Content uploaded by Joost Pluijms
Author content
All content in this area was uploaded by Joost Pluijms on Mar 20, 2014
Content may be subject to copyright.
Translating Key Methodological Issues into
Technological Advancements
When Running In-Situ Experiments in
Sports: An Example from Sailing
by
Joost P. Pluijms, Rouwen Cañal-Bruland, Serge Kats and
Geert J. P. Savelsbergh
Reprinted from
International Journal of
Sports Science
& Coaching
Volume 8 · Number 1 · 2013
Translating Key Methodological Issues into
Technological Advancements
When Running In-Situ Experiments in
Sports: An Example from Sailing
Joost P. Pluijms1, Rouwen Cañal-Bruland1, Serge Kats4and
Geert J. P. Savelsbergh1,2,3
1MOVE Research Institute Amsterdam, Faculty of Human Movement
Sciences, VU University Amsterdam, van der Boechorststraat 9, 1081 BT,
Amsterdam, The Netherlands
E-mail: j.p.pluijms@vu.nl
2Research Institute for Biomedical Research into Human Movement and
Health, Faculty of Science and Engineering, Manchester Metropolitan
University, UK;
3Academy for Physical Education, University of Professional Education,
The Netherlands;
4InnoSportLab The Hague: Innovation in Sailing, The Netherlands
ABSTRACT
In recent years there has been an increasing interest in capturing and
understanding skilled performance by studying complex perceptual-motor
skills. In this context, we identify and discuss key methodological issues that
are particularly relevant when aiming to translate sport scientific knowledge
into practical guidelines for coaches and athletes. These issues are: the
representative performance environment (including fidelity of stimuli and
type of response), generalisability, and experimental control. After a short
introduction of the methodological issues, we first review and critically
discuss to what degree past research studying complex perceptual-motor
skills in sports has or has not sufficiently taken these issues into account.
Second, we illustrate an examination of expertise in sailing as an example
of how to address the key issues when performing experiments in-situ. We
conclude that the presented example illustrates how the collaboration
between coaches, athletes and sports scientists advances the
methodological and technological developments to capture skilled
performance in complex sports.
Key words: Expertise, Field Experiments, Sailing, Sport Technology,
Visual Search Behaviour
International Journal of Sports Science & Coaching Volume 8 · Number 1 · 2013 89
Reviewers: Duarte Araujo (Technical University of Lisbon, Portugal)
Richard Shuttleworth (Australian Institute of Sport, Australia)
INTRODUCTION
Skilled athletes outperform their less skilled counterparts due, amongst other things, to
superior anticipation skills in changing environments [1]. In recent years there has been an
increasing interest in capturing and understanding skilled performance by studying complex
perceptual-motor skills [2, 3]. In this regard, one of the major aims of sport scientists is to
translate experimental research into practical guidelines for coaches and athletes alike [1, 2].
Skilled performance manifests itself in finely tuned perceptual-motor responses
developed over many years of extensive practice [4]. While motor skills, defined as “a skill
for which the primary determinant for success is the quality of movement that the performer
produces” [5, p. 4], have been strongly related to the superiority of skilled athletes over their
lesser skilled counterparts, the reliance on different perceptual information pick-up that
guides the action has only received attention over the last three decades [e.g., 1, 6, 7-12].
Based on this research, it is undisputed that expert performers are also superior at picking-
up the most appropriate perceptual information – at the right time – to help inform and guide
the most suitable motor responses. One of the key challenges for developing athletes though
is that the visual environment typically contains information both relevant and irrelevant to
performance [11, 13, 14], thereby increasing the importance of efficient information
selection processes.
In this article, we identify and discuss key methodological issues that in our view are
particularly relevant when aiming to translate sport scientific knowledge into practical
guidelines for coaches and athletes. After shortly introducing and highlighting the
importance of the issues, we first review and critically discuss to what degree past research
studying complex perceptual motor skills in sports has or has not adequately addressed these
issues. Secondly, based on our current research project into sailing we illustrate how we have
addressed these key issues in in-situ experiments. Finally, we argue that the presented pilot
work provides a clear example of how the collaboration between coaches, athletes and sports
scientists sparks the methodological development to capture skilled performance in sailing.
DIFFERENCES IN LABORATORY AND FIELD EXPERIMENTS
Laboratory and field experiments are the two key research settings which have been used to
further our understanding of how perceptual-motor skills are executed by skilled performers
[15]. While it is well established that each research setting has its advantages and drawbacks
[16], it is interesting to note that over the past two decades approximately two-thirds of all
sports science research in Australian research institutes and universities has been conducted
in laboratory settings [17]. Recent research, however, indicates significant differences
between laboratory studies and representative experimental settings particularly for the
measures of visual search behaviour and movement behaviour [18-20]. For example, Dicks
et al. [18] demonstrated that the visual search behaviour of goalkeepers during soccer penalty
kicks significantly differs between video simulation and in-situ conditions [21]. These
findings emphasize how experimental conditions can markedly alter an athlete’s behaviour.
With this in mind, we identify and discuss key methodological issues that are relevant when
comparing these two types of experimental environments, particularly when aiming to
translate sport scientific knowledge into practical guidelines for coaches and athletes: (i)
representative performance environment [15, 22, 23] (ii) generalisability [24-27], and (iii)
experimental control [18, 28].
