Athletic injuries are costly both from medical and athletic performance perspectives with injury downtime of players threatening the competitive status of the team. On the active playing field, athletes must effectively and efficiently extrapolate meaningful information from their dynamic environment. Focusing visual attention on important cues can lead to effective anticipation and therefore better decisions in competition1. Distributed visual attention is a key feature of perceptual-cognitive training programs used by athletes to process sports-related visual scenes at the perceptual-cognitive level.
Perceptual-cognitive skill has been proposed as critical component of elite athlete performance and is trainable2. Perceptual-cognitive skill is the ability to locate, identify, and process environmental information so as to integrate existing knowledge and current motor capabilities to select and execute appropriate actions3. Improving perceptual-cognitive skill enhances recognition of complex human movement4 allowing athletes to integrate simultaneous environmental stimuli for increased environmental awareness5.
Three-dimensional multiple object tracking (3D-MOT) has been used to train perceptual-cognitive skill and train athletes to absorb and process complex movements and distribute attentional resources throughout the visual field6.
3D-MOT training has produced robust effects on attention, working memory, and visual information processing speed7. These effects of 3D-MOT on perceptual-cognitive skill have been hypothesized as an effective intervention to reduce athlete injury incidence potentially through increased awareness of player movement assisting athletes with avoidance of injury-threatening collisions2.
The purpose of this study was to quantify athlete injuries and the mechanisms of injury, contact versus non-contact, over time in those athletes who received 3D-MOT training compared to those athletes who did not.
We hypothesized those athletes trained in 3D-MOT would demonstrate a significantly lower injury incidence over the course of a National Collegiate Athletic Association (NCAA) competitive season in both contact and non-contact injuries compared to those athletes who did not receive 3D-MOT training.
The participants from a sample of convenience of four NCAA Division III collegiate athletic teams included male ice hockey players (n=18), female ice hockey players (n=16), male lacrosse players (n=32), and female lacrosse players (n=13) (Table 1). Players were randomly assigned to either a 3D-MOT (3D-MOT; n=38) training intervention or a control group (C; n = 41). Prior to participation all subjects signed an informed consent and all procedures were approved by the University’s Institutional Review Board.
Injury Data Collection
University athletic trainers (ATs) worked with the four participating sports teams and attended university-sanctioned practices and games. When an injury occurred, ATs evaluated the athlete and completed a detailed report through the university’s electronic health record (SportsWearOnLineTM, Computer Sports Medicine, Inc., Stoughton, MA) which allowed the medical staff to document injuries as part of their daily medical practice throughout the season. The initial injury report described date of injury, event type, injury mechanism, and injury location. After the initial injury evaluation, ATs had the ability to return to view and update the athlete’s health record as necessary. We relied on the medical expertise of the ATs or team physician to appropriately identify and treat specific diagnoses. Data were from varsity-level games and practices, including team conditioning sessions. Individual training sessions including conditioning or weight lifting, as well as off season training, were excluded.
Injury. A reportable injury was an injury that occurred as a result of participation in an organized varsity-level practice or game and required attention from an athletic trainer or team physician. Multiple injuries could be included as the result of one injury event8.
Injury Mechanism. Injury mechanism was defined as the manner in which the athlete sustained his or her injury. The mechanism of injury was provided by the athletic trainer and was classified as contact or non-contact. A contact injury was defined as an injury sustained by an athlete due to extrinsic contact to the location of injury with another player, the field, or an object on the field. A non-contact injury was defined as an injury sustained by an athlete not due to extrinsic contact to the location of injury with another player, the field, or an object on the field.
Event Type. Event type was the specific event (ie, practice, game) in which the injury was reported to have occurred8.
Athlete-Exposure. A reportable athlete-exposure (AE) was defined as one athlete participating in one NCAA sanctioned practice or game in which he or she was exposed to the possibility of athletic injury, regardless of the time associated with that participation. Only athletes with actual playing time in games were included in game exposures8.
Pre-Elevated Baseline. The pre-elevated baseline time period was defined as the time from the beginning of athletes’ competitive season through completion of the 15th 3D-MOT training session.
Post-Elevated Baseline. The post-elevated baseline time period was defined as the time from the completion of 15th 3D-MOT training session until the conclusion of the athletes’ season.
Three-Dimensional Multiple Object Tracking (3D-MOT) Training
Subjects completed 25 sessions of 3D-MOT training using the NeuroTracker (CogniSens Athletics Inc., Montreal, CN) system. This training was administered 1-4 times per week over 12 weeks during the athletes’ respective competitive seasons. In cases where participants were unable to complete two training sessions within one week they were allowed to complete additional sessions in following weeks for up to four sessions per week.
