COSYNE 2012 Workshops February 27 & 28, 2012 Snowbird, Utah

Monday, February 27 Organizer(s) Location

1. Coding and Computation in visual short-term memory. W.J. Ma Superior A

2. Neuromodulation: beyond the wiring diagram, M. Higley Wasatch B adding functional flexibility to neural circuits.

3. Sensorimotor processes reflected in spatiotemporal Z. Kilpatrick Superior B of neuronal activity. J-Y. Wu

4. Characterizing neural responses to structured K. Rajan Wasatch A and naturalistic stimuli. W.Bialek

5. Neurophysiological and computational mechanisms of D. Freedman Magpie A categorization. XJ. Wang

6. Perception and decision making in rodents S. Jaramillio Magpie B A. Zador

7. Is it time for theory in olfaction? V.Murphy Maybird N. Uchida G. Otazu C. Poo

Workshop Co-Chairs Email Cell Brent Doiron, Pitt [email protected] 412-576-5237 Jess Cardin, Yale [email protected] 267-235-0462

Maps of Snowbird are at the end of this booklet (page 32).

COSYNE 2012 Workshops February 27 & 28, 2012 Snowbird, Utah

Tuesday, February 28 Organizer(s) Location

1. Understanding heterogeneous cortical activity: S. Ardid Wasatch A the quest for structure and randomness. A. Bernacchia T. Engel

2. Humans, neurons, and machines: how can N. Majaj Wasatch B psychophysics, physiology, and modeling collaborate E. Issa to ask better questions in biological vision J. DiCarlo

3. Inhibitory synaptic plasticity T. Vogels Magpie A H Sprekeler R. Fromeke

4. Functions of identified microcircuits A. Hasenstaub Superior B V. Sohal

5. Promise and peril: genetics approaches for systems K. Nagal Superior A neuroscience revisited. D. Schoppik

6. Perception and decision making in rodents S. Jaramillio Magpie B A. Zador

7. Is it time for theory in olfaction? V. Murphy Maybird N. Uchida G. Otazu C. Poo

Workshop Co-Chairs Email Cell Brent Doiron, Pitt [email protected] 412-576-5237 Jess Cardin, Yale [email protected] 267-235-0462

Schedule Each workshop group will meet in two sessions from 8 to 11 am and from 4:30 - 7:30 pm. Workshop summaries and schedules are available starting on page 3 of this booklet.

Transportation Marriott Downtown to Snowbird: Free shuttle provided for registered attendees (leaves @ 5pm on Sunday, 27-Feb). Snowbird to Salt Lake City Airport: Shuttle can also be arranged at Snowbird, or online at: https://store.snowbird.com/products/index.php?product_category_idx=2.

Further information about transportation to/from Snowbird is available at: http://www.snowbird.com/about/accessibility.html For further information on transportation or other logistics please contact Denise Soudan ([email protected]).

Lift tickets Discounted workshop rates Snowbird Chairlifts only: $62 Snowbird Tram & Chairlifts: $72 Pick up at the Cliff ticket window (level 1 of the Cliff Lodge next to the ski rental shop) or at the ticket window on the top level of the Snowbird Center (the plaza deck).

Meals included with registration Breakfast (Day 1 and Day 2) - The Cliff Ballroom Dinner (Day 2) - The Cliff Ballroom Coffee breaks during morning and afternoon sessions

3 Monday, Feb 27 1. Coding and computation in visual short-term memory

Organizer Wei Ji Ma, Baylor College of Medicine

Visual short-term memory (VSTM) is essential for detecting changes between visual scenes, integrating information across eye fixations, and planning reaching movements. Like other forms of short-term memory, it is generally thought to be resource-limited. As the number of objects to be remembered increases, observer performance on VSTM tasks typically decreases rapidly.

A key theoretical debate centers around whether VSTM resource is a discrete or a continuous quantity, and related to this, how the encoding precision of stimuli in VSTM should be modeled. In recent years, converging efforts in psychophysics, systems neuroscience, and mathematical modeling have furthered our understanding of VSTM coding and limitations. Single-neuron recordings in parietal and frontal areas have revealed neural signatures of limited resources, new developments in fMRI are making the decoding of the contents of VSTM possible, and models have reconceptualized VSTM tasks like change detection as probabilistic inference problems. The workshop will present a broad overview of recent developments in the study of VSTM.

