Abstract
Domain-specific knowledge guides our attention and thus influences our perception. Prior change-blindness research has shown that expert athletes can spot meaningful scene changes more quickly than novices, but less is known of whether this expertise is modulated differentially between open and closed sporting activities. We presented 81 individuals (20 gymnasts, 19 rock climbers, 22 parkour practitioners, and 20 control participants) with alternating sequences of images that corresponded to the habitual training landscapes of each group (gymnasiums, rock cliffs, and urban environments, respectively). We included contextual and non-contextual scenic changes to evaluate whether athletes were generally aware of their environments, or whether their observation strategies only targeted sport-related environmental elements. Among these three athletic endeavors, we found that gymnasts were faster at detecting changes in their environment, irrespective of whether or not these changes were contextual to the sports involved. Expert rock climbers presented a domain-specific expertise that was improved even further for contextual changes. Parkour practitioners presented the fastest reaction times in the urban environment and some of the best reaction times for all types of changes. These results confirm that an ability to read the environment is an integral aspect of practice in open-skilled sports, while skills of athletes in closed-skilled sports are more closely related to motor skill repetitions in constant environments. Thus, open skill training may benefit athletes’ guidance of attention. Our finding that parkour practitioners appeared to have developed the widest perceptual abilities was probably linked to these athletes’ extremely wide range of practice environments and with the constant demands of this sport to find solutions in random natural environments that that are not purposely designed for the sport.
Introduction
It is commonly accepted that sport activities shape not only physical but also cognitive skills. Similarly, regular physical activity can sharpen our perception and attention. Apart from the benefits of physical exercise for improved brain functioning (Crush & Loprinzi, 2017), sport activities have particular cognitive benefits due to the importance of perception demands and motor strategies they involve (e.g., Alesi et al., 2016). Regarding observational skills particularly, researchers have shown that athletes exhibited better recognition of actions related to their motor experience than actions not previously experienced (Bläsing & Sauzet, 2018). Better observation abilities regarding pictures of football game situations were also observed among professional footballers as compared to untrained individuals (Werner & Thies, 2000). Naturally, these better performances have been associated with the training effects of identifying and analyzing situation games. These studies raise the question of whether sport practice leads only to improved perception skills for sport-related actions, or whether improved perception extends to inanimate pictures of the athlete’s general surrounding contextual environment.
On this matter, it can be argued that environment perception has different importance to the performance of athletes of varying sport types. Indeed, some athletes evolve their skills from within stable or constant environments designed for their sport, and, in these instances, the athlete’s goal is to develop and magnify a defined set of specific complex skills. It has been well established, for more than half a century, that these stable sport conditions are best for performing planned motor sequences that do not require recalibration to changing environments (Poulton, 1957). This environmental stability can apply to sport activities with closed skills such as track and field or gymnastics disciplines. On the other hand, other sports activities may involve varying environments that require more of the athlete’s adaptation capacity. One of the best examples of disciplines of this kind, requiring accurate vision and the ability to read environmental nuances for high sport performance is rock climbing. Expert climbers must combine complex motor skills with accurate and adaptive environmental perception (Whitaker et al., 2019).
An emerging physical activity or sport over recent decades has been linked to climbing within an urban environment: parkour. In fact, parkour does not involve much climbing movement but is, rather, a plyometric activity that evolves on a horizontal plane by running and jumping to overcome obstacles. But this sport shares with rock climbing a constant visual search for new environmental features that can be incorporated into the athlete’s performance. Indeed, one of the basic elements of parkour is to explore an infinite variety of environments that favor the athlete’s faculty of adapting to every obstacle. Therefore, parkour can help develop particular skills associated with reading the environment, similarly to rock climbing. Parkour has been shown to develop specific muscular performances as other activities considered to have similar muscular demands, such as track and field or gymnastics (Grosprêtre & Lepers, 2016). These sport demands have lead traceurs (parkour practitioners) to exhibit a specific neuromuscular profile (Grosprêtre et al., 2018), and these physical adaptations have been attributed, in turn, to the particularity of the environment in which traceurs evolve their skills. Thus, traceurs should exhibit high perceptual capacity for their general environment, similar to that of climbers, and, in fact, previous research has shown that experience in parkour can modify traceurs’ ability to perceive wall height accurately (Taylor et al., 2011).