The 3D-MOT software was displayed on a 65 inch 3D television (Sony Inc., XBR65X930D) positioned at each athlete’s eye level. The subjects wore active 3D glasses (Sony Inc., TDG-BT500A) and were positioned 173 cm away from the screen for each training session. One training session consisted of 20 3D-MOT trials as described by Parsons et al7. At the beginning of each trial, eight spheres were presented to the participant within a three-dimensional cube. Four spheres were identified for tracking (targeted spheres) and then blended with another four distractors restoring homogeneity of the objects. At the center of the visual field a fixation spot was illustrated and participants were instructed to focus on the fixation spot while tracking the four targeted spheres. The eight spheres randomly moved in a linear trajectory through the true 3D space for 6-8 seconds dependent on the participants previously measured selective attention span, crossing over each other and deflecting following collisions with another sphere or boundaries of the virtual space. At the end of the trial, all eight spheres stopped and were labeled one through eight for identification of the targets. The participant verbally reported the numbers correlating to the target spheres to an experimenter who recorded the answers on a keyboard. Following identification, feedback was given regarding the number of correct target spheres identified by highlighting the original target spheres. If all four target spheres were correctly identified, the program increased the speed of the spheres’ movements for the next trial; however, if targets were not all correctly identified, the program reduced the speed. The speed changes were based on an adaptive staircase in which initial trial speeds varied more widely than later trial speeds ensuring the participant’s optimal training zone was quickly achieved and maintained7.
Participant’s 3D-MOT training gains were measured by speed threshold (m/s) obtained from core training sessions. The first 15 3D-MOT training sessions of the protocol were conducted with the athlete sitting. Subjects see a rapid increase in speed threshold when sitting during 3D-MOT training however, subjects see a less dramatic increase in speed threshold when the training is conducted with the subject standing, suggesting speed thresholds are directly affected by whether a subject is tested standing up2. Sessions 16 through 20 were conducted with the subject standing providing a more challenging training situation. Sessions 23 through 28 were conducted with the subjects standing while the ice hockey athletes stick handled and lacrosse athletes cradled, further increasing the complexity of the 3D-MOT task. The 3D-MOT training protocol is described in Table 2.
Completion of 15 3D-MOT training sessions has shown relative speed threshold gains over initial baseline of approximately 40% allowing an athlete to reach an elevated baseline, a training level by which an athlete has made considerable gains in speed threshold over their initial baseline (e.g., +40%)2. Athletic injuries were coded as having occurred during the pre-elevated baseline time period or during the post-elevated baseline time period for investigation into the efficacy of 3D-MOT as an intervention for injury incidence reduction. The pre-elevated baseline time period included all injuries sustained from the beginning of the competitive season through completion of the athlete’s 15th 3D-MOT training session. The post-elevated baseline time period included all injuries from the completion of the athlete’s 15th training session until the conclusion of the season.
We first analyzed participants’ mean 3D-MOT core session speed thresholds using paired samples t-tests. We compared the athletes’ first mean core session speed threshold (session 1) with the last mean core session of the pre-elevated baseline in which the athletes were sitting (session 15). We then compared the athletes’ first mean core session (session1) speed threshold with the mean core session during the protocol in which the athletes were standing (session 16), and then compared the athletes’ first mean core session speed threshold (session 1) with their last mean core session speed threshold of the protocol in which the athletes were standing and stick handling or cradling (session 21).
Injury data were analyzed using mixed model repeated measures analysis of variance (RM-ANOVA) to determine whether total number of injuries changed over time and between groups. Time was the within subjects factor and group (3D-MOT, Control) was the between factors. We performed two additional RM-ANOVA to explore whether the intervention decreased the number of injuries over time based on injury mechanism (i.e., contact, non-contact). Lastly, a RM-ANOVA was run on only those athletes who had an injury during the pre-elevated baseline time period to test if those athletes who were injured reduced the number of injuries over time and if injuries differed between the groups (3D-MOT, Control). All analyses were performed using SPSS statistical software. The level of significance was set at p ≤ 0.05. All 95% confidence intervals (CIs) that did not include 1.00 were considered different.
The athletes trained in 3D-MOT significantly improved their mean speed threshold. The athletes’ mean speed threshold significantly improved from the first core session (session 1) to the last core session of the pre-elevated baseline in which the athletes were sitting (session 15) (p=.000). The mean speed threshold significantly improved from the first core session (session 1) to the core session in which the athletes were standing (session 16) (p=.000) and significantly improved from the first core session (session 1) to the last core session of the protocol in which the athletes were standing and stick handling or ball cradling (session 21) (p=.000). Mean 3D-MOT speed thresholds are illustrated in Figure 1.
The total number of injuries significantly decreased over time (length of season) (p=0.002). When comparing the 3D-MOT intervention group to the control group, there was no significant difference in total number of injuries (p=0.293). When examining group by time (length of season), there was no significant interaction for total number of injuries (p=0.221). For those athletes injured during the pre-elevated baseline time period, the total number of injuries also significantly decreased over time (length of season) (p=0.002). When comparing the groups (3D-MOT, Control), there was no significant difference in number of injuries (p=0.204) of those injured during the pre-elevated baseline time period. When examining group by time (length of season), there was no significant interaction for total number of injuries of those injured at baseline (p=0.111).