4 Monday, Feb 27 Coding and computation in visual short-term memory

Morning Session

8:20-8:30am Introduction

8:30-9:00am Ed Awh. Discrete resource limits during encoding, selection and storage

9:00-9:30am George Alvarez. The structure and content of working memory representations

9:30-10:00am Coffee Break

10:00-10:30am Wei Ji Ma. Visual short-term memory limitations from variable, continuous resources

10:30-11:00am Earl Miller. Neural dynamics for working memory and cognitive capacity

Afternoon Session

4:30-5:00pm Martin Pare. A change detection approach to study visual working memory in the macaque monkey

5:00-5:30pm Frank Tong. Decoding reveals the contents of visual working memory in early visual areas

5:30-6:00pm Chris Sims. Information theory and visual working memory: A new (old) theoretical foundation for modeling capacity limits

6:00-6:30pm Coffee Break

6:30-7:00pm Robert Jacobs. A probabilistic clustering theory of the organization of visual short-term memory

7:00-7:30pm General discussion

5 Monday, Feb 27

2. Neuromodulation: Beyond the wiring diagram, adding functional flexibility to neural circuits

Organizer Michael J. Higley, Yale University

Flexibility is a critical property of neural circuits, allowing adaptive changes in activity over a range of time scales and enabling context-dependent behavior. While technological advances have enhanced our understanding of the brain’s “wiring diagram”, increasing attention is being paid to processes that confer dynamic flexibility to established circuits. One such example is neuromodulation, a ubiquitous nervous system phenomenon that spans the phylogenetic spectrum of research models.

In contrast to classical fast neurotransmitters that directly excite or inhibit postsynaptic neurons, neuromodulators alter the intrinsic membrane properties of cells and modify synaptic transmission. The actions of modulators like acetylcholine, dopamine, and norepinephrine are critical for cognitive functions including working memory, focused attention, and action selection. Additionally, perturbation of modulatory systems is strongly implicated in psychiatric and neurological illnesses such as schizophrenia, depression, Parkinson’s and Alzheimer’s Diseases.

Despite overwhelming evidence for their general importance, the specific cellular and network mechanisms by which neuromodulators influence behavior remain elusive. However, a surge in technological, experimental, and computational advances have opened up new avenues towards understanding the actions of neuromodulators in vitro and in vivo. In this one-day workshop, we bring together neuroscientists working across multiple levels of analysis to discuss recent findings from cellular, circuit, behavioral, and computational perspectives. Key focal questions for discussion will include:

(1) What are the relevant time scales for neuromodulatory actions? (2) Is neuromodulation best understood at the cell, circuit, or organismal level? (3) Can activity in distinct neuromodulatory systems be causally linked to specific behaviors? (4) What new methodologies are being developed for the study of neuromodulation? (5) How are neuromodulatory processes best incorporated into computational models?

6 Monday, Feb 27

Neuromodulation: Beyond the wiring diagram, adding functional flexibility to neural circuits

Morning Session

8:15-8:30am Michael Higley. Introduction

8:30-9:00am Daniel Durstewitz. Dopamine modulation of prefrontal cortical attractor states revealed by multi-single unit recordings

9:00-9:30am Jeremy Seamans. What do we really know about phasic signaling in the mesocorticolimbic pathway?

9:30-10:00am Coffee Break

10:00-10:30am Xiao-Jing Wang. Complexity of neuromodulation in the prefrontal cortex

10:30-11:00am Randy Bruno. Modulating local circuit dynamics versus long range inputs

Afternoon Session

4:30-5:00pm Alexander Thiele. Cholinergic and glutamatergic control of attention in V1: a double dissociation

5:00-5:30pm Martin Sarter. Cholinergic double-duty: top-down control of cortical glutamatergic-cholinergic transients to optimize cure- detection

5:30-6:00pm Michael Higley. Co-release of GABA and acetylcholine: bidirectional control of cortical interneurons

6:00-6:30pm Coffee Break

6:30-7:00pm Karl Deisseroth. Optogenetic neuromodulation: tools and applications

7:00-7:30pm General Discussion and Wrap-up

Monday, Feb 27 3. Sensorimotor processes reflected in spatiotemporal patterns of neuronal activity

Organizers: Zackeray Kilpatrick, University of Pittsburgh Jian-Young Wu, Georgetown University

Abstract Cortical sensory and motor processes are often subserved by spatiotemporal patterns of neuronal activity. Analysis of such activity often considers coarse-grained statistics of the discharges of neuronal networks involved. Recent advances in voltage sensitive dyes, optical imaging, and multi-electrode recording allow population activity to be studied both in vivo and in slice preparations with excellent spatiotemporal resolution. The genesis of coherent neural activity can then be explored using external stimuli, receptor antagonists, and even optogenetic manipulation. Results from these studies then guide experimental and computational models that consider the role dynamics play in functions such as movement preparation, visual processing, and working memory.

