Previous paragraphs, and the previous chapter, have introduced the idea humans will be able to communicate directly with external electronic or bio-electronic cognitive representations, whether of humans or other types of psyche. This proposition, which underlies many of the predictions contained in the remainder of the book, requires some further explanation.
The direction of much current research, some of it referred to in the previous chapter, and the almost unanimous opinion of technologically expert futurist writers (distinguished from science fiction writers) is indeed that it will be possible to bypass primary sensory input devices (eyes, ears, skin, the afferent nervous system) and for an external cognitive being or device to have two-way interaction with sensory processing modules in the brain or even with purely cognitive sections of the brain by imitating the data-streams the brain expects. See for instance Kurzweil 6. Various signalling techniques may be used, including wireless or magnetic fields, or even cables. On a mechanical level, nanobots could receive and transmit from within capillaries in the brain; and the science of direct brain implants in humans is already under development for therapeutic purposes (eg Parkinson's disease), although evidently it is still at an early stage.
Apart from the mechanics, two other pre-conditions for brain-to-robot communication are that robots' brains should have sufficient processing capacity to match the human brain, even if only in certain respects, and that the 'wiring diagram' of the brain should have been sufficiently deciphered to allow human thought processes to be recreated or modelled in non-human cognitive assemblies (this last is the province of AI, or Artificial Intelligence). It is not really in doubt any longer that these two conditions will be fulfilled, the only question is when. As to the former, the operation of Moore's Law (the doubling of the number of electronic components on a chip every 24 months), the matching progress of miniaturization, and continuing increases in computation speeds would result in a computer which is smaller, faster and with more computational capacity than the human brain within less than ten years. Artificial Intelligence researchers are busy at work both analyzing the methods of the human brain with a battery of investigative techniques and also constructing equivalent methods of computation which achieve the same results as a human brain without necessarily copying its working methods. 2020 seems to be an outside estimate of the moment at which the Turing test will be satisfied by a computer (that it should be indistinguishable from a human brain to an outside observer).
2020 would also seem to be a reasonable estimate of the timescale for at least primitive control by the human brain of remote cognitive representations, presumably still primarily electronic, although perhaps with an increasing proportion of bio-electronic 'tissue'.
There are difficulties in reaching satisfactory names for 'remote cognitive representations'. Conventionally they have been called robots; but the more 'human' they become, the less satisfactory that term comes to seem. The word 'avatar' is also not too satisfactory, since it carries with it the sense of being an artificial construct, whereas in most situations what is wanted, and will be provided, is a more or less faithful (if partial) version of the original. The word clone is also overlaid with a lot of extraneous meanings by now. So for the rest of this book, we will continue to use the term Remote Cognitive Representation (but will gradually switch to abbreviating it to RCR) to describe a device or construct which faithfully represents all or part of an individual in a 'real' or virtual environment.
It is important to see that Remote Cognitive Representations (RCRs) will become the preferred method of interacting with other people (other RCRs no doubt) in a wide variety of situations, and to distinguish them from avatars as used in gaming or other imaginary (and often deceptive) situations. Right here we will skip over the difficulties of identification that will be raised by Remote Cognitive Representations. These are not different in kind from those that already exist as regards people, and they will be solved by the same types of method.
Apart their use in Virtual Internet Communities (VICs), Remote Cognitive Representations will be useful for business meetings between robots or in virtual 'rooms', for queuing, for attending conferences, if such still exist, for going to art galleries, concerts, plays, sporting events (and possibly for competing in some of them), for attending educational classrooms or lecture theatres.
It will be seen that the Virtual Internet Community is in fact not to be thought of as just a playful Internet social environment – as Remote Cognitive Representations become more powerful, the VIC will become the norm for many types of human private or business forum.
It is also clear that VICs allow an individual to become far more efficient, since she can be represented in multiple social settings simultaneously. The Remote Cognitive Representation can of course be given an amount of autonomy appropriate to a particular setting, so that the 'owner' becomes aware of the RCR's sensory and cognitive states only in pre-determined circumstances, or of course at the behest of the owner. 95% of shopping, for instance, does not require a decision process from the owner and could easily be multi-tracked with other activities (child-minding, say).
So far, the RCR has been discussed as if its capabilities merely reflect a sub-set of its owner's chacteristics; but in reality the RCR will soon come to be capable of more, for instance by holding sets of data which cannot be accommodated in (or are not needed by) the owner's 'home' cognitive space. Remote Cognitive Representations will also come to be able to pool the experience of groups of other owners and/or their RCRs. For this type of unit, we will use the expression Remote Cognitive Collective (abbreviated RCC). Uses might include multi-university research projects, in which a Remote Cognitive Collective could house the current state of research knowledge as held in the personal cognitive spaces of multiple researchers, or business teams, so that a marketing strategy meeting could take place in an RCC, combining the current knowledge and skills of the individual members of the team, together with externally-acquired sets of market data which would be too voluminous to be held individually.
It is fascinating to speculate on the protocols which would be necessary to govern the awareness by individuals of the state of a Remote Cognitive Collective, and the rules to govern their active participation in what we must still call a meeting, although it wouldn't seem much like a meeting in the 'real world'. From one point of view it would be easier to communicate with the mental states of one's peers in such a meeting, since the RCC would hold data about the current cognitive state of participants in highly organised forms.
Among groups of collaborators who know each other well, it is not clear whether a participant would want or need to receive visual, auditory or tactile sensory input (albeit bypassing the personal sensory interface) from the Remote Cognitive Collective. A meeting might take place on a purely intellectual level – and it might often be very quick! To some extent, the 'rules of engagement' will depend heavily on a resolution of the 'language' issue referred to in earlier chapters. That is to say, if it emerges that all linguistic constructs have (or could have) non-linguistic representations in the brain, then meetings would be silent, or rather, would not employ words at any level. If, as is perhaps more likely, it emerges that certain classes of linguistic concept are dealt with as words even at deeper levels of the brain (this might for instance be true of names, numbers and time) then exchanges at a linguistic level will still be needed.
Remote Cognitive Collectives might develop as purely passive fora, to be inhabited as occasion demands by multiple individuals; but they will also probably come to have initiative and even perhaps personality on their own account. We could call this an 'active RCC', with autonomous tasks ranging from simple activities such as data collection to the representation of the collective will of the group as a quasi-human actor in relationships with other Remote Cognitive Representations, RCCs or individuals. In terms of an on-line gaming environment, a team could have a joint atavar, or a team of atavars communicating through a collective room.
One way or another, the collective psyche which currently exists among groups of individuals at an unconscious level, as described by Jung and Neumann, will come to exist more transparently in the Remote Cognitive Collective. There are no theoretical limits to the 'size' of an RCC, and at the extreme it would be – will be – possible for large numbers of people to express their will (vote, for instance) through an RCC.
6 Kurzweil, R (2005) The Singularity Is Near, Viking (Penguin Group) USA