REPRESENTATIVE PERFORMANCE ENVIRONMENT
In sailing there is a high level of inter-individual dynamics and the ability of sailors to create
90 In-Situ Experiments in Sports
uncertainty in the environment for fellow competitors is paramount to successful
performance. Sailors can, for example, alter their approach line, gain speed in a squall or
vary their heel angle. However, whereas the issue of engaging opponents is typically not
considered in laboratory settings, studies in the field provide the opportunity to capture the
relationship between the performance environment, including fellow competitors, and
athletes’ actions [29]. Notably, the dynamic situation on the water affords ever-changing
opportunities for action [30]. Hence, a competitive performance environment might give
structure to an athlete’s skilled performance, that is, athlete’s perceptual-motor skills to cope
with uncertainty of situational constraints, such as fellow competitors and changing
environmental characteristics [29]. Athletes in representative experimental designs need to
be provided with opportunities to establish functional relationships between perception (e.g.,
visual information) and action (movement). As Pinder et al. emphasise: “Since information
regulates actions, an important principle is that the critical perception and action processes
that are coupled in a competitive performance environment should be maintained in the
design of experimental task constraints” [31, p. 796]. Following Brunswik, a representative
design is defined as “the arrangement of constraints in an experimental design so that they
represent the behavioral setting to which the results are intended to apply” [32, p. 148]. In a
representative performance environment achievement is based on comparable information
sources (i.e. perceptual cues or other stimuli) – and athlete’s responses remain the same – to
those in a competitive performance environment. Achievement refers to the adaptation of an
organism to a specific environment [33]. To clarify the effect of a representative performance
environment on experimental results, different types of stimuli and responses, used in field
and lab experiments, are discussed.
First, the stimuli used across different research settings can vary, for example, the timing
of the information or the quality of the stimulus presentation can alter markedly. Participants
in competitive environmental settings see, feel, and hear relevant stimuli continuously and
unobtrusively [13, 14], while participants in laboratory settings often react (and start their
action) from the moment that a perceptual stimulus is presented on a video or computer
screen [e.g., 34]. A stimulus may be any form of external information, in particular the
direction of the non-kicking leg of a penalty taker in soccer [9, 10], the setter’s initial contact
with a volleyball [35], or the back foot contact of a bowler in cricket [36, 37]. Aconsiderable
amount of the research literature on perceptual-motor skills has used video or computer
simulations to examine expert-novice differences across a broad spectrum of sports such as
soccer [e.g., 9, 10], tennis [e.g., 21, 38, 39], field hockey [e.g., 40] and sailing [e.g., 41].
However, all of these studies are potentially limited by an inability to truly simulate the
visual information experienced by athletes in a competitive performance environment [18,
42, 43]. Stimuli on a video or computer screen might not adequately reflect those stimuli that
trigger the action of an athlete in the field. For example, three-dimensional information
guiding the perception of depth is undoubtedly lost or decreased. Consequently, participants
might not show the visual search and movement behaviours that they would in a competitive
performance environment. In other words, achievement in a lab setting is not based on
comparable information sources to those in a competitive environment. Therefore, it can be
questioned whether the findings from lab research sufficiently transfer to the field [18].
Second, a review of perception-action research reveals that the type of response
significantly alters the expert-novice difference typically reported in studies on expertise. In
most recent studies within laboratory settings responses are measured in different ways:
verbally [e.g., 8, 21, 44], with pen-and-paper responses [39], by a button press [e.g., 45, 46],
with a simulated joystick movement [e.g., 9], or with simulated responses [e.g., 21, 47]. In
International Journal of Sports Science & Coaching Volume 8 · Number 1 · 2013 91
sports, however, movements are not constrained in such a rigid way. Movements naturally
vary from trial to trial and may still lead to similar outcomes (e.g., a successful free-throw in
basketball) [3]. Laboratory studies often reduce complex perceptual-motor skills to a simpler
level and there is an on-going debate that these studies may not adequately simulate the
athlete-environment interaction [e.g., 23, 32]. In particular, participants in laboratory studies
were frequently limited in their motor degrees of freedom [48], but also in their perceptual
degrees of freedom, thereby not being able to perceive or respond to the situation as in a
competitive performance environments [49]. Therefore, several authors support the notion
that representative perception-action coupling is essential when conducting scientific sports
experiments and thus call for representative task designs [23, 50]. Based on the framework
of Pinder et al. [31] an in-situ design should be based on comparable information sources
(i.e., fidelity of stimuli), and athletes’ responses (i.e., fidelity of response) should remain the
same compared to those in a competitive performance environment.
GENERALISABILITY
Closely connected to the issues related to the type of stimulus and type of response, the
generalisability of conclusions from laboratory studies are often questioned in applied sports
science [e.g., 18]. Generalisability refers “to whether results of a study can legitimately be
generalised to a specified broader population” [51, p. 133] and whether the results are “not
confined to the particulars of time and place” [26, p. 236]. In general, if the objective is to
translate sports scientific knowledge into practical guidelines that generalise to a specified
broader population (e.g., sailors with equivalent skill level), then results that are more
generalisable are viewed as more desirable than results that are less generalisable [19].
However, experimental findings from many laboratory studies in sports science cannot be
generalised to competitive settings and to the specified broader population. In response to
this criticism some researchers [24, 27] argue that the set-up of an experiment does not
always require settings and participants that project competitive environmental conditions.
They point out that the advantage of laboratory experiments lies within the simplification of
competitive situations to only theoretical meaningful aspects, eliminating variables that are
not relevant and making generalization more likely. However, Mook [27] acknowledges the
need for generalisability under circumstances in which researchers aim to derive predictions
for functional behaviour, such as in applied (sports) settings. During athlete-environment
interactions in an applied sports science experiment, achievement (i.e., the adaptation of an
athlete to the environment) should be corresponding with functional behaviour as in a
competitive performance environment [32, 50, 52]. Furthermore, based on Brunswik’s ideas,
Pinder et al. recently stated that “just as participants of an experiment must be representative
of those to which the study wishes to generalise, the experimental task constraints must also
represent the environmental (performance) constraints to which they are to be generalised”
[32, p.148].
