One of the many highlights of the recent FQXi conference on the Physics of the Observer was the session on consciousness. Consciousness is quite possibly the most enigmatic aspect of human existence. It is at the core of who we are as individuals (and, some argue, as a species) and yet we don't really know quite what it is, let alone whether or not it has a physical basis.
This fall I am teaching a course on the nature of time (based partly on the 2011 FQXi conference in Bergen and Copenhagen) and early on I introduce the notion of a system. For the purposes of the class, I begin very simply: a system is anything that has measurable properties or characteristics. Properties and characteristics are used to help distinguish two systems from one another or one system in a certain state from that same system in another state (where a state is any configuration of properties and characteristics). For the moment I'll just say that they are anything that can be measured for the purposes of distinguishing states. I will also refrain from defining the concept of measurement since that could be an entire blog post unto itself.
At any rate, this raises the important question: is consciousness a system? The human body is a system; an automobile is a system; the galaxy is a system. Even language is a system which means systems can be abstract concepts as long as they have measurable properties and characteristics. Finding measurable properties and characteristics of consciousness is one of the goals of Giulio Tononi who, more than a decade ago, first introduced Integrated Information Theory (IIT) as a means of tackling these deep problems of consciousness.
I first met Tononi at the 2014 FQXi conference in Vieques which featured several terrific talks on IIT (and consciousness in general) from the likes of Tononi, Larissa Albantakis, Christoph Koch, and Chris Adami. One of the scheduled conference excursions was a kayak outing on a bioluminescent bay. The road out to the kayak launch was bone-jarringly rough (and was not helped by our guide's penchant for speed and the old van's completely useful shocks and struts). By the time we arrived I was on the verge of losing what little food I had eaten that day. Tononi drew the short straw and ended up as my kayaking companion. Ever gracious, he did much of the paddling and allowed me some time to recoup on the relatively still waters of the bay. We had a wide-ranging conversation about consciousness and, in particular, some of the disorders he had encountered as a psychiatrist over the years. It was a singularly memorable experience.
David Chalmers (who was also at the conference this year) has argued that any attempt to define consciousness in purely physical terms will eventually run into the so-called hard problem which is the problem of explaining how and why we have phenomenon-based experiences, i.e. how sensations, for example, can acquire properties such as taste. IIT, by contrast, begins simply by assuming consciousness exists, i.e. it takes it as an axiomatic truth. Questioning its existence isn't going to get us anywhere. In other words, I could assume that my entire life is nothing but a dream, but that won't stop the IRS from trying to collect my taxes every year. So IIT takes consciousness as being self-evident.
In truth this is how most of science works. We have to start somewhere when developing theories to explain the world and so we create hypotheses, propose axioms, and develop propositions which are then tested and analyzed. We may find that some are correct and we may find that some are not correct. But, again, we have to start somewhere.
Tononi starts with experience. He thinks of consciousness as having an experience. That's what defines it. As he noted in his talk at this year's conference, you can't "squeeze" consciousness out of the brain. In this sense, his approach would seem to circumvent Chalmers' argument that purely physical definitions will ultimately lead to the hard problem by starting with an axiom rather than a physically measurable phenomenon (as a note, Chalmers appears to be personally agnostic concerning IIT).
Of course any discussion of consciousness is bound to have a strong overlap with any discussion about existence, i.e. "what exists." Tononi seems to think there are gradations to the concept of existence. In other words, existence isn't binary. Existence is based on causal power. Anything with maximal causal power definitively exists. Consider the following example which is one I routinely have used when thinking about the nature of information. A painter locks himself (or herself) in a room and paints an absolute masterpiece. Just as the painting is finished, a fire engulfs the building, burning it to the ground, taking the painter and the painting with it. Did the painting exist?
I originally came up with this scenario in order to ask about the nature of the information contained in the painting and, in particular, in the aesthetic appreciation of the painting. The painting itself adds information to the world and burning it simply rearranges that information. But the aesthetic appreciation of that painting also adds information to the world. What happens to that information when the painting is burned if it (the information) is not conveyed to anyone else?
From the standpoint of IIT, this is akin to asking if the painting ever existed in the first place. As Masafumi Oizumi pointed out in his talk, IIT basically says that it is necessary (though not sufficient) for a system to produce information in order to produce consciousness. All of the raw materials of the painting were already in the building when it burned. Creating the painting had no lasting effect; it had no causal power. By the standards of IIT it never existed. As Tononi pointed out, functionally equivalent systems are not necessarily phenomenally equivalent systems and vice-versa. If we were to consider two equivalent rooms in the burned building, both with the same art supplies and each containing an artist, but assume that only one of the artists painted a masterpiece before the building burned to the ground, the two rooms are phenomenally equivalent in the record they left in the world and yet, since the artists each took different actions prior to the fire, the two rooms are not functionally equivalent.
At any rate, Tononi says that if consciousness does exist then its existence must be based on its maximal causal power. This, of course, raises all sorts of interesting sociological and psychological questions about people who go unnoticed by society, but the reason Tononi makes this assertion is because it provides a means by which the theory can be measured--integrated information, [equation]\Phi,[/equation]
the details of which are beyond this blog post. Suffice it to say that the panel discussion at the conference was lively and interesting. But as FQXi's fearless leader Max Tegmark pointed out, while IIT may or may not be correct, it is at least testable. As such, Christoph Koch of the Allen Brain Institute (who was at the 2014 conference but not at this year's) has said it is "the only really promising fundamental theory of consciousness."[1] That seems reason enough to study it.
[1] Carl Zimmer, "Sizing Up Consciousness by its Bits", New York Times, September 20, 2010.