A better way to test for consciousness?

ConsScale Pyramid

No doubt we need better ways to test for consciousness, as much as we need better definitions. From the point of view of science, Consciousness is a controversial and quite elusive phenomenon and trying to build good tests is actually a crucial part of the quest to understand it. Assessment and definition are always inseparable (otherwise, we wouldn’t know what we are testing).

Effectively testing for consciousness in humans is something we take for granted, specially when we put our trust in anesthesiologists right before going into surgery, or when ER physicians perform a Glasgow coma test to assess the level of consciousness of a patient. Assessing the level of consciousness in human subjects is usually a non problematic or challenging task (except from locked-in syndrome patients and the like). However, when we consider the vast number of other sorts of organisms, such as other animals, plants and machines things get much more complicated. The problems is essentially the same for all creatures, and scientifically speaking it involves the very same problem. However, there is a huge difference when we speak about humans: for humans we do assume consciousness as a legitimate feature of the living organism.

So, what do we do if we want to come up with a universal test for consciousness? One that might be applicable to virtually any creature possible, including biological organisms, artificial machines or cyborgs? What features or what measurements do we need to do? In other words, what is consciousness made of, so we can measure it?

ConsScale Summary

As neatly described by Musser in his aeon essay (Consciousness Creep. Our machines could become self-aware without our knowing it) we are making efforts, and hopefully some progress, into building new ways to test for consciousness. ConsScale is an example of this quest for both understanding and measuring consciousness.

Musser presents in his essay several of the tests that have been recently proposed, explaining the vision and position of the authors, including my own. It’s interesting to see how different approaches for testing imply different assumptions about what consciousness is. Tononi’s approach is based on the Information Integration theory, ConsScale is based on cognitive development, Haikonen stresses the importance of inner talk, Schwitzgebel raises the question of consciousness in groups (super-organisms). Perhaps we need to look for a new approach able to deal with all these aspects within the same framework.

ConsScale. A Scale for Measuring Machine Consciousness

Measuring Machine Consciousness

ConsScale is a tool for assessing the functional level of consciousness of a creature. It has been specifically designed for the evaluation of Machine Consciousness implementations.

Now a ConsScale microsite is available where you can explore the conceptual levels of consciousness defined in the scale, learn how agents can also be rated using a quantitative score, and use the online calculator to rate your own implementations:

http://conscious-robots.com/consscale/

ConsScale is a framework for characterizing the cognitive power of a creature. ConsScale includes the definition of an ordered list of cognitive levels arranged across a developmental path. The arrangement of the levels is inspired on the ontogeny and phylogeny of consciousness in biological organisms.

The basic assumption is that there exist different kinds of minds, and they can be characterized in terms of ConsScale criteria. Using ConsScale, characterization and assessment of consciousness can be performed using three related tools:

– the ConsScale conceptual levels of consciousness (levels),
– the CQS (ConsScale Quantitative Score) (CQS), and
– the ConsScale radar graph representation (Calculator).

In order to assess the level of artificial consciousness of an agent using ConsScale, its architectural components have to be identified and its cognitive skills tested. Using this information as input, the scale can be used to obtain both a qualitative and a quantitative measure of consciousness:

ConsScale Process
ConsScale Process