This paper is an empirically based contribution on the communication of corrections in synchronized swimming. Our claim is that marking is a socially orgnanized skill that can be found in sports corrections. Conversation Analysis provides the framework to locate the pathways of communication modalities in real professional trainings. Through video-aided ethnographic work, which includes observation and interviewing the Spanish Olympic team for four months, we captured standardized communication patterns. We analyzed the video of the training sessions with ELAN software for micro-interactions. Our results show that the modalities of speech, marking, gesture and gaze appear in synchronized swimming. There is an epistemological asymmetry between the trainer and the swimmers as experts in different domains. Still, we found instances of distributed marking through analysis of gaze behavior. Marking for others in synchronized swimming has a cognitive function that goes beyond individual reflexivity and recall. Corrections in sports training are a product of socially managed turn taking.
Professional environments such as Olympic sports trainings
require continuous adjustments and corrections. How does a
trainer correct the swimmers of an Olympic synchronized
swimming team? Our research question calls to mind the
need for acknowledging the social root of expert
communication. A review of recent theories on the cognitive
abilities of agents
The gap in explaining real world environments applies to
sports trainings such as Olympic Synchronized swimming.
We look at the conversational components of interaction in
training as distributed elements for communication and
thought. Conversation Analysis (CA)
Our analysis is multimodal because corrections go beyond
speech, including other actions such as facial expression
Turn taking indicates the mutual comprehension of what is
being communicated: not only engaging, but also keeping
the conversation going shows the dynamic nature of human
communication. Verbal corrections are prevalent among
dancers as embodied demonstrations that contrast right and
wrong performances
Studies in language, movement and perception
Marking is a common modality to dancers and athletes,
and also musicians
The ethnography of dance rehearsals introduced a third
distinction, related to gaze direction, which is marking for
self or marking for others. Gaze to one’s body indicates the
need for recall or reflection, while gaze to other participant
bodies stands in for coordination or communication. The
separation of the functional, expressive and directive
dimensions allows for a careful analysis of distributed
cognition
We present a CA analysis of a video ethnography of the Spanish Olympic synchronized swimming team. We structured the excerpts by Turn Constructional Units (TCU) as defined by Sacks et al (1978) to analyze the interactions taking place between the trainer and the duet of the Spanish Olympic team. We consider the trainer as well as the swimmers experts in their domains, with superior knowledge about their activity. We provide detailed video analysis of multimodal interactions within the team during training sessions. Detailed knowledge of the observational field was possible through note taking and video capture at the High Performance Center for 4 months in 2012. The observations and daily interviews took place in the spring of 2012, with 2 cameras.
In order to work with the video, we used ELAN®
software (Figure 1). The Max Planck Institute originally
developed ELAN for Psycholinguistics for the analysis of
micro-gestures and interactions, which makes it suitable for
transcription of movement. Our unit of analysis is the turn
taking among the trainers and the swimmers during training,
as part of Activity Relevant Episodes
From this first standardized pattern we selected phase T4a of the training process, where the trainer is correcting one of the participants of the duet (S2) on the correct position of her hip. This snippet is relevant because of its central location, sufficient duration and rich multimodal patterns. The Jeffersonian transcript conventions provide a literal transcription of the interactions line per line; its primary aim is to help the reading of the original interaction in its natural setting (Figure 3). ELAN snapshots complete the accounts for all modalities of communication.
talk Speech is in bold (translated from Catalan) [] Overlapping talk = Latching WHAT Louder voice (()) Described phenomena () String of talk for which no audio could be achieved Multimodal details have been transcribed to the following conventions. Actions (including gesture and marking) are described in the following line and are synchronized with talk thanks to a series of landmarks. ** Delimitates one participant’s actions descriptions (S1) ** ** Delimitates one participant’s actions descriptions (S2) ---> Gesture or action described continues across subsequent lines >--- Gesture or action described begins before excerpt’s beginning ++ Delimitates one participant’s actions descriptions (T) Gaze Description of direction of gaze S1 Participant is identified when she is not the current speaker
In the interaction we analyze here, the trainer (T) is giving instructions to the duet (S1 and S2) from the Spanish Olympic team. The episode lasts 3 minutes and 25 seconds, and produces 15 seconds of the Olympic choreography. The interaction begins with S1 and S2 in the center of the swimming pool. After a 30 second performance, the trainer corrects, by the side of the pool, one of the members of the duet (S2) because her hip is too low. This dynamic is repeated four times, until the trainer considers that the swimmers understood what she wants and thus they move on to the next phase of the choreography.
