How do dogs negotiate the flyball box?- Descriptive analysis of the box turn dynamics

How do dogs negotiate the flyball box?- Descriptive analysis of the box turn dynamics | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article How do dogs negotiate the flyball box?- Descriptive analysis of the box turn dynamics Scott Blake, Gino Bonilla Lemos, Roberta Blake This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6870857/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Studies pertaining to flyball biomechanics are scarce despite the high frequency of injury in flyball dogs. The objective of this study was to describe the phases of the box turn and the technique employed by canine athletes during flyball runs. 181 videos of 32 dogs performing the box turn were analysed. Data observed from video evaluation included approach gait, velocity of approach, approach technique, take-off technique, sequence of limbs at contact and landing, elevation of the CoM, trunk angle during take-off, etc. Results showed a high variability inter-dogs, but an intra-dog consistence in obstacle performance. The average horizontal (v x ) approach velocity was 7.27 ± 0.79 m/s and the mean dimensionless velocity (Froude number) was 3.34 ± 0.47. The approach speed was faster on running approach (7.37 ± 0.81ms − 1 ) than in the decelerated approach (7.1 ± 0.7 ms − 1 ), where a distinct braking phase was observed. Regarding turn style, 105 (58%) runs were completed using a angled turn, where dogs avoided impacting the box straight, whilst the remaining 75 (41.4%) runs used a direct turn style. Given the high variation in velocity and technique, we could not identify clear patterns, but the phases of the obstacle could be clearly described. We could conclude that individual dog technique is very consistent, and we described some patterns associated with the box angle. Further studies in quantitative kinematics and kinetics are needed to clarify the effects of the different box angles, used in the sport, in the musculoskeletal system. Canine flyball kinematics sports injuries Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Introduction A standard flyball competition involves teams of four dogs racing against each other in the form of a relay, along a straight course comprised of two 51ft (due to its US origins) lanes. Hurdles are distanced at 6, 16, 26 and 36ft from the start line, with races timed via an electronic timing gate placed at the start/finish line. Dog height is defined as the wither’s height (the highest point of the shoulders) and is essential for normalising kinematic variables, thereby allowing fair comparisons between dogs of different sizes. Maximum hurdle heights vary depending on governing body but range from 31 to 38 cm and are determined by the height of the smallest competitor within a team. The race itself can be broken down into three basic elements; The run (both to and from the start/finish line), hurdles, and box turn, with each dog encouraged to complete all three in the fastest possible time, and handlers attempting to have each dog in sequence pass nose to nose at the start line to maximise their chance of winning. In the top tiers of the sport, races are won or lost by milliseconds, through a combination of each dog’s athleticism and handler skill. The current competition record for a team completing a run is 14.182 seconds held by Touch and Go of Utah, USA, whilst in the UK, the current Crufts record of 14.53 seconds, which was set in 2022. Interestingly, this is a full 0.68 seconds faster than the previously held UK record and was accomplished using the maximum angulation of flyball box possible (89°) which had not previously used in major competition. Speeds achieved by top level dogs mean they can complete their run in less than four seconds, which, in a straight line, would equate to 8.59 metres per second (ms − 1 ). Although not as fast as a greyhound’s top speed of 17 ms − 1 (Usherwood and Wilson 2005 ), these velocities do not reflect the critical deceleration required to execute a 180° turn effectively. The sport is open to any dog, with very little specialist equipment required, which may explain why it continues to attract new members and competitors across the globe. Exact numbers of participants are hard to come by, but in the UK alone there are over 400 registered teams, the majority of which are overseen by the British Flyball Association (BFA), whose current membership includes over 6,000 dogs. The largest flyball association in the United States has over 12,000 members and governing bodies are registered as far afield as South Africa, Australia, and Hong Kong. As any sport, flyball comes with its risks for dogs. Blake et al., ( 2023 ) has identified box angle, dog age, and speed to complete the course as factors increasing the odds of injury, and injuries to the forelimbs and back are most common within the sport (Montalbano et al. 2019 ; Pinto et al. 2021 ; Blake et al. 2023 ) The aim of this study was to identify, understand and describe the key kinematics elements of a dog’s movement during a flyball turn. To the authors’ knowledge there are currently no definition of the phases of a flyball turn neither basic kinematics descriptions of a turn. Therefore, it was the primary purpose of this study to quantitatively describe the kinematics of dogs during a flyball box turn and to create a comprehensive description of it. After quantifying the mechanics of dog’s turning on the flyball box, we additionally sought to examine the relationship between continuous variables and associations between categorial variables to establish patterns of dynamics of a flyball box turn. Material and Methods Study Design The data for this descriptive study were captured in a biomechanics set up over a span of three days. Study Population A convenience sample of dogs was recruited from those that regularly train and compete in flyball, via the BFA in the UK. To be included in the study dogs had to have been actively training or competing in flyball during the last 12 months and be free from injury or orthopaedic conditions. For ethical reasons, dogs were allowed to use their own specific box angle which minimised the risk of injuries which could be caused by using box angles they are not used to. Dogs were also allowed to have the jumps set up as per their usual team jump height according to the BFA rules (hurdle height is determined by the smallest dog on the team, and is typically 4 inches (10 cm) below that dog's shoulder height). Lastly, owners were also asked to confirm how long each dog usually took to complete a flyball run in seconds. A total of thirty-two dogs took part in data collection. There were a variety of breeds and as such the sample was deemed to be representative of typical flyball participants. Dogs used their usual box angle of either 45 degrees (n=7); 50 degrees (n=5); 60 degrees (n=9); 70 degrees (n=6) and 83 degrees (n=5) and were allowed to turn to their usual turn side, to the right or to the left. Demographics for each box angle group can be seen in Table 1 and Supplementary table S1.. Experimental set-up A UK standard flyball course was created in an indoor sports hall, consisting of a 4ft x 51ft run, with n=4 hurdles positioned at 6,16,26 and 36 feet from the start/finish line according to the British Flyball Association regulations. Hurdle height was set at 6 inches, which is the minimum height a dog is expected to negotiate during a BFA sanctioned competition . A running surface, constructed from 10mm Tuffspun® matting commonly used in both training and competition was used to denote the flyball lane (Figure 1). Handlers prepared their dogs as they would in flyball competitions, allowing acclimation to the research environment and dogs were directed by their usual handler throughout the study. Prior to the start of data collection owners were requested to warm their dog up in the same manner as for a competition, in any way the handler deemed suitable. Each dog was encouraged to complete the course for a maximum of seven times, aiming at least three successful runs from each dog. Therefore, if a dog did not retrieve the ball three times within seven attempts, data collection was stopped to avoid the dog becoming fatigued. The data collection has yielded 181 valid runs for analysis. Dogs were free to be withdrawn from the study at the owner’s discretion if the dog appeared reluctant or incapable of continuing with the exercise. All flyball boxes were manufactured by Guard Dog Products UK Ltd and were constructed using heavy duty plywood to the standard dimensions set out in British Flyball Association guidelines. Contact surfaces were covered by a 5 mm nonslip rubberised surface (Janxo products UK). Each box had a different contact surface angle: 45°; 50°; 60°; 70° or 83° (Figure 2) but all used the same interchangeable ball release mechanism. Kinematics data collection and analysis A 2D camera (Miqus M3 Colour, Qualysis, Gothenburg, Sweden) capturing 240 frames per second (fps) was attached to a tripod and connected to a laptop and placed perpendicular to the box, to capture 2D movement during the box turn (Appelgrein et al. 2018). In addition, n=6, 3D cameras (Miqus M3 Colour, Qualysis, Gothenburg, Sweden) were placed around the flyball box for the purposes of trialling 3D capture for future research. Data were recorded from the moment each dog passed the last hurdle approximately 5m from the flyball box and finished when they had completed their turn and passed the same hurdle in the direction of the finish line. Direction of turn was noted at each run to assist with data processing. For the initial kinematic description, elements of the flyball contact were separated into similar phases used to describe a horse jumping. Approach was defined as the distance between landing from the last hurdle and take off to the box contact, whilst take off was deemed to be the moment the forelimbs left the ground, until hindlimbs left the ground. Despite variability in the number of approach strides, gait patterns were standardised by analysing only the final stride prior to take-off. Contact/turn was measured as the contact interaction of the dog with the box, from the contact of the first forelimb until the last limb contact with the box, including the turn, whilst landing was considered to be the phase from the first contact with the ground after the interaction with the box until all four limbs had fully contacted the ground contact (Clayton 1989). In instances where a dog’s paw contacted the box while another remained on the ground, to clearly demarcate the transition between phases was defined by the first box contact, regardless of other limbs still be contacting the ground. Temporal and linear kinematics data were obtained from a motion analysis software (Quintic Biomechanics, Birmingham, UK). Centre of mass (CoM) was estimated using a point on the 11 th rib at the level of the costochondral junction (Voss et al ., 2011) which allowed for calculation of the CoM horizontal displacement at the box contact and of the approach speed, from last jump to take-off. Horizontal (v x ) approach velocity calculation was adapted from previous works described for A-frame negotiation in agility (Appelgrein et al. 2019), using horizontal displacement values. Values were calculated using the landing from the last jump until take-off (Figure 3). Once approach velocity was calculated it was also normalised for withers height using the (Alexander and Jayes 1983),employed to normalise approach velocity by accounting for differences in withers height, thus enabling fair comparisons between dogs. Trunk inclination angle was measured based on previously published research (Miró et al. 2020a) using the apparent locations of withers and tuber sacrale to measure the angle between the horizontal plane and line of the back at maximum elevation of the body during take-off (Figure 3). Each video was viewed to confirm successful completion of a run. Complete runs were then smoothed using a Butterworth low-pass filter, fourth order with a cutoff frequency of 10 Hz. Complete videos with full runs were then subsequently divided into four categories: Approach, take-off, box contact and landing. Data collected regarding the technique used by dogs to negotiate the obstacle at each run was determined by the authors after an initial review of the video data. Information noted regarding the approach phase gait and lead was classified according to Hildebrand (1989). Approaches were classified qualitatively as 'running' when no significant deceleration was observed, and as 'decelerated' when a visible braking phase occurred prior to take-off. For take-off the information noted was forelimb take off stance (single or two forelimbs in stance at take-off) and first forelimb to take off. For the box contact, information was noted regarding which forelimb initially made contact, and if all four limbs ultimately made contact or if the hindlimbs remained on the ground (floor contact). Furthermore, it was recorded if the turn was to the right or left and if the technique was ‘head-on’ contact meaning the dog made no attempt to angle its body prior to contact with the box, or, as known within the flyball community, ‘swimmer’s turn’ whereby the dog angled its body in the direction of its turn prior to box contact (Figure 4). To validate results two independent researchers verified the numbers of head-on contact, versus swimmer’s turn. Upon landing after completing the box turn, the forelimb which landed first was noted to establish trail and lead forelimb. Statistical analysis All statistical analysis were performed with SPSS (IBM Corp. Released 2021. IBM SPSS Statistics for Mac, Version 28.0. Armonk, NY: IBM Corp) with confidence level set as 95% (p<0.05). Statistical analysis was performed for a total of 181 flyball runs. The independent variable was the box angles, with analysis of 38 runs at 45° box angle; 24 runs at 50° box angle; 45 runs at 60° box angle; 40 runs at 70° box angle; and 34 runs at 83° box angle. The continuous dependent variables obtained were joint angulation at different phases of the obstacle (approach, take-off, contact, landing), approach speed and Froude number, take-off and landing distances, CoM displacement, trunk angle at take-off, as well height of first forelimb contact with the box. Furthermore, calculations of the box height as percentage of dog withers height were calculated for subsequent data analysis. As this study design was between-subjects, to assess differences across box angles, independent samples tests were used. Differences between trailing and leading limbs were also assessed and these data were deemed related, due to the fact that the limbs were from the same subjects and submitted to the same condition, and the intention was to analyse how the difference in kinetics between limbs could affect the dog’s (experimental unit) risk of injury so two related samples tests were used for these. The data for all parameters were considered continuous and were summarized as mean for each box angle. Parametric data is presented as mean±SD; and non-parametric data is presented as median, IQR= Q1, Q3. To compare the variables between flyball box angles, data from all runs of each dog was averaged, and parametric data were analysed with one-way analysis of variance (one-way ANOVA) with distribution of variables as assessed by visual inspection of a boxplot and Shapiro Wilk test. Homogeneity of variances was tested with Levene’s test and post-hoc tests applied were Tukey’s when equal variances were