EXPERIMENTAL CONTROL
The issue of generalisability also has a direct impact on experimental control [18]. On the
one hand, it is well known that laboratory tasks commonly allow greater experimental
control when compared to field studies. In favour of laboratory experiments, Lucas validly
argues that “laboratory experimental conditions allow researchers to examine only elements
of the situation that are relevant to a certain theory under test; other elements that may mask
or vary with predicted effects are eliminated” [26, p. 246]. However, in field studies
experimental control is often difficult to achieve, especially while studying perceptual-motor
92 In-Situ Experiments in Sports
skills influenced by, and highly attuned to, unpredictable environmental factors. Therefore in
in-situ designs sport scientists are challenged to identify and measure all relevant
environmental factors, thereby creating the opportunity to categorise trials by environmental
conditions afterwards. The preference for simplified research designs in laboratory settings
often emphasizes experimental control and therefore disregards the importance of these
environmental characteristics [18, 28]. To support generalisability and to improve
experimental control, considerable technical sophistication and careful descriptions of
seemingly simple phenomena in the field (i.e., on the water) are required [52].
In summary, the key methodological issues of representative performance environment
(including fidelity of stimuli and type of response), generalisability and experimental control
have a tremendous impact on how to set up experimental research which seeks to translate
research findings into practice. Representative experimental settings, based on the ideas of
Brunswik and the insights of Gibson, provide full visual information for participants, and in
doing so participants are able to perceive and process varying types of stimuli
uninterruptedly, and can respond without dealing with unnecessary environmental or task
constraints [33, 53]. In order to study complex perceptual-motor skills, experimental
conditions should provide participants with opportunities for action and perception in a
representative environment while trying to measure all relevant performance and
environmental characteristics. In the following, we illustrate an initial attempt, based on our
current research project into sailing, on how to adequately implement these issues into
technological advancements when running in-situ experiments.
AN EXAMPLE: A COMPLEX EVENT IN SAILING
If one aims to capture and understand skilled performance in complex sports, several key
factors need to be taken into account. For instance, in sailing, sailors have to deal with wind
shifts, waves at sea, a changing tide, fellow competitors, and the optimisation of boat settings
[54]. The technical skills are therefore an essential component of a sailor’s ability to sail as
fast as possible at all times. Furthermore, decision-making at the right time is necessary in
order to sail the optimal route [41]. The fastest course to reach the finish line is variable,
since the wind and waves usually change in direction and speed along the course. These
unpredictable environmental factors add to the level of perceptual-motor complexity in
sailing.
If the aim of a research project is to inform or translate information into practice, then
experimental designs in the field should replicate as closely as possible the perception-action
coupling produced in the performance domain in order to extract a full picture of expertise
[32, 42]. Therefore athletes must be given the opportunity to detect information-rich areas in
the field so that they can direct their attention appropriately and extract meaningful
information from these areas [11]. In line with this argument, empirical findings have
indicated that context-specific expertise effects are exhibited more clearly under
representative (in-situ) experimental conditions than in simulation laboratory settings [20].
Moreover, in addition to empirical findings also current theoretical frameworks, such as the
two-visual systems model by Milner and Goodale [55], support the notion that perception
and action should be coupled in an unrestricted manner [23]. So far, however, sailing has
predominantly been studied by obtaining data using computer simulation scenarios in which
participants were required to react verbally, or by pressing keyboard buttons during dynamic
sailing tasks [see 41, 56, for an exception see 52]. Despite their experimental control, these
studies would have been more representative if the authors had included more realistic
movement behaviours such as when sailing in a competitive performance environment.
International Journal of Sports Science & Coaching Volume 8 · Number 1 · 2013 93
In the remainder of this paper, we present and discuss our current approach to effectively
capture complex perceptual-motor skills in sailing in light of the methodological criticisms
raised in this paper. To this end, we illustrate some pilot work that outlines how the close
collaboration between coaches, athletes and sports scientists may advance the
methodological development to adequately capture skilled performance in sailing. As an
example, our approach to measure complex perceptual-motor skills in the field is
demonstrated based on a complex sailing manoeuvre – dubbed a windward mark rounding –
in a genuine single-handed Olympic boat.
During a sailing regatta many small events within a couple of hundred milliseconds occur
that can make the difference between winning and losing. The findings of Araújo and
colleagues [41, 52] suggest that expertise in perceptual-motor skills may be of great
significance in sailing, especially during these critical events. For Davidson [54], an event
refers to a critical moment to win time relative to an opponent, for example by performing a
neat start, or a perfect rounding of a mark. Throughout this paper, the term mark is used to
refer to “a floating buoy or other fixed-position indicator of the end of one leg of a racing
course and start of the next. Boats must typically turn 2-3 rad (120-180°) as they round a
mark” [57, p. 1077]. Aleg is defined as a length of water between one mark and the next. To
demonstrate our approach, data is systematically collected during a relevant but relatively
short event in a sailing regatta: the rounding of the windward mark. This particular event was
identified as highly relevant and important following formal discussions and questionnaires
that were completed by interviewing elite coaches and sailors in the Netherlands.
The aim in rounding the windward mark is to turn the boat from a close-hauled course
onto a downwind course while staying as close to the mark as possible. The term windward
is used to refer to the side of a boat, or direction of sailing, towards the wind. The rounding
of a windward mark is not dependent on wind direction. The rounding may be divided into
three main parts and is considered to be a complex perceptual-motor skill: i) the last straight
line toward the windward mark, starting immediately after the last tack; ii) the actual
rounding of the mark, starting when the boat turns, indicated by letting the mainsheet out,
steering away from the wind, and moving the body outside, causing the boat to heel
windward; and iii) the period of time beginning after the rounding as soon as the course is
straight, lasting for at least two seconds [54, 57]. For the helmsman this is a challenging
manoeuvre to master because of the small rudder and powerful rig, which means that a good
technique is imperative. The task of the helmsman varies according to the boat class. The
boat trim and heel should all be employed to assist turning the boat. Heeling may be defined
as tilting a boat away from the wind due to its sideways pressure on sails. This will allow the
sailor to maintain maximum tactical options on exiting the mark.
