Working memory is a crucially important aspect of intelligence. Working memory basically refers to the brain's ability to temporarily store information. Many brain disorders often have impaired working memory processes. New neurotechnology methods like transcranial magnetic stimulation have the ability to amp up certain aspects of the brain's working memory storage capacity. Now researchers have created a new computer model (pdf) of how visuospatial working memory functions at a biophysical level. Visuospatial working memory is how a person remembers an image for a short term period.
The computer model is a fairly simplified one that is merely representational of two specific brain regions that are thought to be involved in visuospatial working memory tasks. The brain regions that were modeled are the dorsolateral prefrontal cortex and the intraparietal sulcus (IPS) in the posterior parietal cortex. The intraparietal sulcus is specifically believed be more correlated with the storage of the visual memories themselves, while the dorsolateral prefrontal cortex has a secondary role (sort of a top level control). The computer model of the IPS, for instance, consisted of about 1280 neurons in total with a surfeit of excitatory pyramidal cells over inhibitory interneurons.
Their computer model predicted that people who had greater activation of the dorsolateral prefrontal cortex would be able to retain more information in their working memory. The model made a few other predictions and they performed fMRI brain scans on patients who were performing a task to correlate the model with actual brain behavior. Excitatory input from the dorsolateral prefrontal cortex is able to boost the holding capacity of visual memory in the posterior parietal cortex.
The caveats for this research are that this is currently a fairly simplified model. The fMRI brain scan's resolution is not not that high, so it can only correlate things to a certain extent. Also I tend to think that explanations for macroscopic neurological behavioral outputs can be fuzzy in nature. Plus there is quite a lot of variation in individual capacities and many other brain regions that may interact that were not modeled. I think, though, that future computer simulations will become increasingly accurate in their portrayal of actual brain activity.
Future computer brain simulations will also likely be coupled with modeling the physics of specific brain manipulations (such as transcranial magnetic stimulation). Simulations may model the induced currents in the brain as the result of pulsed magnetic fields or perhaps the physics of ultrasonic neuromodulation and their resulting effect on neuronal changes. Some brain manipulations can depend on the specific state of neuronal activity. So the alteration of brain chemistry may be different depending on how a region is being activated/deactivated in response to a specific task.
More detailed computer simulations may allow researchers to overcome current bottlenecks to human intelligence. Future scientists may have to carefully weigh each brain manipulation regimen in order to maximize positive impact. The current size of an individual's brain most likely puts constraints on the optimization of specific brain processes. Without some massive reengineering or increase of neuron matter, it may be difficult to squeeze much more computational capacity out of a person's brain. I think what would tend to happen is that if you tried to improve one brain process, it could have unintended negative effects on other aspects of intelligence or behavior. Being able to elucidate the mechanisms of cognition, however, will certainly be a boon for a variety of brain disorders that suffer from more severe functional deficits. Eventually, perhaps neuromorphic brain implants could enable more storage capacity than regular neurons in a given brain region and actually enhance general intelligence itself.
Here's the press release for this new paper.
Fredrik Edin, Torkel Klingberg, Pär Johansson, Fiona McNab, Jesper Tegnér and Albert Compte. Mechanism for top-down control of working memory capacity. PNAS, Online early edition 30 March - 3 April 2009
source: brain stimulant