assumed and Games-Howell when equal variances were not assumed, using Dunn’s (1964) procedure with a Bonferroni correction for multiple comparisons and results report SPSS adjusted p-values. For the non-parametric data, Kruskal-Wallis H test was used. For Kruskal-Wallis, pairwise comparisons were performed using Dunn’s (1964) procedure with a Bonferroni correction for multiple comparisons. The results report SPSS adjusted p-values. To compare between trailing and leading limbs, carpal extension at contact was parametric, and therefore analysed by paired t-test (two-sided p), and carpal extension at landing was non-parametric and as such a Wilcoxon’s test was performed. Furthermore, to analyse if any of the approach and contact techniques influenced the kinematic variables, and to test if dogs’ height would influence obstacle technique, Mann Whitney U test was used if data were non-parametric and an independent t-test was used if data were parametric. Correlations tested individual runs and included approach speed and box height as % of withers height, dog height and trunk angle, trunk angle/CoM displacement, CoM displacement/take-off distance, 1 st forelimb contact height/trunk angle and box height/take-off speed. Pearson’s product-moment correlation was used when data was parametric and Spearman’s rank-order correlation when non-parametric data was detected. Strength thresholds of the correlations were defined as Akoglu (2018), with <+0.2/ – 0.2 considered poor, +0.21/-0.21 to + 0.5/-0.5 considered fair, +0.51/-0.51 to +0.7/-0.7 considered moderate, and +0.71/-0.71 to + 0.99/-0.99 considered very strong. Categorical data analysed data for individual runs and included: approach technique (running or “decelerated”); approach gait (transverse or rotary gallop or half-bound); lead of the gallop at approach; FL take off technique (first forelimb to take off and if the FL stance was single or two FL); ); FL contact to the box pattern; contact technique (box or floor); turn technique (swimmer’s turn or head-on; FL landing pattern (trail forelimb- first FL to land). To analyse categorical data, firstly frequencies were calculated, then to analyse any relevant association cross tabulation and chi-square tests of association were used. Post-hoc with Bonferroni correction were carried out, to find where associations were significant. Values p<0.05 were considered significant. Cramer’s V/phi effect size measurement was used to measure the strength of Dai et al. (2021) any associations, with strength level set according to Phi is a measure for the strength of an association between two categorical variables in a 2 × 2 contingency table. It is calculated by taking the chi-square value, dividing it by the sample size, and then taking the square root of this value. It varies between 0 and 1 without any negative values. Cramer's V is an alternative to phi in tables bigger than 2 × 2 tabulation. Cramer's V varies between 0 and 1 without any negative values. Similar to Pearson's r, a value close to 0 means no association. However, a value bigger than 0.25 is named as a very strong relationship for the Cramer's V, with values bigger than 0.15 considered strong, larger than 0.10 moderate, and larger than 0.05 considered weak (Akoglu 2018). Correlations tested included dog height and trunk angle, trunk angle/CoM displacement, CoM displacement/take-off distance, 1 st forelimb contact height/trunk angle and box height/take-off speed. Pearson’s product-moment correlation was used when data was parametric and Spearman’s rank-order correlation when non-parametric data was detected. Strength thresholds of the correlations were defined as Akoglu (2018), with <+0.2/ – 0.2 considered poor, +0.21/-0.21 to + 0.5/-0.5 considered fair, +0.51/-0.51 to +0.7/-0.7 considered moderate, and +0.71/-0.71 to + 0.99/-0.99 considered very strong. Results Data from a total of 245 flyball runs was collected, however 64 runs were excluded where the dog did not successfully retrieve the ball, leaving 181 runs for analysis. Approach Once the final hurdle was negotiated, all dogs approached the box in a gallop gait, defined as a 4-beat gait with hindlimbs contacting the floor in succession before forelimb contact. Around half of dogs approached on the right lead limb (n=14, 45%); with the other half on the left (n=17, 55%), which was determined by the limbs landing pattern between the last hurdle and take-off. With few exceptions, the approach gallop lead was very consistent within each dog, with minimal intra-dog variation, showing a possible laterality preference for landing from the last jump which then determined the gallop lead towards the box. The majority of runs (n=133, 74.1%) were performed with a 2-suspension phase rotary gallop, whilst n=27, 15.2% of the runs were performed at transverse gallop. Surprisingly, 10.7% (corresponding to three dogs/18 runs) had a half-bound pattern, where the hindlimb stances were so close in time and positioning that even with a frame capture rate of 240 fps it was not possible to define whether the gallop was rotary or transverse (Abourachid 2003). Despite variations in stride count, the overall gait pattern was found to be consistent across all successful runs. At total of 114 (63%) runs saw dogs using a running contact technique where there was no discernible braking phase prior to take off, whilst the remainder (n=67, 37%) used a decelerated contact, displaying a braking phase prior to take-off. The average horizontal (v x ) approach velocity was 7.27±0.79 m/s and the mean dimensionless velocity (Froud number) was 3.34±0.47. A test was carried out for difference in speed between dogs performing running or decelerated approach and if the direction of turn (left/right) influenced approach speed. The approach speed was faster on running approach (7.37±0.81ms -1 ) than on decelerated approach (7.1±0.7 ms -1 ), a statistically significant difference of 0.27 ms -1 (95% CI, 0.035 to 0.51 ms -1 ), t (178)=2.262, p =0.025 (Figure 5). Subsequently, differences in frequency of box contact or decelerated contact were tested, in relation to the height of the box (50.8cm) as a percentage of dog wither height, however no statistical differences were identified (U = 1.836, p=0.066). Lastly, a test was carried out to understand whether the height of dogs at the withers was different with decelerated contact or running contact at approach, however no significant differences were identified (U=1.234, p=0.217). Take-off An interesting feature observed during the take-off (TO) stride is that some dogs (n=7, 21%) presented a single forelimb (trail forelimb) stance during take-off, with what would be the lead forelimb extending directly to contact with the box without floor contact. This was another feature that was consistent, with all seven dogs performing this technique in each of their runs. In summary, four distinct patterns of approach gait and take-off were observed: (1) rotary gallop with double forelimb stance at take-off; (2) rotary gallop with single forelimb (trail) stance at take-off; (3) transverse gallop with double forelimb stance at take-off; (4) transverse gallop with single forelimb (trail) stance at take-off. The analysis concentrated on the final stride prior to take-off, where the gait pattern was most consistent, with minimal variations observed in preceding strides. Lastly in a small number of dogs, a half-bound with both forelimbs at stance at take-off was seen, but these were not consistent enough to be deemed as a valid approach technique. An example take-off sequence can be seen in Figure 6. At point of take-off, 49.7% (n=90) dogs started TO with their right forelimb, 44.2% (n=80) with the left forelimb, and 6.1% (n=11) used both forelimbs simultaneously. There was a fair statistically significant weak positive correlation between dog height and increased trunk angle at take-off (r(179)=0.291, p<0.001), by Spearman’s rank correlation. (See Figure 7). A strong statistically significant association was identified (X 2 (2)=8.825, p<0.012, (Cramer’s V/Phi = 0.221) between the direction of turn (left/right) and the first limb to initiate take-off. Post hoc tests revealed dogs turning left were more likely to initiate take off using the left forelimb (n=50, 62.5%), (p<0.05) whilst those turning right were most likely to use the right forelimb (n=51, 56.7%),(p<0.05) (Figure 8). A fair statistically significant negative correlation was identified between trunk angle and CoM displacement (r(179)=-0.379, p<0.001), by Pearson’s product-moment correlation (Figure 9). There was a fair statistically significant negative correlation between CoM displacement and take-off distance (r(179) =-0.217, p=0.004) (Figure 10), however there was no relationship between dog height and take-off distance (r(176)=-0.025, p=0.736), by Pearson’s product-moment correlation. Box turn We have identified three distinct elements on the box turn: the contact, the turn and the push off. An example of the phases of contact, turn and push off can be seen in Figure 6. A total of 161 runs (89%) were seen where dogs initially contacted the box with both forelimbs at the same time, whilst in n=8 (4.4%) and n=11 (6.1%) runs, dogs made initial contact with either the left or right limb respectively. The large majority of runs (n=169, 93.4%) involved dogs placing all four limbs on the box during contact, whilst in the remaining n=12, (6.6%) dogs kept their hindlimbs on the ground during contact with the box. The relationship between dog height and height of first forelimb contact with the box was assessed, however results were not significant (r(173)=0.005, p=0.948). No significant instances of double hitting were observed during the box turn; any ambiguous cases were resolved using a frame-by-frame video watching. Regarding turn style, 105 (58.3%) runs were completed using a swimmer’s turn, where dogs avoided impacting the box head-on, whilst the remaining 75 (41.7%) runs used a head-on turn style. There was a very strong statistically significant association between box angles (X 2 (4)=23.585, p<0.001, (Cramer’s V/Phi=0.362) and dogs contacting the box head-on, or performing a swimmer’s turn. At 45° and 60° the number of dogs impacting the box head-on was similar to those completing a swimmer’s turn (47.5% and 51.1% versus 52.6% and 48.9%). However, post-hoc analysis showed that at a 50° angle the majority of dogs used a head-on approach (66.7% versus 33.3%) (p<0.05) and at 83°, the majority of dogs (91.2%) (p<0.05) of dogs used a swimmer’s turn (Figure 11). Association between turn direction and the first forelimb to contact the box after take-off was tested, however the results were non-significant (X 2 (2)=4.323, p=0.115). Association between all of the box angles and whether all four limbs were placed on the box, or the hind limbs remained on the floor was also investigated, which found a strong statistically significant association between the angulation of all boxes and dogs placing all four limbs on the box during turning and those whose hindlimbs remained on the floor (X 2 (4)=11.101, p<0.025), (Cramer’s/Phi =0.248). At 45° dogs were much more likely to have their hindlimbs remain on the floor during a turn than any other angulation (81.5%) (p<0.05) (Figure 12). for 181 runs across five different flyball box angulations. 'Box contact' is defined as full contact with the box, while 'floor contact' indicates that the hindlimbs remain on the ground. The height of dogs at the withers was not statistically different whether or not dogs would contact the box with all four paws or leave the hindlimbs remaining on the floor during contact, (U=-0.281,p=0.779). There was a fair statistically significant positive correlation between first forelimb contact height at the box and trunk angle (r(179)=0.23,p=0.002), by Pearson’s product-moment correlation (Figure 13). Landing Upon landing, 74 runs (40.9%) showed dogs landing with their left forelimb first, 73 (40.3%) were where a dog landed with the right forelimb first, and 24 (13.3%) showed dogs landing with both forelimbs touching the ground simultaneously. There was no association between the direction of turn and the first forelimb to touch the ground during landing (X 2 (2)=3.576, p=0.167). An example of the landing phase can be seen in Figure 6. Discussion The data collected within this experiment has provided a better understanding of the complex nature of a flyball turn, however the objective of developing a description that encompasses all elements is challenging. Even the name ‘swimmer’s turn’ is a misnomer, as a swimmer will contact the poolside head-on when initiating a turn, bringing their legs underneath them to rotate through either the vertical or horizontal plane(Chow et al. 1984 ), whereas a flyball dog will attempt to rotate its torso prior to making contact with the box, in effect initiating the turn at take-off. We hereby propose that, technically, what is miscalled swimmer’s turn by the flyball community, is now referred as “angled turn”, and the head-on technique to be technically called “straight turn”, and as such, these terms will be utilised for the remainder of the manuscript. Biomechanically, the standard definition of canine jumping is also too limited, as it only describes jumping from a static stance which includes hindlimbs posturing, take off, flight and landing (Maitre et al., 2007 ).Although the definition of an equine jump takes into account approach (Clayton 1989 ), it nonetheless still falls short of describing all of the elements of a flyball turn. What makes the creation of an all-encompassing descriptor impossible however is the degree of variability within the approach. Albeit intra-dog variability is low, there are at least four distinct approach styles, with observations of a possible fifth. Similar canine studies have encountered the same challenge, with at least five different techniques identified for negotiating weave poles in agility(Eicher et al. 2021 ) whilst most recently attempts to define techniques employed by dogs negotiating the dog walk obstacle in agility were not possible due to the high levels of variability (DiMichele et al. 2024 ). With regards to the approach, although gait pattern varied inter-dog, the majority of dogs used a running contact technique, as an extension of the suspension phase of gallop, in a similar fashion to dogs negotiating an A-frame obstacle in agility (Appelgrein et al. 2018 ). Extending the gallop into a take-off allows the hindlimbs to be used to elevate the CoM at take-off as they move under the body prior to hindlimb retraction. In dogs that had a distinct braking phase prior to TO, previous research suggests that dogs significantly decrease speed once jump height reaches above 76% of their withers height (WH) (Birch et al. 2015 ), however there was no relationship between dog height and running or decelerated contact. In jumping, successful obstacle clearance relies on dogs assessing and executing an optimal combination of forward velocity and distance to the obstacle (Miró et al., 2020; Pfau et al., 2011 ) which cannot be adapted after take-off. A faster approach will reduce the time available to judge optimum take-off distance, and not allow for generation of optimum vertical forces during the final touchdown of the approach stride. The same adaptations have also been noted in horses (Fercher 2017 ), where the obstacle structure is assessed on approach, with slower horses adopting a more appropriate trunk angle at take-off allowing them to achieve a better overall height of CoM (Powers and Harrison 2000 ). In agility, when performing