The pilot work is illustrated briefly in the next section by integrating recent technology to
capture skilled performance of a sailor throughout the duration of a windward mark
rounding.
INTEGRATING TECHNOLOGY TO CAPTURE PERFORMANCE
The rounding of a windward mark is a complex perceptual-motor skill. To unravel the
performance of a sailor during a windward mark rounding it is necessary to utilise a variety
of testing equipment. The key task is to combine different types of technology in a smart and
simple manner to effectively evaluate the overall performance of a sailor on the water. The
following set-up functions as an example to systematically collect data while sailors round
the windward mark. The set-up can be best considered as consisting four sections which
interact to evaluate overall performance: i) visual search behaviour of the sailor, ii)
94 In-Situ Experiments in Sports
movement behaviour of the sailor, iii) environmental conditions (e.g., wind), and iv)
performance of the boat. These variables approximate the competitive performance
environment that is under investigation [32]. For this set-up, an Olympic single-handed
sailing boat was used: the Laser. The Laser is the world’s numerically largest class of single-
handed sailing dinghy for adults and has been an Olympic class for men since 1996 [54].
VISUAL SEARCH BEHAVIOUR
Visual search behaviour is recorded with a mobile eye tracker attached to a pair of safety
glasses, in this case using the Applied Science Laboratories (ASL) Mobile Eye Tracker
(version II). The Mobile Eye samples point of gaze at 30 frames per second. With a mobile
eye tracker it is possible to record the visual search characteristics of sailors in their
performance environment: mean number of fixation points, mean number of areas fixated on,
and mean fixation duration for each trial [for more information, see e.g., 18]. For example,
these visual search characteristics have been addressed in several other situations similar to
rounding a windward mark, e.g., in speedskating on an Olympic Oval and when negotiating
bends during car driving [58-61]. Based on these studies we hypothesise that high skilled
sailors (i) have more control over their boat handling thereby allowing them to gaze more
outside the boat in general, (ii) gaze more to the tangent point just before - and during - the
actual rounding similar to findings from car driving and speedskating studies [58-61], and
(iii) have less fixation points of longer duration during the mark rounding as occuring in
studies in a wide range of sports [e.g., 9, 10]. The eye tracking glasses are relatively
lightweight and unobtrusive. The recording device is placed in a small waterproof backpack
that is worn on the chest of the participant so that it does not hamper the participant in his or
her movements (Figure 1).
International Journal of Sports Science & Coaching Volume 8 · Number 1 · 2013 95
Figure 1. A Sailor Rounding the Windward Mark and Adjusting His Trimming
Lines
Marked by letters A-B: (A) the mobile eye tracker; (B) the recording device
on the chest of the participant worn in a waterproof rucksack.
The eye image and scene image are interleaved and saved on a DVCR tape (60 minutes),
which are analysed with EyeVision software on a laptop computer using a licence key. In
figure 2 several fixation locations during the rounding of a windward mark are displayed.
The locations of gaze are utilised to calculate percentage-viewing time for each fixation
location [e.g., 9, 10, 18, 40].
MOVEMENT BEHAVIOUR
The movement behaviour of a sailor is recorded using a waterproof GoPro video camera
mounted on the bow of a Laser; an example of the perspective of the GoPro is shown in
Figure 1. Recording can be performed at 30 or 60 frames per second, providing the
opportunity to obtain the timing of key movements in reference to the point where sailors
were positioned right above the mark during the rounding of a windward mark (Figure 3) [3,
42]. The images in Figure 3 are screenshots captured and adapted from a DVD instruction
video to illustrate a regular windward mark rounding [62]. This video was shot from a coach
boat nearby. Some examples of key movements during a mark rounding as identified in the
literature [54, 62] are (i) the release of the outhaul and kicking strap (these are trimming
lines) during the approach phase of a windward mark, (ii) heeling to the windward side
during the actual rounding, and (iii) pulling up of the centreboard in the exit phase. The
GoPro video – synchronised with the corresponding GPS track – provides the opportunity to
obtain the timing of key movements in reference to a point (t=0) where sailors were
positioned right above the mark during the rounding of a windward mark (Figure 3). From
this point on key movements were identified 4500 ms before and 4500 ms after t=0. On
average, key movements were performed within these 9000 ms. Movement responses are
coded in milliseconds in reference to t=0 and averaged across other trials that were
performed in the same wind condition. Wind speed was obtained by multiple iPhone
96 In-Situ Experiments in Sports
Figure 2. Fixation Locations of the Visual Search Behaviour of a Sailor
Marked by letters A-H: (A) virtual inner curve of the rounding mark; (B)
interception of the mark and boom; (C) interception of the horizon and
next downwind mark; (D) in front of the boat; (E) upwind area; (F)
downwind area; (G) other boat through sail; (H) trimming: cunningham.
applications, such as Navionics Europe (version 2.5.6), WindGURU (version 2.0) and
Windfinder (version 2.0). These applications delivered us information about possible tides,
wind speed, gust, and direction per hour. We hypothesise that high skilled sailors are more
attuned to differences in environmental conditions (e.g. low versus strong wind) and are
therefore more consistent in reproducing successful movements [41]. On the water,
participants are able to respond without any limitation to every stimulus available in the
environment [18, 23, 32].
ENVIRONMENTAL CONDITIONS
To control for unpredictable environmental factors such as wind, a wind wand (Tacktick)
mounted either on the mast or bow of the sailing boat provides wind information. The
measurement of wind speed is a key factor in our attempt to capture environmental conditions
in sailing. More specifically, multiple trials from sailors with different skill levels are categorised
into one of the following three categories: (i) 0-5 knots, (ii) 5-8 knots, and (iii) 8-12 knots. As a
result we can compare trials – from the same wind condition - across participants. These
measures, and others like wave strength and strength of currents help us to effectively evaluate
performance across different participants. The Technical Support Division of the VU University
in collaboration with InnoSportLab The Hague is currently in the process of developing a tool
to measure wave strength in metres and currents in metres per second. In future studies this tool
is most likely to be integrated to capture relevant environmental conditions. So far, we used
multiple iPhone applications like Navionics Europe, WindGURU and Windfinder.
