Aspen Center for Physics: Physics of Behavior

From Ilya Nemenman: Theoretical Biophysics @ Emory
Jump to: navigation, search
Aspen, CO,
May 27 - June 17, 2012
Application/Registration Deadlines


Mission Statement

Traditionally the interface between physics and biology has been defined by a simple fact that complex molecules underlie biological phenomena, and such molecules are a natural subject for physical studies. Now we understand that the role of physics in biology is much broader, as physical constraints define the strategies and the biological machinery that living systems use to shape their behavior in the dynamic, noisy, and resource-limited physical world. Neither the physical constraints imposed by the world, nor the tasks that biological systems trying to survive in it perform depend strongly on their structure. Hence studying behavior of living systems has a potential for a unified understanding, which could bring together various subfields of biological physics, from single molecules, to neural computation.

A classic example of such a system is bacterial chemotaxis -- many bacteria will direct its movements towards certain chemicals. The behavioral strategy of bacterium E. coli approaches an optimal solution for navigating up a concentration gradient in light of well-defined physical constraints due to diffusion. This broad view of chemotactic behavior in E. coli has revealed a deeper understanding of some general design principles of biochemical signaling networks, such as how they can produce large output gain over a wide dynamic range, while remaining robust to fluctuations in protein levels and microscopic rates. The study of chemotaxis has clearly demonstrated that the search for physical principles at the systems-level complement reductionist studies at the molecular level, and that some principles emerge only when studying behavior of the network holistically.

Can a similar understanding emerge for other, more complex living systems? For the proposed workshop we will define "behavior" broadly as an input-output relation between a stimulus and an action of the system. To expose participants to a diverse range of systems and to facilitate the development of theories that span living systems, we will invite experts who quantify behavior of a wide range of model organisms. We will address the behavior of molecular circuits through organisms, focusing on exploring motion as possibly optimal adaptive responses given the physical and the statistical structure of environment. Importantly, since constraints on behavior are imposed by the environment, which in turn is influenced by the behavior, we expect to have a presence of ecologists as well.

We expect that the workshop will be structured around the following themes:

Navigation and foraging
Strategies for navigation and search within local environments and across large distances have been studied since the earliest days of ethology. Evidence exist that navigation strategies adapt optimally subject to relevant physiological and environmental constraints. The modern advent of small, lightweight sensors and fast cameras have led to quantitative data movements of individual organisms from single cells to humans. Developing quantitative approaches to the analysis of behaviors that evolve over long time scales poses a particular challenge due to their potentially high-dimensional structure.
Active Sensing
The strategies with which organisms deploy their array of sensors profoundly influence their experience of the natural world. Biological sensors are noisy and number-limited and thus no organism has the capacity to capture sensory signals with arbitrary precision. Under selection pressure to optimally sense available signals, organisms actively collect information from their environment to guide behavioral decisions. Examples include eye movements, echolocation, whisking and locomotion through odor plumes.
Locomotion and rhythmic behavior
For an animal, movement is not translation or rotation of its body relative to an external coordinate system, but rather transformations of shape as measured in its own intrinsic coordinates. The kinematics of the body and the dynamics of neural control both shape cyclic movements that animals make while walking, flying, swimming, chewing, etc. There has been little work relating these movement primitives to larger-scale functional consequences on interactions between the organism and the surrounding world, and the discussion is overdue.
Adaptive responses
All behavior is in response to external stimulation, and the fidelity of the response is limited by the ability of an organism to learn from its environment. Such responses may be deterministic, but often stochastic components of the behavior serve as a bet-hedging strategy against inaccurate learning, cheaters, or inability to respond. Work is needed to achieve a holistic understanding of deterministic and stochastic response strategies in a single context of adaptation to the surrounding world.


Ila Fiete, UT Austin
Ilya Nemenman, Emory U
Leslie Osborne, U Chicago
William Ryu, U Toronto
Greg Stephens, Princeton U


General Thoughts

This gathering at the Aspen Center for Physics is a workshop, but not a conference. Thus the program will be light and fluid (in fact the organizers are not permitted to schedule dense programs by the Center rules). However, if it becomes clear that we need to schedule additional talks or other events, we will do so.

