More philosophy >

The Layered Mind

As we discover more about the human brain and the processes involved in thinking, we are coming to see it as a layered structure: Thought itself comprises dynamically structured patterns of symbolic information. These symbolic structures are superimposed on the electrochemical activity of the brain, which in turn is superimposed on the physical nerve cells.

There is an obvious parallel with the processing of information in a networked, multi-processor computer system: meaning is coded into software apps and host-to-host communications. A complex "protocol stack" underlies this activity, with each layer handling the data at the appropriate level and passing it on to the next layer up or down. In this way the stacks break the symbolic data down into something that the hardware can handle and pass around as required. Because of course the hardware itself underlies the whole thing.

Computer specialists analyse network connectivity stacks in terms of "layer models". It does not seem unreasonable to attepmpt the same approach with the processes of human thought and, in a parallel manner, describe the communications between nerve cells using a layer model.

The OSI Seven-Layer Model

Communications theory has a venerable reference model for data communications, known as the OSI seven-layer model. It a theoretical model, in practice almost never implemented as seven layers, but it provides a standard logical breakdown of the various functions involved in data transmission.

In terms of sentient brain function it looks something like this:

7: Application LayerData representation of information with symbolic meaning *
6: Presentation Layer  Encoding of data into a physical signal
5: Session Layer(manages end-to-end communication channels) +
4: Transport Layer)
3: Network Layer} Signal format and transmission over the connections
2: Data Link Layer)
1: Physical LayerPhysiology of nerve activity and electrochemical connections

* The app data within the Application layer is individually defined outside this specification.

+ The Session layer is not applicable to the brain: a nerve connection is always available. (Sleep is too blunt an instrument to apply to individual connections in this way. Rather, it represents a temporary damping within the physical layer.)

The first six layers are all definable in terms of the laws of physics and the world which they support.

Layer 7 is conceptually a bridge between two worlds, those of physical reality and of symbolic information. However the data representation used is quite arbitary, provided that it is common to both ends of the link. The nature of the symbols lies outside the ISO model, meaning that the latter is concerned only with the physical world including a crude, physical view of information. Any meaning is left to the individual processes active in the hosts at each end of the link.

The UDP Four-layer Model

The Internet is traditionally built on a simpler, four-layer model. Variations exist and UDP is probably the most appropriate.

4: Application layerData representation and encoding *
3: UDP layerSignal format
2: Internet (IP) layer(Network routing) +
1: Network Interface layer  Physical links and signal transmission. o

* The app data within the Application layer is individually defined outside this specification.

+ The IP layer is not applicable to the brain: a nerve connection is hard-wired.

o UDP is hardware-independent (one may regard the hardware layer as rolled into Layer 1)

The better match to nerve signalling in the brain is immediately obvious. However at the same time it is quite independent of the physiology which underlies those signals. It is thus more readily comparable to human culture rather than limited to brain function, which is perhaps why it was chosen for the Internet. But it still misses that top layering, any recognition of symbolic meaning.

A Neural Five-Layer Model

Taking these models a step higher, we may identify the symbolic meaning of the data as being, of necessity, encoded in the processes on each host. The mind is concerned primarily with these meanings and the neural data patterns underlie these thoughts. That is to say, the physical patterns rendered for example in a brain scan or an EEG trace are distinct from the actual meanings of the neural patterns themselves. One may know the meaning but have no idea of the underlying pattern, or vice versa.

Moreover, sentience also includes the presence of a subjective experiential quality - a quale - associated with each thought. These qualia are not open to physical study and neither can they be deduced from any information-based model of thought, they appear to be entirely subjective. That is to say, in terms of the stack model they comprise yet another layer. Find out more in Towards a Theory of Qualia.

Based on the above approach, here is a possible model for human thought:

5: Experience layer  Subjective quality of the experience
4: Meaning layerSymbolic meaning or context of the information
3: Transport layerInformation-to-signal format encoding
2: Link layerSignal transmission over the connections
1: Wetware LayerPhysiology of nerve activity and electrochemical connections

Conclusion

Analytical models from the world of information technology can provide novel and useful approaches to the analysis of brain function and the mechanisms of human thought.

A subsidiary purpose of this discussion has been to try and highlight how subjective experience may be readily incorprated into a rational theory of mind, whether one approaches the subject as a scientist or as a philosopher. I am always astonished at how many zealous atheists deny the existence of qualia, the very raw material from which they happily deduce the existence of the physical world around them. I suspect that such people may be afraid that acknowledgement would open the door to mysticism and religious dogma. Perhaps this essay may help to provide some ressurance that their own inner experiences are as respectable as the neural energies which correlate with that experience. Just as app data lies beyond the domain of the network layer but is rationally structured none the less, so too may qualia lie beyond the domain of neurology yet be amenable to some other rational model. It is OK not to be as robotic as a worm.

Updated 9 May 2017