International Journal of Sports Science & Coaching Volume 8 · Number 1 · 2013 97
Figure 3. Some Key Movements During the Rounding of a Windward Mark
in Time
Marked by letters A-G: (A) cleaning up the mainsheet; (B) releasing the
outhaul and cunningham; (C) releasing the kicking strap and mainsheet;
(D) sailing the boat to windward side, moving body to behind; (E) moving
body to inside; (F) possibly releasing the cunningham more; (G) putting
centreboard up; Upper right: The track of a windward mark rounding with
the letters of the corresponding key movements displayed.
PERFORMANCE OF THE BOAT
The Pi Garda logger, which houses an internal battery, a three-axis accelerometer, and a high
performance GPS, is used to log data from on-board sensors as well as the boat’s position
and speed via a 5 Hz GPS; that is, the GPS receives a fix every 200 ms. The Pi Garda logger
uses NMEA (National Marine Electronics Association) communication protocols to work
with other systems, such as the Tacktick systems. The Pi Garda logger is positioned at the
starboard side of the mast. The Pi Garda logger and the Pi Toolbox software is used to track,
measure, and evaluate critical elements of sailboat performance (e.g., boat speed, heel angle
or course over ground).
The video data from the ASL eye tracker and GoPro are synchronised (by UTC time) and
coupled to the GPS track within the Pi Toolbox software package (Figure 4). The software
searches the nearest video frame for any given moment in time; using this method it is
possible to load and synchronise videos, of any type of video recording equipment, for
comparison with any other data feed. As a result, it is possible to simultaneously review
synchronised data variables: GoPro video footage, mobile eye video, GPS tracks, boat speed,
GPS course over ground, heel angle, and rudder angle. Synchronisation is achieved via a
UTC/GMT iPhone application (Emerald & Sequoia LLC; version 1. 5) which uses Network
Time Protocol (NTP). The application uses these NTP servers on the Internet to obtain a
more accurate time than is typically available from a device’s internal clock. A large clock
display shows the corrected time obtained from the NTP servers in UTC time; the average
round-trip-time (rtt) to the server varies between 50 and 200 ms and is processed in the
98 In-Situ Experiments in Sports
Figure 4. A Screenshot of the Pi Toolbox Software Suite with Synchronised
Data from Seven Sources of Information
Marked by letters A-G: (A) GoPro video; (B) mobile eye video; (C) GPS
track of multiple windward mark roundings; (D) boat speed in knots; (E)
GPS course over ground in degrees; (F) heel angle in degrees; (G) rudder
angle in degrees.
corrected time. A red, yellow or green light above the UTC time indicates the quality of the
server connection in the iPhone application. Agreen light is on when quality is sufficient. We
used the 3G networks for our Internet connection; average Internet speed varied between 144
and 284 Kilobytes per second. UTC time was presented several times during a measurement
(e.g., every 20-30 minutes), both to the scene camera of the mobile eye tracker and GoPro to
minimize errors when videos were analysed subsequently in relation to other data feeds. In
addition, the export of multivariate data from Pi Toolbox to other software packages such as
Microsoft Excel or IBM PASW Statistics is convenient. Pi Toolbox is a suitable software
option to inform coaches and athletes immediately after a measurement in the field. This tool
significantly improves translation from research to practice [2].
WHY DO WE FOLLOW THIS APPROACH?
With the above procedure we set up an initial attempt to capture and understand skilled
performance in sailing to translate sport scientific knowledge into practical guidelines for
coaches and athletes. We sought to address key methodological issues outlined previously in
an attempt to adequately capture complex perceptual-motor skills in sailing. In our research
project we made sure that sailors were able to perform a complex sailing skill in a
representative performance environment. That is, we measured sailors’ performances in the
field, or one should better say, on the water. Given our aim to translate our knowledge
directly into practical guidelines for athletes and coaches, moving to the water is imperative
because laboratory studies may fail to adequately sample the environmental characteristics
[23]. More specifically, no video simulation or similar methods are used and three-
dimensional visual information is fully provided [18, 42]. Using an in-situ approach, sailors
did not have to wait for perceptual stimuli before an action was initiated [21, 63], as is the
case in laboratory studies. Actions of sailors in this research design are based on similar
stimuli that trigger sailor’s action in the field. Hence, the entire natural flow of information
from the onset of the trial until the end of the trial can be used to form the basis of an action.
By measuring movement behaviour on the water (i.e., the type of response) the sailor’s
actions are captured during the whole trial and remain the same compared with a competitive
environment [23, 32]. That is, sailors are neither limited in their motor nor perceptual degrees
of freedom [49]. Therefore in our project sailors are able to adapt their actions to
environmental changes, thereby giving us the opportunity to gain deeper insights into the
complex perceptual-motor skills of expert sailors.
Secondly, laboratory experiments are considered to be less generalisable, because the
transfer of research findings from one sample to a specified larger population may not be
applicable [26]. Our approach is an attempt to generalise research findings in sailing. That
is, one of our aims is to generalise specific findings to a specified broader population (e.g.,
sailors with equivalent skill levels) and to immediately provide feedback of these results to
coaches and athletes to improve performance [27, 64].