1.Focused Attention
2. Structured Understanding 3. Prompting action
Olympic swimmers see themselves on film, since the
camera is part of the work process on a daily basis (see
Figure 4). The athletes have both underwater and above
water cameras. The trainer (T) stops the action (the
swimmers’ performance) and immediately shows on the
screens what has gone wrong. The swimmers (S1, on the
left side of the image, is the most experienced swimmer, and
S2, on the right side of the image, the youngest) swim to the
side of the pool and wait for the trainer to show them the
corrections. The trainer selects and views the snippet where
the correction needs to be made based on her expert criteria,
looking at the camera and pressing the rewind button, in
figure 5. This directed action of attention comes with the
manipulation of cognitive tools
01 S1 S1 S2 T 02 03 04
T +presses camera button to forward video+
+Looks at the camera ---> (1.5) * grabs her bottle and drinks --->* * stares at the screen by the swimming pool ---> ** stares at the screen by the swimming pool ---> >---Upward hand gesture --->
+Looks at S2 --->+ T Do you see it? Ok? Look here
+Looks back to camera --->
Pointing to video screen--->+
Figure 6 includes lines 03 to 10: this is the second phase
of the MIS. S2 in line 03 hesitates and pauses while small
marking. In line 04, the trainers’ previous verbal info is
reinforced by 0,3 seconds of showing the step again on the
screen. She manipulates the camera while S2 looks at her
mistake again. Both the trainer and S2 use embodied
modalities: the former’s cognitive tool is the manipulation
of technology and the latter relies on her own body as a tool
for understanding. The trainer structures the information
that the swimmers perceive, disciplining their perceptions
by applying professional vision
S2 Yes but: (2.0) ** small marking right hand for her legs --->** T +Looks at S2--->+
+gaze to camera --->+ + rewinding the video twice with no sound --->+ *with the second rewind, stops drinking --->* *Taking into account that all the time you close S1 marking with both hands for legs, crossing hands --->* S1 *looks at her hands--->* turns to the left -->* 06 T YOU CLOSE, you close:
+FF the video twice with no sound + S1 * looks at screen ---> S2 **grabs the bottle--->** 07 T but you don´t raise the pelvis = +Marking her right arm for S2 leg+ +Looking at S2--> 08 T = You mustn’t lower your pelvis=
+Gaze to her right hand---> +Marking right arm 2 fingers--->+ 09 T = You must close from here= +Rotating right hand 2 fingers, then again--->+ +looks at S2--->+ S2 ** looks at T---> **nods--->** 10 T = You must close (So) in fact this element = +Raising right arm and 2 fingers--->+ +Gaze to her right hand--->+
Then, the trainer shows the wrong position four more
times, correcting verbally (loudly in 06), gesturing and
marking as illustrations for emphasis (upward hand gesture
as a rising leg in 09-10). Held gestures enhance salience and
observability
The third phase of the MIS (27-33s.), which is not
included in this paper, closes with the trainer verbalizing the
strategic objective of obtaining the maximum score in the
Olympic Games. She links this specific correction to a
shared and overreaching goal such as getting a higher
punctuation. Such an utterance sets a common goal for all
the participants. The turn design in phase 2 and 3 of the MIS
shows an asymmetric distribution of local knowledge, or
epistemic asymmetry
In the next phase (34-42 seconds, not included in this paper) S2 follows the trainer’s instructions and performs the corrected move twice, without getting out of the water. Both the trainer and S1 align their bodies in the same angle and direction to watch the performance. S2’s body becomes their common object of perception. S1 self-selects and evaluates S2 by describing what S2 is doing wrong: She is charging her leg. S2 puts her weight on her hip without tensing her leg.
The trainer agrees verbally, and when S2 gets out of the water, in excerpt 3 (Figure 7; 43-73 sec.), she loudly corrects S2, rephrasing in more precise terms what S1 just said, saying what S2 is not doing right (holding her body with her leg) (lines 1-2). The correcting sequence is similar to excerpt 2, with speech and marking as main modalities. Again, when taken as features in turn design, the analysis of gaze enables us to understand the form of participation characteristic of the setting. S1 and S2 look back at the trainer while she talks. T looks back while speaking when wanting to stress the content of what is being said. (2.0) S2 looks at T---> 1 T S2! You are not holding your body with the leg + Vertical right hand towards her own body --->+
S1 *looks at T---> 2 T It is not like that, you turn your hip.
