The quest to understand consciousness is rife with speculation, historical bias, and confusion, but more recently there have been substantial efforts to bring quantitative order to understanding the relationship of consciousness to information.
We propose that further insight may be gained by examining consciousness as a software layer that handles the results of big data optimization: correlation extraction, compression, and filtering. Brains continually receive a huge amount of sensory data, originating in the external world and from the host organism’s internal functions. To remain viable and to participate in the larger landscape of Darwinian selection, an extant organism will necessarily make the best use of it capabilities. That should include efficient data processing and real-time decision-making, as well as projection and longer-term interpretation.
A reasonable hypothesis is therefore that an experience of self, and consciousness, is simply the distillation of information for decision-making. If I ‘feel sad’, that condition is the real-time outcome of the flow of data evaluation taking place in the brain. That experience may or may not meet a threshold for action or trigger decisions on actionable items. Similarly, if I am aware of my ‘self’ that is an evaluation state, a weighing of data and probabilities impacting the organism – it’s efficient, perhaps optimal in its simplicity.
The mystery is where and how does this take place in biological brains? Perhaps we can gain insight by examining optimization and feedback processes across science. Can we use computational science to construct model minds that are further abstractions of neural networks, incorporating sensory input nets and ‘actions’ that influence the external sources of stimulation? By asking what an optimal layer between these two things (the ‘conscious layer’) might look like – perhaps we can find clues to what happens in the real world.