Consciousness Blog 12 July 2015
Recent experiments carried out at Duke University Medical Center, North Carolina, have shown that interconnected mammal brains can out-perform single such brains at computational tasks requiring cooperation at the neuronal level.
Described in Scientific Reports, July 2015, the experiments, carried out by teams including Miguel Pais-Vieira, Mikhail Lebedev and Miguel A L Nicolelis, implanted electrodes in monkeys and rats in brain regions that are present in all mammals, and trained the connected animals in reward-based tasks that required concerted action between groups of two to four conspecific individuals. The electrodes are used both to deliver stimuli to the animals and to record their subsequent neuronal behaviour.
Dubbed Brainets by the researchers, the connected brains delivered above average results on tasks which included discrete classification, image processing, storage and retrieval of tactile information, and weather forecasting. Say the researchers: "Brainets consistently performed at the same or higher levels than single rats in these tasks. Based on these findings, we propose that Brainets could be used to investigate animal social behaviors as well as a test bed for exploring the properties and potential applications of organic computers."
Of particular importance is the fact that enhanced cooperative behaviour results only when the participants are awake and alert; when subjects are anaesthetized, performance is no better than chance. This suggests that cortical motor and judgmental processes are accessing remote as well as local neuronal behaviours in reaching their conclusions; in other words, they are happy to reach across brain-to-brain interfaces in forming their conclusions.
The laboratory says it has demonstrated that brain-to-brain interfaces can utilized to directly transfer tactile or visuomotor information between pairs of brains in real time. Other parallel studies have described transmission of hippocampus representations between rodents, transmission of visual information between a human and a rodent, and transmission of motor information between two humans.
It is not perhaps a surprise that if more neurons are devoted to a task they deliver a faster result; and perhaps it is even not a surprise that neurons find it so easy to recognize and function together with their homologues in other conspecifics. Maybe it even could have been predicted that such cooperation could obtain between species. Maybe. A neuron is a neuron, after all.
Predicted or not, these results are dramatic in terms of forecasting the cognitive behaviour of future models of brain function whether based on biological, bio-electronic or purely electronic mechanisms. It is no longer possible to question whether a human brain will be capable of working with an electronic assembly which mimics evolved biological neuronal behaviour. If a remote neural assembly has the external characteristics of a biological neurone, it will be accepted as such by the 'home' neurone. Of itself, this is perhaps not such dramatic news: brains are already controlling electro-mechanical devices such as artificial hands; and vice versa – brains accept suitably coded messages from external non-biological devices. But active cooperation is a step onwards from point-to-point messaging.
If a small group of neurones in one brain can cooperate with a similar small group of neurones in another biological brain, as represented by electronic impulses, who can doubt that the same could obtain between major sections of whole brains? Of course, a bundle of electrodes is a very primitive means of communication, which will no doubt quickly be supplanted by directed wireless messages or some other new communications technology; but the principle has now been established.
Inevitably, this leads on towards the feasibility of establishing human consciousness in external, remote cognitive assemblies, and all the problems inherent in the existence of multiple consciousnesses for one individual. The technical feasibility no longer remains in doubt. We just need 100 years to straighten out the practicalities!
Read more in Chapter Eleven of Agent Human by Michael Bell, The Future Of Groups And Consciousness