Studying spatiotemporal activity in the brain also allows one to consider the functional role of interactions between multiple brain areas. For example, there has been a lot of recent interest in the role thalamocortical interactions play in sleep and hippocampal- cortical interactions play in learning and memory. These rich data sets offer a plethora of new opportunities for computational modeling to fill in the missing pieces of experimental observation. This workshop will bring together a number of leading researchers to discuss results at the forefront of this multi-disciplinary field.

Monday, Feb 27

Sensorimotor processes reflected in spatiotemporal patterns of neuronal activity

Morning Session

8:30-9:00am Bard Ermentrout. The interactions of noise with spatiotemporal patterns

9:00-9:30am Jian-young Wu. Spatiotemporal patterns of visually evoked cortical activity

9:30-10:00am Coffee Break

10:00-10:30am David Senseman: Karhunen-Loève (KL) Decomposition of Cortical Waves: What it means and what it doesn't mean

10:30-11:00am Paul Bressloff. Neural field model of binocular rivalry waves

Afternoon Session

4:30-5:00pm Nicholas Hatsopoulos. A potentiel role of motor cortical waves in spatio-temporal coordination

5:00-5:30pm Carina Curto. Spatially structured networks from sequences

5:30-6:00pm Brian Lustig. Voltage sensitive dye imaging reveals shifting spatiotemporal spread of whisker-induced activity in rat barrel cortex

6:00-6:30pm Coffee Break

6:30-7:00pm Zachary Kilpatrick. Response of coherent neural activity to external stimuli

7:00-7:30pm Ila Fiete. Spontaneous self-organization of networks to support spatiotemporal activity sequences through spike- time dependent plasticity

Monday, Feb 27

4. Characterizing Neural Responses to Structured and Naturalistic Stimuli

Organizers Kanaka Rajan and William Bialek, Princeton University

The efficiency by which the nervous systems of organisms capture features of the natural world through neural responses, can be thought of as resulting from an adaptation to the statistical properties of their environment. Recent years have seen substantial growth in the development of methods for the statistical analysis of neural data, yielding results on the optimized structure of receptive fields of visual neurons, for instance. These results not only highlight the efficient coding principle, they also emphasize the idea that biological resources have to be allocated optimally so as to transmit and represent information in the most efficient manner possible. While the concept is generally accepted, we lack a systematic framework to understand the mechanisms underlying the efficient processing of natural sensory information. We can express this paucity in terms of the following questions.

1) How do we develop useful descriptions of the multivariate statistics of natural or synthetic structured sensory stimuli? 2) How do we understand the nonlinearities in the information processing mechanisms at the level of populations of neurons? 3) How can we utilize our knowledge of the emerging relationship between the two to develop better tools for analyzing multidimensional neural data?

It is the perfect time to initiate a discussion on these issues because of the recent advances in parameterizing natural stimuli, designing generative models for correlated stimulus ensembles, and developing analysis methods for population responses, many of which have been made by our invited speakers. We believe that bringing together theorists and experimentalists in this way, will help elucidate a list of goals for the next 3-5 years and forge new collaborations.

Monday, Feb 27

Characterizing Neural Responses to Structured and Naturalistic Stimuli

Morning Session

8:00-8:30am Jonathan Pillow. Estimating flexible, hierarchical models of neural responses with Empirical Bayes

8:35-9:05am Maria N Geffen. Encoding of natural sounds in the primary auditory cortex

9:30-10:00am Coffee Break

10:05-10:35am William Bialek. Keynote

10:40-11:10am Olivier Marre. Reading the population code of the retina

Afternoon Session

4:35-5:05pm Tatyana Sharpee. Minimal models of neural responses to natural stimuli

5:10-5:40pm Jonathan Victor. How high-order image statistics shape cortical visual processing

6:00-6:30pm Coffee Break

6:35-7:05pm Gasper Tkacik. Higher-order interactions and critical behavior in networks of neurons responding to natural stimuli

7:05-7:35pm Vijay Balasubramanian. Retinal adaptive decorrelation of spatial stimuli with varying correlation structures

Monday, Feb 27

5. Neurophysiological and Computational Mechanisms of Categorization

Organizers: David Freedman, University of Chicago Xiao-Jing Wang, Yale University

This workshop is focused on bringing together researchers working to understand the brain mechanisms of visual categorization and category-based decision making. This has been an extremely active area of research over the past decade, and the researchers that have agreed to participate in the proposed workshop have been central to many of the recent advancements in our growing understanding of this fundamental perceptual and cognitive process. The goal of the workshop is to bring together experimentalists (working with both non-human and human model systems) and those using computational/theoretical approaches. This should make for lively discussion and will also broaden the appeal of the workshop to a wide audience.

