Bernstein Netzwerk Computational Neuroscience

Skip to content. | Skip to navigation

Personal tools
Bernstein Netzwerk Computational Neuroscience
 
You are here: Home News and Events Research News The neural code for short term memory
Document Actions

The neural code for short term memory

Scientists from Berlin and Munich have discovered how to read out information from cellular short term memory (June 2008).

When the brain processes information, neurons emit a quick succession of electrical impulses forming a spatial and temporal pattern. This neuronal information transmission happens within a few milliseconds, but nevertheless represents information that has been gathered and experienced over a longer period of time. Christian Leibold, Richard Kempter, Dietmar Schmitz (Bernstein Centers Munich and Berlin) and their colleagues have now shown which cellular mechanisms can form the basis of this compression of event series, using electrophysiological experiments and theoretical modeling.

 

kurzzeitgedaechtnis_400

Top: While the rat strives through its territory, the timing of the place cells misaligns cumulatively with respect to the theta oscillations. Thereby the heading direction of the rat is coded within each oscillation cycle. Bottom: When the rat rests, memories from the last few seconds that are saved in the synapses are read out.

The cellular mechanism that is assumed to underlie short term memory is the so-called ‘synaptic facilitation´. Repeated transmission of a signal from one neuron to another increases the efficiency of the synapse linking the two cells. Even if this strengthening of the synapse is not permanent, it persists for several seconds – the synapse “memorizes” the event. ‘Memories that are saved in the synapse in this way must also be read out by the rest of the brain,’ explains Leibold. Leibold and his colleagues discuss this issue using the example of spatial navigation in rats.

A rat that is familiar with its territory has established so-called “place cells” that are active when the rat visits a particular area.  If, for example, place cells of two areas A and B are active at the same time, the rat is located at the intersection of both areas. As long as the rat is moving, the place cells in the hippocampus produce a joint oscillation. They preferentially emit signals in the so-called “theta rhythm”- comparable to an audience applauding rhythmically after a concert. This rhythm serves a reference to measure the exact moment of neuronal discharges. The longer the rat remains in one location, the more the rhythm of the respective place cells differs from the theta rhythm. Thus, the rat “knows” at any moment not only where it is located, but also for how long it has already been in this area.

The place cells of the most recently visited places are activated in reverse sequence.This phase shift can be explained by 'synaptic facilitation' as the scientists from Berlin and Munich could prove. When the rat passes an area, the respective cell in the hippocampus repeatedly receives signals from an upstream brain region. The transmission efficiency of the synapse increases with each signal and the strength of the signals increases. Due to the augmented signal strength, the hippocampus cell fires its neural impulses more rapidly than before and thereby gets out of rhythm.

If the rat rests after its walk or feeds, it recapitulates – unconsciously – the passed trail once again. During these rest periods, the places visited before are replayed in reverse sequence. Possibly, also this “reverse replay” is based on synaptic facilitation. Even several seconds after the rat has passed the trail from A via B to C, the synapses still contain traces of this “memory” – the synapses of the place cell C are strongest, while the ones of place cell A have nearly decayed to the normal level. When the rat rests, the place cells are stimulated and reveal this “memory”. They forward signals with corresponding differences in signal strength. Once again, the signal strength has an impact on the exact moment of the next signal.

The conversion of signal strength into a temporal code is supported by neural oscillations. However, in resting periods, no theta rhythm exists, but fast field potential changes, so-called „sharp wave ripples“. For a long time, sharp wave ripples have been assumed to play an important role in the process of memory consolidation. How events can be  recalled from short term memory during these sharp waves is now shown by the scientists’ work.


Contact persons
Source

 Prof. Dr. Christian Leibold
Abteilung Biologie II- Neurobiologie
Ludwig-Maximilians-Universität München

Leibold, C., Gundlfinger, A., Schmidt, R., Thurley,K., Schmitz, D. & Kempter, R. (2008). Temporal compression mediated by short-term synaptic plasticity. Proc Natl Acad Sci USA. 105(11):4417-22.

Dr. Richard Kempter
Institut für Biologie - Theoretische Biologie
Humboldt-Universität zu Berlin

Thurley, K., Leibold, C., Gundlfinger, A., Schmitz, D. & Kempter, R. (2008). Phase precession through synaptic facilitation. Neural Comput. 20(5):1285-1324.

Prof. Dr. Dietmar Schmitz
Neuroscience Research Center
Charité Campus Mitte