Coordination between motor units in an expiratory muscle in the Bengalese Finch across timescales Open Access
Conn, Rachel (Spring 2022)
Abstract
Motor behaviors result from the impressive coordination of many neural signals
and the eect of those signals on muscles in the body and its environment. Recent
work implicates sub-behavioral timescale spike patterns for the control of motor
behaviors. Whether sub-behavioral spike patterns are important for the coordination
between motor units within a muscle remains an open question. We asked
whether motor units within a songbird expiratory muscle exhibit coordination at various
timescales during an anesthetized breathing behavior. We recorded from four
motor units in the EXP of a Bengalese Finch while recording air sac pressure. We
segmented spike times of all motor units into breath cycles dened by the air sac
pressure waveform. We estimated the mutual information between the spike patterns
of all pairwise combinations of motor units for three combinations of timescales: spike
counts of each motor unit, the spike counts of one motor unit and the spike timings of
another, and the spike timings of each motor unit. All pairs of motor units exhibited
coordination in their spike counts, the spike timings of some motor units exhibited
coordination with the spike counts of some other motor units. We were too data
limited to draw conclusions about whether motor unit pairs were coordinated in their
spike timings. We noticed that the spike patterns of each motor unit varied greatly
throughout the data collection, so we split the data set into two consecutive halves
and repeated spike count mutual information analysis for three example motor unit
pairs for each consecutive half separately. We found that the coordination between
each example motor unit pairs changed from the rst half to the second half of the
data collection. These results illustrate variability in the coordination between motor
units across timescales, emphasize the importance of developing novel experimental
techniques to increase data sample sizes and computational techniques to analyze
under-sampled data, and highlights the propensity for neural data to exhibit nonstationarities
even during a consistent behavior.
Table of Contents
1 Introduction 1
1.1 Quantitative theories of motor control . . . . . . . . . . . . . . . . . 2
1.1.1 Motor cortex: The representational view . . . . . . . . . . . . 3
1.1.2 Motor cortex: The dynamical systems view . . . . . . . . . . . 6
1.2 Bridging motor control theories with what is known about biomechanics 8
1.2.1 Muscles: The size principle . . . . . . . . . . . . . . . . . . . . 9
1.2.2 Muscles: Hill-type models . . . . . . . . . . . . . . . . . . . . 11
1.3 Motor control via sub-behavioral, millisecond timescale changes in
spike patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.4 Experimental Approach . . . . . . . . . . . . . . . . . . . . . . . . . 14
2 Methods 16
2.1 Surgical Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2 Data Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.1 EMG Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.2 Pressure Data: Identifying Breath Cycle Onset Times . . . . . 20
2.2.3 Pressure Data: Omitting False Cycles and Outliers . . . . . . 21
2.3 Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3.1 Overview of Mutual Information and the KSG Method . . . . 23
2.3.2 Validating Stability, Estimating Error, and Parameter Selection
for the MI Estimates . . . . . . . . . . . . . . . . . . . . . . . 28
2.3.3 Mutual Information to Identify Coordination Between the Activity
of Pairs of Motor Units . . . . . . . . . . . . . . . . . . 30
2.3.4 Error Propagation for Conditional MI Estimates . . . . . . . . 32
3 Results 36
3.1 Motor units exhibit unique spiking properties that vary throughout the
data collection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.2 Spike patterns of dierent motor units are coordinated in their spike
counts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.3 The spike timings of some motor units may be coordination with the
spike counts of others, but no coordination between the spike timings
of pairs of motor units was detected. . . . . . . . . . . . . . . . . . . 40
3.4 The structure of motor unit coordination may change throughout the
data collection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4 Discussion 49
4.1 Motor unit coordination at sub-behavioral and mixed timescales . . . 49
4.1.1 Case 1: These motor units are coordinated at mixed and/or
sub-behavioral timescales, but we were unable to detect enough
coordination. . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.1.2 Case 2: Lack of coordination at mixed and subcycle timescales 56
4.2 Motor unit coordination across timescales . . . . . . . . . . . . . . . 58
4.3 A Challenge to the idea of a static code for motor control . . . . . . . 61
4.3.1 Nonstationarities in the coordination between motor units . . 62
4.3.2 Addressing non-stationarity concerns for motor control . . . . 64
4.4 Conclusions and future directions . . . . . . . . . . . . . . . . . . . . 68
Appendix A Proof of Upperbound on Error for Spike Timing MI Es-
timates 72
A.1 Mixed Timescales: Timing and Count . . . . . . . . . . . . . . . . . . 72
A.2 Subcycle Timescales: Timing and Timing . . . . . . . . . . . . . . . . 77
Bibliography 82
About this Master's Thesis
| School | |
|---|---|
| Department | |
| Subfield / Discipline | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Keyword | |
| Committee Chair / Thesis Advisor | |
| Committee Members |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Coordination between motor units in an expiratory muscle in the Bengalese Finch across timescales () | 2022-03-14 17:10:47 -0400 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|