An experimental paradigm commonly used to assess the way observers perceive and encode visual material consists of altering an aspect of previously seen visual images. When participants are asked to notice a change that has occurred between an original and a modified image, they often fail to identify the change. The inability to report the change instantaneously is called “change blindness.” It particularly occurs when a blank picture is inserted between the two images (R. A. Rensink et al., 1997). The ability to detect the change is known to depend on the extent to which the changed detail of the image is meaningful to the observer (Simons, 1996) and thus can capture their attention. The selection of meaningful elements within an image depends on the individual characteristics of the observer, such as their expertise in sport. For example, the change blindness paradigm has been used to compare perceptual capacity in American football players according to their expertise, by presenting them alternating sequences of football-related images (Werner & Thies, 2000). Werner and Thies (2000) altered the pairs of images in two ways: by modifying an aspect of the displayed football action (semantic change) or by modifying a part of the picture irrelevant to the displayed action (non-semantic change). They found that expert athletes were faster than novices to detect the changes in each football image, and athletes were even faster when presented with semantic changes. Change detection times were similar between novices and expert football players for neutral images (urban traffic scenes).
Various research studies suggest that sports experience guides the allocation of cognitive resources towards the meaningful parts of a visual scene, but to what extent this improved resource allocation is sports domain-specific, especially when sports share common characteristics, remains to be explored. Therefore, the aim of the present study was to assess whether athletes’ observations and perceptions of the environment were specific to the usual environment in which they train or might generalize to other environments. We approached this question by comparing groups of athletes from different sport activities: (a) rock climbers, (b) traceurs, and (c) gymnasts. We tested these athletes and a fourth group of non-athletes (control recreational participants) in a change blindness paradigm. We presented all participants with environmental images from the same three types of environments that corresponded to the habitual training landscapes of the athletes’ three sports groups (i.e., rock cliffs, an urban environment and a gymnasium, respectively). According to prior findings of domain-specific expertise in detecting environmental changes, we hypothesized that each group would be faster in detecting changes in images of their sport’s respective practice environment. By comparing the various athletes’ perception times with those of a control group of recreationally active participants, we aimed to determine whether sport practice would lead to a better reading of the environment than would recreational physical activity; and, by comparing these athletic groups to each other, we sought to determine whether any advantage of sport expertise would be specific to the training environment of the athlete’s sport. We also implemented a secondary parameter of change with regard to the contextual aspect of the environment. Contextual changes were changes to those parts of the image that related to the corresponding sport practice (e.g., altering an environmental aspect used to perform a sport movement). Non-contextual environmental changes involved only parts of the background scene. By comparing the participants’ responses to contextual/non-contextual changes, we sought to provide clues of whether the athletes were generally aware of their environment or whether their observation strategies only targeted sport-related elements. Comparisons of athletes from different sport activities permitted a comparison of the relationships athletes develop with their practice environment in sports with closed ability (gymnastics) and sports with open ability (parkour and climbing).
Method
Participants
Eighty-one young healthy participants took part in the present experiment. None of them reported having any neurological or cognitive difficulties or any visual deficiency that would prevent their ability to use computer-based programs. Each participant reported having normal vision (natural or corrected) with no alteration in color vision (no color blindness). After being fully informed about the investigation, all participants gave written informed consent to participate in the study. The research protocol was approved by the institutional ethics committee of the university, and it was performed in accordance with the latest version of the Declaration of Helsinki.