We will be emphasizing informal interactions among the participants. Thus presentations will be half an hour long, and will have ample discussion time during and following them. We will also reserve time for organizer-facilitated group discussions. Since physics is an experimental subject, there will likely also be a preference for talks by experimentalists over theorists. We also encourage you to talk about new, exciting things that you are working on currently, rather than on well-understood results that have been absorbed by the community already.

Aspen is in the middle of beautiful mountains, so the schedule leaves ample time for recreation. In our experience, some of the best discussions often happen on trails. We will schedule time for group hikes, trying to force the participants away from computers, and into close proximity with each other. Finally, we understand that getting away from busy labs for 3 weeks may be problematic -- and so the schedule will have sufficient time to sustain our e-presence at our homes.

Tentative Schedule

This is the current planned schedule as of May 24. Please expect changes.

10:00 - 11:30 -- Introduction of new participants. Newly arriving participants, if they choose to do so, can give informal introductions of themselves to the rest of the workshop. Length - 10 minutes each. No slides, or any other formal arrangements. To facilitate discussions, we would like to ask you to address the following questions in your introductions:
(a) What are you working on, and how it is related to your "shining star" goal for the next five-ten years?
(b) What's your absolutely most exciting discovery of the last year?
(c) If you are a theorist, which kind of data would you love to get?
(d) If you are an experimentalist, what don't you understand in what you measure?
12:30 - 2:00 -- lunch
afternoon -- unscheduled
10:00 - 12:15 -- Double Feature. Two 30 minute talks separated by a short coffee break. We will try to arrange both talks on related subjects. Speakers: please prepare your talks to last 30 minutes, and the length will double due to questions. Try to be provocative and talk about new things that you don't yet quite understand.
12:30 -- Lunch
afternoon -- unscheduled
5:00 - open ended -- Weekly Aspen Center for Physics BBQ and volleyball.
9:00 -- Group hikes.
12:30 - 2:00 -- Lunch for those staying on campus.
2:30 - 4:45 -- Double Feature. Two 30 minute talks separated by a short coffee break. We will try to arrange both talks on related subjects. Speakers: please prepare your talks to last 30 minutes, and the length will double due to questions. Try to be provocative and talk about new things that you don't yet quite understand.
10:00 - 12:15 -- Double Feature. Two 30 minute talks separated by a short coffee break. We will try to arrange both talks on related subjects. Speakers: please prepare your talks to last 30 minutes, and the length will double due to questions. Try to be provocative and talk about new things that you don't yet quite understand.
12:30 - 2:00 -- Lunch.
3:00-4:00 -- Aspen Center for Physics Colloquium.
morning - Unscheduled
12:30 - 2:00 -- Lunch.
2:00 - open ended -- Group hikes.

Tentative Speakers

The schedule below is based upon the visit schedules provided to the organizers by the Center. If you cannot present on the day we pencilled you in, or if you want to present, but are not listed, please contact the organizers ASAP.

Week 1
Tuesday. The Worm -- David Biron and Chris Fang-Yen
Wednesday. Sensory neuroscience and behavior -- Rob de Ruyter van Steveninck and Mark Nelson
Thursday. Erin Rericha, and Greg Stephens
Week 2
Tuesday. Biomechanics -- Callum Ross and Itai Cohen
Wednesday. Complex neural behavior -- Richard Hahnloser and Leslie Osborne
Thursday. From microbes to psychology -- Tom Shimizu and Joshua Goldberg
Notice that June 8-11 is also the McKnight Neuroscience meeting at one of the nearby hotels, and this may lead to program changes.
Week 3
Tuesday. Sensory Neuroscience and Behavior -- David Kleinfeld and Joshua Shaevitz
Wednesday. Microorganisms -- Douglas Weibell and Robert Austin
Thursday. Neural control of behavior -- Tatyana Sharpee and Will Ryu
Personal tools