Finally, experimental control in field studies is often complicated to achieve [7, 18]. As
addressed already, laboratory tasks commonly have the benefit of greater experimental control
when compared to representative experimental settings [24, 26]. This is considered a thorny
issue, but should not distract sport scientists from the measurement of complex perceptual-
motor skills in the field. In the near future technical tools to measure wavelength, wave height,
and the strength of currents have the potential to measure environmental conditions for each
trial. This allows us to analyse visual search and movement data across trials and participants,
categorised by relevant environmental conditions such as wind strength. In this regard, sports
scientists might benefit from better embracing new technological developments and close
International Journal of Sports Science & Coaching Volume 8 · Number 1 · 2013 99
collaborations with other disciplines [1]. The recent technological advances in eye-tracking
systems (i.e., wireless digital recording and improved data collection technology) and remote
monitoring sensor systems are examples that provide a practical option to better capture and
understand visual search behaviour of athletes in in-situ research designs [43].
For the analysis of gaze data, Dicks et al. [18] recently suggested that the averaging of
data in statistical analyses may have disguised important individual differences in
performance [for an alternative approach, see 65], and therefore question the existence of an
optimal perceptual strategy. Our approach allows us to shed further light on this issue by
comparing our findings with previous research results performed in laboratory settings, and
to identify both individual and group differences in sailors’ gaze strategies.
Elite sailing coaches have an interest in examining complex events in the field, but are
commonly eager to study the broad picture, for example, to capture performance during a
whole leg or even a whole sailing regatta. As Williams and Kendall conclude: “An elite
coach is concerned primarily with sports performance, whereas a sports science researcher is
focused on increasing sports science knowledge (both applied and theoretical), based on
sound research questions” [17, p. 194]. To support the collaboration between sport science
and practice, it is of great importance to report useful and concise feedback to experts in the
field after the research has been conducted [66]. Additionally, findings from sport scientific
research could be accepted and adopted more by athletes, coaches and staff at whom they are
targeted [2].
CONCLUSION
This paper has argued that our presented approach may be an adequate method to examine
skilled performance within changing environments (such as in sailing), both from a
methodological and practical perspective. By translating key methodological issues into
technological advancements we aim to improve in-situ experiments in the field. Using, for
example, waterproof and cold-resistant high-tech equipment, we could use this approach for
the future design of experimental and practice tasks to capture and understand the skilled
performance of other Olympic sports, such as kiteboarding, BMX racing, mountain biking,
or snowboard cross. These sports have received little attention from a perception-action point
of view. Note that we use our technology as a means to an end, not to an end itself. From a
long-term perspective, a better understanding of skilled performance in the field, and the role
of perception-action coupling in varying sports could support the talent identification of
young athletes [18]. In addition, the knowledge gathered by our approach could contribute
to the development of evidence-based perceptual training methods [e.g., 67].
We are currently pursuing our research project in sailing, in close collaboration with
(elite) coaches and sailors to examine performance during specific sailing events. These
experiments are designed to add substantially to our understanding of key predictors of
skilled performance during sailing. An important practical implication when implementing
these issues in in-situ experiments is that it allows us to optimise the scientific guidance of
athletes, coaches and staff to the desired performance at world championships and Olympics
by translating sports scientific knowledge into practice.
ACKNOWLEDGEMENTS
We gratefully acknowledge the close collaboration with the InnoSportLab The Hague:
Innovation in Sailing for supporting this project. We also thank the sailors and coaches who
provided their time to participate in our study, the Technical Support Division of the Faculty
of Human Movement Science at the VU University for their technical support, and David
100 In-Situ Experiments in Sports
Mann for helpful comments and suggestions on an earlier version of this manuscript.
REFERENCES
1. Williams, A.M., Ford, P.R., Eccles, D.W. and Ward, P., Perceptual-Cognitive Expertise in Sport and Its
Aquisition: Implications for Applied Cognitive Psychology, Applied Cognitive Psychology, 2011, 25, 432-
442.
2. Bischop, D., An Applied Research Model for the Sport Sciences, Sports Med, 2008, 38, 253-263.
3. Oudejans, R.R.D. and Coolen, B.H., Human Kinetics and Event Control: On-Line Movement Registration
as a Means for Experimental Manipulation, Journal of Sports Sciences, 2003, 21, 567-576.
4. Abernethy, B. and Russell, D.G., Expert-Novice Differences in an Applied Selective Attention Task, Journal
of Sport Psychology, 1987, 9, 326-345.
5. Schmidt, R.A. and Wrisberg, C.A., Motor Learning and Performance: A Situation-Based Learning
Approach, Human Kinetics, Champaign, IL, 2008.
6. Abernethy, B. and Russell, D.G., The Relationship between Expertise and Visual Strategy in a Racquet Sport,
Human Movement Science, 1987, 6, 283-319.
7. Ericsson, K.A. and Williams, A.M., Capturing Naturally Occurring Superior Performance in the Laboratory:
Translational Research on Expert Performance, Journal of Experimental Psychology, 2007, 13, 115-123.
8. Ranganathan, R. and Carlton, L.G., Perception-Action Coupling and Anticipatory Performance in Baseball
Batting, Journal of Motor Behaviour, 2007, 39, 369-380.
9. Savelsbergh, G.J.P., Van Der Kamp, J., Williams, A.M. and Ward, P., Visual Search Anticipation and
Expertise in Soccer Goalkeepers, Journal of Sports Sciences, 2005, 20, 79-82.
10. Savelsbergh, G.J.P., Williams, A.M., Van Der Kamp, J. and Ward, P., Visual Search, Anticipation and
Expertise in Soccer Goalkeepers, Journal of Sports Sciences, 2002, 20, 279-287.
11.Williams, A.M., Davids, K. and Williams, J.G., Visual Perception and Action in Sport, E & FN Spon,
London, United Kingdom, 1999.
12. Williams, A.M., Ward, P. and Chapman, C., Training Perceptual Skill in Field Hockey: Is There a Transfer
from the Laboratory to the Field?, Research Quarterly for Exercise and Sport, 2003, 74, 98-103.
13. Williams, A.M. and Grant, A., Training Perceptual Skill in Sport, International Journal of Sport Psychology,
1999, 30, 194-220.
