Artificial neural networks (ANNs) constitute a class of flexible nonlinear models designed to mimic biological neural systems. In this entry, we introduce ANN using familiar econometric terminology and provide an overview of ANN modeling approach and its implementation methods. † Correspondence: Chung-Ming Kuan, Institute of Economics, Academia Sinica, 128 Academia Road, Sec. 2, Taipei 115, Taiwan; ckuan@econ.sinica.edu.tw. †† I would like to express my sincere gratitude to the editor, Professor Steven Durlauf, for his patience and constructive comments on early drafts of this entry. I also thank Shih-Hsun Hsu and Yu-Lieh Huang for very helpful suggestions. The remaining errors are all mine.

Artificial neural networks

The development of more effective rehabilitative interventions requires a better understanding of how humans learn and transfer motor skills in real-world contexts. Presently, clinicians design interventions to promote skill learning by relying on evidence from experimental paradigms involving simple tasks, such as reaching for a target. While these tasks facilitate stringent hypothesis testing in laboratory settings, the results may not shed light on performance of more complex real-world skills. In this perspective, we argue that virtual environments (VEs) are flexible, novel platforms to evaluate learning and transfer of complex skills without sacrificing experimental control. Specifically, VEs use models of real-life tasks that afford controlled experimental manipulations to measure and guide behavior with a precision that exceeds the capabilities of physical environments. This paper reviews recent insights from VE paradigms on motor learning into two pressing challenges in rehabilitation research: 1) Which training strategies in VEs promote complex skill learning? and 2) How can transfer of learning from virtual to real environments be enhanced? Defining complex skills by having nested redundancies, we outline findings on the role of movement variability in complex skill acquisition and discuss how VEs can provide novel forms of guidance to enhance learning. We review the evidence for skill transfer from virtual to real environments in typically developing and neurologically-impaired populations with a view to understanding how differences in sensory-motor information may influence learning strategies. We provide actionable suggestions for practicing clinicians and outline broad areas where more research is required. Finally, we conclude that VEs present distinctive experimental platforms to understand complex skill learning that should enable transfer from therapeutic practice to the real world.

/pdf/learning-and-transfer-of-complex-motor-skills-in-virtual-c17bnijzu1.pdf

Learning and transfer of complex motor skills in virtual reality: a perspective review

We rarely experience difficulty picking up objects, yet of all potential contact points on the surface, only a small proportion yield effective grasps. Here, we present extensive behavioral data alongside a normative model that correctly predicts human precision grasping of unfamiliar 3D objects. We tracked participants' forefinger and thumb as they picked up objects of 10 wood and brass cubes configured to tease apart effects of shape, weight, orientation, and mass distribution. Grasps were highly systematic and consistent across repetitions and participants. We employed these data to construct a model which combines five cost functions related to force closure, torque, natural grasp axis, grasp aperture, and visibility. Even without free parameters, the model predicts individual grasps almost as well as different individuals predict one another's, but fitting weights reveals the relative importance of the different constraints. The model also accurately predicts human grasps on novel 3D-printed objects with more naturalistic geometries and is robust to perturbations in its key parameters. Together, the findings provide a unified account of how we successfully grasp objects of different 3D shape, orientation, mass, and mass distribution.

/pdf/predicting-precision-grip-grasp-locations-on-three-35iul13a0y.pdf

Predicting precision grip grasp locations on three-dimensional objects.

Grasping is one of the fundamental actions we perform to interact with objects in real environments, and in the real world we rarely experience difficulty picking up objects. Grasping plays a fundamental role for interactive virtual reality (VR) systems that are increasingly employed not only for recreational purposes, but also for training in industrial contexts, in medical tasks, and for rehabilitation protocols. To ensure the effectiveness of such VR applications, we must understand whether the same grasping behaviors and strategies employed in the real world are adopted when interacting with objects in VR. To this aim, we replicated in VR an experimental paradigm employed to investigate grasping behavior in the real world. We tracked participants' forefinger and thumb as they picked up, in a VR environment, unfamiliar objects presented at different orientations, and exhibiting the same physics behavior of their real counterparts. We compared grasping behavior within and across participants, in VR and in the corresponding real world situation. Our findings highlight the similarities and differences in grasping behavior in real and virtual environments.

