Short-term memory is based on synchronized brain oscillations

January 31, 2012
A monkey has to carry out a classic memory task: The animal is shown two consecutive images and then has to indicate whether the second image was the same as the first one. Credit: Stefanie Liebe, MPI for Biological Cybernetics

Scientists have now discovered how different brain regions cooperate during short-term memory.

Holding information within one's memory for a short while is a seemingly simple and everyday task. We use our short-term memory when remembering a new telephone number if there is nothing to write at hand, or to find the beautiful dress inside the store that we were just admiring in the shopping window. Yet, despite the apparent simplicity of these actions, short-term memory is a complex cognitive act that entails the participation of multiple . However, whether and how different brain regions cooperate during memory has remained elusive. A group of researchers from the Max Planck Institute for in Tübingen, Germany have now come closer to answering this question. They discovered that oscillations between different brain regions are crucial in visually remembering things over a short period of time.

It has long been known that brain regions in the frontal part of the brain are involved in , while processing of visual information occurs primarily at the back of the brain. However, to successfully remember visual information over a short period of time, these distant regions need to coordinate and integrate information.

In each of the two brain regions (IPF and V4) brain activity shows strong oscillations in a certain set of frequencies called the theta-band. Credit: Stefanie Liebe, MPI for Biological Cybernetics

To better understand how this occurs, scientists from the Max Planck Institute of Biological Cybernetics in the department of Nikos Logothetis recorded electrical activity both in a visual area and in the frontal part of the brain in monkeys. The scientists showed the animals identical or different images within short intervals while recording their brain activity. The animals then had to indicate whether the second image was the same as the first one.

The scientists observed that, in each of the two brain regions, brain activity showed strong oscillations in a certain set of frequencies called the theta-band. Importantly, these oscillations did not occur independently of each other, but synchronized their activity temporarily: "It is as if you have two revolving doors in each of the two areas. During working memory, they get in sync, thereby allowing information to pass through them much more efficiently than if they were out of sync," explains Stefanie Liebe, the first author of the study, conducted in the team of Gregor Rainer in cooperation with Gregor Hörzer from the Technical University Graz. The more synchronized the activity was, the better could the animals remember the initial image. Thus, the authors were able to establish a direct relationship between what they observed in the brain and the performance of the animal.

The study highlights how synchronized brain oscillations are important for the communication and interaction of different brain regions. Almost all multi-faceted cognitive acts, such as visual recognition, arise from a complex interplay of specialized and distributed neural networks. How relationships between such distributed sites are established and how they contribute to represent and communicate information about external and internal events in order to attain a coherent percept or is still poorly understood.

Explore further: Have we met before? Scientists show why the brain has the answer

More information: Stefanie Liebe, Gregor M Hoerzer, Nikos K Logothetis & Gregor Rainer (2012) Theta coupling between V4 and prefrontal cortex predicts visual short-term memory performance. Nature Neuroscience, 29 January 2012, doi: 10.1038/nn.3038

Related Stories

Examining the brain as a neural information super-highway

June 2, 2011

An article demonstrating how tools for modeling traffic on the Internet and telephone systems can be used to study information flow in brain networks will be published in the open-access journal PLoS Computational Biology ...

Brain state affects memory recall

June 13, 2011

Lost your keys? Your brain might be in a better state to recall where you put them at some times than at others, according to new research from UC Davis. A paper describing the work is published June 13 in the journal Proceedings ...

Recommended for you

A turbo engine for tracing neurons

April 27, 2017

Putting a turbo engine into an old car gives it an entirely new life—suddenly it can go further, faster. That same idea is now being applied to neuroscience, with a software wrapper that can be used on existing neuron tracing ...


Adjust slider to filter visible comments by rank

Display comments: newest first

1.8 / 5 (5) Jan 31, 2012
Wow - this seems so clearly intuitive that I'm totally depressed that we're anywhere near understanding the brain in my lifetime,,,sigh. We need smarter people working on these issues but instead they're staving to death in Africa, working as labor slaves in China, or getting shot by Syrian dictators, rotting in prison in the USA,,ad nauseum.
Jan 31, 2012
This comment has been removed by a moderator.
1 / 5 (4) Jan 31, 2012
In dense aether model the long term memory is working on the Hamilton principle: the prolonged travel of neural signal between some neurons will modify their synapses in such a way, this path becomes permanent. In this sense the brain is behaving like switchboard, which is hardwiring itself just using of itself.


Note than in quantum theory every wave is behaving like more dense place of space-time. The brain wave is gradually burning its own path trough brain, i.e. more dense brain network just with own existence. It's because the intrinsic property of neurons network is to make new synaptic connections along the axons, which is used often and removing them along the path, which are used sparingly in similar way, like the muscles, which are swelling, when they're used often. In this way the neural net is implementing its own learning.
5 / 5 (1) Jan 31, 2012
Keep your off-topic cold-fusion and aether shit out of here Rawa/Callippo

1 / 5 (2) Jan 31, 2012
I'm with you on the Hamilton principle, Callippo. At least for short term memory. As for cold fusion, that's a big nonsequitor. We have working free-energy machines. Not efficient or refined, but working. Sadly I understand they're part of the 6000 patents the US has classified as dangerous, so goons will appear at your free energy plant. Must suck to live in a gilded cage like that. Then again you president now feels free to shoot you in the face or put you behind bars now for no reason. I guess cold fusion plays second fiddle there even if it was powering time-traveling Deloreans.
not rated yet Feb 01, 2012
Callipo, ...just out of curiosity, are you familiar with snarks


, NON-Hamiltonian dynamics, and the Icosian game ?

The game's object is finding a Hamiltonian cycle along the edges of a dodecahedron such that every vertex is visited a single time, no edge is visited twice, and the ending point is the same as the starting point ?

1 / 5 (1) Feb 01, 2012
are you familiar with snarks
Yes, I do consider this model a bit schematic, but the concept of Hamiltonian is correct here. The brain need to find the most effective path between two neurons, which do fullfill the another conditions (usually represented with preferred paths of another neurons). The most effective way here is the way, in which the neural wave need to overcome as low membrane potential, as possible. The membrane potential along axon is defined of calcium ion potential inside of synapses, which are working here as a multibits of memory. The another way, how to decrease the potential of path is the increasing of concentration of neurotransmitters around synapses, which works for medium term memory. If you need even longer memory, then another synapses are build along axon. In this way the paths of neural signals are gradually becoming more dense along preferred directions.
not rated yet Feb 01, 2012
Keep your off-topic cold-fusion and aether shit out of here Rawa/Callippo

1 / 5 (1) Feb 05, 2012
Well, if you don't like the aether wave theory of human consciousness, you can read about "liquid computing" model, which has been developed by Swiss neuroscientist Henry Markram together with Graz University of Technology http://www.spring...t43j.pdf
In this model the brain works like a pond in which stones are thrown. The waves caused by this perturbation don't disappear immediately, but rather overlap with each other and collect information about how many stones were thrown in and how big they were. The main difference is just that the waves in the brain spread in a network of neurons and at very high speed. Anyway, I don't understand, why it's necessary to keep this model separately from aether theory, which is using fluid models too.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.