Chaire André Aisenstadt Chair 2007

John Rinzel

(Center for Neural Science and Courant Institute of Mathematical Sciences,
New York University)

Une série de conférences / A series of lectures

Le lundi 17 septembre 2007 / Monday, September 17, 2007

16 h 30 / 4:30 pm

Conférence André-Aisenstadt Lecture
Cette conférence s'adresse à un large auditoire. / Suitable for a general audience.

Dynamics of Visual Bistable Perception: e.g., the Necker Cube

The neural bases for various behaviors are thought to involve competing sub-populations of neurons.  Alternate actions are required for repetitive motor behavior such as respiration and locomotion and in executing the outcome of a decision-making task.  When visualizing an ambiguous scene  (such as the Necker cube) one may perceive ongoing temporal alternation between the possible interpretations. In binocular rivalry each eye views different images but perception alternates randomly between them, with a time scale of seconds.

Various dynamical models lead to alternating mutual exclusivity with neuronal competition implemented as reciprocal inhibition between neuronal populations.  Slow negative feedback sets the basic time scale for switching. We will describe two mechanistic frameworks for the switching behavior.  If the negative feedback is strong enough it can overcome dominance and alternations occur intrinsically and periodically. In an alternate, attractor-based, framework negative feedback is relatively weaker and alternations are induced by noise operating on a bistable system.  Concepts from dynamical systems are applied to understand the underlying mathematical structure.

Centre de recherches mathématiques
Université de Montréal
Pavillon André-Aisenstadt, 2920, chemin de la Tour
Salle / Room 1360 

Une réception suivra la conférence au Salon Maurice-l'Abbé,
Pavillon André-Aisenstadt (Salle 6245).

A reception will follow the lecture in Salon Maurice-l'Abbé,
Pavillon André-Aisenstadt (Room 6245).

Le mardi 18 septembre 2007 / Tuesday, September 18, 2007

9 h 10 / 9:10 am

Conférence dans le cadre de l’Atelier sur les neurosciences mathématiques
Lecture within the Workshop on Mathematical Neuroscience

Timing computations in the auditory brain stem

Sound localization involves precise temporal processing by neurons in the auditory brain stem. The first neurons in the auditory pathway to receive input from both ears can distinguish interaural time differences (ITDs) in the sub-millisecond range.  These cells in the mammalian medial superior olive (MSO) have specialized biophysical features: two dendrites, each receiving input from only one side; very short membrane time constant (less than one ms);  specialized ionic channel properties, including a low-voltage activated potassium current, I-KLT.  This I-KLT contributes to phasic firing (one spike in response to a step of current), precise phase-locking, and extremely timing-sensitive coincidence detection.  We will describe the temporal feature-selecting properties of MSO cells based on biophysical (HH-like) modeling, in vitro (gerbil) electrophysiology and application of concepts from dynamical systems theory and coding theory.

Centre de recherches mathématiques
Université de Montréal
Pavillon André-Aisenstadt, 2920, chemin de la Tour
Salle / Room 6214

Le mardi 25 septembre 2007 / Tuesday, September 25, 2007

11 h / 11:00 am

Conférence dans le cadre de l’Atelier sur la décomposition des réseaux biochimiques
Lecture within the Workshop on Deconstructing Biochemical Networks 

Modeling the rhythmic dynamics of developing spinal cord

Many developing neural systems exhibit spontaneous rhythmic activity: episodes of many neurons firing (say for 10s of seconds) separated by long silent phases. In collaboration with expermentalists (M O'Donovan lab, NIH), we have formulated and studied a set of models that describe the activity patterns in the chick spinal cord, where silent phases can be very long, 10 mins or so. The behavior is network-mediated; a neuron model if isolated does not oscillate episodically.  Slow (synaptic) depression accumulates in the functionally excitatory network to terminate an episode and then recovers between episodes. In early development the usually inhibitory GABA/glycine synapses are effectively excitatory and their driving forces (chloride concentrations) can vary slowly and contribute to the synaptic depression.  We use mean-field models as well as cell-based networks of spiking neurons to understand the dynamics and to design experiments and then analyze results.  The structural framework of the models (including bistability on the fast time scale) allows for fast/slow analysis of the emergent rhythmicity.

Centre de recherches mathématiques, Université de Montréal
Pavillon André-Aisenstadt, 2920, chemin de la Tour
Salle / Room 6214