A new paper offers a broad challenge to a certain kind of ‘grand theory’ about the brain. According to the authors, Federico E. Turkheimer and colleagues, it is problematic to build models of brain function that rely on ‘strong emergence’.
Emergence refers to the idea that a system can exhibit behavior or properties that none of its individual parts possess. Such a behavior or property ’emerges’ from the whole system, in other words, from the parts and their interactions. To give an example, a single person could run around and kick a ball, but a single person could never play a game of soccer – soccer emerges from a group of people.
It seems very plausible that the brain is an emergent system – that complex functions emerge from the interactions between lots of neurons. But Turkheimer et al. say that we need to distinguish between two different kinds of emergence, strong and weak:
A system is said to exhibit strong emergence when its behaviour, or the consequence of its behaviour, exceeds the limits of its constituent parts. Thus the resulting behavioural properties of the system are caused by the interaction of the different layers of that system, but they cannot be derived simply by analysing the rules and individual parts that make up the system.
Weak emergence on the other hand, differs in the sense that whilst the emergent behaviour of the system is the product of interactions between its various layers, that behaviour is entirely encapsulated by the confines of the system itself, and as such, can be fully explained simply though an analysis of interactions between its elemental units.
I have to say that I don’t quite follow this distinction. Does anyone really believe that the brain is such a strongly emergent system that we could never, even in principle, ‘explain it though an analysis of interactions between its elemental units’? Apart from the units and their interactions, what else is there – unless we invoke dualism?
I think that what Turkheimer et al. really mean by ‘strong emergence’ is a theory which posits strong ‘top-down’ influences in the brain, such that we can’t understand the ‘lower’ levels without understanding the ‘higher level’ causes. The ‘top down’ nature of the Free Energy Principle/Bayesian Brain model seems to be what makes it strongly emergent, in the authors’ view:
The Bayesian computational model of brain function, also called the “free energy principle” (FEP) is… a paradigmatic exemplar of strong emergence (Lestienne, 2014). In this model, brain-environment interactions of an agent are represented as a loop in which the primary sensory inputs are first processed with prior knowledge of the most probable cause of these signals in a top-down fashion; the brain then combines prior and sensory information and calculates the posterior percept…
…This largely Bayesian hypothesis formulates perception as a constructive process based on internal models. As FEP is operated by a set of rules that are treated independently of underlying neurobiology and only loosely constrained (inspired) by metabolic anatomical/neural constraints, FEP can be considered strongly emergent.
Similarly, Integrated Information Theory as a model of consciousness is a strongly emergent theory because it holds that “emergent phenomena are more accurate descriptions of underlying reality” than reductionist approaches.
So what’s the problem with strong emergence? Turkheimer et al. are a little vague on this point, but by my reading, they question whether strong emergence provides any kind of real understanding or has any predictive power:
The paradigm of strong emergence seems not to have moved far from the perennial philosophical puzzle of emergent phenomena floating inconsistently over some unspecific physical substrate. The whole of the emergent phenomena still cannot be reduced or explained by its parts; thus, it follows that no change in its components can have a predictable effect on the whole.
This cartoon, which the authors reproduce with permission, seems to sum up their view of strong emergence:
Turkheimer et al. go on to describe how a weaker form of emergence is a more appropriate model and they highlight recent work (some of it their own) that seeks to model brain function from the single-cell level up to behavior by taking coupled oscillators (representing pairs of neurons) as the fundamental units.
In my view, this is a provocative and interesting paper, but ‘strong emergence’ seems to be a bit of a strawman, here. I don’t know much about Integrated Information Theory, but the Bayesian Brain model, as I understand it, is based on a purely mechanistic model of the brain. It does feature lots of top-down information transmission (i.e. signals from ‘higher’ to ‘lower’ brain areas), but not in any mysterious sense. The top-down signals are modelled in just the same way as bottom-up signals.
Then again, I’m no expert on recent work on the Bayesian Brain, and perhaps it has strayed into more strongly emergent territory recently?