Each participant completed a survey on which they indicated their current and past sport experience. This survey was completed after participants completed the experimental tasks so as not to influence the participants’ experimental results with this question. This survey allowed us to subdivide the 81 participants into the four comparison groups: a recreationally active control group (CON), gymnasts (GYM), rock climbers (RC), and traceurs (i.e., parkour practitioners or PK). The CON group was composed of 20 individuals (M age = 26.9 years, SD = 7.9; seven females) who engaged in no significant amount of prior or current physical activity (M hours/week =1.2, SD = 1.4), and no particular experience in either parkour, gymnastics or rock climbing. The GYM group was composed of 19 individuals (Mage = 26.5 years, SD = 3.6; six females) who practiced one or several gymnastics routines (floor, rings, bars, etc.) at various levels from local to regional with significant experience (M hours/week = 7.1, SD = 4.2 for M years = 7.9, SD = 3.4). None of the GYM participants reported current or past experience in either parkour or rock climbing. The RC group was composed of 20 individuals (M age = 28.2 years, SD = 5.3; four females) who practiced rock climbing (M hours/week = 5.9, SD = 4.1 for M years = 7.5, SD = 2.4). All RC participants reported practicing climbing outdoors on natural cliffs at least five times per year, regularly practicing indoors on artificial walls, and having no or very negligible experience in parkour. The PK group was composed of 22 individuals (M age = 26.3 years, SD = 6.1; four females) who regularly practiced parkour (M hours/week = 6.7, SD =3.5 for M years = 8.6, SD = 4.5). All PK participants reported practicing parkour outdoors in urban environments, and 40% of them reported practicing indoors less than 20% of the time. When indoors, traceurs reported practicing in a parkour-specific room, with different equipment than the equipment used in gymnastics or rock climbing. Two out of 22 participants reported also practicing rock climbing, but for less than two hours/week for one and four years. Some traceurs reported to frequent gymnasiums three to five times a year, which was considered negligible.
Stimulus Protocol
To assess the participants’ perceptual capacity for detecting environmental changes, we employed, the change blindness paradigm (Loussouarn et al., 2011; R. Rensink, 2009; Werner & Thies, 2000). This task consists of detecting small changes between two apparently identical images displayed alternatively at high frequencies. Each of these image pairs represents the same picture, but with the two images distinguished by a single modifying detail as manipulated by the software program, Photoshop CS7 (Adobe, San Jose, Costa Rica). Each image was 1400 pixels wide and 1000 pixels high. We randomly distributed the spatial distributions of the image changes for each picture, and we calibrated area of the image change to be between 10% and 15% of the total image area. The presentation of the stimuli and the collection of the participants’ responses were achieved using E-prime 2 software (Psychology Software Tools, Pittsburgh, USA).
We created a total of 84 color image pairs, from pictures of the several different target environments available on the internet. Pictures were from foreign countries and were not taken from habitual or well-known practice places of the participants so as to avoid familiar environments. Each pair of images contained only one changed detail and was displayed once. We created four categories of images. The first category, hereafter identified as NEUTRAL, was composed of 12 photographs of various indoor and outdoor scenes not related to any sport practice (garden, beach, etc.). The second category, GYM, corresponded to a gymnastics environment, and was composed of 24 pictures of gymnastics gymnasiums. The third category, RC, was composed of 24 rock climbing images composed of either natural or artificial cliffs (half of each). The fourth category, PK, was composed of 24 parkour images created from large plan photographs of urban environments, considered particularly appropriate for parkour practice. Figure 1 gives examples of image pairs used in each condition.
Examples of Pairs of Images in Each Sport Condition.