The total number of non-contact injuries significantly decreased over time (length of season) (p=0.013). When comparing the 3D-MOT group and the control group, there was no significant difference in the number of non-contact injuries (p=0.125). When examining group by time (length of season), there was no significant interaction for total number of non-contact injuries (p=0.452). Overall, the total number of contact injuries significantly decreased over time (p=0.016). When comparing the groups (3D-MOT, Control), there was no significant difference in the number of contact injuries (p=.962). When examining group by time (length of season), there was no significant interaction for total number of contact injuries (p=0.225). Descriptive data for all analyses are illustrated in Table 3.
The aim of the present study was to compare the injury incidence of NCAA Division III men’s and women’s ice hockey and lacrosse athletes trained in 3D-MOT to that of a control group. Our secondary aim was to explore whether 3D-MOT training influenced injury mechanism (i.e., contact, non-contact). To our knowledge, this investigation is the first to examine changes in injury incidence over time in collegiate athletes trained in 3D-MOT using NeuroTracker, further exploring whether perceptual-cognitive training reduces contact and non-contact injury incidence over the course of a competitive season.
Peripheral vision training has allowed athletes to decrease their response time necessary to see and respond to stimuli presented in their peripheral visual field9. Faubert and Sidebottom2 hypothesized that perceptual-cognitive training programs increase awareness of player movement in the peripheral visual field to assist athletes in avoidance of injury threatening situations. We found that all athletes, regardless of the 3D-MOT training intervention, experienced fewer injuries as their respective seasons progressed, and that regardless of injury mechanism, both contact and noncontact injuries decreased over the length of the athletes’ seasons.
Both 3D-MOT trained athletes and untrained athletes experienced a decrease in total number of injuries over the length of the season, with no effect by group. A decrease in injury incidence over the length of an athletic season was similarly found in a 16-year analysis of injury rates for 15 collegiate NCAA sports including men’s and women’s ice hockey and lacrosse. This 16-year analysis determined preseason injury rates (6.6 injuries per 1000 A-Es) were found to be higher than both in-season (2.3 injuries per 1000 A-Es) and post season (1.4 injuries per 1000 A-Es) injury rates10.
Improved sport specific conditioning may have created a protective factor mitigating the risk of injuries as the athletes improved their physical conditioning over the course of their competitive seasons. At the beginning of the season, athletes may be poorly conditioned placing them under higher stress during practice and competition, potentially resulting in an excess of injuries10. Over time, it is likely that in-season conditioning and training exert a large enough workload on an athlete so that favorable physiological changes occur. This may result in fewer injuries occurring later in an athlete’s competitive season due to the potential for improved sports specific conditioning generating a protective effect against injury. Accordingly, our study found that significantly more injuries occurred earlier in the season than later in the season.
Effective cognitive abilities including psychomotor networks and spatial awareness are imperative for sport performance11. Adept visual information processing can provide vital sport performance advantages minimizing athlete response times and enhanced decision-making2, 12. Game intelligence has been linked to various cognitive abilities including pattern recall and recognition, advanced visual cue utilization, and visual search strategies13. Obtained in sport-specific training, learned sequential procedures have shown to improve visual search strategy, a highly demanding cognitive task14. Repetitive training, such that occurs during athletic practices and competitions, may induce neuroplasticity of the prefrontal cortex and lead to improved efficiency of information processing and multitask performance11, 15. The training of various perceptual cognitive skills during regular sport participation over the course of a competitive season may have caused the decreased injury incidence over time due to the improved psychomotor activity and spatial awareness obtained by athletes.
In collegiate athletics, there is a serious need for effective interventions to reduce and prevent athlete injuries. Sport vision training programs are developed under the logic that practicing challenging visual, perceptual, and sensorimotor tasks will improve vision leading to faster responses and improved athletic performance while potentially reducing the risk of athlete injury16. Sport vision training often isolate certain components of visual performance16. 3D-MOT training may train an athlete’s ability to maintain attentional focus on a subgroup of objects simultaneously moving through a three-dimensional environment13, a skill essential to an athlete’s anticipatory response17 and previous research17, 18 indicates 3D-MOT training with NeuroTracker can transfer to sport performance abilities19. Though our study did not find our 3D-MOT training protocol to be effective in reducing athletic injury incidence, more research is needed to refine 3D-MOT training methods to determine the extent to which improved perceptual-cognitive skill can reduce injuries. Conclusion:
Motion perception training with 3D-MOT did not decrease injury incidence in NCAA Division III men’s and women’s ice hockey and lacrosse athletes. Collegiate NCAA Division III men’s and women’s ice hockey and lacrosse athletes experience fewer injuries as their seasons’ progress. Additionally, contact and non-contact injuries decrease over the course of a collegiate NCAA Division III hockey and lacrosse season. Additional research is needed to refine 3D-MOT training methods to determine the extent to which improved perceptual-cognitive skill can contribute to injury risk reduction.