wrap jumps, dogs anticipate the turn and slow down, similar to quadrupedal curve movement (Walter and Carrier 2009 ; Tan and Wilson 2011 ; Wynn et al. 2015 ), which is similar to dogs performing a decelerated contact in boxes with smaller angle. Regarding trunk angle, it is increased in smaller dogs, which may link to the instances of back injuries described in flyball dogs. While certain breeds may show distinct genetic traits, our data do not support broad generalisations regarding breed-specific injury risks. A further hypothesis is that some injuries may be due to the two accelerative phases of a flyball run both to and from the box, combined with the need to achieve decelerative forces prior to the box turn. Dogs will typically lower their centre of mass (CoM) to initiate acceleration, starting with the spine angled downwards cranially before rotating the torso head-end up as power is generated (pitch-axis), in order to offset the GRF being directed in front of the CoM during the first part of stance. This creates greater angular excursions of the spine than that seen during gallop (Walter and Carrier 2007 ) and requires large increases in joint work at both the hip and hock, as well as power production via the hip extensor muscles to generate the amount of torque required (Walter and Carrier 2007 ). In addition, the CoM shifts caudally, generating larger GRFs at the hindlimbs. As speed increases, so does the amplitude of flexion and extension of the spine, generated in part by the rapid shortening of the epaxial muscles to generate propulsive force (Schilling and Hackert, 2006 ; Walter and Carrier, 2007 ). Albeit there was an association between the forelimb that initiates take-off and the direction of turn, whereby a dog that turns to the left will initially lift the contralateral forelimb, there is no continuing association when the dog contacts the box, as most runs resulted in the dogs contacting with both forelimbs simultaneously. This is contrary to what was expected, whereby dogs that performed an angled turn would generate an angle of rotation at take-off, allowing one forelimb to land in advance of the other. Regarding the association between turn direction and forelimb used to initiate take-off, there was no association between limb used at take-off versus lead or trail limb at landing, which would be relevant to injury within a specific limb. Unlike horses, the canine distal forelimb does allow passive rotation, which would enable the dog to maintain a ground contact point with their forelimb’s paw(s) whilst rotating the anterior trunk prior to TO, to generate increased vertical and lateral ground reaction forces and achieve centripetal acceleration (Chateau et al. 2013 ). Dogs are also more capable of combining a jump with a turn, due to greater flexibility within the spine, allowing them to generate a greater degree of lean angle than equines (Parkes and Witte 2015 ). However, the results are more comparable to curve running in quadrupeds (Söhnel et al. 2020 )whereby the inner leg stabilizes the frontal plane’s movement, with a longer stance duration helping to reduce forces whilst the outer leg generates and controls rotation in the horizontal plane(Crevier-Denoix et al. 2013 ; Alt et al. 2015 ). At present it is not fully understood how rotation is initiated, so the assumption is made that torque forces are applied at TO, which would be the most efficient way of creating a turn angle. During take-off, the forelimbs displayed a strut-like function typical of jumping, as indicated by the use of one or both forelimbs to lift the body and transform horizontal velocity into horizontal velocity, which has been identified in both jumping horses and agility dogs(Clayton and Barlow 1989 ; Söhnel et al. 2020 ). However, it is important to distinguish that some dogs in the present study used only trail limb contact with the ground during the TO, with the lead limb going straight to the box to become the inner limb during the turn. In agility jumps, during take-off, forelimbs produce net breaking impulses only, and it is expected that when dog’s take-off with a single forelimb, this limb will have increased braking forces acting upon it with the corresponding decelerative impulses necessary to change the momentum from horizontal to vertical. Displacement of the CoM in relation to the 60° box angle was the second lowest overall, whilst height of forelimb contact on the box was greatest in the 60° group, suggesting that the taller dogs within this group were not needing to displace their CoM to a large degree as they would naturally impact higher on the box, which resulted in a disproportionately lower trunk angle versus the other box angles. What is evident is that because the flyball box is a standard height of 50.8cm, taller dogs may have a competitive advantage because of the reduced requirement for elevating the CoM, saving both time and energy(Daniels and Burn 2018 ). In equine jumping, between three to five times the propulsive force of canter needs to be generated via the hindlimbs to lift the CoM to heights above 0.8m, but below this the CoM remains horizontal, with the horse lifting the limbs as opposed to the trunk at an obstacle (Schambardt et al. 1993 ). It might be assumed therefore that taller dogs employ a similar technique as they approach the box, which does not require a braking phase to allow the limbs to be flexed prior to take off, so velocity and therefore impact forces would be higher. We have identified the two turning techniques amongst the dogs: straight (41.7%, n = 75) and angled turn (58.3%, n = 105), with the technique being consistent within the same dog. The data showed that at angles between 45° and 60° between 47% and 67% of runs involved dogs contacting the box straight whilst the remainder executed an angled turn. At angles of 70° and 83° the percentage of dogs performing an angled turn increased to 61% and 91% respectively. Albeit the dogs using the 83° angle may have been more experienced in general, it may be that angles closer to the vertical encourage an angled turn. Although there is no existing literature with which to compare, in the authors opinion it may be that the dog understands that there is a risk of injury if it impacts the higher angles in a straight fashion, and therefore alters its technique accordingly, similar to changing duty factor to reduce peak vertical GRF during bend running (Self Davies et al. 2019 ). Whereas our data suggest a potential association between turn technique and injury risk, further research is required to confirm this hypothesis. The large majority of dogs (89%), landed with both forelimbs on the box simultaneously, with all of the contact at 83° being both paws, which was contrary to expectations. An angled turn is actively encouraged in the sport, as it is believed to reduce the impact forces through the forelimbs at contact, as the four-limb contact could help to distribute the forces at impact. Albeit no scientific data currently exists to confirm this assumption in dogs, in gymnasts it is known that a synchronous landing in two feet reduces the overall forces on individual limbs when compared with asynchronous landing (Seegmiller and McCaw 2003 ), therefore, it could be anticipated that placing all four limbs on the box would reduce the impact forces through the forelimbs, figurewhich would limit forces through the musculoskeletal system. Data collected through the previous surveys showed that injury is more likely to occur in the forelimb on the inside of the turn, so it was expected to see a greater number of runs involving a single forelimb impacting the box first. Results therefore suggest the injuries previously noted may not be result of impact or rotational forces during box impact alone or may be as a result of another element of the turn. During landing, both forelimbs touch down almost synchronously, and there was no relationship between the turn direction and the limb to land first, which could denote individual preference rather than biomechanical constrictions. The forelimbs shift the body in the new heading direction after a 180° turn generating higher lateral forces and impulses in the hindlimbs, similar to those seen in circling horses (Hobbs et al., 2011 ). In contrast to hindlimbs, canine forelimbs produce zero net forward acceleration in the landing phase (Söhnel et al. 2020 ), similar to run initiation from standing still (Walter and Carrier 2009 ). However, after landing, considerable vertical force still needs to be generated during this period as the upper body is already in a lowered position from weight capturing. Thus, it would seem that the forelimbs support suitable initial body configuration for initiating running by providing braking and lateral forces. Our results demonstrate clear associations between approach speed and turn execution; however, additional studies are needed to explore further correlations. Albeit the data collected was contrary to many of the original assumptions, they are consistent with data that shows that some quadrupeds will have an individual, preferred technique for jumping, that does not deviate with age, but does improve with experience. For example, studies have shown many jumping parameters present in young, untrained horses are retained as they age, even after training (Santamaría et al. 2004 ), but that the variability of the same parameters decreases with age and experience (Becker and Lewczuk 2022 ). Dogs turning on a flyball box are therefore likely to be showing the same characteristics regards a natural jump technique that differs between individuals but is retained by the same dog over a long period of time. This would mean that injury risk is more unique to an individual dog as opposed to extrinsic factors such as the box. As both the dogs age and experience increase and the jump technique becomes less variable, the musculoskeletal system will be subject to greater repetitive strain in specific anatomical locations (Rocheleau et al. 2023 ) caused by the more precise nature of the turn. There were some limitations to this study. The use of 2D data collection as opposed to 3D also limited the number of results that could be collated but did provide insight into the direction of future research. 3D data capture would be beneficial in capturing the majority of a flyball turn but would still be limited by the excessive number of markers that it would require in relation to the dog’s movement, which would create an unacceptable risk of injury to the dog. When applied correctly, the markers used in this study did not noticeably interfere with the dogs’ natural movement patterns. Further studies are needed to ensure the box turn is analysed in a 3D fashion, particularly an overview over the box would be beneficial to define better paw’s placement and style of turn patterns. In conclusion, turning in flyball is an orchestrated, motorically complex manoeuvre which takes training and practise to complete quickly. A 180° change of direction at speed requires the dog to judge its flight arc in relation to the box surface, reduce its velocity to zero whilst retrieving a ball, and generate propulsive forces through the hindlimbs whilst rotating its body in readiness to land squarely, in preparation to re-accelerate back to the start/finish line. The results detailed above have provided valuable insight into the previously unexplored area of turning technique in flyball dogs. We identified that flyball dogs use different but individual techniques for negotiating the box turn, and an assumption could be made that each individual becomes more skilled at their own technique as they gain more experience and practise. In conclusion, our study offers a comprehensive characterisation of flyball turning techniques, highlighting the interplay between speed, deceleration, and turn execution. Declarations The data was acquired according to modern ethical standards and according to guidelines set by The Animal (Scientific Procedures) Act 1986, although this was considered a non-regulated procedure. Research was approved by the Hartpury University Ethics Committee, approval number ETHICS2021-54. A written informed consent was also obtained from the owners of the participants of the study confirming their dog was capable of participating in the research. Acknowledgments The authors would like to thank Skaiste-Justa Masiukaite for her support during data collection. The authors would also like to thank Jeannette Shelley of the British Flyball Association for her supporting role. Our special thanks go out to all of the BFA community and committee, who have supported this project at every step, including taking time out of their busy lives to provide dogs for data collection. Funding The authors declare that no funds, grants, or other support were received during the preparation of this manuscript. Competing Interests The authors have no relevant financial or non-financial interests to disclose. Authors Contributions Scott Blake : conceptualisation, methodology, formal analysis, investigation, writing-original draft, project administration; Gino Bonilla Lemos : investigation, data curation, writing- review and editing; Roberta Blake : methodology, investigation, writing-original draft, resources. Data availability The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. Ethics Approval The data was acquired according to modern ethical standards and according to guidelines set by The Animal (Scientific Procedures) Act 1986, although this was considered a non-regulated procedure. Research was approved by the Hartpury University Ethics Committee, approval number ETHICS2021-54. A written informed consent was also obtained from the owners of the participants of the study confirming their dog was capable of participating in the research. Authors Contributions Scott Blake : conceptualisation, methodology, formal analysis, investigation, writing-original draft, project administration; Gino Bonilla Lemos : investigation, data curation, writing- review and editing; Roberta Blake : methodology, investigation, writing-original draft, resources. References Abourachid A (2003) A new way of analysing symmetrical and asymmetrical gaits in quadrupeds. C R Biol 326:625–630. https://doi.org/10.1016/S1631-0691(03)00170-7 Akoglu H (2018) User’s guide to correlation coefficients. Turk J Emerg Med 18:91–93. https://doi.org/10.1016/j.tjem.2018.08.001 Alexander RMcN, Jayes AS (1983) A dynamic similarity hypothesis for the gaits of quadrupedal mammals. J Zool 201:135–152. https://doi.org/10.1111/j.1469-7998.1983.tb04266.x Alt T, Heinrich K, Funken J, Potthast W (2015) Lower extremity kinematics of athletics curve sprinting. J Sports Sci 33:552–560. https://doi.org/10.1080/02640414.2014.960881 Appelgrein C, Glyde M, Hosgood G, et al (2018) Reduction of the A-Frame Angle of Incline does not Change the Maximum Carpal Joint Extension Angle in Agility Dogs Entering the A-Frame. Veterinary and Comparative Orthopaedics and Traumatology 31:077–082. https://doi.org/10.3415/VCOT-17-04-0049 Appelgrein C, Glyde M, Hosgood G, et al (2019) Kinetic Gait Analysis of Agility Dogs Entering the A-Frame. Veterinary and Comparative Orthopaedics and Traumatology 32:097–103. https://doi.org/10.1055/s-0038-1677492 Becker K, Lewczuk D (2022) Variability of Jump Biomechanics Between Horses of Different Age and Experience Using Commercial Inertial Measurement Unit Technology. J Equine Vet Sci 119:104146. https://doi.org/10.1016/j.jevs.2022.104146 Birch E, Boyd J, Doyle G, Pullen A (2015) The effects of altered distances between obstacles on