An overview of adaptive model theory: solving the problems of redundancy, resources, and nonlinear interactions in human movement control

Abstract

Adaptive model theory (AMT) is a computational theory that addresses the difficult control problem posed by the musculoskeletal system in interaction with the environment. It proposes that the nervous system creates motor maps and task-dependent synergies to solve the problems of redundancy and limited central resources. These lead to the adaptive formation of task-dependent feedback/feedforward controllers able to generate stable, noninteractive control and render nonlinear interactions unobservable in sensory-motor relationships. AMT offers a unified account of how the nervous system might achieve these solutions by forming internal models. This is presented as the design of a simulator consisting of neural adaptive filters based on cerebellar circuitry. It incorporates a new network module that adaptively models (in real time) nonlinear relationships between inputs with changing and uncertain spectral and amplitude probability density functions as is the case for sensory and motor signals.

Consciousness as recursive, spatiotemporal self-location

“…DS takes as input homeostatic needs. (Adapted from Miall et al 1993;Neilson & Neilson 2005) Figure 3 figure. Panel B shows the circuit consisting of DS-CS comparator and DS-FM comparator diminishing in importance, as does Inverse Controller to FM; while the CS-FM reciprocal connectivity strengthens. Also, the this-here-now Event Schema ("SL Event Schema") constitutes an ever-greater proportion of both CS and FM.…”

“…And biological organisms can be viewed most basically as energy regimes (over half the body's weight is devoted to cellular elements that manufacture energy), so that 'any organism that can capture prey, escape predators and achieve goals with a minimum demand on metabolic energy, has an evolutionary advantage' (Neilson and Neilson, 2005). Thus, whatever the somatic and neurological changes engineered by the need to meet specific environmental challenges, the essential ongoing challenge is always to operate with maximum energy efficiency.…”

“…Simpler versions of this architecture often omit a principled representation of the current state, which must be coherently represented in order to be compared to desired state (DS), particularly when the CS of this particular processing module is not a simple biological marker such as glucose level, but the current spatiotemporal self-location of the agent in relation to the environment. The ongoing coherence of this Current-State-of-SpatiotemporalOrientation representation is maintained by the continuous structuring input from a memory buffer (Merfeld, 2001;Bridgeman et al, 1994;Neilson and Neilson, 2005) containing the selflocating event schema (SL EVENT SCHEMA). As explained, the this-here-now self-locating event schema, is an abstracted generalization over ongoing interactive experience which serves to organize ongoing phenomenal experience (Current State) in terms of an extended spatiotemporal manifold, or more fully, in terms of a self-moving causal agent continuously located spatially and temporally within such a manifold.…”

“…When connected to a memory system that stores sensory-input to motor-output conjunctions, the Inverse Controller is capable of retrieving motor output likely to produce required sensory inputs (Held, 1968), of taking as input a perceptual representation of the required adjustment and producing as output the motor movement that will result in perceptual feedback incorporating that adjustment so that Current State (CS) becomes more closely aligned with Desired State (DS). The Inverse Controller, in other words, translates the desired perceptually-represented goal of a movement into the set of actions (a motor command) needed to achieve that movement (Neilson and Neilson, 2005).…”

“…Current State (CS) is constituted by three inputs: (1) continuous feedforward updating from the Forward Model; (2) periodic feedback from the sensory systems (as gated by expected feedback represented by the FM); and finally (3), CS is consistently structured as the representation of an interaction event by the structuring input from a memory buffer (Merfeld, 2001;Bridgeman et al, 1994;Neilson and Neilson, 2005) consisting of the self-locating event schema (SL EVENT SCHEMA). Based on brain imaging studies consistently identifying the posteromedial cortex as the neural source for conscious cognition (Laureys, 2004), and the direct innervation of this area by the wake/sleep arousal system sourced from the postero-lateral hypothalamus via the basal forebrain (Saper, Scammell and Lu, 2005), it is postulated that this structure-providing SL EVENT SCHEMA memory buffer is independently activated by the sleep/wake innervation system, thus providing a self-activating base to the interactional configuration of waking experience.…”