Latash presented a control scheme that bears simultaneously on two fundamental challenges faced by biological movement systems: (a) the temporary assembly of the many components of the body into task-specific organizations appropriate for task performance, and (b) the preservation of the functional output of the assembled organization in the face of perturbations. According to his proposition, the realization of (a) is supported by a control variable that sets up task-appropriate body configurations by shifting thresholds for muscle activation—a style of control consistent with the Equilibrium Point (EP) Hypothesis. The realization of (b) may involve the organization of a multi-muscle synergy that stabilizes the desired value of a particular performance variable—a strategy consistent with the Uncontrolled Manifold hypothesis. Importantly, Latash suggested that the organization of synergies might, in fact, follow naturally from the EP-style of control with referent configurations. It is our contention that crucial for such a proposition and its gener alization is the determination of informational constraints required for modulating the identified control variables according to task demands. A referent configuration corresponds to a body configuration in which all the involved muscles are at their threshold of activation via the tonic stretch reflex. By shifting such thresholds at the level of individual joints, configurations of joints, and the body as a whole (see Feldman, 1998; Feldman & Levin, 1995), the control variable sets a global attractive state to which the movement system relaxes in the context of body-environment interactions. This attractive state supposedly corresponds to a task-appropriate and stable body posture associated with minimal muscle activation. The promise is that any deviation from this state, elicited by external perturbations for example, naturally results in activation of appropriate muscles synergies to maintain the task-appropriate body (or limb) configuration. Latash suggests that the referent configuration is set by the controller according to task requirements based on perception of what the movement system has to accomplish as a whole. In an abstract sense, the controller creates energy gradients that direct the movement system toward the desired goal state (see Kugler, Kelso, & Turvey, 1980). Latash, however, offers no suggestions as to the nature of the task-specific informational variables necessary to shape such a gradient. Arguably, the primary variables in question must be (a) carried in the patterns of soft tissue deformation, (b) specific to positions and motions of the limb segments relative to