the jump kinematics and apparent joint angulations of large agility dogs. The Veterinary Journal 204:174–178. https://doi.org/10.1016/j.tvjl.2015.02.019 Blake SP, Melfi VA, Tabor GF, Wills AP (2023) Injury Risk Factors Associated With Training and Competition in Flyball Dogs. Top Companion Anim Med 53–54:100774. https://doi.org/10.1016/j.tcam.2023.100774 Chateau H, Camus M, Holden-Douilly L, et al (2013) Kinetics of the forelimb in horses circling on different ground surfaces at the trot. The Veterinary Journal 198:e20–e26. https://doi.org/10.1016/j.tvjl.2013.09.028 Chow JW, Hay JG, Wilson BD, Imel C (1984) Turning techniques of elite swimmers. J Sports Sci 2:241–255. https://doi.org/10.1080/02640418408729720 Clayton HM (1989) Terminology for the description of equine jumping kinematics. J Equine Vet Sci 9:341–348. https://doi.org/10.1016/S0737-0806(89)80073-5 Clayton HM, Barlow DA (1989) The effect of fence height and width on the limb placements of show jumping horses. J Equine Vet Sci 9:179–185. https://doi.org/10.1016/S0737-0806(89)80046-2 Crevier‐Denoix N, Falala S, Holden‐Douilly L, et al (2013) Comparative kinematic analysis of the leading and trailing forelimbs of horses cantering on a turf and a synthetic surface. Equine Vet J 45:54–61. https://doi.org/10.1111/evj.12160 Dai J, Teng L, Zhao L, Zou H (2021) The combined analgesic effect of pregabalin and morphine in the treatment of pancreatic cancer pain, a retrospective study. Cancer Med 10:1738–1744. https://doi.org/10.1002/cam4.3779 Daniels KAJ, Burn JF (2018) A simple model predicts energetically optimised jumping in dogs. Journal of Experimental Biology. https://doi.org/10.1242/jeb.167379 DiMichele JK, Pechette Markley A, Shoben A, Kieves NR (2024) Evaluation of variability in performance and paw placement patterns by dogs completing the dog walk obstacle in an agility competition. PLoS One 19:e0299592. https://doi.org/10.1371/journal.pone.0299592 Eicher LD, Markley AP, Shoben A, et al (2021) Evaluation of Variability in Gait Styles Used by Dogs Completing Weave Poles in Agility Competition and Its Effect on Completion of the Obstacle. Front Vet Sci 8:. https://doi.org/10.3389/fvets.2021.761493 Fercher C (2017) The Biomechanics of Movement of Horses Engaged in Jumping Over Different Obstacles in Competition and Training. J Equine Vet Sci 49:69–80. https://doi.org/10.1016/j.jevs.2016.10.002 Hildebrand M (1989) The Quadrupedal Gaits of Vertebrates. Bioscience 39:766–775. https://doi.org/10.2307/1311182 HOBBS SJ, LICKA T, POLMAN R (2011) The difference in kinematics of horses walking, trotting and cantering on a flat and banked 10 m circle. Equine Vet J 43:686–694. https://doi.org/10.1111/j.2042-3306.2010.00334.x Maitre P, Poujol L, Lequang T, et al (2007) Jumping in dogs: concurrent assessment of four limbs with a portable electronic walkway. Comput Methods Biomech Biomed Engin 10:105–106. https://doi.org/10.1080/10255840701479214 Miró F, López P, Vilar JM, et al (2020a) Comparative Kinematic Analysis of Hurdle Clearance Technique in Dogs: A Preliminary Report. Animals 10:2405. https://doi.org/10.3390/ani10122405 Miró F, López P, Vilar JM, et al (2020b) Comparative Kinematic Analysis of Hurdle Clearance Technique in Dogs: A Preliminary Report. Animals 10:2405. https://doi.org/10.3390/ani10122405 Montalbano C, Gamble L-J, Walden K, et al (2019) Internet Survey of Participant Demographics and Risk Factors for Injury in Flyball Dogs. Front Vet Sci 6:. https://doi.org/10.3389/fvets.2019.00391 Parkes RS V., Witte TH (2015) The foot–surface interaction and its impact on musculoskeletal adaptation and injury risk in the horse. Equine Vet J 47:519–525. https://doi.org/10.1111/evj.12420 Pfau T, Garland de Rivaz A, Brighton S, Weller R (2011) Kinetics of jump landing in agility dogs. The Veterinary Journal 190:278–283. https://doi.org/10.1016/j.tvjl.2010.10.008 Pinto KR, Chicoine AL, Romano LS, Otto SJG (2021) An Internet survey of risk factors for injury in North American dogs competing in flyball. Can Vet J 62:253–260 Powers PNR, Harrison AJ (2000) A study on the techniques used by untrained horses during loose jumping. J Equine Vet Sci 20:845–850. https://doi.org/10.1016/S0737-0806(00)80115-X Rocheleau PJ, Dycus DL, Lotsikas PJ (2023) Internet-Based Survey on Diagnosis and Treatment Recommendations for Medial Shoulder Syndrome and Instability in Dogs Santamaría S, Bobbert MF, Back W, et al (2004) Evaluation of consistency of jumping technique in horses between the ages of 6 months and 4 years. Am J Vet Res 65:945–950. https://doi.org/10.2460/ajvr.2004.65.945 Schambardt HC, Merkens HW, Vogel V, Willekens C (1993) External loads on the limbs of jumping horses at take-off and landing. Am J Vet Res 54:675–80 Schilling N, Hackert R (2006) Sagittal spine movements of small therian mammals during asymmetrical gaits. Journal of Experimental Biology 209:3925–3939. https://doi.org/10.1242/jeb.02400 Seegmiller JG, McCaw ST (2003) Ground Reaction Forces Among Gymnasts and Recreational Athletes in Drop Landings. J Athl Train 38:311–314 Self Davies ZT, Spence AJ, Wilson AM (2019) Ground reaction forces of overground galloping in ridden Thoroughbred racehorses. Journal of Experimental Biology 222:. https://doi.org/10.1242/jeb.204107 Söhnel K, Rode C, de Lussanet MHE, et al (2020) Limb dynamics in agility jumps of beginner and advanced dogs. Journal of Experimental Biology. https://doi.org/10.1242/jeb.202119 Tan H, Wilson AM (2011) Grip and limb force limits to turning performance in competition horses. Proceedings of the Royal Society B: Biological Sciences 278:2105–2111. https://doi.org/10.1098/rspb.2010.2395 Usherwood JR, Wilson AM (2005) No force limit on greyhound sprint speed. Nature 438:753–754. https://doi.org/10.1038/438753a Walter RM, Carrier DR (2009) Rapid acceleration in dogs: ground forces and body posture dynamics. Journal of Experimental Biology 212:1930–1939. https://doi.org/10.1242/jeb.023762 Walter RM, Carrier DR (2007) Ground forces applied by galloping dogs. Journal of Experimental Biology 210:208–216. https://doi.org/10.1242/jeb.02645 Wynn ML, Clemente C, Nasir AFAA, Wilson RS (2015) Running faster causes disaster: trade-offs between speed, manoeuvrability and motor control when running around corners in northern quolls ( Dasyurus hallucatus ). Journal of Experimental Biology 218:433–439. https://doi.org/10.1242/jeb.111682 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6870857","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":470447131,"identity":"21905dae-5f3a-42fa-ba58-722d27e0aa1d","order_by":0,"name":"Scott Blake","email":"","orcid":"","institution":"Hartpury University","correspondingAuthor":false,"prefix":"","firstName":"Scott","middleName":"","lastName":"Blake","suffix":""},{"id":470447132,"identity":"85f9e4ce-042f-44db-91d4-bd2d4210cf29","order_by":1,"name":"Gino Bonilla Lemos","email":"","orcid":"","institution":"Universidade Federal de Pelotas","correspondingAuthor":false,"prefix":"","firstName":"Gino","middleName":"Bonilla","lastName":"Lemos","suffix":""},{"id":470447133,"identity":"9748537a-35a8-4bdd-95a1-b207c0cbcb68","order_by":2,"name":"Roberta Blake","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA90lEQVRIiWNgGAWjYFACHghlwM7AzMBQARdOIKAFKG/ADNJyhmQtjG1EaDFv7z34uPKHHYM5M/Njg4/z7PINDjA//MDYloZTi8yZc8mGZxKSGSyb2YwTZ25LttxwgM1YgrEtB6cWCYkcM8mGBGYGg8MMxod5tzEbGBxgMAO6sAKfFvOfDQn1QC3snw//nVMP1ML+jZAWM8aGhMNALTzGyYwNh4FaeEC24HEYz7lkyYa04zyWzTzFhj3HjhtIHuYplkg4h9v7Euy9Bz822FTLmbO3b5b4UVNtwHe8feOHD2XJOLXAAA+CCYxRvBE5CkbBKBgFo4AwAAAlhkoeT2Kd5wAAAABJRU5ErkJggg==","orcid":"","institution":"Anglia Ruskin University","correspondingAuthor":true,"prefix":"","firstName":"Roberta","middleName":"","lastName":"Blake","suffix":""}],"badges":[],"createdAt":"2025-06-11 10:38:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6870857/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6870857/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":85021994,"identity":"973cf6f9-31d1-43b1-9060-3a0004ce2e1b","added_by":"auto","created_at":"2025-06-20 05:10:31","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":112778,"visible":true,"origin":"","legend":"\u003cp\u003eEquipment set-up and flyball run configuration. Panel A shows the camera view. Distances between obstacles are marked (6 ft, 16 ft, 26 ft, and 36 ft). Not to scale.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/a90f3d50ca12be3acf499e5e.png"},{"id":85021995,"identity":"ebb13a38-9e72-474e-b4cd-a398d1b6c0b5","added_by":"auto","created_at":"2025-06-20 05:10:31","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":326776,"visible":true,"origin":"","legend":"\u003cp\u003eFlyball boxes used for the experiment, showing their angulations\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/9e486cbe3c3ce072c58ecb44.png"},{"id":85021997,"identity":"639629b7-df6e-463c-9def-f36934e8456b","added_by":"auto","created_at":"2025-06-20 05:10:31","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":147827,"visible":true,"origin":"","legend":"\u003cp\u003eExample of variables collected during the experiment used to identify the key elements of a dog’s movement during a flyball turn. The approach speed is represented as vx. Note that the prop shown in front of the box was allowed only during warm up, not during the data collection.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/f80e3d30c82a3e70c877b3dc.png"},{"id":85024078,"identity":"75ac3b1b-8275-46df-96a6-b4a64813eede","added_by":"auto","created_at":"2025-06-20 05:35:31","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":406214,"visible":true,"origin":"","legend":"\u003cp\u003eHead on box turn (A) versus swimmer’s turn (B)\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/7b0dda74db8252c5203ecff2.png"},{"id":85024059,"identity":"36d3e082-4bd5-4827-a9d5-7af91881224b","added_by":"auto","created_at":"2025-06-20 05:34:35","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":23847,"visible":true,"origin":"","legend":"\u003cp\u003eMean approach speed (m/s) during decelerated (d) and running (r) contact for 32 flyball dogs. The boxplot displays the first (bottom) and third (top) quartiles, with the median shown as a band and the mean indicated by an ‘x’. Whiskers represent the minimum and maximum values. Significant differences are denoted by * for p\u0026lt;0.05 and ** for p\u0026lt;0.01.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/53290203f8f4eb4cb475bf25.png"},{"id":85022000,"identity":"c90992c5-7ec9-4ca3-9c12-112f94c7fe28","added_by":"auto","created_at":"2025-06-20 05:10:31","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":881213,"visible":true,"origin":"","legend":"\u003cp\u003eSequence of still frames showing a single dog during take-off, contact, turn, push-off, and landing using an 83° flyball box.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/a71d518a47d537c41c0387b3.png"},{"id":85024074,"identity":"a1845947-4e5b-4125-bfb1-f9b86a690e44","added_by":"auto","created_at":"2025-06-20 05:35:22","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":50951,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between trunk angle at take-off (°) and withers height (cm) (n = 32 dogs; multiple observations per dog are shown). Linear regression: y = 0.45x + 12.3 (R² = 0.40, p \u0026lt; 0.05).\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/9c50d3dd901cf2da2730a020.png"},{"id":85022002,"identity":"a360d9eb-f044-4297-8028-f50e09ed8334","added_by":"auto","created_at":"2025-06-20 05:10:31","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":16517,"visible":true,"origin":"","legend":"\u003cp\u003eAssociation between the first forelimb (right or left) to initiate take-off and turn direction in flyball dogs (data represent the number of trials for 32 dogs).\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/d0118d78ff6f934afb8de76e.png"},{"id":85024060,"identity":"346404a0-53ab-431b-8a68-78ea44f3cb30","added_by":"auto","created_at":"2025-06-20 05:34:35","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":49886,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between centre of mass displacement (m) and trunk angle (°) at take-off (n = 32 dogs; multiple observations per dog are shown). Linear regression: y = -0.35x + 45.2 (R² = 0.38, p \u0026lt; 0.05).\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/c386c271b949434859e027fd.png"},{"id":85024065,"identity":"28430505-3c4c-4311-baa6-b149ca6aad96","added_by":"auto","created_at":"2025-06-20 05:34:42","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":46025,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between take-off distance (m) and centre of mass displacement (m) during take-off (n = 32 dogs; multiple observations per dog are shown). Linear regression: y = -0.21x + 1.05 (R² = 0.22, p \u0026lt; 0.05).\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/4e22e80212f9dcc53d3e7078.png"},{"id":85021999,"identity":"3ea20a9a-e9fc-4609-9adb-317ae8ce21bb","added_by":"auto","created_at":"2025-06-20 05:10:31","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":24413,"visible":true,"origin":"","legend":"\u003cp\u003eContact technique (“Heads on” versus “swimmers’ turn”) for 32 dogs across five different flyball box angulations. The sample size for each angulation is indicated.\u003c/p\u003e","description":"","filename":"11.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/c393985eebedc9154ca1ba08.png"},{"id":85024061,"identity":"52aed6f2-26ae-4f1d-a898-485adc3cec84","added_by":"auto","created_at":"2025-06-20 05:34:35","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":24421,"visible":true,"origin":"","legend":"\u003cp\u003eLimbs technique at contact technique (‘Box contact’ vs. ‘Floor contact’)\u003c/p\u003e\n\u003cp\u003efor 181 runs across five different flyball box angulations. 'Box contact' is defined as full contact with the box, while 'floor contact' indicates that the hindlimbs remain on the ground.\u003c/p\u003e","description":"","filename":"12.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/d757aedcd3b30a7a82dde241.png"},{"id":85022005,"identity":"f09b2296-eff6-4a32-b62b-b4bd45b512d9","added_by":"auto","created_at":"2025-06-20 05:10:31","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":49619,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between height of forelimb contact (m) and trunk angle (°) during take-off (n = 32 dogs). Linear regression: y = -0.18x + 30.7 (R² = 0.25, p \u0026lt; 0.05). 'Box contact' is defined as full contact with the box, while 'floor contact' indicates that the hindlimbs remain on the ground.\u003c/p\u003e","description":"","filename":"13.png","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/5f604fcbafc31600fc52d067.png"},{"id":102136368,"identity":"b0ce2708-fee4-4b6b-8f48-cbdfd7feaec9","added_by":"auto","created_at":"2026-02-08 08:55:21","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3021872,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6870857/v1/0bfa7d15-f4c6-4c54-8f49-cd816056a89d.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"How do dogs negotiate the flyball box?- Descriptive analysis of the box turn dynamics","fulltext":[{"header":"Introduction","content":"\u003cp\u003eA standard flyball competition involves teams of four dogs racing against each other in the form of a relay, along a straight course comprised of two 51ft (due to its US origins) lanes. Hurdles are distanced at 6, 16, 26 and 36ft from the start line, with races timed via an electronic timing gate placed at the start/finish line. Dog height is defined as the wither\u0026rsquo;s height (the highest point of the shoulders) and is essential for normalising kinematic variables, thereby allowing fair comparisons between dogs of different sizes. Maximum hurdle heights vary depending on governing body but range from 31 to 38 cm and are determined by the height of the smallest competitor within a team. The race itself can be broken down into three basic elements; The run (both to and from the start/finish line), hurdles, and box turn, with each dog encouraged to complete all three in the fastest possible time, and handlers attempting to have each dog in sequence pass nose to nose at the start line to maximise their chance of winning.