Memory

Stored information plays a critical role in the lives of many animals. Knowing a nest location, landmarks for navigation in a home range, where food and water have been found the past, and how previous social interactions with another animal have turned out, all are critical pieces of information in shaping future behavior.

One of the most interesting aspects of the study of memory is that it has essentially the same functional structure and underlying biochemistry across the animal kingdom. This suggests that memory or its evolutionary predecessor evolved early in the history of the animals. Studies of learning and memory in a sea slug, Aplysia, have resulted in a model of learning and memory that seems to be broadly applicable to all animals. Eric Kandel of Columbia University was awarded the Nobel Prize in 2000 for his work on learning and memory and for developing Aplysia as a model system.

Memory starts as a biochemical response in the brain following sensory input. The first change comes at the level of the synapse--junctions between nerve cells. Acetylcholine, a well known neurotransmitter, plays a critical synaptic role in the initial formation of memory. We can think of short-term memory as resulting from a transient change in neurostransmitter levels at synapses.

Treatments for memory-deficit problems, like Alzheimer's, in humans often involve enhancing the retention of acetylcholine in brain synapses. Aricept (donepezil hydrochloride), a drug marketed to prevent memory problems, acts by inhibiting the action of the enzyme in the brain that breaks down acetylcholine (acetylcholinesterase). People with moderate levels of dementia can show marked improvement with this sort of treatment.

At a behavioral level, what happens next depends on the animal's experience after the initial learning phase. If the animal moves on and doesn't have immediate reinforcement of the learned event, likely that memory quickly fades and is irrecoverable. Short-term memory is highly subject to amnesia effects. A traumatic event can obliterate short-term memory, as can a variety experimental treatments, such as carbon dioxide or nitrogen narcosis or chilling. This makes sense, as retention relies on neurotransmitter levels, which can be quite sensitive to an animal's general internal physiological state.

Evolutionarily, the transient nature of short-term memory is a huge advantage in allowing the animal to sort learning that only has immediate importance from learning that will be useful in the days or weeks ahead. We can think of the rapid turnover of information in short-term memory as a way of continuously clearing the mental "chalkboard" of potentially extraneous information.

Some information, though, has critical future importance and needs to be moved to long-term memory. Oftentimes reinforcement--re-experiencing the item to be learned--plays a critical role in the formation of long-term memory. Another key behavioral aspect of the formation of long-term memory is the formation of associations. Often associations are sensory; for example, humans talk about odors which are evocative of complex sets of memories from the past. This combination of reinforcement and association can be thought of as helping to build a network of neurons which hold this key information. Association is probably a key element in accessing long-term memory. A single memory which is not linked to other remembered items is quite difficult to access. Forming long-term memories requires the synthesis of proteins in neurons. If protein synthesis is inhibited or disrupted, long-term memories will not be formed.

Between short-term and long-term memory lies middle-term memory, or the consolidation phase. This is a time of transition between the neurotransmitter/synaptic basis for short-term memory and the protein based long-term memory. At least in the honey bee, short-term memory and middle-term memory are independent--both are formed during learning, but only middle-term memory is the gateway to forming long term memories.

While treatments, like the Aricept mentioned above, can delay the onset of serious memory deficit problems in humans, death of brain cells is also a key issue in human dementia. Once the cells have died, treatment at the neurotransmitter level is futile. Treatments currently under development are aimed at preventing cell death. There is outstanding promise, with a combination of a treatment that manipulates acetylcholine levels and one that prevents cell death, for managing human brains to prevent dementia.

Related issues:

• where in the brain does memory reside?
• what is the role of consolidation, strength of memory, and forgetting in managing memories?

Menzel R, Muller U 1996 Learning and memory in honeybees: From behavior to neural substrates. ANNU REV NEUROSCI 19: 379-404

Milner, B., Squire, L.R., and Kandel, E.R.1998. Cognitive neuroscience and the study of memory. Neuron 20:445-468