Rose: Consciousness 
Updates
FiguresRose: Consciousness
Chapter 04
Modules (sections 4.2.2. and 4.2.7)
A major overview of systems theory can be found in the edited volume by Callebaut and Ruskin-Gutman (2005). The ubiquity of modular systems in Nature is emphasized, in particular from an evolutionary perspective. The section on mental and cultural modularity includes a chapter by Velichkovsky which stresses the need for a multi-level hierarchical approach to cognition.
The Foreword by Herbert Simon re-iterates the importance of the spontaneous development of modular architectures, and reminds us of Simon and Ando’s (1961) original definition of ‘nearly-decomposable’ systems. First put forward in 1956, their idea is that dynamic systems possess internal sub-structure, in that some of the variables that comprise the system come to cluster together spontaneously into subsystems. Interactions within each subsystem are strong, but are much weaker between subsystems. Over short time-spans the values of the variables within a subsystem are thus highly linked in their dynamic behaviour. But this does not happen between subsystems, which are nearly autonomous and can vary largely independently from each other. In the long term however the opposite is true; each subsystem has had ample time to reach its internal equilibrium end-state, but the weaker interactions between subsystems eventually have a cumulative effect. The exact location of each subsystem’s end-state is then dominated by the (slower) dynamic of the whole system. In other words, over the long run each variable is moved roughly the same amount as its fellows within the same subsystem. Each subsystem can therefore be treated as (reduced to) a single variable within the longer-term dynamics of the global system.
Intriguingly, this notion — of weak causal interactions between modules but strong within — presages Rolls’ suggestion that synaptic connections must be stronger within neural networks than they are between them (Consciousness, p. 188), A similar conception is Edelman’s depiction of the separation of brain tissue into (more or less discrete) ‘neuronal groups’ (see section 7.3.4). Importantly, note that this whole approach views interactions as only of one type, varying in strength. It contrasts with the various metaphors of ‘communication’ between modules discussed in sections 4.3.4-4.3.7, in which it is the type of information transmitted (or the type of operation performed) that defines the module’s boundaries and properties (or more provocatively: it is the intentionality or function of the interaction that matters; cf. also section 12.2.3.1).
In addition, it is worth repeating that both within the brain/mind (section 10.3.1) and in general (Simon and Ando, 1961) dynamic systems have longer time-constants at higher levels!
Emergence (sections 2.2.2.1, 4.2.3, 4.2.4, and Side-box 4.1)
Check out an important new volume on the reality of emergence (Clayton and Davies, 2006). The message here is that multiple levels develop throughout Nature, all the way from those studied by physics to religion! There is emergence of new hierarchical structures and causal powers. Several chapters address consciousness directly.
Empirical evidence (sections 4.2.7.2 and 4.3.6)
Under homuncular functionalism, modular systems are posited to self-organize at multiple levels. Yet it is not yet clear how many levels there are (section 3.5.5.1). It is becoming clear that at least one level exists between that of the cortical area and that of the whole brain, in the form of ‘networks’ of connected cortical regions scattered around the brain and mediating particular functional states. Thus numerous brain scanning studies have found that around 3-6 brain areas or regions typically light up under any given task condition.
At one level, widespread regions and functions are incorporated, such as when consciousness is focussed either on oneself or on the outside world. The former state includes medial regions of the cerebral hemispheres and the frontal pole, while the latter involves several parts of the lateral surfaces of the brain. This dichotomy can manifest itself even in the absence of overt behavioural changes and can be revealed even without specific stimulation (Fox et al., 2005) as well as with naturalistic inputs for processing (e.g. Golland et al., 2006). For more examples and details of this level see the web update to section 11.3.
Within the externally-focussed regions a sub-division has then been resolved between those parts involved in top-down attention and those in bottom-up. The former are bilateral and include the parietal lobes and the ‘frontal eye fields’ (see the web update to sections 8.3 and 8.4), while the latter lie more ventrally in the right hemisphere frontal and parieto-temporal regions. This schism also appears in the spontaneous activity of the brain (Fox et al., 2006).
We can thus detect at least two levels within these networks of regions/areas. Whether the functional processes within each of these can be classified as ‘modules’ according to the terms defined and discussed within Chapter 4 remains an issue for future analysis (and an essay topic!).
References
Callebaut, W. and Ruskin-Gutman, D. (2005) eds. Modularity: understanding the development and evolution of natural complex systems. MIT Press, Cambridge MA.
Abstracts from the original conference are available at
http://www.kli.ac.at/workshops-c.html?stuff/workshops/wtb-00
Clayton, P. and Davies, P. (2006) eds. The Re-emergence of Emergence: the emergentist hypothesis from science to religion. Oxford University Press, New York.
Fox, M.D., Snyder, A.Z., Vincent, J.L., Corbetta, M., Van Essen, D.C. and Raichle, M.E. (2005) The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proceedings of the National Academy of Sciences of the USA 102, 9673-9678.
Fox, M.D., Corbetta, M., Snyder, A.Z., Vincent, J.L. and Raichle, M.E. (2006) Spontaneous neuronal activity distinguishes human dorasl and ventral attention systems. Proceedings of the National Academy of Sciences of the USA 103, 10046-10051.
Golland, Y., Bentin, S., Gelbard, H., Benjamini, Y., Heller, R., Nir, Y., Hasson, U. and Malach, R. (2006) Extrinsic and intrinsic systems in the posterior cortex of the human brain revealed during natural sensory stimulation. Cerebal Cortex (in press; epub May 12).
Simon, H.A. and Ando, A. (1961) Aggregation of variables in dynamic systems. Econometrica 29, 111-138.
View print version