\u003c/p\u003e \u003cp\u003eIn the top tiers of the sport, races are won or lost by milliseconds, through a combination of each dog\u0026rsquo;s athleticism and handler skill. The current competition record for a team completing a run is 14.182 seconds held by Touch and Go of Utah, USA, whilst in the UK, the current Crufts record of 14.53 seconds, which was set in 2022. Interestingly, this is a full 0.68 seconds faster than the previously held UK record and was accomplished using the maximum angulation of flyball box possible (89\u0026deg;) which had not previously used in major competition. Speeds achieved by top level dogs mean they can complete their run in less than four seconds, which, in a straight line, would equate to 8.59 metres per second (ms\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e). Although not as fast as a greyhound\u0026rsquo;s top speed of 17 ms\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (Usherwood and Wilson \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), these velocities do not reflect the critical deceleration required to execute a 180\u0026deg; turn effectively. The sport is open to any dog, with very little specialist equipment required, which may explain why it continues to attract new members and competitors across the globe. Exact numbers of participants are hard to come by, but in the UK alone there are over 400 registered teams, the majority of which are overseen by the British Flyball Association (BFA), whose current membership includes over 6,000 dogs. The largest flyball association in the United States has over 12,000 members and governing bodies are registered as far afield as South Africa, Australia, and Hong Kong.\u003c/p\u003e \u003cp\u003eAs any sport, flyball comes with its risks for dogs. Blake et al., (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) has identified box angle, dog age, and speed to complete the course as factors increasing the odds of injury, and injuries to the forelimbs and back are most common within the sport (Montalbano et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Pinto et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Blake et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eThe aim of this study was to identify, understand and describe the key kinematics elements of a dog\u0026rsquo;s movement during a flyball turn. To the authors\u0026rsquo; knowledge there are currently no definition of the phases of a flyball turn neither basic kinematics descriptions of a turn. Therefore, it was the primary purpose of this study to quantitatively describe the kinematics of dogs during a flyball box turn and to create a comprehensive description of it. After quantifying the mechanics of dog\u0026rsquo;s turning on the flyball box, we additionally sought to examine the relationship between continuous variables and associations between categorial variables to establish patterns of dynamics of a flyball box turn.\u003c/p\u003e"},{"header":"Material and Methods","content":"\u003ch3\u003e\u003cem\u003eStudy Design\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eThe data for this descriptive study were captured in \u0026nbsp;a biomechanics set up over a span of three days.\u003c/p\u003e\n\u003ch3 id=\"_Toc166564559\"\u003e\u003cem\u003eStudy Population\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eA convenience sample of dogs was recruited from those that regularly train and compete in flyball, via the BFA in the UK. To be included in the study dogs had to have been actively training or competing in flyball during the last 12 months and be free from injury or orthopaedic conditions. For ethical reasons, dogs were allowed to use their own specific box angle which minimised the risk of injuries which could be caused by using box angles they are not used to. Dogs were also allowed to have the jumps set up as per their usual \u0026nbsp;team jump height according to the BFA rules (hurdle height is determined by the smallest dog on the team, and is typically 4 inches (10 cm) below that dog\u0026apos;s shoulder height). Lastly, owners were also asked to confirm how long each dog usually took to complete a flyball run in seconds.\u003c/p\u003e\n\u003cp\u003eA total of thirty-two dogs took part in data collection. There were a variety of breeds and as such the sample was deemed to be representative of typical flyball participants. Dogs used their usual box angle of either 45 degrees (n=7); 50 degrees (n=5); 60 degrees (n=9); 70 degrees (n=6) and 83 degrees (n=5) and were allowed to turn to their usual turn side, to the right or to the left. Demographics for each box angle group can be seen in Table 1 and Supplementary table S1..\u003c/p\u003e\n\u003ch3 id=\"_Toc166564560\"\u003e\u003cem\u003eExperimental set-up\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eA UK standard flyball course was created in an indoor sports hall, consisting of a 4ft x 51ft run, with n=4 hurdles positioned at 6,16,26 and 36 feet from the start/finish line according to the British Flyball Association regulations. Hurdle height was set at 6 inches, which is the minimum height a dog is expected to negotiate during a BFA sanctioned competition . A running surface, constructed from 10mm Tuffspun\u0026reg; matting commonly used in both training and competition was used to denote the flyball lane (Figure 1). Handlers prepared their dogs as they would in flyball competitions, allowing acclimation to the research environment and dogs were directed by their usual handler throughout the study. Prior to the start of data collection owners were requested to warm their dog up in the same manner as for a competition, in any way the handler deemed suitable. Each dog was encouraged to complete the course for a maximum of seven times, aiming at least three successful runs from each dog. Therefore, if a dog did not retrieve the ball three times within seven attempts, data collection was stopped to avoid the dog becoming fatigued. The data collection has \u0026nbsp;yielded 181 valid runs for analysis. Dogs were free to be withdrawn from the study at the owner\u0026rsquo;s discretion if the dog appeared reluctant or incapable of continuing with the exercise.\u003c/p\u003e\n\u003cp\u003eAll flyball boxes were manufactured by Guard Dog Products UK Ltd and were constructed using heavy duty plywood to the standard dimensions set out in British Flyball Association guidelines. Contact surfaces were covered by a 5 mm nonslip rubberised surface (Janxo products UK). Each box had a different contact surface angle: 45\u0026deg;; 50\u0026deg;; 60\u0026deg;; 70\u0026deg; or 83\u0026deg; (Figure 2) but all used the same interchangeable ball release mechanism.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e\u003cem\u003eKinematics data collection and analysis\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eA 2D camera (Miqus M3 Colour, Qualysis, Gothenburg, Sweden) capturing 240 frames per second (fps) was attached to a tripod and connected to a laptop and placed perpendicular to the box, to capture 2D movement during the box turn (Appelgrein et al. 2018). In addition, n=6, 3D cameras (Miqus M3 Colour, Qualysis, Gothenburg, Sweden) were placed around the flyball box for the purposes of trialling 3D capture for future research. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eData were recorded from the moment each dog passed the last hurdle approximately 5m from the flyball box and finished when they had completed their turn and passed the same hurdle in the direction of the finish line. Direction of turn was noted at each run to assist with data processing.\u003c/p\u003e\n\u003cp\u003eFor the initial kinematic description, elements of the flyball contact were separated into similar phases used to describe a horse jumping. Approach was defined as the distance between landing from the last hurdle and take off to the box contact, whilst take off was deemed to be the moment the forelimbs left the ground, until hindlimbs left the ground. Despite variability in the number of approach strides, gait patterns were standardised by analysing only the final stride prior to take-off. Contact/turn was measured as the contact interaction of the dog with the box, from the contact of the first forelimb until the last limb contact with the box, including the turn, whilst landing was considered to be the phase from the first contact with the ground after the interaction with the box until all four limbs had fully contacted the ground contact (Clayton 1989). In instances where a dog\u0026rsquo;s paw contacted the box while another remained on the ground, to clearly demarcate the transition between phases was defined by the first box contact, regardless of other limbs still be contacting the ground.\u003c/p\u003e\n\u003cp\u003eTemporal and linear kinematics data were obtained from a motion analysis software (Quintic Biomechanics, Birmingham, UK). Centre of mass (CoM) was estimated using a point on the 11\u003csup\u003eth\u003c/sup\u003e rib at the level of the costochondral junction (Voss \u003cem\u003eet al\u003c/em\u003e., 2011) which allowed for calculation of the CoM horizontal displacement at the box contact and of the approach speed, from last jump to take-off. \u0026nbsp; Horizontal (v\u003csub\u003ex\u003c/sub\u003e) approach velocity calculation was adapted from previous works described for A-frame negotiation in agility \u0026nbsp;(Appelgrein et al. 2019), using horizontal displacement values. Values were calculated using the landing from the last jump until take-off (Figure 3). Once approach velocity was calculated it was also normalised for withers height using the\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cimg src=\"data:image/png;base64,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\" width=\"831\" height=\"44\"\u003e\u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(Alexander and Jayes 1983),employed to normalise approach velocity by accounting for differences in withers height, thus enabling fair comparisons between dogs. Trunk inclination angle was measured based on previously published research (Mir\u0026oacute; et al. 2020a) \u0026nbsp;using the apparent locations of withers and tuber sacrale to measure the angle between the horizontal plane and line of the back at maximum elevation of the body during take-off (Figure 3).\u003c/p\u003e\n\u003cp\u003eEach video was viewed to confirm successful completion of a run. Complete runs were then smoothed using a Butterworth low-pass filter, fourth order with a cutoff frequency of 10 Hz. Complete videos with full runs were then subsequently divided into four categories: Approach, take-off, box contact and landing.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eData collected regarding the technique used by dogs to negotiate the obstacle at each run was determined by the authors after an initial review of the video data. Information noted regarding the approach phase gait and lead was classified according to Hildebrand (1989). Approaches were classified qualitatively as \u0026apos;running\u0026apos; when no significant deceleration was observed, and as \u0026apos;decelerated\u0026apos; when a visible braking phase occurred prior to take-off. For take-off the information noted was forelimb take off stance (single or two forelimbs in stance at take-off) and first forelimb to take off. For the box contact, information was noted regarding which forelimb initially made contact, and if all four limbs ultimately made contact or if the hindlimbs remained on the ground (floor contact). Furthermore, it was recorded if the turn was to the right or left and if the technique was \u0026lsquo;head-on\u0026rsquo; contact meaning the dog made no attempt to angle its body prior to contact with the box, or, as known within the flyball community, \u0026lsquo;swimmer\u0026rsquo;s turn\u0026rsquo; whereby the dog angled its body in the direction of its turn prior to box contact (Figure 4). To validate results two independent researchers verified the numbers of head-on contact, versus swimmer\u0026rsquo;s turn. Upon landing after completing the box turn, the forelimb which landed first was noted to establish trail and lead forelimb.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e\u003cem\u003eStatistical analysis\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eAll statistical analysis were performed with SPSS (IBM Corp. Released 2021. IBM SPSS Statistics for Mac, Version 28.0. Armonk, NY: IBM Corp) with confidence level set as 95% (p\u0026lt;0.05). Statistical analysis was performed for a total of 181 flyball runs.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe independent variable was the box angles, with analysis of 38 runs at 45\u0026deg; box angle; 24 runs at 50\u0026deg; box angle; 45 runs at 60\u0026deg; box angle; 40 runs at 70\u0026deg; box angle; and 34 runs at 83\u0026deg; box angle. The continuous dependent variables obtained were joint angulation at different phases of the obstacle (approach, take-off, contact, landing), approach speed and Froude number, take-off and landing distances, CoM displacement, trunk angle at take-off, as well height of first forelimb contact with the box. Furthermore, calculations of the box height as percentage of dog withers height were calculated for subsequent data analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs this study design was between-subjects, to assess differences across box angles, independent samples tests were used. \u0026nbsp; Differences between trailing and leading limbs were also assessed and these data were deemed related, due to the fact that the limbs were from the same subjects and submitted to the same condition, and the intention was to analyse how the difference in kinetics between limbs could affect the dog\u0026rsquo;s (experimental unit) risk of injury so two related samples tests were used for these.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe data for all parameters were considered continuous and were summarized as mean for each box angle. Parametric data is presented as mean\u0026plusmn;SD; and non-parametric data is presented as median, IQR= Q1, Q3.\u003c/p\u003e\n\u003cp\u003eTo compare the variables between flyball box angles, data from all runs of each dog was averaged, and parametric data were analysed with one-way analysis of variance (one-way ANOVA) with distribution of variables as assessed by visual inspection of a boxplot and Shapiro Wilk test. Homogeneity of variances was tested with Levene\u0026rsquo;s test and post-hoc tests applied were Tukey\u0026rsquo;s when equal variances were assumed and Games-Howell when equal variances were not assumed, using Dunn\u0026rsquo;s (1964) procedure with a Bonferroni correction for multiple comparisons and results report SPSS adjusted p-values. For the non-parametric data, Kruskal-Wallis H test was used. For Kruskal-Wallis, pairwise comparisons were performed using Dunn\u0026rsquo;s (1964) procedure with a Bonferroni correction for multiple comparisons. The results report SPSS adjusted p-values. To compare between trailing and leading limbs, carpal extension at contact was parametric, and therefore analysed by paired t-test (two-sided p), and carpal extension at landing was non-parametric and as such a Wilcoxon\u0026rsquo;s test was performed. Furthermore, to analyse if any of the approach and contact techniques influenced the kinematic variables, and to test if dogs\u0026rsquo; height would influence obstacle technique, Mann Whitney U test was used if data were non-parametric and an independent t-test was used if data were parametric.