Monday, Feb 27

Neurophysiological and Computational Mechanisms of Categorization

Morning Session

8:00-8:05am Welcome

8:05-8:30am David Freedman. Neurophysiological mechanisms of visual categorization and decision making

8:35-9:00am Jamie Fitzgerald. Generalized associative representations in lateral intraparietal area

9:05-9:30am Earl Miller. Categories and adaptive coding in the prefrontal cortex

9:30-10:00am Coffee Break

10:00-10:25am Matthew Chafee. Executive control of categorization in prefrontal cortical networks . 10:30-11:00am Yale Cohen. Neural correlates of auditory categorization

Afternoon Session

4:30-4:55pm Nikolaus Kriegeskorte. Primate IT: a continuous object space that emphasizes behaviourally-relevant categorical divisions

5:00-5:25pm F. Gregory Ashby. Human category unlearning

5:30-5:55pm Vincent Ferrera. Modeling the effects of reward on category learning

6:00-6:30pm Coffee Break

6:35-7:00pm Xiao-Jing Wang. Multiple neural processes underlying categorization: a computational model

7:00-7:30pm General discussion

Monday, Feb 27

6. Perception and decision making in rodents

Organizers: Santiago Jaramillo and Anthony M. Zador, Cold Spring Harbor Laboratory

Recent technological advances in electrophysiology, microscopy and optogenetics provide unprecedented opportunities to dissect neural circuits in the awake preparation. How do these neural circuits underlie perception, attention, decision making, and other cognitive processes? Here we focus on recent advances in developing rodent models of these behaviors. These models combine decades of work in rodent behavior with new approaches originally developed to study primate behavior.

By bringing together laboratories on the cutting edge of these approaches, we will address the following questions:

1. How can novel imaging/physiological/optogenetic technologies be applied to uncover mechanisms in the awake rodent? Are some behavioral paradigms (e.g., head-fixed vs three-port two-alternative choice) more amenable to various technologies than others?

2. What cognitive processes can be adequately modeled with existing rodent behavioral paradigms? What challenges remain in developing new paradigms?

3. How well do insights gained from studying neural mechanisms in rodents apply to humans and non-human primates?

Monday, Feb 27

Perception and decision making in rodents

Morning Session

8:00-8:20am Introductory remarks

8:25-8:45am Cris Niell. Visual processing and behavioral state in the mouse

8:55-9:20am Matteo Carandini. Vision and superstition in man and mouse

9:20-9:50am Coffee Break

9:50-10:15am Nao Uchida. Dissecting computation in the dopamine reward circuit: An approach using a mouse model

10:25-10:50am Mike DeWeese. Neural correlates of selective auditory attention in rodents

Afternoon Session

4:30-4:55pm Zach Mainen. The dynamics of commitment and temptation in the premotor cortex

5:05-5:30pm Bence Ölveczky. Exploring the capacity of rodents for complex motor sequence learning using a fully automated training system

5:40-6:10pm Coffee Break

6:10-6:30pm Clay Reid. Functional specialization of mouse higher visual cortical areas

6:40-7:10pm Anne Churchland. Multisensory decision making in rodents and humans

Monday, Feb 27 7. Is it time for theory in olfaction?

Organizers: Venkatesh Murthy and Naoshige Uchida, Harvard University. Gonzalo Otazu, Bernstein Center for Computational Neuroscience. Cindy Poo, Champalimaud Neuroscience Programme.

There is a certain mystery about the sense of smell, in part because of the difficulty in quantifying many dimensions of this sense. Modern olfactory research underwent a paradigm shift after the identification of odorant receptors in several species, and the appreciation of the pattern of the projections from the sensory organ to the brain. The circuits in the early olfactory system are being unraveled at a rapid phase, in addition, neural activity at various stages of sensory processing is also proceeding apace. Technological innovations have now enabled a large variety of experiments to be done, resulting in rich and complex data sets. These are heady times for olfaction research. We feel that it may be an appropriate time to step back and look at the larger picture to both put experimental data in perspective, as well as to help select the right experiments to do, among the very large number possibilities. Therefore, the goal of this workshop is bring together theorists and experimentalists to consider: whether there are experimentally testable hypotheses from theory? What kinds of theories are expected to advance our understanding of olfaction? How and whether olfaction research can play a role in advancing our understanding of basic principles of how the brain works.

We hope that this workshop will energize the field by creating an informal setting in which theorists and experimentalists could have inspired and lively discussions. We believe that this workshop will appeal to neuroscientists with a wide range of interests, ranging from those who are interested in olfactory coding and circuits, to those who are interested in computational goals and theoretical frameworks for sensory systems.