14. Abernethy, B., Training the Visual-Perceptual Skills of Athletes: Insights from a Study of Motor Expertise,
American Journal of Sports Medicine, 1996, 24, 89-92.
15. Xiao, Y. and Vicente, K.J., A Framework for Epistemological Analysis in Empirical (Laboratory and Field)
Studies, Human Factors, 2000, 42, 87-101.
16. Abernethy, B., Thomas, K.T. and Thomas, J.T., Chapter 17 Strategies for Improving Understanding of Motor
Expertise [or Mistakes We Have Made and Things We Have Learned!!], Advances in Psychology, 1993, 102,
317-356.
17. Williams, S.J. and Kendall, L.R., AProfile of Sport Science Research (1983 - 2003), Journal of Science and
Medicine in Sport, 2007, 10, 193-200.
18. Dicks, M., Button, C. and Davids, K., Examination of Gaze Behaviours under in-Situ and Video Simulation
Task Constraints Reveals Differences in Information Pickup for Perception and Action, Attention, Perception
& Psychophysics, 2010, 72, 706-720.
19. Mann, D.L., Abernethy, B. and Farrow, D., Action Specificity Increases Anticipatory Performance and the
Expert Advantage in Natural Interceptive Tasks, Acta Psychologica, 2010, 135, 17-23.
20. Mann, D.T.Y., Williams, A.M., Ward, P. and Janelle, J.M., Perceptual-Cognitive Expertise in Sport: A Meta-
Analysis, Journal of Sport & Exercise Psychology, 2007, 59, 2162-2186.
21. Farrow, D. and Abernethy, B., Do the Expertise and the Degree of Perception-Action Coupling Affect Natural
Anticipatory Performance?, Perception, 2003, 32, 1127-1139.
International Journal of Sports Science & Coaching Volume 8 · Number 1 · 2013 101
22. Schmuckler, M.A., What Is Ecological Validity? A Dimensional Analysis, Infancy, 2001, 2, 419-436.
23. Van Der Kamp, J., Rivas, F., Van Doorn, H. and Savelsbergh, G.J.P., Ventral and Dorsal System
Contributions to Visual Anticipation in Fast Ball Sports, International Journal of Sport Psychology, 2008,
39, 100-130.
24. Berkowitz, L. and Donnerstein, E., External Validity Is More Than Skin Deep: Some Answers to Criticisms
of Laboratory Experiments, American Psychologist, 1982, 37, 245-257.
25. Green, L.W. and Glasgow, R.E., Evaluating the Relevance, Generalization and Applicability of Research:
Issues in External Validation and Translation Methodology, Evaluation & The Health Professions, 2006, 29,
126-153.
26. Lucas, J.W., Theory-Testing, Generalization, and the Problem of External Validity, Social Theory, 2003, 21,
236-253.
27. Mook, D.G., In Defence of External Validity, American Psychologist, 1983, 38, 379-389.
28. Dicks, M., Davids, K. and Araújo, D., Ecological Psychology and Task Representativeness: Implications for
the Design of Perceptual-Motor Training Programmes in Sport, in: Hong, Y. and Bartlett, R., eds., The
Routledge Handbook of Biomechanics and Human Movement Science, Routledge, London, 2008, 129-139.
29. Passos, P., Araújo, D., Davids, K. and Shuttleworth, R., Manipulating Constraints to Train Decision Making
in Rugby Union, International Journal of Sports Science and Coaching, 2008, 3, 125-139.
30. Fajen, B.R., Riley, M.A., and Turvey, M.T., Information, Affordance, and the Control of Action in Sport,
International Journal of Sport Psychology, 2008, 40, 79-107.
31. Pinder, A., Renshaw, I., Davids, K., and Kerhervé, H., Principles for the Use of Ball Projection Machines in
Elite and Developmental Sport Programmes, Sports Medicine, 2011, 41, 793-800.
32. Pinder, A., Davids, K., Renshaw, I., and Araújo, D., Representative Learning Design and Functionality of
Research and Practice in Sport, Journal of Sport & Exercise Psychology, 2011, 33, 146-155.
33. Brunswik, E., Perception and the Representative Design of Psychological Experiments, University of
California Press, Berkeley, 1956.
34. Paull, G., Case, I. and Grove, R., Computer Controlled Interactive Video Simulation for Motor Skills
Research, Behavior Research Methods, Instruments & Computers, 1997, 29, 534-541.
35. Starkes, J.L., Edwards, P., Dissanayake, P. and Dunn, T., A New Technology and Field Test of Advance Cue
Usage in Volleyball, Research Quarterly for Exercise & Sport, 1995, 66, 162-167.
36. Müller, S. and Abernethy, B., Batting with Occluded Vision: An in-Situ Examination of the Information Pick-
up and Interceptive Skills of High- and Low-Skilled Cricket Batsmen, Journal of Science and Medicine in
Sport, 2006, 9, 446-458.
37. Müller, S., Abernethy, B. and Farrow, D., How Do World-Class Cricket Batsmen Anticipate a Bowler’s
Intention?, The Quarterly Journal of Experimental Psychology, 2006, 29, 457-478.
38. Rowe, R., Horswill, M.S., Kronvall-Parkinson, M., Poulter, D.R. and Mckenna, F.P., The Effect of Disguise on
Novice and Expert Tennis Players’Anticipation Ability, Journal of Applied Sport Psychology, 2009, 21, 178-185.
39. Huys, R., Cañal-Bruland, R., Hagemann, N., Beek, P.J., Smeeton, N.J. and Williams, A.M., Global
Information Pickup Underpins Anticipation of Tennis Shot Direction, Journal of Motor Behavior, 2009, 41,
158-170.