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCorrelations tested individual runs and included approach speed and box height as % of withers height, dog height and trunk angle, trunk angle/CoM displacement, CoM displacement/take-off distance, 1\u003csup\u003est\u003c/sup\u003e forelimb contact height/trunk angle and box height/take-off speed. Pearson\u0026rsquo;s product-moment correlation was used when data was parametric and Spearman\u0026rsquo;s rank-order correlation when non-parametric data was detected. Strength thresholds of the correlations were defined as Akoglu (2018), with \u0026lt;+0.2/ \u0026ndash; 0.2 considered poor, +0.21/-0.21 to + 0.5/-0.5 considered fair, +0.51/-0.51 to +0.7/-0.7 considered moderate, and +0.71/-0.71 to + 0.99/-0.99 considered very strong.\u003c/p\u003e\n\u003cp\u003eCategorical data analysed data for individual runs and included: approach technique (running or \u0026ldquo;decelerated\u0026rdquo;); approach gait (transverse or rotary gallop or half-bound); lead of the gallop at approach; \u0026nbsp;FL take off technique (first forelimb to take off and if the FL stance was single or two FL); ); FL \u0026nbsp;contact to the box pattern; contact technique (box or floor); \u0026nbsp;turn technique (swimmer\u0026rsquo;s turn or head-on; FL landing pattern (trail forelimb- first FL to land). To analyse categorical data, firstly frequencies were calculated, then to analyse any relevant association cross tabulation and chi-square tests of association were used. Post-hoc with Bonferroni correction were carried out, to find where associations were significant. Values p\u0026lt;0.05 were considered significant. Cramer\u0026rsquo;s V/phi effect size measurement was used to measure the strength of Dai et al. (2021) any associations, with strength level set according to Phi is a measure for the strength of an association between two categorical variables in a 2 \u0026times; 2 contingency table. It is calculated by taking the chi-square value, dividing it by the sample size, and then taking the square root of this value. It varies between 0 and 1 without any negative values. Cramer\u0026apos;s V is an alternative to phi in tables bigger than 2 \u0026times; 2 tabulation. Cramer\u0026apos;s V varies between 0 and 1 without any negative values. Similar to Pearson\u0026apos;s r, a value close to 0 means no association. However, a value bigger than 0.25 is named as a very strong relationship for the Cramer\u0026apos;s V, with values bigger than 0.15 considered strong, larger than 0.10 moderate, and larger than 0.05 considered weak (Akoglu 2018).\u003c/p\u003e\n\u003cp\u003eCorrelations tested included dog height and trunk angle, trunk angle/CoM displacement, CoM displacement/take-off distance, 1\u003csup\u003est\u003c/sup\u003e forelimb contact height/trunk angle and box height/take-off speed. Pearson\u0026rsquo;s product-moment correlation was used when data was parametric and Spearman\u0026rsquo;s rank-order correlation when non-parametric data was detected. Strength thresholds of the correlations were defined as Akoglu (2018), with \u0026lt;+0.2/ \u0026ndash; 0.2 considered poor, +0.21/-0.21 to + 0.5/-0.5 considered fair, +0.51/-0.51 to +0.7/-0.7 considered moderate, and +0.71/-0.71 to + 0.99/-0.99 considered very strong.\u0026nbsp;\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eData from a total of 245 flyball runs was collected, however 64 runs were excluded where the dog did not successfully retrieve the ball, leaving 181 runs for analysis.\u003c/p\u003e\n\u003ch3 id=\"_Toc166564564\"\u003e\u003cem\u003eApproach\u0026nbsp;\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eOnce the final hurdle was negotiated, all dogs approached the box in a gallop gait, defined as a 4-beat gait with hindlimbs contacting the floor in succession before forelimb contact. Around half of dogs approached on the right lead limb (n=14, 45%); with the other half on the left (n=17, 55%), which was determined by the limbs landing pattern between the last hurdle and take-off. With few exceptions, the approach gallop lead was very consistent within each dog, with minimal intra-dog variation, showing a possible laterality preference for landing from the last jump which then determined the gallop lead towards the box. The majority of runs (n=133, 74.1%) were performed with a 2-suspension phase rotary gallop, whilst n=27, 15.2% of the runs were performed at transverse gallop. Surprisingly, 10.7% (corresponding to three dogs/18 runs) had a half-bound pattern, where the hindlimb stances were so close in time and positioning that even with a frame capture rate of 240 fps it was not possible to define whether the gallop was rotary or transverse (Abourachid 2003). Despite variations in stride count, the overall gait pattern was found to be consistent across all successful runs. At total of 114 (63%) runs saw dogs using a running contact technique where there was no discernible braking phase prior to take off, whilst the remainder (n=67, 37%) used a decelerated contact, displaying a braking phase prior to take-off. The average\u0026nbsp;horizontal (v\u003csub\u003ex\u003c/sub\u003e) approach velocity was 7.27\u0026plusmn;0.79 m/s and the mean dimensionless velocity (Froud number) was 3.34\u0026plusmn;0.47.\u003c/p\u003e\n\u003cp\u003eA test was carried out for difference in speed between dogs performing running or decelerated approach and if the direction of turn (left/right) influenced approach speed. The approach speed was faster on running approach (7.37\u0026plusmn;0.81ms\u003csup\u003e-1\u003c/sup\u003e) than on decelerated approach (7.1\u0026plusmn;0.7 ms\u003csup\u003e-1\u003c/sup\u003e), a statistically significant difference of 0.27 ms\u003csup\u003e-1\u003c/sup\u003e (95% CI, 0.035 to 0.51 ms\u003csup\u003e-1\u003c/sup\u003e), \u003cem\u003et\u003c/em\u003e(178)=2.262, \u003cem\u003ep\u003c/em\u003e=0.025 (Figure 5).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSubsequently, differences in frequency of box contact or decelerated contact were tested, in relation to the height of the box (50.8cm) as a percentage of dog wither height, however no statistical differences were identified (U = 1.836, p=0.066).\u003c/p\u003e\n\u003cp\u003eLastly, a test was carried out to understand whether the height of dogs at the withers was different with decelerated contact or running contact at approach, however no significant differences were identified (U=1.234, p=0.217).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTake-off\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAn interesting feature observed during the take-off (TO) stride is that some dogs (n=7, 21%) presented a single forelimb (trail forelimb) stance during take-off, with what would be the lead forelimb extending directly to contact with the box without floor contact. This was another feature that was consistent, with all seven dogs performing this technique in each of their runs. In summary, four distinct patterns of approach gait and take-off were observed: (1) rotary gallop with double forelimb stance at take-off; (2) rotary gallop with single forelimb (trail) stance at take-off; (3) transverse gallop with double forelimb stance at take-off; (4) transverse gallop with single forelimb (trail) stance at take-off. The analysis concentrated on the final stride prior to take-off, where the gait pattern was most consistent, with minimal variations observed in preceding strides. Lastly in a small number of dogs, a half-bound with both forelimbs at stance at take-off was seen, but these were not consistent enough to be deemed as a valid approach technique.\u003c/p\u003e\n\u003cp\u003eAn example take-off sequence can be seen in Figure 6.\u003c/p\u003e\n\u003cp\u003eAt point of take-off, 49.7% (n=90) dogs started TO with their right forelimb, 44.2% (n=80) with the left forelimb, and 6.1% (n=11) used both forelimbs simultaneously.\u003c/p\u003e\n\u003cp\u003eThere was a fair statistically significant weak positive correlation between dog height and increased trunk angle at take-off (r(179)=0.291, p\u0026lt;0.001), by Spearman\u0026rsquo;s rank correlation. (See Figure 7).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA strong statistically significant association was identified (X\u003csup\u003e2\u003c/sup\u003e(2)=8.825, p\u0026lt;0.012, (Cramer\u0026rsquo;s V/Phi = 0.221) between the direction of turn (left/right) and the first limb to initiate take-off. Post hoc tests revealed dogs turning left were more likely to initiate take off using the left forelimb (n=50, 62.5%), (p\u0026lt;0.05) whilst those turning right were most likely to use the right forelimb (n=51, 56.7%),(p\u0026lt;0.05) (Figure 8).\u003c/p\u003e\n\u003cp\u003eA fair statistically significant negative correlation was identified between trunk angle and CoM displacement (r(179)=-0.379, p\u0026lt;0.001), by Pearson\u0026rsquo;s product-moment correlation (Figure 9).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThere was a fair statistically significant negative correlation between CoM displacement and take-off distance (r(179) =-0.217, p=0.004) (Figure 10), however there was no relationship between dog height and take-off distance (r(176)=-0.025, p=0.736), by Pearson\u0026rsquo;s product-moment correlation.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBox turn\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe have identified three distinct elements on the box turn: the contact, the turn and the push off. An example of the phases of contact, turn and push off can be seen in Figure 6.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA total of 161 runs (89%) were seen where dogs initially contacted the box with both forelimbs at the same time, whilst in n=8 (4.4%) and n=11 (6.1%) runs, dogs made initial contact with either the left or right limb respectively. The large majority of runs (n=169, 93.4%) involved dogs placing all four limbs on the box during contact, whilst in the remaining n=12, (6.6%) dogs kept their hindlimbs on the ground during contact with the box. The relationship between dog height and height of first forelimb contact with the box was assessed, however results were not significant (r(173)=0.005, p=0.948). No significant instances of double hitting were observed during the box turn; any ambiguous cases were resolved using a frame-by-frame video watching.\u003c/p\u003e\n\u003cp\u003eRegarding turn style, 105 (58.3%) runs were completed using a swimmer\u0026rsquo;s turn, where dogs avoided impacting the box head-on, whilst the remaining 75 (41.7%) runs used a head-on turn style. There was a very strong statistically significant association between box angles (X\u003csup\u003e2\u003c/sup\u003e(4)=23.585, p\u0026lt;0.001, (Cramer\u0026rsquo;s V/Phi=0.362) and dogs contacting the box head-on, or performing a swimmer\u0026rsquo;s turn. \u0026nbsp;At 45\u0026deg; and 60\u0026deg; the number of dogs impacting the box head-on was similar to those completing a swimmer\u0026rsquo;s turn (47.5% and 51.1% versus 52.6% and 48.9%). However, post-hoc analysis showed that at a 50\u0026deg; angle the majority of dogs used a head-on approach (66.7% versus 33.3%) (p\u0026lt;0.05) and at 83\u0026deg;, the majority of dogs (91.2%) (p\u0026lt;0.05) of dogs used a swimmer\u0026rsquo;s turn (Figure 11).\u003c/p\u003e\n\u003cp\u003eAssociation between turn direction and the first forelimb to contact the box after take-off was tested, however the results were non-significant (X\u003csup\u003e2\u003c/sup\u003e(2)=4.323, p=0.115). Association between all of the box angles and whether all four limbs were placed on the box, or the hind limbs remained on the floor was also investigated, which found a strong statistically significant association between the angulation of all boxes and dogs placing all four limbs on the box during turning and those whose hindlimbs remained on the floor (X\u003csup\u003e2\u003c/sup\u003e(4)=11.101, p\u0026lt;0.025), (Cramer\u0026rsquo;s/Phi =0.248). \u0026nbsp;At 45\u0026deg; dogs were much more likely to have their hindlimbs remain on the floor during a turn than any other angulation (81.5%) (p\u0026lt;0.05) (Figure 12).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003efor 181 runs across five different flyball box angulations. \u0026apos;Box contact\u0026apos; is defined as full contact with the box, while \u0026apos;floor contact\u0026apos; indicates that the hindlimbs remain on the ground.\u003c/p\u003e\n\u003cp\u003eThe height of dogs at the withers was not statistically different whether or not dogs would contact the box with all four paws or leave the hindlimbs remaining on the floor during contact, (U=-0.281,p=0.779).\u003c/p\u003e\n\u003cp\u003eThere was a fair statistically significant positive correlation between first forelimb contact height at the box and trunk angle (r(179)=0.23,p=0.002), by Pearson\u0026rsquo;s product-moment correlation (Figure 13).\u0026nbsp;\u003c/p\u003e\n\u003ch4\u003e\u003cem\u003eLanding\u003c/em\u003e\u003c/h4\u003e\n\u003cp\u003eUpon landing, 74 runs (40.9%) showed dogs landing with their left forelimb first, 73 (40.3%) were where a dog landed with the right forelimb first, and 24 (13.3%) showed dogs landing with both forelimbs touching the ground simultaneously. There was no association between the direction of turn and the first forelimb to touch the ground during landing (X\u003csup\u003e2\u003c/sup\u003e(2)=3.576, p=0.167).\u003c/p\u003e\n\u003cp\u003eAn example of the landing phase can be seen in Figure 6.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe data collected within this experiment has provided a better understanding of the complex nature of a flyball turn, however the objective of developing a description that encompasses all elements is challenging. Even the name \u0026lsquo;swimmer\u0026rsquo;s turn\u0026rsquo; is a misnomer, as a swimmer will contact the poolside head-on when initiating a turn, bringing their legs underneath them to rotate through either the vertical or horizontal plane(Chow et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e1984\u003c/span\u003e), whereas a flyball dog will attempt to rotate its torso prior to making contact with the box, in effect initiating the turn at take-off. We hereby propose that, technically, what is miscalled swimmer\u0026rsquo;s turn by the flyball community, is now referred as \u0026ldquo;angled turn\u0026rdquo;, and the head-on technique to be technically called \u0026ldquo;straight turn\u0026rdquo;, and as such, these terms will be utilised for the remainder of the manuscript. Biomechanically, the standard definition of canine jumping is also too limited, as it only describes jumping from a static stance which includes hindlimbs posturing, take off, flight and landing (Maitre et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2007\u003c/span\u003e).Although the definition of an equine jump takes into account approach (Clayton \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1989\u003c/span\u003e), it nonetheless still falls short of describing all of the elements of a flyball turn. What makes the creation of an all-encompassing descriptor impossible however is the degree of variability within the approach. Albeit intra-dog variability is low, there are at least four distinct approach styles, with observations of a possible fifth. Similar canine studies have encountered the same challenge, with at least five different techniques identified for negotiating weave poles in agility(Eicher et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) whilst most recently attempts to define techniques employed by dogs negotiating the dog walk obstacle in agility were not possible due to the high levels of variability (DiMichele et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWith regards to the approach, although gait pattern varied inter-dog, the majority of dogs used a running contact technique, as an extension of the suspension phase of gallop, in a similar fashion to dogs negotiating an A-frame obstacle in agility (Appelgrein et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Extending the gallop into a take-off allows the hindlimbs to be used to elevate the CoM at take-off as they move under the body prior to hindlimb retraction.