Some specific topics for discussion include: • What computational roles do specific neural circuits in the olfactory system serve? How do inhibitory, recurrent, or feedback circuits accomplish tasks such as decorrelation, orthogonalization, gain control, normalization, predictive sensing, pattern completion, pattern separation etc. • How is odor information encoded? What are the relevant time scales and coding schemes at different stages of odor information processing? • How do our knowledge of chemical space and spatiotemporal dynamics of naturally occurring odors help us understand design principles and computational goals of the olfactory system? What are some attractive frameworks to understand olfaction? Can we use olfaction as a model system for understanding general principles of how the brain works?

Monday, Feb 27

Is it time for theory in olfaction?

Morning Session: Neural Circuits

8:00-8:30am Thom Cleland. Standing on the shoulders of theory: the computational architecture of early olfaction

8:30-9:00am Andreas Schaeffer. Inhibition, oscillations and discrimination

9:00-9:30am Florin Albeanu: Gain control in the olfactory bulb: roles of short axon cells and cortical feedback

9:30-10:00am Coffee Break

10:00-10:30am Matt Wachowiak. Bottom-up and top-down control of odor representations during active sensing

10:30-11:00am Ben Strowbridge. Inhibitory circuits in the olfactory bulb

Afternoon Session: Randomness and Recurrence of Connectivity

4:30-5:00pm Venki Murthy. Feedback control of circuits and codes in the olfactory bulb

5:00-5:30pm Kevin Franks. Properties and features of recurrent circuits in piriform cortex

5:30-6:00pm Cindy Poo. Intracortical Circuits in Piriform Cortex

6:00-6:30pm Coffee Break

6:30-7:00pm Haim Sompolinsky. Sparse coding in olfactory cortex: The role of randomness and plasticity

7:00-7:30pm Larry Abbott. Deciphering the variety of network architectures in olfaction

Tuesday, Feb 28

1. Understanding heterogeneous cortical activity: the quest for structure and randomness.

Organizers: Salva Ardid, Alberto Bernacchia, and Tatiana Engel, Yale University

Neural activity in the cortex is highly heterogeneous and variable. For a long time neuroscience research has focused on studying regularities in neural activity, such as cortical maps or feature selectivity of single neurons. Under this tradition, selected neurons with properties of interest have been probed with stimuli highly optimized for those neurons and deviations from regularities have been considered as noise. Only recently investigators have moved on to explore the sources of heterogeneity in neural populations and its possible computational consequences. Nowadays experimental techniques allow simultaneous recording from large neural assembles, which minimizes the preselection bias and exposes neural heterogeneity in full. These data suggest that neural representations are very intricate and a multitude of signals is mixed on the levels of individual cells and neural populations. The complexity of such data sets requires novel analysis methods as well as new theoretical models. Yet there is still no consensus on what is the appropriate framework to characterize the heterogeneity of neural signals and whether it is structure or randomness what matters.

This workshop brings together experimentalists and theoreticians, who seek to understand the heterogeneity of cortical activity, the role it plays in neural computation, and how it emerges from synaptic connectivity. Neural heterogeneity in sensory, motor and association areas will be addressed by the speakers who span the range of methodologies (in vitro- and in vivo physiology, and computation). The debates will be centered on the following questions. What are the potential biases introduced by data acquisition and analysis methods that may skew our view of neural heterogeneity? Can random connectivity explain heterogeneous activity and if so to what extent is the cortical connectivity disordered? What are the advantages and flaws of random and structured network architectures for information processing? The workshop is aimed at those interested in how the complexity of neural signals is related to computation and to the underlying synaptic connectivity.

Tuesday, Feb 28 Understanding heterogeneous cortical activity: the quest for structure and randomness

Morning Session

8:00-8:15am Alberto Bernacchia. Introduction

8:15-8:50am Larry Abbott. Adding intrinsic heterogeneity to a sparsely connected spiking model.

8:50-9:25am Christian Machens. Demixing the information contained in neural population responses.

9:25-9:50am Coffee Break

9:50-10:25am David Sussillo. Integration and gating of sensory information is achieved by a single cortical circuit with orthogonal mixed representations

10:25-11:00am Peter Janssen. Functional heterogeneity in posterior parietal cortex

Afternoon Session

4:30-5:05pm Jeffery D. Schall. From salience to saccades: Multiple- alternative gated stochastic accumulator model of visual search

5:05-5:40pm Daniel Durstewitz. Prefrontal cortical ensemble encoding of working memory, task rules, and context

5:40-6:10pm Coffee Break

6:10-6:45pm Haim Sompolinsky. Random Connectivity: A Design Principle for Cortical Circuits?

6:45-7:45 Discussion and wine.

Tuesday, Feb 28

2. Humans, neurons, and machines: how can psychophysics modeling and physiology collaborate to ask better questions in biological vision?

