PHASE 1: IRCAM

Respiration & HRV ~ Movement Models ~ Study #1

Respiration & HRV

In an initial series of experiments, we examined temporal relationality of heart rate with breath, for which we used both ECG and respiration belt sensors. During sessions with two different performers, we identified reproducible patterns of variation in the heart, which relate to holding the breath, deep or shallow panting, slow inhales and exhales, and various other combinations of duration and amplitude in respiration. Importantly, we repeated the breathing structures with each dancer lying supine, sitting up, standing, walking, and eventually improvising movement, and still observed familiar patterns of relation between ECG and respiration data throughout, but with a base increase or decrease in heart rate related to the level of the dancer’s physical exertion.

Below are a series of graphs based on one session of the breathing structure outlined above, during which the performer was lying in supine position on her back. The breathing structure was also recorded with a second dancer, and many shorter tests were conducted with additional volunteers, all of which rendered consistent results.

RESULTS FROM THE RESPIRATION SENSOR 

Both respiration sensors measure the expansion and contraction of the rib cage.
Pink line = Biopic Respiration Belt Transducer
Blue Line = Fabric Respiration Sensor

breathing_lying_respiration

RESULTS FROM THE ECG SENSOR

breathing_lying_heartrate

RESULTS OF RESPIRATION & ECG SENSOR SCALED FOR OVERLAY

breathing_lying_heartrate_respiration

A CLOSER LOOK AT BREATH & HEART ACTIVITY DURING:

“NORMAL” BREATHING

Observations: During the initial two minutes of “normal breathing”, the heart rate remained around 80bmp, with minimal variability between 65-80bpm.

1_breathing_lying_normal_0-120

FAST-DEEP BREATHING

Observations: During two minutes of fast, deep panting the heart rate ascended steadily from 65bpm to 120bmp over the course of the first 30seconds, and sustained this high rate with minimal variability for the remaining duration.

2_breathing_lying_fastdeep_120-240

HOLD BREATH & RECOVER

Observations: With a breath hold for two minutes, the heart rate dropped from 120bpm down to 70bpm over the course of the initial 20 seconds, and sustained this low with extremely minimal variability for 90 more seconds.  Near the end of this graph, there is a gradual rise in heart rate, before a sudden release, and gasp for air.

3_breathing_lying_hold_240-360

SLOW-DEEP BREATHING

Observations: During two minutes of slow, deep breathing, the heart rate presented a wave-like pattern of variability, rising and falling between 80bpm and 60bpm with each inhale and exhale.

5_breathing_lying_slowdeep_480-600

FAST INHALE – SLOW EXHALE

Observations: During two minutes of fast, deep inhales followed by long, full exhales the heart rate presented variability between 85bpm and 60bmp with each breath.  In this pattern, the rise and drop of the heart rate happens very quickly, followed by a sustained low rate during the exhale.

8_breathing_lying_fastinslowout_780-920

Conclusions

The results in the graphs above demonstrate temporal patterns of correlation between breath and heart activity that have been observed consistently during the outlined experiment.  These patterns will guide the choreographer, composer, and performers to craft intentional arcs in heart activity – and therefore musical tempo – over the course of a performance.

Variations for further exploration

  1. Lying down on back – eyes closed
  2. Sitting down cross-legged – eyes closed
  3. Standing, neutral – eyes open
  4. Walking (continuous, even pace) – eyes open
  5. Movement improvisations – experiment with eyes open and closed
    1. Standing, upper body gestures only
    2. Standing, lower body gestures only
    3. Full body movement on spo
    4. Full body movement, travelling (all vertical)
    5. Full body movement, level change
    6. Full body movement on floor

 

Movement Models

Purpose
Observe and measure with an ECG the ability of dancers to achieve intentional arcs in heart activity during movement improvisation, based on a prescribed score.

Instructions
-Dancer wears ECG, and improvises movement.
-Dancer receives prompts to accelerate or decelerate heart rate, based on a prescribed score.
-ECG data is recorded to create a graph that illustrates the heart activity over time.
-ECG data from improvisation is overplayed to compare with the model score.
-Each structure is repeated multiple times to access reproducibility of results.

Outcomes & Observations

Please Note: The ECG data in the graphs below is more noisy than in other images on the website, as these experiments were carried out very early in the research process before improvements to the software had been implemented.