\u003c/p\u003e \u003cp\u003eIn dogs that had a distinct braking phase prior to TO, previous research suggests that dogs significantly decrease speed once jump height reaches above 76% of their withers height (WH) (Birch et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), however there was no relationship between dog height and running or decelerated contact. In jumping, successful obstacle clearance relies on dogs assessing and executing an optimal combination of forward velocity and distance to the obstacle (Mir\u0026oacute; et al., 2020; Pfau et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) which cannot be adapted after take-off. A faster approach will reduce the time available to judge optimum take-off distance, and not allow for generation of optimum vertical forces during the final touchdown of the approach stride. The same adaptations have also been noted in horses (Fercher \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), where the obstacle structure is assessed on approach, with slower horses adopting a more appropriate trunk angle at take-off allowing them to achieve a better overall height of CoM (Powers and Harrison \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). In agility, when performing wrap jumps, dogs anticipate the turn and slow down, similar to quadrupedal curve movement (Walter and Carrier \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Tan and Wilson \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Wynn et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), which is similar to dogs performing a decelerated contact in boxes with smaller angle. Regarding trunk angle, it is increased in smaller dogs, which may link to the instances of back injuries described in flyball dogs. While certain breeds may show distinct genetic traits, our data do not support broad generalisations regarding breed-specific injury risks. A further hypothesis is that some injuries may be due to the two accelerative phases of a flyball run both to and from the box, combined with the need to achieve decelerative forces prior to the box turn. Dogs will typically lower their centre of mass (CoM) to initiate acceleration, starting with the spine angled downwards cranially before rotating the torso head-end up as power is generated (pitch-axis), in order to offset the GRF being directed in front of the CoM during the first part of stance. This creates greater angular excursions of the spine than that seen during gallop (Walter and Carrier \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) and requires large increases in joint work at both the hip and hock, as well as power production via the hip extensor muscles to generate the amount of torque required (Walter and Carrier \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). In addition, the CoM shifts caudally, generating larger GRFs at the hindlimbs. As speed increases, so does the amplitude of flexion and extension of the spine, generated in part by the rapid shortening of the epaxial muscles to generate propulsive force (Schilling and Hackert, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Walter and Carrier, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2007\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAlbeit there was an association between the forelimb that initiates take-off and the direction of turn, whereby a dog that turns to the left will initially lift the contralateral forelimb, there is no continuing association when the dog contacts the box, as most runs resulted in the dogs contacting with both forelimbs simultaneously. This is contrary to what was expected, whereby dogs that performed an angled turn would generate an angle of rotation at take-off, allowing one forelimb to land in advance of the other.\u003c/p\u003e \u003cp\u003eRegarding the association between turn direction and forelimb used to initiate take-off, there was no association between limb used at take-off versus lead or trail limb at landing, which would be relevant to injury within a specific limb. Unlike horses, the canine distal forelimb does allow passive rotation, which would enable the dog to maintain a ground contact point with their forelimb\u0026rsquo;s paw(s) whilst rotating the anterior trunk prior to TO, to generate increased vertical and lateral ground reaction forces and achieve centripetal acceleration (Chateau et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Dogs are also more capable of combining a jump with a turn, due to greater flexibility within the spine, allowing them to generate a greater degree of lean angle than equines (Parkes and Witte \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). However, the results are more comparable to curve running in quadrupeds (S\u0026ouml;hnel et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)whereby the inner leg stabilizes the frontal plane\u0026rsquo;s movement, with a longer stance duration helping to reduce forces whilst the outer leg generates and controls rotation in the horizontal plane(Crevier-Denoix et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Alt et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). At present it is not fully understood how rotation is initiated, so the assumption is made that torque forces are applied at TO, which would be the most efficient way of creating a turn angle. During take-off, the forelimbs displayed a strut-like function typical of jumping, as indicated by the use of one or both forelimbs to lift the body and transform horizontal velocity into horizontal velocity, which has been identified in both jumping horses and agility dogs(Clayton and Barlow \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1989\u003c/span\u003e; S\u0026ouml;hnel et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). However, it is important to distinguish that some dogs in the present study used only trail limb contact with the ground during the TO, with the lead limb going straight to the box to become the inner limb during the turn. In agility jumps, during take-off, forelimbs produce net breaking impulses only, and it is expected that when dog\u0026rsquo;s take-off with a single forelimb, this limb will have increased braking forces acting upon it with the corresponding decelerative impulses necessary to change the momentum from horizontal to vertical.\u003c/p\u003e \u003cp\u003eDisplacement of the CoM in relation to the 60\u0026deg; box angle was the second lowest overall, whilst height of forelimb contact on the box was greatest in the 60\u0026deg; group, suggesting that the taller dogs within this group were not needing to displace their CoM to a large degree as they would naturally impact higher on the box, which resulted in a disproportionately lower trunk angle versus the other box angles. What is evident is that because the flyball box is a standard height of 50.8cm, taller dogs may have a competitive advantage because of the reduced requirement for elevating the CoM, saving both time and energy(Daniels and Burn \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). In equine jumping, between three to five times the propulsive force of canter needs to be generated via the hindlimbs to lift the CoM to heights above 0.8m, but below this the CoM remains horizontal, with the horse lifting the limbs as opposed to the trunk at an obstacle (Schambardt et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e1993\u003c/span\u003e). It might be assumed therefore that taller dogs employ a similar technique as they approach the box, which does not require a braking phase to allow the limbs to be flexed prior to take off, so velocity and therefore impact forces would be higher.\u003c/p\u003e \u003cp\u003eWe have identified the two turning techniques amongst the dogs: straight (41.7%, n\u0026thinsp;=\u0026thinsp;75) and angled turn (58.3%, n\u0026thinsp;=\u0026thinsp;105), with the technique being consistent within the same dog. The data showed that at angles between 45\u0026deg; and 60\u0026deg; between 47% and 67% of runs involved dogs contacting the box straight whilst the remainder executed an angled turn. At angles of 70\u0026deg; and 83\u0026deg; the percentage of dogs performing an angled turn increased to 61% and 91% respectively. Albeit the dogs using the 83\u0026deg; angle may have been more experienced in general, it may be that angles closer to the vertical encourage an angled turn. Although there is no existing literature with which to compare, in the authors opinion it may be that the dog understands that there is a risk of injury if it impacts the higher angles in a straight fashion, and therefore alters its technique accordingly, similar to changing duty factor to reduce peak vertical GRF during bend running (Self Davies et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Whereas our data suggest a potential association between turn technique and injury risk, further research is required to confirm this hypothesis.\u003c/p\u003e \u003cp\u003eThe large majority of dogs (89%), landed with both forelimbs on the box simultaneously, with all of the contact at 83\u0026deg; being both paws, which was contrary to expectations. An angled turn is actively encouraged in the sport, as it is believed to reduce the impact forces through the forelimbs at contact, as the four-limb contact could help to distribute the forces at impact. Albeit no scientific data currently exists to confirm this assumption in dogs, in gymnasts it is known that a synchronous landing in two feet reduces the overall forces on individual limbs when compared with asynchronous landing (Seegmiller and McCaw \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2003\u003c/span\u003e), therefore, it could be anticipated that placing all four limbs on the box would reduce the impact forces through the forelimbs, figurewhich would limit forces through the musculoskeletal system. Data collected through the previous surveys showed that injury is more likely to occur in the forelimb on the inside of the turn, so it was expected to see a greater number of runs involving a single forelimb impacting the box first. Results therefore suggest the injuries previously noted may not be result of impact or rotational forces during box impact alone or may be as a result of another element of the turn.\u003c/p\u003e \u003cp\u003eDuring landing, both forelimbs touch down almost synchronously, and there was no relationship between the turn direction and the limb to land first, which could denote individual preference rather than biomechanical constrictions. The forelimbs shift the body in the new heading direction after a 180\u0026deg; turn generating higher lateral forces and impulses in the hindlimbs, similar to those seen in circling horses (Hobbs et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). In contrast to hindlimbs, canine forelimbs produce zero net forward acceleration in the landing phase (S\u0026ouml;hnel et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), similar to run initiation from standing still (Walter and Carrier \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). However, after landing, considerable vertical force still needs to be generated during this period as the upper body is already in a lowered position from weight capturing. Thus, it would seem that the forelimbs support suitable initial body configuration for initiating running by providing braking and lateral forces. Our results demonstrate clear associations between approach speed and turn execution; however, additional studies are needed to explore further correlations.\u003c/p\u003e \u003cp\u003eAlbeit the data collected was contrary to many of the original assumptions, they are consistent with data that shows that some quadrupeds will have an individual, preferred technique for jumping, that does not deviate with age, but does improve with experience. For example, studies have shown many jumping parameters present in young, untrained horses are retained as they age, even after training (Santamar\u0026iacute;a et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2004\u003c/span\u003e), but that the variability of the same parameters decreases with age and experience (Becker and Lewczuk \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Dogs turning on a flyball box are therefore likely to be showing the same characteristics regards a natural jump technique that differs between individuals but is retained by the same dog over a long period of time. This would mean that injury risk is more unique to an individual dog as opposed to extrinsic factors such as the box. As both the dogs age and experience increase and the jump technique becomes less variable, the musculoskeletal system will be subject to greater repetitive strain in specific anatomical locations (Rocheleau et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) caused by the more precise nature of the turn.\u003c/p\u003e \u003cp\u003eThere were some limitations to this study. The use of 2D data collection as opposed to 3D also limited the number of results that could be collated but did provide insight into the direction of future research. 3D data capture would be beneficial in capturing the majority of a flyball turn but would still be limited by the excessive number of markers that it would require in relation to the dog\u0026rsquo;s movement, which would create an unacceptable risk of injury to the dog. When applied correctly, the markers used in this study did not noticeably interfere with the dogs\u0026rsquo; natural movement patterns. Further studies are needed to ensure the box turn is analysed in a 3D fashion, particularly an overview over the box would be beneficial to define better paw\u0026rsquo;s placement and style of turn patterns.\u003c/p\u003e \u003cp\u003eIn conclusion, turning in flyball is an orchestrated, motorically complex manoeuvre which takes training and practise to complete quickly. A 180\u0026deg; change of direction at speed requires the dog to judge its flight arc in relation to the box surface, reduce its velocity to zero whilst retrieving a ball, and generate propulsive forces through the hindlimbs whilst rotating its body in readiness to land squarely, in preparation to re-accelerate back to the start/finish line. The results detailed above have provided valuable insight into the previously unexplored area of turning technique in flyball dogs. We identified that flyball dogs use different but individual techniques for negotiating the box turn, and an assumption could be made that each individual becomes more skilled at their own technique as they gain more experience and practise. In conclusion, our study offers a comprehensive characterisation of flyball turning techniques, highlighting the interplay between speed, deceleration, and turn execution.