Dynamics of Memory: What is the evidence?

The overall scientific goal of the conference is to discuss experimental approaches that would help to evaluate and assess different proposed theoretical frameworks comprising mechanistic explanations for how memories are stored, retrieved and maintained in the brain.

The most firmly established mechanism for memory is based on the idea of attractor dynamics. Classical attractor neural networks store static patterns of neuronal activity in their synaptic connection matrices.

Positive feedback in these networks causes an initial pattern of activity to evolve in time in such a way as to approach the most similar of the stored patterns, which can then sustain itself without the necessity of an external input. This type of dynamics implements a process of categorization of transient inputs, and five of the proposed speakers (Treves, O’Keefe, Moser, Munk and Friedrich) will discuss evidence suggesting that such a categorization process is taking place in the cerebral cortex and hippocampus. Because categorization can also be achieved by feed-forward networks which do not use attractor dynamics, an interesting topic for discussion will be whether the available evidence directly supports the attractor hypothesis and, if not, what kind of experimental data would be necessary to do so.

A challenge for the original formulation of attractor networks was their reliance on fixed-point attractors, i.e., self-sustained patterns of activity associated with stationary firing rates for the neurons in the network. Although some neurons display this type of stationary self-sustained activity, neuronal activity profiles recorded in-vivo are very often time-varying. In principle, the attractor framework can accommodate time varying activity, i.e., there can be attractive trajectories of activation in the space of activity in the network. However, our theoretical understanding of how to produce such patterns of activity is much more limited.

Three of our speakers (Durstewitz, Laurent and Abeles) will discuss physiological evidence in favor of time-varying attractors in the rodent prefrontal cortex and in the locust olfactory system. Recently, it has been suggested that complex recurrent networks may produce memory function in the absence of attractors. The recurrent network in this framework, called a Liquid State Machine, carries information in its instantaneous time-varying activity about the recent history of inputs to the network. Although the robustness and feasibility of this theoretical paradigm is still an active topic of discussion, one of our speakers (Maass – an initial proponent of these ideas) will discuss recordings from the visual cortex which show evidence in favor of this framework. Several reasons make this the right time to bring together theoreticians and experimentalists to discuss the dynamics of memory. First, the existence of a diversity of experimental techniques – especially, but not exclusively, the ability to record the simultaneous activity of large populations of neurons – which are capable of providing evidence sufficiently precise so as to discard proposed theoretical alternatives. Second, the progressive infiltration of theoretical ideas into experimental journals and forums has made it possible for experimentalists to become familiar and interested in quantitative approaches which, until recently, were only discussed among theoreticians. Because of this, we feel like this conference will be of great interest to systems neuroscientists, both those using experimental as well as theoretical approaches.

This activity is a one-time satellite event to the FENS Forum of European Neuroscience.

Alfonso Renart, Champalimaud Neuroscience Programme, Lisbon, Portugal
Albert Compte, IDIBAPS, Barcelona, Spain

…

Moshe Abeles, Hebrew University, Israel

Matthew Chafee, University of Minnesota, USA

Albert Compte, IDIBAPS, Spain

Daniel Durstewitz, BCCN Heidelberg-Mannheim, Germany

Rainer Friedrich, FMIBR, Switzerland

Mark Goldman, UC Davis, USA

Gilles Laurent, MPI Frankfurt, Germany

Wolfgang Maass, Graz Univ. of Technology, Austria

Matthias Munk, MPI Tuebingen, Germany

Alessandro Treves, SISSA, Italy

Misha Tsodyks, Weizmann, Israel

IF YOUR INSTITUTION COVERS YOUR REGISTRATION FEE: Please note that, in case your institution is paying for the registration via bank transfer, you will have to indicate your institution details and choose “Transfer” as the payment method at the end of the process.

We are aware of a number of current scams targeting participants at CRM activities concerning registration or accommodation bookings. If you are approached by a third party (eg travellerpoint.org, Conference Committee, Global Travel Experts or Royal Visit) asking for booking or payment details, please ignore them.

Please remember:
i) CRM never uses third parties to do our administration for events: messages will come directly from CRM staff
ii) CRM will never ask participants for credit card or bank details
iii) If you have any doubt about an email you receive please get in touch