Information processing in the lateral-line system of fish [Elektronische Ressource] / Julie Goulet

Julie Goulet

Sujets

• Physik

Informations

• Publié par: technische_universitat_munchen
• Publié le: 01 janvier 2010
• Nombre de lectures: 190
• Langue: English
• Poids de l'ouvrage: 33 Mo

Extrait

Information Processing in the
Lateral-Line System of Fish
Julie Gouletcover illustration:
Copyrightc Pepin van Roojen
The Pepin Press, Art Nouveau
(Amsterdam, 2008) # 186
AThis dissertation has been written in LT XE
using the memoir class. Typesetting was done
Awith pdfLT X.E
Funding has been provided by the Bernstein
Center for Computational Neuroscience
(BCCN) { Munich.Physik Department
Technische Universität München
Information Processing in the
Lateral-Line System of Fish
Julie Goulet
Vollst andiger Abdruck der von der Fakult at fur Physik der Technischen Universit at
Munc hen zur Erlangung des akademischen Grades eines Doktors der Naturwissen-
schaften genehmigten Dissertation.
Vorsitzender Univ.-Prof. Dr. A. Bausch
Prufer der Dissertation 1. Dr. J. L. van Hemmen
2. Jun.-Prof. Dr. J. Engelmann
Universit at Bielefeld
Die Dissertation wurde am 08.02.2010 bei der Technischen Universit at Munc hen
eingereicht und durch die Fakult at fur Physik am 23.02.2010 angenommen.Preface
he nervous system enables animals to get information and react to their environ-Tment. Sensory systems are the doorways through which animals get information
about their surroundings. How they handle this information is crucial and permits
animals to create an inner representation of the outside world. In neuroscience
therefore studying sensory systems is always an important source of information on
how the brain works.
In addition to the traditional human \ ve senses" (touch, vision, hearing, smell,
and taste), there are many other sensory systems that deal exclusively with body-
internal processes. For instance, the vestibular system which deals with equilibrium
and position in space. Other systems specialize in order to analyze the physical
information contained in a particular physical quantity available in the environment
of an animal. Many more sensory systems have evolved in other animals.
Aquatic vertebrates use a special sensory system to detect minute water move-
ments: the lateral line system. The lateral-line system decodes the information in
the water surrounding the sh to enable the sh to identify and localize the source
of the perturbation. This thesis deals with how a sh can analyze and extract the
important information contained in the water motion around its body to create a
representation of its environment.
Analyzing water motion is complicated. Unlike light for vision, a moving object
in water travels in a continuous medium at smaller velocity. Important parameters
such as distance, viscosity, boundary layers must be taken into account.
Organization of this thesis
After reviewing the actual knowledge of the anatomy and the physiology of the
sh lateral-line system in chapter 1, a major part of this thesis will show which
quantities are measured by the lateral-line detectors (neuromasts) and that a minimal
model based on the Euler equation can explain accurately the stimuli to the sh
lateral-line ( chap. 2 and 3). This model will be compared with experimental
measurements (chap. 3). We will then characterize (chap. 4) the response of the
vPreface
lateral-line neuromasts on the skin of the sh to white noise. Using information
theory, we will try to characterize which features of the stimulus (displacement,
velocity, acceleration) play a role in spikes generation. Chapter 5 will be devoted
to how the position of an object can be encoded and decoded by the lateral line
periphery and in the brain of the sh. In chapter 6, we continue with the question
of integration by showing how the lateral input can form a map compatible with the
retinotopic map to be integrated in a multimodal representation in the optic tectum.
Finally, chapter 7, is an in depth discussion of the results presented in this thesis.
Thank you
Many people have been of help during my time at the Physik Department. First I
would like to thank Prof. J. Leo van Hemmen, my thesis supervisor. His love and
motivation for physics and neuroscience are most impressive. He convinced me to
come to Munich and do my doctoral studies here, in this wonderful city. Scienti cally,
I also have a great debt to Prof. Pierre Depommier from the Universite de Montreal.
He encouraged me to undertake a scienti c career, and his knowledge of sciences and
culture helped me to nd my place in the sciences.
I also want to thank my biological collaborator: Prof. Jacob Engelmann, a good
