchers have questioned the utility of each sampling method under different contexts. Our study compared computerized simulations of both pinpoint and one-zero sampling to continuous recordings. Two separate computer simulations were generated, one for response frequency and one for response duration, with three different response frequencies (high, medium, or low) and response durations (short, medium, and long) in each simulation, respectively. Similarly, three different observation intervals (5, 50, and 500 s) were used to record responses as both pinpoint and one-zero sampling methods in the simulations. Under both simulations, pinpoint sampling outperformed one-zero sampling, with pinpoint sampling producing less statistical bias in error rates under all frequencies, durations, and observation intervals. As observation intervals increased, both mean error rates and variability in error rates increased for one-zero sampling, while only variability in error rate increased for pinpoint sampling. The results suggest that pinpoint sampling techniques are effective for measuring both frequency (event) and duration (state) behaviours, and that pinpoint sampling is a less statistically biased behavioural observation method than one-zero sampling.

About pace: How variations in method and definition affect quantification of pacing in bears?

Repetitive pacing behavior is exhibited by many species in zoos and is particularly prevalent in carnivores with large home ranges, such as bears. Pacing can be a behavioral indicator of poor welfare, however, understanding this behavior can be challenging. As many bears that pace are singly housed, efforts to systematically examine and ameliorate pacing may be strengthened by multi-institutional studies. However, there is currently no standardized method to quantify pacing, which makes cross-institutional analyses of causal factors and intervening measures challenging. The purpose of this study was to compare multiple sampling methods and definitions for quantifying pacing in bears to understand how they affect outcome measures. We analyzed video recordings of two grizzly and two black bears pacing, using three sampling methods (continuous, instantaneous 30-s interval, instantaneous 1-min interval), and three definitions of pacing (AB-two repetitions of the path, ABA-three repetitions, ABAB-four repetitions). A generalized linear mixed model revealed that continuous and instantaneous 30-s interval methods captured more pacing than instantaneous 1-min methods, and definitions captured a decreasing amount of pacing from AB to ABA to ABAB. AB also captured the highest number of pacing bouts. The importance of comparability across institutions is growing, and a standard methodology and definition for recording pacing would be useful. We suggest that the combination of instantaneous sampling and the ABA definition presents a good balance between capturing the right data and being flexible enough for a variety of institutions to implement.

Technical contribution: use of continuous recording video monitoring of maintenance and pain behaviors in piglets after surgical castration to validate six continuous time sampling periods for behavior scoring

Castration is a painful procedure routinely performed on piglets. Specific periods relative to castration and time sampling rules are used widely to quantify deviations in piglet's behavior associated with castration rather than assessing behavior for the entire trial period. However, very limited work has evaluated time sampling recording rules to quantify behavioral changes to piglets undergoing castration. Therefore, the objective of this study was to validate the accuracy of six continuous time sampling periods (5, 10, 15, 20, 30, and 45 min), to quantify piglet behavior post-castration when compared with a full 60-min observation period. Sixteen Yorkshire-Landrace x Duroc piglets were surgically castrated. Data were collected using continuous observation (recording rule) of each individual male piglet per litter for 60-min post-castration. The 60-min continuous behavioral data set was then subdivided into six data subsets for each defined continuous time sampling period (5, 10, 15, 20, 30, and 45 min). Data from each continuous time sampling period and the full period data were analyzed using both a generalized linear mixed model and linear regression analysis. For the generalized linear mixed model, 30- and 45-min continuous time sampling periods were not different (P > 0.05) when compared with the full observation period data for all behaviors. For the linear regression analysis, affiliative interaction, sitting, walking, huddled up, prostrated, scratching, spasms, and trembling behaviors met the pairwise comparison accuracy criteria: [1) the coefficient of determination (R2) was > 0.90, 2) the intercept did not differ from 0 (P > 0.05), and 3) the slope did not differ from 1 (P > 0.05)] at the 45-min continuous time sampling period compared to full observation period. Results from this study suggest that a 45-min continuous time sampling period would be necessary to accurately investigate piglet behavior during the acute pain sensitivity time post-castration when considering both maintenance and pain-associated behaviors.

Novel predators and anthropogenic disturbance influence spatio-temporal distribution of forest antelope species

Understanding the effects of anthropogenic disturbance on species' behaviour is crucial for conservation planning, considering the extent of habitat loss. We investigated the influence of anthropogenic disturbances including agriculture, urbanisation, protected areas, and the presence of novel predators, on the temporal and spatial behaviour of sympatric forest antelope (Tragelaphus scriptus, Philantomba monticola, Sylvicapra grimmia, and Cephalophus natalensis) in an anthropogenic matrix containing forest fragments in the Indian Ocean Coastal Belt of South Africa. We integrated land-use types with camera-trap data and compared activity patterns using circular statistics and occupancy modelling. Antelope species overlapped in temporal and spatial distribution and exhibited diurnal activity for 50% of the time. All species exhibited nocturnal activity for ∼25-33% of all observations, except for C. natalensis. Nocturnal activity varied between species and land-use types. The predators Canis familiaris, C. mesomelas and Caracal caracal negatively influenced occupancy of P. monticola, S. grimmia and C. natalensis. Humans negatively influenced temporal activity of P. monticola, and spatial distribution of T. scriptus and S. grimmia. C. familiaris had an overall negative influence on S. grimmia. We found spatial, and to a lesser extent temporal, segregation between species. However, plasticity in activity patterns existed, which varied in response to anthropogenic effects and novel predators.