Of mice and men

Researchers may have used genetics to improve the memory of mice, but it is nonsense to talk of an 'intelligence gene', argues Dr Stuart Derbyshire

Teams of researchers from Princeton University, New Jersey, and another from the University of Tokyo have discovered improved memory and learning ability in two strains of genetically modified mice. The results have caused much excitement, with some commentators talking about an 'intelligence gene' - raising the possibility of enhancing the normal intelligence of 'other mammals', including humans.

But while this work is undoubtedly a step towards enhancement of some of the basic information-processing capacities that all mammals share, it is still a long stretch from these studies to a facilitation of human memory, let alone intelligence.

Memory in humans and all other mammals is intricately tied in with a piece of the brain called the hippocampus. If you were to be so unlucky as to lose your hippocampus on both sides of your brain, you would never again be able to place an event into long-term memory. Memories never develop beyond the moment of hippocampal destruction and the patient forever lives in the past and immediate present.

Knowing this important site for memory, researchers have turned their attention towards the mechanisms of the hippocampus. This is what the current studies investigated. It is known that information is shuffled around and maintained inside the hippocampus dependent upon the properties of receptors (sites at which actions take place) and their interaction with neurotransmitters (chemicals that allow communication between receptor sites). The important receptor is NMDA and the important neurotransmitter is glutamate. When NMDA binds to glutamate a current is evoked which opens a channel allowing calcium ions to enter the cell. This is the magical moment when the cells of the hippocampus talk in harmony and strengthen their connections, known technically as long-term potentiation (LTP) and more commonly as the beginning of memory. The researchers at Princeton hoped to improve the memory of mice through the increase of an NMDA subtype that has an extended LTP - while the researchers in Tokyo deleted the genetic code for a non-NMDA hippocampal receptor that inhibits LTP, thereby inadvertently extending overall LTP.

Both groups demonstrated improved memory in their transgenic strains through a standard behavioural test. The mice were released into a shallow pool of water and swam around until stumbling across a hidden platform and standing on it. This was done on multiple occasions and the time taken to reach the platform was recorded. All the mice reached the platform more quickly on later trials, indicating learning. The transgenic mice, however, performed better than the normal mice, showing further improved learning due to the genetic manipulation.

So will we be able to create humans with super memories? Perhaps. I have no doubt that humans and mice share important properties relating to the hippocampus and its basic function. But even so, memory in mice and memory in humans are far removed. Simple associative learning - discovering that A follows B, etc - is an aspect of memory and, if you are a mouse surviving from one moment to the next, it might be terrific to have your associative learning enhanced. But humans do not survive in this manner - we have developed ourselves and our society and can operate on a more general, abstract level. Being able to understand connections between A and B might be a prerequisite for abstract memories but that does not mean a better recognition of concrete relationships will improve abstract memory. It might even be detrimental. Humans use memory in a transformative manner to go beyond simple relationships and to develop abstract connections.

The ability to draw abstract relationships and pursue meaningful goals is the hallmark of human intelligence that is sorely lacking in the mouse world - even in the newly created super mouse world. In the mouse, memory is merely an extension of their basic biological function with zero intelligent content. Our intelligence remains under the influence of basic information-processing skills (as evidenced by its severe compromising when processing is lost to dementia), but intelligence is more than the simple additive effects of processing environmental associations. For humans, memory is not a natural extension of biology but is a part of the transformation and interweaving of innate biological processes into higher intelligent function.

The transformation of our biological information-processing capacity, which we share with mice, takes place in the sociocultural world, which we do not share with mice. During development, formal schooling and other cultural interventions, the subordination of biological instinct to volitional control is nurtured and encouraged. Memory ceases to be slavishly dictated by internal and external events and becomes a tool that can be used in the pursuit of reasoned goals. A genetic and neural influence can remain, but the cause of intelligence lies outside of genetics and pharmacology.

Dr Stuart Derbyshire is head of neuro-imaging at the Neuroenteric Disease Program, University of California in Los Angeles