Organizers: Najib Majaj, Elias Issa, and Jim DiCarlo McGovern Institute for Brain Research, Dept. of Brain and Cognitive Sciences, MIT

In this workshop, we consider the triad formed by psychophysics, physiology, and modeling. Rarely, does a single lab employ all three techniques, but each vertex in this triad is critical for supporting the other two in solving biological vision problems. However, it can be surprisingly difficult to bridge the gaps between fields: How can neural data lead to implemented models (e.g. computer vision systems) that can be tested, potentially falsified, and show good real-world performance? Or, conversely, how can a built computer vision system be mapped back to neural circuits? A first step in addressing these questions is to operationalize the tasks being solved across communities. A theme of this workshop is that psychophysics is the glue that binds; both neural studies and modeling efforts strive to match human perception and performance. Using psychophysical benchmarks of agreed upon tasks can create a common currency for both physiology and modeling efforts, inspiring interaction across fields.

We have identified three topic areas in biological vision where there is potential for synergy: face processing, invariant object recognition, and crowding/clutter. This workshop will bring together members of these communities to define the psychophysically grounded problem being solved. Participants will have to construct their presentations around two questions: (1) Which of the three methods is weakest/ strongest in the vision problem you are studying? (2) What experiment would you like to see your physiology/psychosphysicst/modeling counterpart do and what is your prediction of the outcome?

The discussion-oriented workshop will engage a variety of topics including current challenges for face detection algorithms, psychophysical evidence against holistic processing of faces, the psychophysical/modeling motivation behind early exploration of face-selective neurons, whether IT neurons can explain human performance in invariant object recognition tasks, the role of learning in building invariance, the cognitive neuroscience of invariant object recognition, statistical models of crowding and and their generalization to object recognition, the psychophysical use of mongrels and metamers, and where in the visual pathway crowding can be physiologically studied.

Tuesday, Feb 28

Humans, neurons, and machines: how can psychophysics modeling and physiology collaborate to ask better questions in biological vision?

Morning Session

8:20-8:30am Welcome.

8:30-9:30am Invariance talks. Najib Majaj, Yann LeCun, Irving Biederman

9:30-9:50am Coffee Break

9:50-10:30am Invariance discussion (led by Tomaso Poggio)

10:30-11:30am Crowding talks. Jeremy Freeman, Eero Simoncelli, Denis Pelli.

Afternoon Session

4:30-5:10pm Crowding discussion (led by Ruth Rosenholtz)

5:10-6:10pm Faces talks. Winrich Freiwald, Erik Learned-Miller, Bosco Tjan

6:10-6:30pm Coffee Break

6:30-7:10pm Faces discussion (led by Elias Issa)

Tuesday, Feb 28 3. Workshop Inhibitory synaptic plasticity

Organizers: Timothy P. Vogels, École Polytechnique Fédérale Henning Sprekeler, Humbolt University Robert Froemke, New York University

“The role of inhibition in the working of the central nervous system has proved to be more and more extensive and more and more fundamental as experiment has advanced in examining it.” (Sir Charles Sherrington, Nobel Lecture, December 12, 1932)

80 years later, this still holds true, especially for inhibitory synaptic plasticity (ISP) which has received little attention until the very recent past, when it became implicated as a major player in shaping the dynamics of cortical function.

Here, we bring together experimental and theoretical scientists working on various facets of ISP ranging from molecular mechanisms to behavioral effects. We hope to summarize the current state of knowledge and discuss possible functions of ISP during the different stages and ages of cortical processing.

Morning Session

8:00-8:05am Welcome

8:05-8:25am Melanie Woodin. STDP of inhibitory GABAergic synapses (Experimental)

8:30-8:50am Nicolas Doyon*/ Yves Dekonick. Plasticity of inhibition via altered chloride homeostasis (Modeling)

8:55-9:15am Arianna Maffei. Inhibitory plasticity dictates the sign of plasticity at excitatory synapses (Experiment)

9:20-9:40am Paul Miller. Inhibitory LTP boosts selectivity in random networks to enhance performance in cognitive tasks (Modeling)

]9:45-10:15am Coffee Break

10:15-10:35am Robert Froemke. ISP re-balances excitation and inhibition (Experimental)

10:40-11:00am Henning Sprekeler. ISP balances excitation and inhibition (Modeling) 22

Tuesday, Feb 28

Afternoon Session

4:30-4:50pm Corrette Wierenga. Two-photon imaging of activity- dependent formation and plasticity of inhibitory synapses (Experimental)

4:55-5:15pm Friedemann Zenke. Implications for ISP in recurrent networks (Modeling)