Apart from the NEUTRAL images, we displayed two types of changes for the sport image pairs. First, for contextualized change (CT), in each of the sport image categories (GYM, PK, RC), half of the changes (12 of 24) were intended to alter an environmental element that related to the corresponding sport practice. In GYM images, we changed one of the pieces of gymnastics equipment that was visible in the picture (mat, bar, beam, etc.). In RC images, we changed one of the climbing holds (artificial or natural). In PK images, we changed an urban element that could serve as an obstacle (bench, wall, barrier, etc.). Second, for non-contextualized change (NCT), across all categories, the other half of the manipulated changes were to picture elements deemed irrelevant to the practice of each sport. For indoor pictures (GYM and RC), this could be represented by a poster on a wall, or detail in the room itself, such as a change in one ceiling lamp or the color of a door. For outdoor pictures (RC and PK), changes concerned elements of the environment that were considered irrelevant to the sport practice (a bush, a car, etc.) or elements of the global scenery (a cloud, a shadow, etc.). For CT and NCT images, we took particular care to avoid overcrowding images, choosing environments containing empty places or with as few people or animals as possible.
Procedure
We tested each participant in a single session of 40 to 60 minutes, individually, and in a quiet room. We used the same computer with a 17’ wide screen for all participants. Participants sat comfortably in front of the computer and were informed that they were to perform computer-based testing, consisting of detecting changes in pairs of pictures in a series as fast as possible. We specified that each pair of images displayed a single change that might appear in any part of the picture. Each trial consisted of displaying the two images of the same pair (the original and the modified one) in an alternating manner during a maximum period of 40 seconds. Each image of each pair was displayed for 240 milliseconds, interspersed with a gray mask lasting 120 milliseconds (see Figure 2). This flickering paradigm avoided motion transients and caused the “change blindness” phenomenon.
Design of Each Trial of the Change Blindness Task.
Participants were instructed to press the space bar immediately once they had detected a change. A grid then appeared on the picture, dividing it into nine identical sub-squares numbered from 0 to 8 (see Figure 2). Participants then had to press the key number corresponding to the sub-square where the change appeared. The next trial began automatically after this response, with a 500 ms pause. If no change was detected during the 40 seconds of image display time (i.e., the participant did not press the space bar), the next trial began automatically without the grid appearance or response to sub-squares.
For each participant group, we first presented a training program consisting of twelve image pairs of the NEUTRAL condition. Next, we randomly presented pairs of each sport type of image (PK/RC/GYM, with each of these sets divided into contextual and non-contextual pairs (CT/NCT). Since the total number of sport image pairs was substantial (72), following the initial training program (12 pairs) and at 1/3 and at 2/3 of the way through the pairs of sport images, participants were allowed a break (i.e., at pair numbers 24 and 48). At each break, a black mask appeared and informed participants that they could rest and press the space bar when they wanted to start again (maximum rest time: five minutes).
Data Analysis
We recorded the search time (i.e., the time between the onset of the changed image and the participant’s space bar press) for each participant. If no change was detected within the 40 seconds allowed to find the change, the trial was considered unsuccessful, and the number of unsuccessful trials was also analyzed. When the participant pressed the wrong number of the sub-square to locate the changed image, the trial was considered “false” and was not analyzed. Only a very small number of trials reflected such errors (less than two per participant across all conditions).
For each participant, we separately averaged the search times for all successful trials for the twelve images of each sport and contextual condition and considered these as dependent measures. We took the number of unsuccessful trials into consideration. We presented all data as means (M) and standard deviations (SD.). We verified the normality and the homogeneity of the data distribution with the Shapiro-Wilk and Levene tests, respectively. We performed a separate data analysis for the search time among the 12 NEUTRAL image pairs and the number of unsuccessful trials with a one-way ANOVA with the factor “group” (CONTROL, PK, GYM, RC). For sport-related images, we used a three-way ANOVA with the factors “group” (CONTROL, PK, GYM, RC), “type of environment” (Gymnasium, Urban, Rock Cliff) and “context” (contextual, non-contextual). Any significant main effect or interaction was followed up with a post-hoc Tukey’s t-test. We performed all statistical analyses using STATISTICA software (8.0 version, Statsoft, Tulsa, Okhlaoma, USA), and we set the level of significance for all inference tests at p < 0.05.
