Grounding higher cognition in sensorimotor and predictive loops

Post date: Oct 20, 2012 8:40:58 AM

How do living organisms (and possibly robots) develop higher cognition from sensorimotor skills?

My tentative answer --in keeping with embodied and motor theories of cognition-- is that the architecture of motor prediction and control of our earlier ancestors was gradually improved to afford cognitive control and executive functions (and in parallel, joint actions and communication in the social domain).

From an evolutionary perspective, the vertebrate brain's architecture developed to meet the needs of interactive behaviour, and has to continuously specify and select among its action alternatives (Cisek & Kalaska, 2010). In our "cognitive leverage hypothesis", we hypothesized that predictive abilities originally developed for motor control have been “exapted” during evolution and have bootstrapped increasingly sophisticated cognitive abilities, including prospection, thinking, and cognitive control.

A crucial step in this process is the reuse of predictive abilities off-line and outside the overt sensorimotor loop; in this view, thinking is essentially, an “internalization” of the sensorimotor loop and is supported by the same predictive-control dynamics (Pezzulo, 2011; Pezzulo and Castelfranchi, 2007, 2009). The close relationships between mental imagery and the systems for motor preparation provide some support to this hypothesis (Cisek & Kalaska, 2010; Koziol et al., 2012).

An example of embodied problem solving. During climbing competitions, before climbing an unknown route, athletes can pre-view it for some minutes; they typically mimic, imagine and plan their future climbing movements (in the picture, they do this overtly, but this is not always the case). This is a complex problem solving, as (route settlers ensure that) the sequence of movements to reach the top of the wall is novel and far from trivial. It depends on goal state (the climbing hold to reach) and previous moves. Second, it incorporates the athlete's embodied knowledge, as length of limbs, strength of fingers, affordances offered by the various kinds of climbing holds, possible or impossible kinematics, all constraint the problem solving process. Part of the athletes' skill is in the ability to anticipate a lot of information (proprioceptive, body posture at critical points, how much force to use, etc.) prior to climbing, as during climbing there is little place for re-planning and errors. In Pezzulo et al., 2010 we reported an advantage of expect climbers in a memory task (i.e., remembering sequences of holds in a route), but only when "climb-ability" constraints were respected (not when the sequences formed a non-climbable route). Photos courtesy of http://risk4sport.com/

We continue conducting conceptual, empirical and computational modeling studies to support these ideas (Pezzulo 2009; Pezzulo et al., 2010).

Selected Pubs:

• Pezzulo, G., Cisek P. (2016) Navigating the Affordance Landscape: Feedback Control as a Process Model of Behavior and Cognition. Trends in Cognitive Sciences 20 (6), 414-424 [link]

• Pezzulo, G. and Castelfranchi, C. (2009). Thinking as the control of imagination: a conceptual framework for goal-directed systems. Psychological Research, 73(4):559–577. [pdf]

• Pezzulo, G. and Castelfranchi, C. (2007). The symbol detachment problem. Cognitive Processing, 8(2):115–131. [more] [pdf]

• Pezzulo, G., Barca, L., Bocconi, A. L., and Borghi, A. M. (2010). When affordances climb into your mind: Advantages of motor simulation in a memory task performed by novice and expert rock climbers. Brain and Cognition, 73(1):68–73. [pdf]

• Pezzulo, G. (2011). Grounding procedural and declarative knowledge in sensorimotor anticipation. Mind and Language, 26(1):78–114. [pdf]

• Pezzulo, G. (2009). Dipra: A layered agent architecture which integrates practical reasoning and sensorimotor schemas. Connection Science, 21(4):297–326. [more] [pdf]

Other Pubs:

• Cisek, P. & Kalaska, J. F. (2010) Neural mechanisms for interacting with a world full of action choices. Annu Rev Neurosci, 33, 269-298

• Koziol, L. F.; Budding, D. E. & Chidekel, D. (2012) From movement to thought: executive function, embodied cognition, and the cerebellum. Cerebellum, 11, 505-525