5:20-5:40pm Robert Liu. The role of experience and physiological state on changes in the evoked inhibitory response of single units to natural vocalizations in adult mice (Experimental)

5:45-6:05pm Mathieu Gibson. Frequency Selectivity Emerging from Spike-Timing-Dependent Plasticity (Modeling)

6:05-6:30pm Coffee Break

6:30-6:50pm Michael Brainard. Sensory motor learning engages inhibitory plasticity in song premotor circuitry (Experimental)

6:55-7:15pm Michela Fagiolini (TBA)

7:20-7:30pm Closing discussion

Tuesday, Feb 28

4. Functions of identified cortical microcircuits

Organizers: Andrea Hastenstaub and Vikaas Sohal, UCSF

The cerebral cortex consists of multiple chemically, anatomically, and genetically distinct neuron types that form precise, laminarly organized microcircuits. These patterns of microcircuit organization follow stereotyped motifs that are conserved throughout neocortex, suggesting that these motifs offer unique computation advantages while at the same time being flexible enough to meet the diverse computational demands of various neocortical regions. Thus, a complete picture of cortical circuit function would necessarily depend on understanding the functions of these microcircuits. However, this has been historically difficult owing to our inability to identify and manipulate specific circuit elements in functioning networks. Recent advances in optical and genetic technology have allowed significant progress in understanding how specific features of cortical microcircuits contribute to their function. The aim of this workshop will be to bring together investigators employing experimental and computational approaches to link the low-level anatomical and physiological details of specific microcircuits, to the more abstract computational capabilities that these circuits enable. Speakers span a range of methodologies (computation, anatomy, in vitro and in vivo electrophysiology, and imaging), and cortical regions (visual, auditory, somatosensory, and prefrontal).

Tuesday, Feb 28

Functions of identified cortical microcircuits

Morning Session

8:00-8:20am Vikaas Sohal. Introduction

8:20-8:55am David McCormick. Gain modulation: a key feature of cortical circuits

8:55-9:30am Brent Doiron. Slow dynamics and high variability in balanced cortical networks with clustered connections

9:30-9:50am Coffee Break

9:50-10:25am Shawn Olsen*/ Massimo Scanziani. Gain control by layer six in cortical circuits of vision

10:25-11:00am Alex Reyes. Shifts in the excitatory-inhibitory balance with changing stimuli

Afternoon Session

4:30-5:05pm Jess Cardin. Regulation of cortical activity by distinct sources of inhibition

5:05-5:40pm Darcy Peterka*/Rafael Yuste. Two-photon connectivity mapping using a novel channelrhodopsin

5:40-6:15pm Alex Kwan*/Yang Dan. Dissection of cortical microcircuits by single-neuron stimulation in vivo

6:15-6:30pm Coffee Break

6:30-6:50pm Andrea Hasenstaub. Conclusion

6:50-7:30pm Open Discussion

Tuesday, Feb 28 5. Promise and Peril: Genetic approaches for systems neuroscience revisited

Organizers: Katherine Nagal and David Schoppik, Harvard University.

Six years ago a COSYNE workshop on "Genetic approaches for systems neuroscience" introduced a powerful new set of tools for answering questions in systems neuroscience. Since that workshop many young investigators from systems backgrounds have moved to genetic model organisms, and found them both more powerful and more limited than expected. In this workshop we propose to revisit some of the promising technology— labeled populations of neurons, optical manipulation of genetically defined neurons— that led many of us to move to simpler organisms, and some of the challenges we have encountered in working with these models. We envision this workshop as a "working meeting" for new PIs, senior post- docs, and mid-stage graduate students interested in asking how neurons work together to generate behavior in worms, flies, fish, and mice. In addition to reporting on novel biological insights from their current research, speakers will discuss their own research trajectories: what led them to their current model organism, what they expected, and what surprised them. Together we hope to address questions central to this emerging field, including:

Behavior What kind of behavioral tasks can GMOs perform? Can the behavioral approaches, and attendant analytical tools that have proved so powerful in primates be ported to simpler organisms, or is the behavior of these organisms qualitatively different? What kinds of behavioral tasks and assays are most appropriate and fruitful for each organism?

Neural coding How does our understanding of population activity change when the primary criteria for identifying neurons are genetic in origin, rather than functional? To what extent are the different functional properties of neurons (sensory receptive fields, temporal response properties, connectivity) specified genetically? What does it mean if the anatomy or physiology of a genetically identified population varies across individuals? How should we think about "noise" in gene expression levels, and its network-level consequences?

Technology What kinds of tools (genetic, behavioral, optical, and analytical) will be most useful for answering systems question in genetic model organisms? What are the biggest gaps in our current toolkit and what are the most promising ways we see to fill these?