Results
Athletes’ Global Change Detection Performances
A summary of the search times and number of unsuccessful trials results for all sports types and all environments is presented Table 1.
Change Blindness Search Times (ST) and Number of Unsuccessful Trials (UN).
Data (means and standard deviations) are displayed in seconds for ST. The four groups are shown in columns (parkour, rock climbing, gymnastics, and control) and the different conditions and types of images are displayed in lines. CT: contextual changes. NCT: non-contextual changes.
First, we investigated whether some athletes showed higher change detection performances, irrespective of the sport specificity of the environment. We found main group differences on search times, when taking into account the whole sport environment (F(3,76)=16.13, p < 0.001). PK athletes’ search times were faster than those of CON participants (p < 0.001) and GYM participants (p < 0.001). RC participants were faster in detecting environmental changes than CON participants (p < 0.001), and GYM participants (p = 0.029). Regarding unsuccessful trials, PK participants exhibited a lower number of unsuccessful trials than CON participants (p = 0.043), RC participants (p = 0.008) and GYM participants (p = 0.036).
We found group differences in search times when we compared participants on the first twelve NEU training images, (F(3,76)=3.62, p = 0.017). Post hoc comparisons revealed a significantly lower search time for PK participants (p = 0.046) and RC participants (p = 0.036) compared to CON participants (see Figure 3A). No significant participant group differences were found regarding the number of unsuccessful NEU trials (F(3,76)=0.11, p = 0.955; see Figure 3B).
Results of the First Phase of the Experiment Regarding Neutral Images.
Sports Domain-Specific Change Detection
Regarding our three way ANOVA test of whether athletes in one sport showed domain-specific environmental perception skills versus generalized environmental perception skills, we found a significant “type of environment” × “group” interaction (F(6,152)=7.51; p < 0.001) that reflected a domain-specific expertise for RC and GYM participants. RC participants exhibited a significantly lower search time in environmental scenes of cliffs compared to images of gymnasiums (p = 0.035) and urban images (p = 0.004). In the GYM group, there was a significantly faster search time for gymnasium as compared to urban images (p < 0.001) and cliff images (p = 0.003). There was no domain-specific advantage for PK and CON participants, who showed similar search times in all environments (see Figure 4).
Search Time for the Four Groups in the Several Types of Image.
Regarding unsuccessful trials, the PK group made fewer errors when presented with urban versus gymnasium environments (p = 0.044). The RC group made fewer unsuccessful trials when presented with cliff than gymnasium images (p = 0.046) and urban images (p = 0.023). There were no domain-specific unsuccessful trial differences between CON and GYM participants (see Figure 4).
Regarding tests of whether changes in one type of environment were more easily detected in the related sport activity, we found a search time “type of environment” × “group” interaction effect, (F(6,152)=7.51; p < 0.001) showing that for urban environment images, PK group search times were faster than CON group times (p < 0.001), RC group times (p = 0.046) and GYM group times (p < 0.001). For cliff environment, RC participants’ search time was faster than that of PK (p = 0.021), CON (p < 0.001) and GYM participants (p < 0.001). Neither GYM participants nor CON participants showed faster search times than the other groups.
Regarding unsuccessful trials, there was a significant sport group by image effect (F(6,152)=2.857; p = 0.012) such that the number of unsuccessful trials to urban images was lower among PK participants than among CON (p = 0.002), GYM (p = 0.001) and RC participants (p = 0.009). Numbers of unsuccessful trials to cliff images were lower for the RC group compared to the CON (p = 0.001), PK (p = 0.006), GYM (p = 0.003) groups (see Figure 4). Again, GYM and CON groups did not exhibit lower unsuccessful trials than the other groups.
