Memory and Emotions: How the Brain Stores Our Experience
Brain.Com Exclusive

Why is it that the most powerful and vivid memories seem to be the ones loaded with emotion? Just imagine running into an old flame on the street. Your head suddenly fills with memories, and you can picture the good times and the bad times in the relationship, as if they had happened only yesterday.

Stephann Hamann and colleagues have observed something similar in the laboratory. Subjects were shown pictures and tested for their ability to remember images they had seen. Pleasant images, such as cute animals, or unpleasant images, such as diseased or mutilated bodies were the easiest to recall, while images considered emotionally neutral, such as chess players or household scenes, were harder to remember. The researchers found that activity in a particular region of the brain called the amygdala was associated with recalling both pleasant and unpleasant images. (1).

The amygdala is a more primitive part of the brain in evolutionary terms and is responsible for injecting the emotional pique into any memory. Animal studies show that the instant Pavlovian-type fear responses depend on this structure. If rats, for example, learn to associate the sound of a tone with a mild electric shock to their feet, they soon become afraid every time they hear the tone— their blood pressure and heart rate rise and they freeze in place in a typical response to fear—even weeks after the shock is no longer given. If the amygdala is damaged, however, the animals never learn to make the association between tone and shock (2).

Researchers believe the emotions that form in the amygdala are communicated directly to the hippocampus along interconnecting neuronal pathways. Whichever memory is being formed at the time, it will be tagged with an accompanying emotion, which somehow serves to intensify the memory itself, although exactly how this happens is a mystery.

The nature of memories seems to lie in how networks, or circuits, of neurons work together. According to Tim Bliss at the National Institute for Medical Research in London, memory is the internal re-creation of an experience through the reactivation of the same networks that were involved in the original experience. To make this reactivation more likely, these networks must somehow be permanently altered following the initial experience. “So now when you activate part of the network the whole network is triggered,” says Bliss.
LTP

So what kind of changes could explain a permanently altered neuronal circuit, rendering a memory trace more easily retrievable? Scientists believe that at least part of the answer lies in the way neurons are able to strengthen their connections following strong electrical stimulation--a phenomenon first discovered in the laboratory more than 30 years ago and known as long-term potentiation, or LTP. After receiving a high frequency of electrical pulses, neurons from the hippocampus in particular, become more easily excitable by subsequent stimuli. This enhancement of signaling potential can last for weeks.

Scientists have shown that LTP is a common property of neurons from other memory-related areas – the cortex and the amygdala --suggesting that LTP could be involved in memory formation in very different parts of the brain.

While LTP has been induced in the laboratory by electrical stimulation, no one has been able observe it directly during learning and memory formation. Still, many lines of evidence point to the likelihood that it is involved. “If the LTP hypothesis is not the case then it would be pretty astonishing,” says Bliss.

Early studies showed that LTP in isolated nerves requires the neurotransmitter glutamate to signal to one of its receptors, the NMDA receptor. If drugs that block the interaction between glutamate and the NMDA receptor, or with another receptor, AMPA, are given to rats during a learning procedure such as escaping from a water maze, learning is inhibited. This is particularly true when the drugs are infused directly into the hippocampus. Likewise, infusion of blockers into the amygdala disrupts learning of the fear response.

High-powered electron microscopy is enabling researchers to actually see LTP in action: Even in the space of just 60 minutes, the branching contact points between nerves, called dendrites, can grow and form new extensions (3).

In addition, AMPA receptors can be seen to appear in parts of dendrites where they were previously absent (4), which could also increase the signaling potential of the contacting neurons.

Molecular biologists have added to this picture by literally knocking-out the genes for NMDA and AMPA receptors in mice. These animals are then unable to learn particular tasks , especially those involving spatial navigation (5).


So would more NMDA boost learning power? Dr Jo Tsien and colleagues at Princeton University showed that this may be a possibility, with the creation of “Doogie” the “smart mouse” genetically engineered to contain higher than normal levels of NMDA receptor. Great fanfare in the press followed, when as a result, the mice had a greater ability to learn new tasks (6).

But the triggering of glutamate receptors is just the first brief event in a whole cascade that leads to altered signaling. This begins with the triggering of new genes and proteins inside each neuron undergoing LTP. By identifying which genes and proteins are involved, scientists hope to be a step closer to understanding the nature of memory formation, and to shed light on human conditions in which memory is affected--including senile dementia, mental retardation, and schizophrenia.

One protein in this largely elusive chain is the enzyme CamKII, which becomes active and sparks other proteins into action. Another is CREB (cyclic adenosine monophosphate (cAMP) response element binding protein), which switches on new genes. Studies in snails and fruit flies have provided a strong case for its role in promoting a withdrawal reflex away from unpleasant sensations, such as chemicals, in invertebrates, but its role in vertebrate memory is still under debate.

The Future

But LTP does not explain everything about memory formation. It is a rapid event occuring in minutes, which does not account for the many years it sometimes takes for consolidation of a memory to occur between the hippocampus and the cortex. The time scale of consolidation is entirely different, and “is more related to regeneration of patterns of activity in the brain at different times,” believes Michael Fanselow of the University of California, Los Angeles. One possibility is that this is a process that only takes place during the dream state of sleep, when the electrical activity of neurons between the hippocampus and cortex enter a similar set of rhythms (7).

According to Tim Bliss, the most exciting possibilities for the future understanding of memory lies in elucidating the changes that take place in whole neural networks. “That’s what is eventually going to crack the problem of what sort of changes occur during learning, and how changes in these circuits are reconstituted during recall,” he says. These changes are likely to involve more than just LTP.

Fred Gage and his colleagues at the Salk Institute in California, for example, have discovered that throughout life, in both humans and rodents, another process is at work – the formation of new nerve cells in the hippocampus. This helps to enhance the process of learning and memory, at least in laboratory mice reared in very stimulating, so-called “enriched” environments--toys, smells and colors included in the cages, and lots of exercise (8). Such findings raise the possibility that new nerve cells could somehow be stimulated to grow and replace those damaged in conditions such as Alzheimer’s disease, and help to maintain memory. As for the rest of us, it gives hope that we could, perhaps, design our own more enriched environments to improve our education and performance throughout life.


1. Hamann SB, Ely TD, Grafton ST and Kilts CD. Amygdala activity related to enhanced memory for pleasant and aversive stimuli. Nature neuroscience 1999;

2: 289-293.2. Maren, S. Long-term potentiation in the amygdala: a mechanism for emotional learning and memory. Trends in Neurological Sciences 1999; 22: 561–566.

3. Toni N, Buchs P-A, Nikonenko I, Bron CR, Muller D. LTP promotes formation of multiple spine synapses between a single axon terminal and a dendrite. Nature 1999; 402: 421-425.

4. Malinow R. Rapid spine delivery and redistribution of AMPA receptors after synaptic NMDA Receptor Activation. Science 1998; 284:1811-1816.

5. Chen C and Tonegawa S. Molecular genetic analysis of synaptic plasticity, activity-dependent neural development, learning, and memory in the mammalian brain. Annual Review of Neuroscience 1997; 20: 157–184.

6. Tang Y-P et al. Genetic enhancement of learning and memory in mice. Nature 1999; 401: 63-69.

7. Poe GR, Nitz DA, McNaughton, B, Barnes CA. Experience-dependent phase-reversal of hippocampal neuron firing during REM sleep. Brain Research 2000; 855: 176–80.8. Eriksson P.S. et al. Neurogenesis in the adult hippocampus. Nature medicine 1998; 4:1313–1317.