+Marks hands & wide arms ---> +
+Gaze to S1---> +Right arm upper right & lower left +Gaze towards her own hands-->+ 3 T And it stays here:: +Gaze towards S2--> + Scissor cutting with hands & 3 lifts right foot+ 4 S1 Yes *opens wide her arms--->* *Gaze towards S2---> +T gaze towards S1--->+ 5 S1 And you leave the leg: and then you see this thing here *Rotates clockwise, straight arms* Turns & collapses arms facing S2*
T looks at S2 +T moves forward with her body, still looking at S1+ 6 So you because you let this hip go * Aligns with S2, opens arms--->* *Rotating body clockwise, angled arms 7 S1 You have to move back and the leg stays, then *pushing her left arm* *closing both arms* 8 **S2 nods** 9 S1 Yes, where you look the hip remains open hand, straight arm *Looking at her arm ---> *Turning counterclockwise, straight right arm* looks at S2-->
In line 04 a third party, S1, breaks the epistemological
asymmetry a second time. There is an interesting shift in the
trainer’s gaze in line 02 that anticipates the turn: the trainer
looks at S2, conveying a communicative intention as in
excerpt 2, but in 02 she looks briefly at S1. This is a
different situation to what we describe in excerpt 2, where
the trainer interrupts or self selects her turn verbally while
the swimmers are still talking or marking. The trainer in 02
leaves a wide linguistic space, a suspension
S1 continues her turn with an “and” (05) which puts her
as the extension of the trainer’s discourse. She proceeds to
first mark, and then verbalize, the specific micro steps of the
wrong trajectory that involves a low hip, a body that stays in
the wrong place and a leg that is in a weird angle. S1’s
utterance and gaze towards S2 positions her in line with the
trainer (05-09). The swimmer marks the same move, this
time in a larger scale, and she verbalizes in detail the point
where S2’s hip is too low. Her large marking has a
communicative function, as indicated by gaze direction. In
09 the swimmer says yes again, looks at S2 and continues
her large marks in the water. During the turn, the trainer is
looking at S1, and S1 is looking at S2, who gazes back at
S1. Most importantly, the trainer’s gaze follows an
exchange between S1 and S2, and S1 adds verbally the use
of a visual cue (Where you look) to help S2 in keeping her in
the right place (line 09). The trainer waits until S1 finished
large marking and showing the step to S2 to ask for
confirmation from S2 (line 10). S1 is thus a cognitive
extension of the trainer: her marking is both part of
cognitive processing and a communication modality. This is
an instance of marking for others with a cognitive function.
Modalities such as gaze and marking in this particular
arrangement can be considered instances of distributed
cognition
We provide a video ethnography of an Olympic synchronized swimming team with a Conversational Analysis (CA) framework. Our unit of analysis is the turn design of corrections in the institutional context of expert sports training. The conversation fragments that we transcribe and analyze with ELAN software for video analysis of small-scale interactions show how the trainer and the swimmers are experts in their specific domain of action. The semantic reference of the transcribed conversation is a swimmer’s position of the hip. The technical nature of the training process makes it a
Multimodal Interactive System (MIS)
The conversation pattern in phase 2 of the MIS points
towards an asymmetric distribution of local knowledge, or
epistemological asymmetry
Still, we captured empirically how the epistemological asymmetry of corrections in synchronized swimming translates into a different turn design in excerpt 2 and particularly in excerpt 3, where swimmer 1 marks for swimmer 2 under the trainer’s gaze. These are instances of marking both as a cognitive mechanism and as a communication modality. Thus, the link between function and gaze direction in line 05 in excerpt 2 and along lines 0509 in excerpt 3 work differently. Swimmer 1 small marks first, and large marks later, completing and selecting the information given by their trainer. She is both a cognitive extension, since she marks in further detail what the trainer is saying, and an interpreter, because she filters out the relevant information coming from the performance of the other swimmer. Her expertise as a synchronized swimmer, which the trainer lacks, lies in the skilled use of her own body as an underwater tool. While the trainer provides multimodal and embodied guidance (through speech and marking) the most expert swimmer provides key marking, together with evaluative and detailed speech that specifies the corrections further for swimmer 2. Corrections and adjustments are not merely an attribute or skill of a particular expert, but part of an interactive activity. Thus, this level of analysis puts forward the distributed nature of marking, since a third party, the most expert swimmer in the duet, acts as the eyes and the body of the trainer. The presence of this episode of marking is an empirical evidence not only for multimodal communication, but also for distributed cognition.
Corrections become multimodal and part of a complex cognition system of agents, environmental cues, and social rules of communication and interaction. Conversation analysis of corrections in synchronized swimming gives us sufficient empirical evidence to explain the mechanisms through which not only athletes, but also dancers, musicians, scientists and physicians think contextually. Distributed marking is a relevant phenomenon that needs to be studied in wider contexts across different professional fields.