Detection of Contextual Versus Noncontextual Changes
Regarding sport practice relationships to participant detection of contextual changes relevant to the domain-specific sport activity, we found a significant “type of environment” × “group” × “context” interaction (F(6,152)=2.30, p = 0.037). Interestingly, only the RC and PK groups showed faster detections of contextual versus non-contextual changes. The RC group had a significantly lower search time (p = 0.035) for contextual changes as compared to non-contextual changes of images of cliffs (see Figure 5), and the PK group was faster for detecting contextual versus noncontextual changes (p = 0.031) for urban images (see Figure 5). No contextual versus non-contextual difference was evident for the GYM group with any type of environment.
Search Times and Unsuccessful Trials for the Four Groups in Their Domain-Specific Type of Image.
We also found a significant “type of environment” × “group” × “context” interaction for number of unsuccessful trials (F(6,152)=3.935, p = 0.001) such that PK participants exhibited significantly fewer unsuccessful trials with contextual as compared to non-contextual changes to images of urban environments (p = 0.009), and RC participants showed the same effect for contextual changes to cliff images (p = 007). There were no contextual versus non-contextual change detection differences among GYM participants for any environment images.
Discussion
The goal of the present study was to explore the specificity of observation capacity among athlete participants in different sports. Although parkour, gymnastics and rock climbing are sports activities that share similar characteristics, we found that participant search times for detecting changes in images of various environments revealed very distinct performance patterns. Gymnasts were faster at detecting changes in images of their gymnasium training environment, irrespective of whether these image changes were contextual to gymnastics (i.e., gymnastics equipment) or involved non-contextual changes in the images (i.e., background scene). Expert rock climbers also presented a domain-specific advantage in their perceptions in that they showed faster change detection searches for contextual changes of direct relevance to rock climbing. Parkour practitioners presented the fastest reaction times for detecting changes in urban images, and these participants also showed faster search times for changes in all types of images.
Domain-Specific Expertise
All athlete groups showed an advantage in detecting changes in images of sport domain-specific environments. Familiarity with their own sports were associated with stronger perceptual skills for detecting image changes that were relevant to their sport, whereas other changes to images of other environments remained unaffected by their sport experience. This finding is consistent with a previous change blindness study on football players that showed a benefit of sport expertise only for domain-related image changes (Werner & Thies, 2000). Of added importance, however, athletes from different sport groups processed visual information differently. Rock climbers and traceurs more rapidly detected changes in images relevant to rock climbing and urban environments, consistent with past literature on the role of sport expertise in visual perception (e.g., Allard & Burnett, 1985; Chase & Simon, 1973; Yarbus, 1967). Gymnasts processed visual gymnasium changes faster than athletes from other sports whether or not the image changes were directly relevant to gymnastics. Several hypotheses may explain the absence of any contextual effect for this group. First, gymnasts spend a lot of time in the same environment, since all gymnastic disciplines are standardized with regard to equipment and gymnasium organization. Consequently, even if some gymnasts practice in different gymnasiums, the global organization of their environment always proceeds similarly, resulting in both general familiarity with gymnasiums and relative unfamiliarity with the high degree of variation that occurs in other types of environments (urban, natural, etc.). Second, given that attention guidance is not a predominant skill that gymnasts need to achieve good performances, they may not be particularly attentive to visual processing of other environments. Gymnasts are known to rely more on kinesthetic than visual information to recall their performances (Ille & Cadopi, 1999).
General Capacity of Observation
Traceurs and climbers showed shorter search times for detecting changes in images than control and gymnast participants, both for neutral images and for images generally. We would argue that both parkour and rock climbing are sports with open abilities, for which accurately reading the sport environment is integral to practice and performance, while gymnastics is a sport with closed abilities, more directly related to repeating the same movement across environments. Gymnasts were only faster than control participants who engaged in recreational activity but no specific support, and this faster search process was only true for detecting changes in images of their specific gymnastics environment. Thus, closed-skill sport practice does not tend to improve general arousal of visual perception ability, but is rather specific to the closed skill practice environment. On the contrary, practice with open skill sports for which environmental changes are constant requires that athletes continually adapt movement to the environmental context seemingly enhancing these athletes’ capacity to detect environmental details and transfer this skill to a variety of environments.