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cspan\u003eThe data was acquired according to modern ethical standards and according to guidelines set by The Animal (Scientific Procedures) Act 1986, although this was considered a non-regulated procedure. Research was approved by the Hartpury University Ethics Committee, approval number ETHICS2021-54. A written informed consent was also obtained from the owners of the participants of the study confirming their dog was capable of participating in the research.\u003c/span\u003e\u003c/p\u003e\u003ch2\u003eAcknowledgments\u003c/h2\u003e\n\u003cp\u003eThe authors would like to thank Skaiste-Justa Masiukaite for her support during data collection. The authors would also like to thank Jeannette Shelley of the British Flyball Association for her supporting role. Our special thanks go out to all of the BFA community and committee, who have supported this project at every step, including taking time out of their busy lives to provide dogs for data collection. \u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThe authors declare that no funds, grants, or other support were received during the preparation of this manuscript.\u003c/p\u003e\n\u003ch2\u003eCompeting Interests\u003c/h2\u003e\n\u003cp\u003eThe authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\n\u003ch2\u003eAuthors Contributions\u003c/h2\u003e\n\u003cp\u003e\u003cstrong\u003eScott Blake\u003c/strong\u003e: conceptualisation, methodology, formal analysis, investigation, writing-original draft, project administration; \u003cstrong\u003eGino Bonilla Lemos\u003c/strong\u003e: investigation, data curation, writing- review and editing; \u003cstrong\u003eRoberta Blake\u003c/strong\u003e: methodology, investigation, writing-original draft, resources.\u003c/p\u003e\n\u003ch2\u003eData availability\u003c/h2\u003e\n\u003cp\u003eThe datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003ch2\u003eEthics Approval\u003c/h2\u003e\n\u003cp id=\"_Toc9976531\"\u003eThe data was acquired according to modern ethical standards and according to guidelines set by The Animal (Scientific Procedures) Act 1986, although this was considered a non-regulated procedure. Research was approved by the Hartpury University Ethics Committee, approval number ETHICS2021-54. A written informed consent was also obtained from the owners of the participants of the study confirming their dog was capable of participating in the research.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eAuthors Contributions\u003c/h2\u003e\n\u003cp\u003e\u003cstrong\u003eScott Blake\u003c/strong\u003e: conceptualisation, methodology, formal analysis, investigation, writing-original draft, project administration; \u003cstrong\u003eGino Bonilla Lemos\u003c/strong\u003e: investigation, data curation, writing- review and editing; \u003cstrong\u003eRoberta Blake\u003c/strong\u003e: methodology, investigation, writing-original draft, resources.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAbourachid A (2003) A new way of analysing symmetrical and asymmetrical gaits in quadrupeds. C R Biol 326:625\u0026ndash;630. https://doi.org/10.1016/S1631-0691(03)00170-7\u003c/li\u003e\n \u003cli\u003eAkoglu H (2018) User\u0026rsquo;s guide to correlation coefficients. Turk J Emerg Med 18:91\u0026ndash;93. https://doi.org/10.1016/j.tjem.2018.08.001\u003c/li\u003e\n \u003cli\u003eAlexander RMcN, Jayes AS (1983) A dynamic similarity hypothesis for the gaits of quadrupedal mammals. J Zool 201:135\u0026ndash;152. https://doi.org/10.1111/j.1469-7998.1983.tb04266.x\u003c/li\u003e\n \u003cli\u003eAlt T, Heinrich K, Funken J, Potthast W (2015) Lower extremity kinematics of athletics curve sprinting. J Sports Sci 33:552\u0026ndash;560. https://doi.org/10.1080/02640414.2014.960881\u003c/li\u003e\n \u003cli\u003eAppelgrein C, Glyde M, Hosgood G, et al (2018) Reduction of the A-Frame Angle of Incline does not Change the Maximum Carpal Joint Extension Angle in Agility Dogs Entering the A-Frame. Veterinary and Comparative Orthopaedics and Traumatology 31:077\u0026ndash;082. https://doi.org/10.3415/VCOT-17-04-0049\u003c/li\u003e\n \u003cli\u003eAppelgrein C, Glyde M, Hosgood G, et al (2019) Kinetic Gait Analysis of Agility Dogs Entering the A-Frame. Veterinary and Comparative Orthopaedics and Traumatology 32:097\u0026ndash;103. https://doi.org/10.1055/s-0038-1677492\u003c/li\u003e\n \u003cli\u003eBecker K, Lewczuk D (2022) Variability of Jump Biomechanics Between Horses of Different Age and Experience Using Commercial Inertial Measurement Unit Technology. J Equine Vet Sci 119:104146. https://doi.org/10.1016/j.jevs.2022.104146\u003c/li\u003e\n \u003cli\u003eBirch E, Boyd J, Doyle G, Pullen A (2015) The effects of altered distances between obstacles on the jump kinematics and apparent joint angulations of large agility dogs. The Veterinary Journal 204:174\u0026ndash;178. https://doi.org/10.1016/j.tvjl.2015.02.019\u003c/li\u003e\n \u003cli\u003eBlake SP, Melfi VA, Tabor GF, Wills AP (2023) Injury Risk Factors Associated With Training and Competition in Flyball Dogs. Top Companion Anim Med 53\u0026ndash;54:100774. https://doi.org/10.1016/j.tcam.2023.100774\u003c/li\u003e\n \u003cli\u003eChateau H, Camus M, Holden-Douilly L, et al (2013) Kinetics of the forelimb in horses circling on different ground surfaces at the trot. The Veterinary Journal 198:e20\u0026ndash;e26. https://doi.org/10.1016/j.tvjl.2013.09.028\u003c/li\u003e\n \u003cli\u003eChow JW, Hay JG, Wilson BD, Imel C (1984) Turning techniques of elite swimmers. J Sports Sci 2:241\u0026ndash;255. https://doi.org/10.1080/02640418408729720\u003c/li\u003e\n \u003cli\u003eClayton HM (1989) Terminology for the description of equine jumping kinematics. J Equine Vet Sci 9:341\u0026ndash;348. https://doi.org/10.1016/S0737-0806(89)80073-5\u003c/li\u003e\n \u003cli\u003eClayton HM, Barlow DA (1989) The effect of fence height and width on the limb placements of show jumping horses. J Equine Vet Sci 9:179\u0026ndash;185. https://doi.org/10.1016/S0737-0806(89)80046-2\u003c/li\u003e\n \u003cli\u003eCrevier‐Denoix N, Falala S, Holden‐Douilly L, et al (2013) Comparative kinematic analysis of the leading and trailing forelimbs of horses cantering on a turf and a synthetic surface. Equine Vet J 45:54\u0026ndash;61. https://doi.org/10.1111/evj.12160\u003c/li\u003e\n \u003cli\u003eDai J, Teng L, Zhao L, Zou H (2021) The combined analgesic effect of pregabalin and morphine in the treatment of pancreatic cancer pain, a retrospective study. Cancer Med 10:1738\u0026ndash;1744. https://doi.org/10.1002/cam4.3779\u003c/li\u003e\n \u003cli\u003eDaniels KAJ, Burn JF (2018) A simple model predicts energetically optimised jumping in dogs. Journal of Experimental Biology. https://doi.org/10.1242/jeb.167379\u003c/li\u003e\n \u003cli\u003eDiMichele JK, Pechette Markley A, Shoben A, Kieves NR (2024) Evaluation of variability in performance and paw placement patterns by dogs completing the dog walk obstacle in an agility competition. PLoS One 19:e0299592. https://doi.org/10.1371/journal.pone.0299592\u003c/li\u003e\n \u003cli\u003eEicher LD, Markley AP, Shoben A, et al (2021) Evaluation of Variability in Gait Styles Used by Dogs Completing Weave Poles in Agility Competition and Its Effect on Completion of the Obstacle. Front Vet Sci 8:. https://doi.org/10.3389/fvets.2021.761493\u003c/li\u003e\n \u003cli\u003eFercher C (2017) The Biomechanics of Movement of Horses Engaged in Jumping Over Different Obstacles in Competition and Training. J Equine Vet Sci 49:69\u0026ndash;80. https://doi.org/10.1016/j.jevs.2016.10.002\u003c/li\u003e\n \u003cli\u003eHildebrand M (1989) The Quadrupedal Gaits of Vertebrates. Bioscience 39:766\u0026ndash;775. https://doi.org/10.2307/1311182\u003c/li\u003e\n \u003cli\u003eHOBBS SJ, LICKA T, POLMAN R (2011) The difference in kinematics of horses walking, trotting and cantering on a flat and banked 10 m circle. Equine Vet J 43:686\u0026ndash;694. https://doi.org/10.1111/j.2042-3306.2010.00334.x\u003c/li\u003e\n \u003cli\u003eMaitre P, Poujol L, Lequang T, et al (2007) Jumping in dogs: concurrent assessment of four limbs with a portable electronic walkway. Comput Methods Biomech Biomed Engin 10:105\u0026ndash;106. https://doi.org/10.1080/10255840701479214\u003c/li\u003e\n \u003cli\u003eMir\u0026oacute; F, L\u0026oacute;pez P, Vilar JM, et al (2020a) Comparative Kinematic Analysis of Hurdle Clearance Technique in Dogs: A Preliminary Report. Animals 10:2405. https://doi.org/10.3390/ani10122405\u003c/li\u003e\n \u003cli\u003eMir\u0026oacute; F, L\u0026oacute;pez P, Vilar JM, et al (2020b) Comparative Kinematic Analysis of Hurdle Clearance Technique in Dogs: A Preliminary Report. Animals 10:2405. https://doi.org/10.3390/ani10122405\u003c/li\u003e\n \u003cli\u003eMontalbano C, Gamble L-J, Walden K, et al (2019) Internet Survey of Participant Demographics and Risk Factors for Injury in Flyball Dogs. Front Vet Sci 6:. https://doi.org/10.3389/fvets.2019.00391\u003c/li\u003e\n \u003cli\u003eParkes RS V., Witte TH (2015) The foot\u0026ndash;surface interaction and its impact on musculoskeletal adaptation and injury risk in the horse. Equine Vet J 47:519\u0026ndash;525. https://doi.org/10.1111/evj.12420\u003c/li\u003e\n \u003cli\u003ePfau T, Garland de Rivaz A, Brighton S, Weller R (2011) Kinetics of jump landing in agility dogs. The Veterinary Journal 190:278\u0026ndash;283. https://doi.org/10.1016/j.tvjl.2010.10.008\u003c/li\u003e\n \u003cli\u003ePinto KR, Chicoine AL, Romano LS, Otto SJG (2021) An Internet survey of risk factors for injury in North American dogs competing in flyball. Can Vet J 62:253\u0026ndash;260\u003c/li\u003e\n \u003cli\u003ePowers PNR, Harrison AJ (2000) A study on the techniques used by untrained horses during loose jumping. J Equine Vet Sci 20:845\u0026ndash;850. https://doi.org/10.1016/S0737-0806(00)80115-X\u003c/li\u003e\n \u003cli\u003eRocheleau PJ, Dycus DL, Lotsikas PJ (2023) Internet-Based Survey on Diagnosis and Treatment Recommendations for Medial Shoulder Syndrome and Instability in Dogs\u003c/li\u003e\n \u003cli\u003eSantamar\u0026iacute;a S, Bobbert MF, Back W, et al (2004) Evaluation of consistency of jumping technique in horses between the ages of 6 months and 4 years. Am J Vet Res 65:945\u0026ndash;950. https://doi.org/10.2460/ajvr.2004.65.945\u003c/li\u003e\n \u003cli\u003eSchambardt HC, Merkens HW, Vogel V, Willekens C (1993) External loads on the limbs of jumping horses at take-off and landing. Am J Vet Res 54:675\u0026ndash;80\u003c/li\u003e\n \u003cli\u003eSchilling N, Hackert R (2006) Sagittal spine movements of small therian mammals during asymmetrical gaits. Journal of Experimental Biology 209:3925\u0026ndash;3939. https://doi.org/10.1242/jeb.02400\u003c/li\u003e\n \u003cli\u003eSeegmiller JG, McCaw ST (2003) Ground Reaction Forces Among Gymnasts and Recreational Athletes in Drop Landings. J Athl Train 38:311\u0026ndash;314\u003c/li\u003e\n \u003cli\u003eSelf Davies ZT, Spence AJ, Wilson AM (2019) Ground reaction forces of overground galloping in ridden Thoroughbred racehorses. Journal of Experimental Biology 222:. https://doi.org/10.1242/jeb.204107\u003c/li\u003e\n \u003cli\u003eS\u0026ouml;hnel K, Rode C, de Lussanet MHE, et al (2020) Limb dynamics in agility jumps of beginner and advanced dogs. Journal of Experimental Biology. https://doi.org/10.1242/jeb.202119\u003c/li\u003e\n \u003cli\u003eTan H, Wilson AM (2011) Grip and limb force limits to turning performance in competition horses. Proceedings of the Royal Society B: Biological Sciences 278:2105\u0026ndash;2111. https://doi.org/10.1098/rspb.2010.2395\u003c/li\u003e\n \u003cli\u003eUsherwood JR, Wilson AM (2005) No force limit on greyhound sprint speed. Nature 438:753\u0026ndash;754. https://doi.org/10.1038/438753a\u003c/li\u003e\n \u003cli\u003eWalter RM, Carrier DR (2009) Rapid acceleration in dogs: ground forces and body posture dynamics. Journal of Experimental Biology 212:1930\u0026ndash;1939. https://doi.org/10.1242/jeb.023762\u003c/li\u003e\n \u003cli\u003eWalter RM, Carrier DR (2007) Ground forces applied by galloping dogs. Journal of Experimental Biology 210:208\u0026ndash;216. https://doi.org/10.1242/jeb.02645\u003c/li\u003e\n \u003cli\u003eWynn ML, Clemente C, Nasir AFAA, Wilson RS (2015) Running faster causes disaster: trade-offs between speed, manoeuvrability and motor control when running around corners in northern quolls ( \u003cem\u003eDasyurus hallucatus\u003c/em\u003e ). Journal of Experimental Biology 218:433\u0026ndash;439. https://doi.org/10.1242/jeb.111682\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Canine, flyball, kinematics, sports injuries","lastPublishedDoi":"10.21203/rs.3.rs-6870857/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6870857/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eStudies pertaining to flyball biomechanics are scarce despite the high frequency of injury in flyball dogs. The objective of this study was to describe the phases of the box turn and the technique employed by canine athletes during flyball runs. 181 videos of 32 dogs performing the box turn were analysed. Data observed from video evaluation included approach gait, velocity of approach, approach technique, take-off technique, sequence of limbs at contact and landing, elevation of the CoM, trunk angle during take-off, etc. Results showed a high variability inter-dogs, but an intra-dog consistence in obstacle performance. The average horizontal (v\u003csub\u003ex\u003c/sub\u003e) approach velocity was 7.27\u0026thinsp;\u0026plusmn;\u0026thinsp;0.79 m/s and the mean dimensionless velocity (Froude number) was 3.34\u0026thinsp;\u0026plusmn;\u0026thinsp;0.47. The approach speed was faster on running approach (7.37\u0026thinsp;\u0026plusmn;\u0026thinsp;0.81ms\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) than in the decelerated approach (7.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.7 ms\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), where a distinct braking phase was observed. Regarding turn style, 105 (58%) runs were completed using a angled turn, where dogs avoided impacting the box straight, whilst the remaining 75 (41.4%) runs used a direct turn style. Given the high variation in velocity and technique, we could not identify clear patterns, but the phases of the obstacle could be clearly described. We could conclude that individual dog technique is very consistent, and we described some patterns associated with the box angle. Further studies in quantitative kinematics and kinetics are needed to clarify the effects of the different box angles, used in the sport, in the musculoskeletal system.\u003c/p\u003e","manuscriptTitle":"How do dogs negotiate the flyball box?- Descriptive analysis of the box turn dynamics","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-20 05:10:26","doi":"10.21203/rs.3.rs-6870857/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c83714fe-eb0b-4f21-b80d-691dfc8db26f","owner":[],"postedDate":"June 20th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-02-08T08:54:51+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-20 05:10:26","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6870857","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6870857","identity":"rs-6870857","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}