I will try to express here two worries that I have had dancing around in my head (and probably body too) for some time now, but which I had not been too explicitly aware of until relatively recently. These preoccupations have arisen from general conversations with people in CCNR-related environments and have occurred with more than one person in different contexts, but it is not aimed at any one person in particular.
While there is a general feeling of people getting tired of this 2-behaviours, 2-environment-types, “2-attractors”, dynamical systems analyses of some minimally cognitive tasks.. most of any such analysis is still in such an early period of its infancy, because of the difficulties in studying multiple timescale, transient, hybrid (discontinuous as well as continuous variables), and non-autonomous (from the situatedness) dynamical systems, that I cannot help but fall into somewhat of a state of despair. On the one hand, I’m thinking to myself that I need another 4 years to understand even such simplest of systems. On the other, I can sort of feel the pressure of others around me suggesting that these are either too simple or too boring or too well understood already, and most likely all of the above.
Similarly, while many around me feel as if, unless an algorithmic version of the strategy of an evolved agent can be extracted from prolonged sessions of observing it interacting with the environment, little has been learned about the system; I remain to be more interested in a geometric/topological understanding of the interaction – from which I can feel many people near me remain skeptical about, despite these people being (more or less) on the dynamical-systems-understanding-of-adaptive-behaviour side of things.
I guess one has to properly explain the motivation for tackling any or both of these issues as clearly as possible in order to be understood, avoid such trivial criticisms, and hopefully even make the feeling of pressure to ‘move on to the interesting stuff’ disappear.