You are here

Lab Story: Glimpses of the learning mind: how knowledge and memory are stored in the brain

Nestled in the quiet and sparsely populated CNS wing, Dr. Balaji’s lab is full of activity. The research in this very recently established lab focuses on the storage and organisation of information in the brain and how this influences learning and behaviour. Dr. Balaji, an alumnus of UCLA (University of California, Los Angeles), explains that while this can be studied in a variety of ways, his lab mainly uses mice as a model system. Currently, attention in this lab is centred on two main areas of research: autobiographical memory and associative learning.

Autobiographical or declarative memory refers to the memories of events that have been experienced and can be consciously recalled when needed. The rationale behind Dr. Balaji’s current research is based on the fact that autobiographical memory, though very “crisp” or sharp and detailed on initial formation, becomes “vague” or fuzzy after a period of time. This process of “crisp memory” becoming “vague” is called “generalisation”. Dr. Balaji explains “We believe that this process of generalisation helps in adaptive learning. If any memory is very rigid, there will be no adaptation to new situations. For instance, we may know that a spear is used for game hunting and that a fish, a good source of nutrition, can be caught by using a net. But relating both these concepts, that is, using a spear to hunt for fish is an example of adaptive learning. This is what Tom Hanks’ character does in the movie ‘Castaway’.”

While experiencing an event, various neurons in the brain are activated and have certain molecular cues associated with them. The interaction between a network of activated neurons creates memory. Aditya, a PhD student at the lab studies these networks of activated neurons via ‘memory traces’. A memory trace, whose definition is currently under debate, can be loosely visualised as a neuronal network actively associated with a memory.

Aditya explains, “We create a “cranial window” on the skulls of the mice, through craniotomy. That is, a surgical opening is made through the skull to implant a glass cover-slip to visualise different brain regions. Ten days prior to memory training sessions, the mice’s brains are injected with genetically modified viruses carrying DNA of certain molecular markers. These molecular markers, which are fluorescent proteins, are expressed and become visible when the neurons containing them are active.”

Aditya further adds “We train the mice and after an hour or so, the molecular markers peak in their fluorescence activity. Hence, we are able to observe a subset of active neurons (within the memory trace) via in vivo imaging (through the cranial window). It has been shown that activation of the same set of neurons is also sufficient for specific memory retrieval.”  The information acquired from his research includes functionally defining and physically visualising a memory trace of stabilized or consolidated memory, i.e. literally seeing where a memory is stored in the brain during and long time after encoding. Aditya muses “While we see which neurons are active during different training paradigms, can we also see which subsets of neurons are active during differential memory retrieval? For encoding memories, the hippocampus (a seahorse shaped structure in the brain thought to be responsible for memory) was considered the only region necessary and cortical regions were considered to be required only for retrieving remote memories. However, recent evidences have demonstrated crucial role of cortical substructure during memory encoding and necessity of hippocampus in remote retrieval. The interaction among brain regions encoding different memories might play a crucial role in problem solving.

The second focus of the lab is associative learning. Associative learning is a process by which the association between two stimuli or a stimulus and a specific behaviour is learned by an organism. “In our lab, we also work on flavour–place association, with an experimental arena modified for mice”, says Dr. Balaji.

Vikram, another PhD student with Dr. Balaji, explains his work on associative and complex learning after busily training a few school students who have joined the lab as summer trainees. “Research in the field of learning and memory has mainly focused on associative learning. For example, the Skinner box was used to study associative learning by giving a mild shock to the animal. The whole box and its properties are stored in the animal’s memory as a context and animal forms a memory of that particular box being associated with a foot shock (contextual fear conditioning).”

Vikram’s current work involves the use of an observational box called the event arena with the floor containing 5 sand wells at specific locations, which have unique flavour pellets buried in them. A few visual cues are also placed in the box to serve as reference points and help the test animals navigate the arena. Training involves exposing animals to particular flavours outside the box, and rewarding them with a food reward when they find the correct well associated with that particular flavour. Generally, it takes mice a few weeks to learn 5 such unique flavour–place associations, but consequently, when 2 new flavour–place associations are added, it takes mice only one day to learn these new configurations. This indicates that their previous experiences helped them learn the new patterns and assimilate the information faster, a process termed as complex learning.

Vikram also tests the efficiency of learning in mice by modifying the flavour – place association ‘map’ in the experiment box. If the changes are gradual, he finds that the learning efficiency is high; but if the change is abrupt, learning efficiency is low. “This means that gradual changes in the ‘mental map’ of the experimental arena in mice brains are stored as small variations of the same memory”, Vikram says. “These are more easily accessible than abrupt changes, which are probably stored as different memories and hence maybe less easily accessible”, he adds.

Post-mortem resolution of neurons and synapses is also important for Vikram’s study to observe the active brain regions during any learning task. He uses a special technique with the apt acronym CLARITY (Clear Lipid-exchanged Acrylamide-hybridized Rigid Imaging/Immunostaining/In situ hybridization-compatible Tissue-hYdrogel). The technique is used to render all brain tissue transparent such that specific genetically-marked neurons can be detected through fluorescence. This actually helps Vikram trace connectivity patterns between neurons in the brain.

While the thought of a physically whole, yet transparent brain is a stunning one, there are more wonderfully unique techniques being developed and used in Dr. Balaji’s lab. Vikram is currently also creating a real time virtual reality system for studying neuronal connectivity patterns in mice. This consists of a stationarily held mouse running on a smooth air cushioned Styrofoam ball while it navigates the “terrain” shown on a screen. Meanwhile, the setup allows a researcher to observe memory traces unfolding in its brain through the cranial window on its head using a multi photon imaging system. Virtual reality, by itself an awesome piece of technology, makes research in this lab doubly cool!

About the lab:

http://www.cns.iisc.ernet.in/balaji/

PH.No: 080 2293 3049; E-mail:balaji@cns.iisc.ernet.in