neurophysiologist. The collaboration with him was bene cial (and I hope will last).
In a more general way, I want to thank Prof. Horst Bleckmann, who invited me to
visit his laboratory in Bonn and supported this collaboration with people from his
laboratory.
Writing this thesis would not have been possible without all my colleagues at the
\T35" chair. In particular, Christine Vossen and Maria Suttner, who shared (for a
time) an o ce with me. I also want to thank the Bernstein Center for Computational-
Neuroscience Munich for funding (Dr Isolde von Bulo w, Dr Dagmar Bergmann-Erb).
I have had the opportunity to learn many things in Munich. I am grateful
to the Studentinnenheim Theresianum where I passed four wonderful year (Ms
Emanuela Romero and Maria Conception Baque), the KHG Leopoldstrasse 11,
my zen meditation group (P. Dr Stefan Bauberger SJ). In particular, the contact
with Stefan Bauberger (also a physicist) really in uenced my vision of the world
as a physical and spiritual entity. I want to thank,all the people I met during this
adventure of writing a PhD in the Bavarian capital. Many are really special to me,
some of them became really good friends (Zoe, Christine, Kerstin, J org, Dominik,
etc.). A special thanks also to Prof. J. Leo van Hemmen, Mr Klaus Hardtke, Prof.
Jacob Engelmann and Dr Moritz Franosch for careful editing.
viPreface
Finally, I want to thank my father, Dr Ronald Goulet, to whom I dedicate
this thesis. He shared with me a love for science in general and computational
neuroscience in particular. He has been a major source of support psychologically
personally, scienti cally and nancially during all my life and particularly during my
studies.
Munich, January 2010
viiContents
Preface v
Contents viii
1 Function, Anatomy and Physiology 1
1.1 Anatomy and Physiology of Detectors . . . . . . . . . . . . . . . . . . 2
1.1.1 Hair Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.2 Mechanics of the Cupula . . . . . . . . . . . . . . . . . . . . . 5
1.2 Anatomy and Physiology of the A erent Nerves . . . . . . . . . . . . 5
1.3 Central Processing of the Lateral Line Projection . . . . . . . . . . . 6
1.3.1 Central Projection of the Lateral-Line System . . . . . . . . . 6
1.3.2 Lateral-Line Maps . . . . . . . . . . . . . . . . . . . . . . . . 7
2 Hydrodynamics of the Stimulus 11
2.1 Equations of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.2 Boundary Layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.1 Neuromasts as Lateral-line Detectors . . . . . . . . . . . . . . 15
2.2.2 Pressure within a Boundary Layer . . . . . . . . . . . . . . . . 16
2.2.3 Velocity Field Within a Boundary Layer . . . . . . . . . . . . 21
2.3 Outer Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.4 Stimulus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.5 E ect of the Fish Body . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.5.1 Fish Body Within a Constant Flow . . . . . . . . . . . . . . 25
2.5.2 Dipole Oscillating Near the Fish Body . . . . . . . . . . . . . 27
2.5.3 Constant Flow and Dipole . . . . . . . . . . . . . . . . . . . . 29
2.5.4 Numerics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.5.5 Numerical Simulation . . . . . . . . . . . . . . . . . . . . . . . 31
3 Minimal Model 35
3.1 Vibrating Sphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.1.1 Comparison Between Modeled and Measured Data . . . . . . 39
3.1.2 Constant Flow . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.2 Translating Sphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
viiiContents
3.2.1 Comparison Between Modeled and Measured Data . . . . . . 42
3.3 Wake Tracking and Detection of Vortex Ring . . . . . . . . . . . . . . 45
3.3.1 Comparison between Modeled and Experimental Data . . . . 45
4 Response to Noise: Coding at High Precision in the Velocity
Regime 49
4.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.2 Encoding White Noise . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.3 Linear Reconstruction of Spike Trains . . . . . . . . . . . . . . . . . . 53
4.4 Time or Rate Coding? . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.5 Linearity of the Reconstruction . . . . . . . . . . . . . . . . . . . . . 60
4.6 Estimating Nonlinearity Using a Covariance Matrix Analysis . . . . . 62
4.7 Extraction of the FI Function . . . . . . . . . . . . . . . . . . . . . . 66
4.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5 Localization: Determining Distance Perpendicular to the Detec-
tors 71
5.1 Distance Determination by SNs . . . . . . . . . . . . . . . . . . . . . 71
5.1.1 Velocity Parallel to the Fish Body . . . . . . . . . . . . . . . . 72
5.1.2 Velocity Perpendicu