ORIGINAL MODEL A: 
60sec              Sustain heart rate at 80bpm
90sec              Raise heart rate to 180bpm
90sec              Lower heart rate to 80bpm

80_180_80-model

IMPROVISATION 1A: Follows the score quite accurately. During the descent there is a upwards arc before the heart rate lowers.
80_180_80-run1

 IMPROVISATION 1A & 2A: Different movement, but similar results in heart rate behaviour.
80_180_80-runs1-2

ADAPTED MODEL A: Based on these observations, one choice would be to adapt the model score (as seen below), such that it correlates more closely to the temporal patterns being observed in of the heart rate of this dancer.  Alternatively, different movement and breath strategies could be explored to adhere to the prescribed pathway.

80_180_80-runs1-2-adj

ORIGINAL MODEL B.
60sec              Sustain heart rate at 180bpm
90sec              Lower heart rate to 80bpm
90sec              Raise heart rate to 180bpm
180_80_180-model

IMPROVISATION 1B: Follows the score quite accurately, but does not reach the lower heart rate goal of 80bpm at 150seconds.

180_80_180-run1

IMPROVISATION 1B & 2B RESULTS: Different movement, but again the dancer does not reach the lower heart rate goal of 80bpm at 150seconds.

180_80_180-runs1-2

ADAPTED MODEL B: Based on the results from Model B, the score could be adapted (as seen below), to reflect the activity of the heart rate of this dancer, or, alternate strategies for heart rate regulation could be explored.
180_80_180-runs1-2-adj

Conclusions

Movement exertion and relaxation can be an effective strategy for heart rate regulation, based on a prescribed score.  The design of intentional arcs in heart activity of dancers will require further performance-based research to develop realistic models and expectations for each performer.  Prescriptive scores may also be used to challenge the potential highs and lows of heart rate to extremes for each performer.

Notes for further inquiry
-Integrate respiration sensor to evaluate relationship of movement and heart rate to breath.
-Create much longer structures, to explore sustained periods of extreme highs or lows, as well as long transition periods
-Evaluate more closely the function of biofeedback as a tool for interoception for the dancer

 

Study #1

This 10-minute solo performance study draws on observations from the experiments regarding respiration and movement scores in order to compose intentional arcs the heart rate of a dancer – and therefore the musical tempo – over the course of a piece. During this solo, the dancer wears the wireless ECG unit, and her real-time heart rate data is processed to inform temporal characteristics and pitch in the electronic music.

The goals of this study include to:

  • stress the potential of the ECG prototype, in terms of reliability, durability, latency, and wearability;
  • evaluate the effectiveness of the beat tracking and QRS classification algorithms in the software to provide clear data during dynmaic, full-bodied movement;
  • choreograph an intentional and reproducible arc in the heart rate of the dancer, including patterns of heart rate variability, via breath and movement; and
  • explore music as a source of biofeedback for the dancer, in terms of tempo and pitch.

The model below is the structure from which choreography and composition was developed. Although in this model the line indicating the heart rate of the danger is straight, MacCallum and Naccarato designed the score with particular attention to patterns of heart rate variability during different sections, as well as the processes of transformation between each temporal state for the dancer and music.

SOLO_MODEL_annotated

Below is a labelled graph that demonstrates the ECG results from one performance of the solo.  The heart rate of the dancer, Teoma Naccarato, follows the score fairly closely, with expected patterns of heart rate variability based on breath and movement.  Importantly, the arcs in heart rate tempo and variability from this solo have been repeated numerous times by Naccarato, as well as by a second dancer, Bekah Edie (scroll down for additional charts).

Solo, Dancer A (Teoma Naccarato) – Run 01

model_teoma.graffle

Solo, Dancer A – Run 02

model_teoma1-20141209-19-19-18

Solo, Dancer A – Run 03

model_teoma.graffle

 –

Solo, Dancer A – Run 04

model_teoma.graffle

 –

Solo, Dancer B (Bekah Edie) – Run 01

model_bekah.graffle

 –

Solo, Dancer B – Run 02

model_bekah.graffle

 –

Solo, Dancer B – Run 03

model_bekah.graffle

 –

Solo, Dancer B – Run 04

model_bekah.graffle