Of note, however, traceurs did not exhibit better image change detection in their usual environments (urban) than in other environments. Although some parkour practitioners may frequently practice gymnastics and rock climbing, this was not the case in the present study (Only two of our parkour participants reported doing so). Thus, these results suggest another hypothesis related to the parkour way of training. Traceurs must attend to all aspects of their environment in order to learn how to identify the best places for practice; they must constantly visually and search for the most appropriate urban environment elements. In contrast to rock climbing and gymnastics, there is no pre-identified practice site for parkour. Rather, it can be practiced anywhere, and they must discover new environments and not base their performance on memories of past places of practices. This constant environmental variation and a multiplicity of practice environments characterize parkour training and may explain why these practitioners did not exhibit faster performances in urban versus other types of environments.
Sports Activities and Affordances
Perception of one’s environment depends on the ability of the observer to act within it ( Witt et al., 2011; Witt, Sugovic, & Taylor, 2012 ). James Gibson (1979) conceptualized this phenomenon as “affordance.” Affordances define the number of actions that an object or an environment permits an individual, and the more a person evolves and interacts in an environment the more affordances the person experiences and the greater the person’s environmental perceptions. This idea is not limited to sports fields. For example, experienced drivers and urban cyclists have been found to have better and faster recognition of road signs than novice drivers (Beanland & Hansen, 2017).
Past researchers showed that expert rock climbers have better visual memory of their holds in the climbing environment when they have had a larger number of affordances (Boschker et al., 2002; Pezzulo et al., 2010). The contextual changes to rock climbing images in the present study can be a marker for these affordances. Interestingly, each group showed faster change detection times and fewer unsuccessful trials when image changes were contextual versus non-contextual only for their own training environment. Traceurs and climbers exhibited faster change detection times and fewer unsuccessful trials in detecting changes in urban and rock cliff images, respectively, again suggesting that environmental affordances play a major role in an observer’s change detection strategy, as the observer becomes more vigilant toward familiar environmental elements. However, in our study, gymnasts were engaged in more closed skills and in more constant environmental elements with fewer affordances. Traceurs, however, may have been influenced by what William Gaver (1991) defined as “hidden affordances”—the possibility of actions an object offers that are not the object’s initial and explicit functionality. As parkour training leads its practitioner to learn how to identify climbing and jumping possibilities among nearly every material a surrounding environment may offer, traceurs may develop a keen sense of visual observation that helps explain why our parkour group was the only group to demonstrate a faster image change detection time than controls in all types of environments.
Conclusion
The present study investigated the observation capacity of several types of athletes. Beyond a global athlete advantage for quickly scanning meaningful scenes, it appears that the type of an athlete’s sport influences these perceptual skills. We found open-skill parkour and rock climbing to be activities that are well suited for developing improved environmental perception largely because of training in highly variable environments. By contrast, gymnastics, a more closed-skill sport that relies on optimizing a specific movement with standardized gymnasium equipment did not seem to be associated with the ability to transfer enhanced domain-specific environmental perception to other environments.
Finally, as a future perspective, eye movement measures could be helpful to further determine the position and trajectory of a fixation point, particularly in athletes (Baker et al., 2019; Williams & Ericsson, 2005; Williams et al., 2004). Therefore, through the integration of eye-tracking, it becomes possible to more accurately examine visual search strategies during change of detection comparing expertise versus novice athletes.
Footnotes
Acknowledgments
The authors would like to thank the subjects for their time and enthusiasm. The authors are particularly grateful to the French Parkour Federation (Fédération de Parkour, FPK) for its help and support.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