Methodology How does the conceptual approach of geneticists and molecular biologists differ from that of systems and computational neuroscientists? What can these two cultures learn from one another?What do we expect to learn from GMOs that we don't know from monkeys, rats, and retinas? What kinds of questions are GMOs best poised to answer? 26

Tuesday, Feb 28

Promise and Peril: Genetic approaches for systems neuroscience revisited

Morning Session

8:00-8:20am David Schoppik. Introduction to the morning session: Approaches to behavior in genetic model organisms

8:30-8:50am Mark Histed. Mouse vision and psychophysics

9:00-9:30am Alex Katsov. Reading the mind, predicting the brain: constraints from spontaneous behavior in GMOs

9:30-10:00am Coffee Break

10:00-10:20am Rex Kerr. Multisensory integration and behavioral choice in C. elegans

10:30-10:50am Vivek Jayaraman. Studying sensorimotor integration in Drosophila: opportunities and challenges

Afternoon Session

4:30-4:50pm Katherine Nagel. Introduction to the afternoon session: Specificity and variability in genetically identified neural circuits

5:00-5:20pm Vanessa Ruta. Tracing of Innate and Adaptive Olfactory Circuits in the Fly Brain

5:30-5:50pm Saskia de Vries. Linking computation to action in the Drosophila visual system

6:00-6:30pm Coffee Break

6:30-6:50pm Emre Aksay. How do circuits remember: insights from the larval zebrafish

7:00-7:20pm Matthieu Louis. Towards a systems understanding of larval chemotaxis: deconstructing behavior to reconstruct neural circuits

Each 20 minute talk will be followed by 10 minutes for questions and discussion

Tuesday, Feb 28 6. Perception and decision making in rodents

Organizers: Santiago Jaramillo and Anthony M. Zador, Cold Spring Harbor Laboratory

By bringing together laboratories on the cutting edge of these approaches, we will address the following questions:

2. What cognitive processes can be adequately modeled with existing rodent behavioral paradigms? What challenges remain in developing new paradigms?

3. How well do insights gained from studying neural mechanisms in rodents apply to humans and non-human primates?

Tuesday, Feb 28

Perception and decision making in rodents

Morning Session

8:10-8:35am Alla Karpova. Evidence for representation resets in the medial prefrontal cortex at the onset of behavioral exploration

8:45-9:10am Tony Zador. Corticostriatal projections mediate auditory decisions

9:20-9:40am Coffee Break

9:40-10:05am Mark Laubach. Simple behavioral tasks for studying reward consumption and reward-based decisions

10:15-10:40am Christof Koch. Project Mindscope at the Allen Institute

Afternoon Session

4:30-4:55pm TBA

5:05-5:30pm David Tank. Cellular resolution imaging in awake behaving mice

5:40-6:10pm Coffee Break

6:10-6:35pm Daniel Huber. Probing cortical circuits in behaving mice . 6:45-7:10pm Pam Reinagel. Visual object discrimination by rats

7:20-7:30pm Closing remarks

Tuesday, Feb 28 7. Is it time for theory in olfaction?

Organizers: Venkatesh Murthy and Naoshige Uchida, Harvard University. Gonzalo Otazu, Bernstein Center for Computational Neuroscience. Cindy Poo, Champalimaud Neuroscience Programme.

Tuesday, Feb 28

Is it time for theory in olfaction?

Morning Session: Computation and neural coding

8:00-8:30am Nathan Urban. Ups and downs of correlation in the olfactory bulb

8:30-9:00am Diego Restrepo. How does the olfactory system deal with 1,400 dimensional input?

9:00-9:30am Alex Koulakov. Sparse coding in the olfactory bulb

9:30-10:00am Coffee Break

10:00-10:30am Alex Pouget. Probabilistic olfaction

10:30-11:00am Mitya Chklovskii. Non-linear predictive coding as a model of early sensory processing

Afternoon Session: Computation, transformation and behavior

4:30-5:00pm Mark Stopfer. Spontaneous olfactory receptor neuron activity determines follower cell response properties

5:00-5:30pm Nao Uchida. Transformation of odor representations in the central olfactory pathway

5:30-6:00pm Don Wilson. Cortical odor object processing

6:00-6:30pm Coffee Break

6:30-7:00pm Chuck Stevens. New insights into the architecture of the vertebrate olfactory system

7:00-7:30pm Zach Mainen. Origins of uncertainty in decision-making

The Cliff Lodge - Level C (Upstairs) Wasatch A, Wasatch B,Superior A, Superior B, Maybird

The Cliff Lodge - Level 8 (Downstairs) Magpie A, Magpie B

Goldon Cliff