Planning of reach-and-grasp movements: effects of validity and type of object information.

Individuals are assumed to plan reach-and-grasp movements by using two separate processes. In 1 of the processes, extrinsic (direction, distance) object information is used in planning the movement of the arm that transports the hand to the target location (transport planning); whereas in the other, intrinsic (shape) object information is used in planning the preshaping of the hand and the grasping of the target object (manipulation planning). In 2 experiments, the authors used primes to provide information to participants (N = 5, Experiment 1; N = 6, Experiment 2) about extrinsic and intrinsic object properties. The validity of the prime information was systematically varied. The primes were succeeded by a cue, which always correctly identified the location and shape of the target object. Reaction times were recorded. Four models of transport and manipulation planning were tested. The only model that was consistent with the data was 1 in which arm transport and object manipulation planning were postulated to be independent processes that operate partially in parallel. The authors suggest that the processes involved in motor planning before execution are primarily concerned with the geometric aspects of the upcoming movement but not with the temporal details of its execution.

Finding final postures.

In this study, a model of movement planning (Rosenbaum, Engelbrecht, Bushe, & Loukopoulos, 1993a, 1993b; Rosenbaum, Loukopoulos, Meulenbroek, Vaughan, & Engelbrecln, 1995), in which movements are generated on the basis of the efficacy of different possible goal postures, was tested. The model predicts which limb segments will be used and how the segments will be combined in reaching. The model's predictions were compared with observations from a study in which seated participants reached for targets in a sagittal plane, using the hip, shoulder, and elbow. Estimates of 4 free parameters-an expense factor for each of the 2 contributing joints, and a 4th parameter that specified the relative weight of spatial accuracy versus effort minimization, were used for fitting the model to the observed joint angles. The model accounted for 96% of the variance of the observed joint angles and did a better job accounting for the data than several alternative models. A key new finding was that balance constraints play a more important role in determining joint contributions than was previously recognized.

Adaptation of a reaching model to handwriting: how different effectors can produce the same written output, and other results.

This report shows how a model initially developed for the control of reaching can be adapted for the control of handwriting. The main problem addressed by the model is how people can produce essentially the same written output with different effectors (e.g., the preferred or nonpreferred hand, the foot, or even the mouth). The model is based on the assumption that writers strive for invariant graphic outputs when they write with different effectors, when they write on surfaces with different orientations, or when they write large or small script; such output invariance is an essential requirement for later recognition of the written result. Given this assumption, the question is how the motor system enables the relevant effectors to generate the necessary pen strokes. The adapted model provides one possible answer to this question. It is first fully working model of multijoint activity underlying writing and related graphic tasks. We describe how the model differs from other models developed in the past, and we review the model's strengths and weaknesses.

Planning reaches by evaluating stored postures.

This article describes a theory of the computations underlying the selection of coordinated motion patterns, especially in reaching tasks. The central idea is that when a spatial target is selected as an object to be reached, stored postures are evaluated for the contributions they can make to the task. Weights are assigned to the stored postures, and a single target posture is found by taking a weighted sum of the stored postures. Movement is achieved by reducing the distance between the starting angle and target angle of each joint. The model explains compensation for reduced joint mobility, tool use, practice effects, performance errors, and aspects of movement kinematics. Extensions of the model can account for anticipation and coarticulation effects, movement through via points, and hierarchical control of series of movements.

Knowledge Model for Selecting and Producing Reaching Movements.

This article presents a new model of reaching control. The aim of the model is to characterize the computations underlying the selection of coordinated motion patterns among the limb segments. When a spatial target is selected, stored postures are evaluated for the contributions they can make to the task, and a special weighted average (the gaussian average) is taken of the postures to find a single target posture. Movement to the target posture is achieved without explicit planning of the trajectory. Rather, the reaching motion is driven by error correction (reducing the discrepancy between the current and target posture) shaped by inertia. The model solves the degrees-of-freedom problem for reaching. It also allows joints to compensate automatically for reduced mobility of other joints and explains established effects of practice, speed-accuracy trade-off, and kinematics. The model can be extended to other tasks and motor subsystems because of the generality of its underlying concepts.

A model for reaching control.

In this paper we propose that reaches are made to target postures which are selected by evaluating stored postures. Target postures are chosen by taking a weighted average of the stored postures, where the weights assigned to the stored postures depend on their effectiveness for the task. Movements from starting postures to target postures are achieved by reducing the distance, in joint space, between the two. The form of the movement depends on drive (assumed to decrease as less distance remains) and inertia. The model predicts the Power Law of learning, compensation for immobility of joints, changes in limb contributions depending on movement speed, asymmetric bell-shaped velocity profiles, velocity-amplitude relations, Fitts' Law, and position-dependent variations in hand-path curvature. Planned extensions of the model may broaden its application-for example, to handwriting.