40. Cañal-Bruland, R., Van Der Kamp, J., Arkesteijn, M., Janssen, R.G., Van Kesteren, J. and Savelsbergh,
G.J.P., Visual Search Behaviour in Skilled Field-Hockey Goalkeepers, International Journal of Sport
Psychology, 2010, 41, 327-339.
41. Araújo, D., Davids, K. and Serpa, S., An Ecological Approach to Expertise Effects in Decision-Making in a
Simulated Sailing Regatta, Psychology of Sport and Exercise, 2005, 6, 671-692.
42. Mann, D.L., Abernethy, B., Farrow, D., Davis, M. and Spratford, W., An Event-Related Visual Occlusion
Method for Examining Anticipatory Skill in Natural Interceptive Tasks, Behaviour Research Methods, 2010,
42, 556-562.
43. Dicks, M., Davids, K. and Button, C., Representative Task Designs for the Study of Perception and Action
in Sport, International Journal of Sport Psychology, 2009, 40, 506-524.
102 In-Situ Experiments in Sports
44. Williams, A.M. and Davids, K., Assessing Cue Usage in Performance Contexts: A Comparison between Eye-
Movement and Concurrent Verbal Report Methods, Behaviour Research Methods, 1997, 29, 364-375.
45. Cañal-Bruland, R., Van Der Kamp, J. and Van Kesteren, J., An Examination of Motor and Perceptual
Contributions to the Recognition of Deception from Others’ Actions, Human Movement Science, 2010, 29,
94-102.
46. Sebanz, N. and Shiffrar, M., Bluffing Bodies: Inferring Intentions from Actions, Psychonomic Bulletin &
Review, 2009, 16, 170-175.
47. Shim, J., Carlton, L.G., Chen, D., Steinberg, G.M., and Frehlich, S.G., Visual Search, Anticipation, and
Reactive Comparison between Highly Skilled and Beginning Tennis Players, Journal of Motor Behavior,
2005, 37, 164-175.
48. Bernstein, N.A., The Co-Ordination and Regulation of Movements, Pergamon Press, Oxford, 1967.
49. Savelsbergh, G.J.P. and Van Der Kamp, J., Information in Learning to Co-Ordinate and Control Movements:
Is There a Need for Specificity of Practice?, International Journal of Sport Psychology, 2000, 31, 467-484.
50. Araújo, D., Davids, K. and Passos, P., Ecological Validity, Representative Design, and Correspondence
between Experimental Task Constraints and Behavioral Setting: Comment on Rogers, Kadar, and Costall
(2005), Ecological Psychology, 2007, 19, 69-78.
51. Mctavish, D.G. and Loether, H.J., Social Research: An Evolving Process, Allyn and Bacon, Boston, MA,
2002.
52. Araújo, D., Davids, K., and Hristovski, R., The Ecological Dynamics of Decision Making in Sport,
Psychology of Sport and Exercise, 2006, 7, 653-676.
53. Gibson, J.J., The Ecological Approach to Visual Perception, Houghton Mifflin, Boston, MA, 1979.
54. Davidson, T., The Laser Book, Laser Sailing from Start to Finish, John Wiley & Sons Ltd, Chichester, United
Kingdom, 2009.
55. Milner, A.D. and Goodale, M.A., The Visual Brain in Action, Oxford University Press, Oxford, 1995.
56. Walls, J., Bertrand, L., Gale, T. and Saunders, N., Assessment of Upwind Dinghy Sailing Performance Using
a Virtual Reality Dinghy Sailing Simulator, Journal of Science and Medicine in Sport, 1998, 1, 61-71.
57. Glossary of Sailing Terms, Journal of Sports Sciences, 2007, 25, 1077-1080.
58. Kandil, F.I., Rotter, A. and Lappe, M., Car Drivers Attend to Different Gaze Targets When Negotiating
Closed vs. Open Bends, Journal of vision, 2010, 10, 1-11.
59. Falkmer, T. and Gregersen, N.P., A Comparison of Eye Movement Behaviour of Inexperienced and
Experienced Drivers in Real Traffic Environments, Optometry and vision science, 2005, 82, 732-739.
60. Vickers, J.N., Gaze of Olympic Speedskaters at Full Speed on a Regulation Oval: Perception-Action
Coupling in a Dynamic Performance Environment, in ICSC: Dynamics in Spatial Interactions, 2006.
61. Wilson, M., Stephenson, S., Chattington, M. and Marple-Horvat, D.E., Eye Movements Coordinated with
Steering Benefit Performance Even When Vision Is Denied, Experimental Brain Research, 2007, 397-412.
62. Van Mackelenbergh, R., Dvd Basic Laser Sailing, 2010, Delta Lloyd, Watersportverbond, SailCenter Almere.
63. Panchuk, D. and Vickers, J.N., Gaze Behaviours of Goaltenders under Spatial-Temporal Constraints, Human
Movement Science, 2006, 25, 733-752.
64. Dhami, M.K., Hertwig, R. and Hoffrage, U., The Role of Representative Design in an Ecological Approach
to Cognition, Psychological Bulletin, 2004, 130, 959-988.
65. Withagen, R. and Van Wermeskerken, M., Individual Differences in Learning to Perceive Length by
Dynamic Touch: Evidence for Variation in Perceptual Learning Capabilities, Attention, Perception &
Psychophysics, 2009, 71, 64-75.
66. Farrow, D., Baker, J. and Macmahon, C., Developing Sport Expertise: Researchers and Coaches Put Theory
into Practice, Routledge, London, 2008.
67. Savelsbergh, G.J.P., Van Gastel, P. and Van Kampen, P., Anticipation of a Penalty Kicking Direction Can Be
Improved by Directing Attention through Perceptual Learning, International Journal of Sport Psychology,
2010, 41, 24-41.
International Journal of Sports Science & Coaching Volume 8 · Number 1 · 2013 103
A preview of this full-text is provided by SAGE Publications Inc.
Content available from International Journal of Sports Science & Coaching
This content is subject to copyright.