A Reservoir Model of Explicit Human Intelligence

A fundamental and distinguishing feature of human intelligence is that we transfer and accumulate knowledge as a society and across generations. We suggest that for communication and accumulation, explicit (but not implicit) intelligence is likely encapsulated by a network of attractors representing abstractions such as objects or actions and direct connections between them, rather than one that utilizes hidden layers of representation. We propose a network model for explicit human intelligence that is composed of attractors and learned connections as elemental units. As attractors are added and more connections are made the network forms an increasingly large but conceptually simple graph, and the contents remain teachable. We suggest that for lower animals the analogous network is used for associative learning, as in classical conditioning, but that a key innovation in humans was a mechanism to switch between an online mode in which the network operates as an input/output mapping device, and an offline mode which is necessary for hypothetical thought. As an arbitrarily connected graph, the network dynamics would be similar to a reservoir, but the reservoir itself would contain and accumulate learned intelligence. Newly created and connected attractors would become part of the reservoir and be available to participate in future learning. A second key innovation was the development of language, which provides random access to the reservoir nodes without the need for demonstration and a mechanism for higher levels of abstraction. For the rapid expansion of societal human intelligence seen in recent millennia, language enabled us to directly excite sequences of attractors in the brains of others and thereby rapidly populate young human brains with a distillation of associations from the accumulated body of human intelligence. We suggest that the proposed network for explicit intelligence works in conjunction with older and likely more complex deep networks that perform sensory, motor, social, and other implicit forms of processing.