Grasping objects in immersive Virtual Reality

Shape is a defining feature of objects. Yet, no image-computable model accurately predicts how similar or different shapes appear to human observers. To address this, we developed a model (ShapeComp), based on over 100 shape features (e.g., area, compactness, Fourier descriptors). When trained to capture the variance in a database of >25,000 animal silhouettes, ShapeComp predicts human shape similarity judgments almost perfectly (r2>0.99) without fitting any parameters to human data. To test the model, we created carefully selected arrays of complex novel shapes using a Generative Adversarial Network trained on the animal silhouettes, which we presented to observers in a wide range of tasks. Our findings show that human shape perception is inherently multidimensional and optimized for comparing natural shapes. ShapeComp outperforms conventional metrics, and can also be used to generate perceptually uniform stimulus sets, making it a powerful tool for investigating shape and object representations in the human brain.

/pdf/an-image-computable-model-of-human-visual-shape-similarity-1am5r8tbjw.pdf

An image-computable model of human visual shape similarity

The plasticity of the human nervous system allows us to acquire an open-ended repository of sensorimotor skills in adulthood, such as the mastery of tools, musical instruments or sports. How novel sensorimotor skills are learned from scratch is yet largely unknown. In particular, the so-called inverse mapping from goal states to motor states is underdetermined because a goal can often be achieved by many different movements (motor redundancy). How humans learn to resolve motor redundancy and by which principles they explore high-dimensional motor spaces has hardly been investigated. To study this question, we trained human participants in an unfamiliar and redundant visually-guided manual control task. We qualitatively compare the experimental results with simulation results from a population of artificial agents that learned the same task by Goal Babbling, which is an inverse-model learning approach for robotics. In Goal Babbling, goal-related feedback guides motor exploration and thereby enables robots to learn an inverse model directly from scratch, without having to learn a forward model first. In the human experiment, we tested whether different initial conditions (starting positions of the hand) influence the acquisition of motor synergies, which we identified by Principal Component Analysis in the motor space. The results show that the human participants’ solutions are spatially biased towards the different starting positions in motor space and are marked by a gradual co-learning of synergies and task success, similar to the dynamics of motor learning by Goal Babbling. However, there are also differences between human learning and the Goal Babbling simulations, as humans tend to predominantly use Degrees of Freedom that do not have a large effect on the hand position, whereas in Goal Babbling, Degrees of Freedom with a large effect on hand position are used predominantly. We conclude that humans use goal-related feedback to constrain motor exploration and resolve motor redundancy when learning a new sensorimotor mapping, but in a manner that differs from the current implementation of Goal Babbling due to different constraints on motor exploration.

/pdf/goal-related-feedback-guides-motor-exploration-and-3xfcr50efm.pdf

Goal-related feedback guides motor exploration and redundancy resolution in human motor skill acquisition

We report an illusion in which the felt weight of an object changes depending on whether a previously manipulated object was lighter or heavier The illusion is not modulated by visual weight cues, yet it transfers across hands

https://journals.sagepub.com/doi/pdf/10.1177/2041669518790275

The Sequential-Weight Illusion.

Humans exhibit spatial biases when grasping objects. These biases may be due to actors attempting to shorten their reaching movements and therefore minimize energy expenditures. An alternative explanation could be that they arise from actors attempting to minimize the portion of a grasped object occluded from view by the hand. We reanalyze data from a recent study, in which a key condition decouples these two competing hypotheses. The analysis reveals that object visibility, not energy expenditure, most likely accounts for spatial biases observed in human grasping.

https://journals.sagepub.com/doi/pdf/10.1177/2041669519827608

Lina K Klein

Papers

Predicting precision grip grasp locations on three-dimensional objects.

Grasping objects in immersive Virtual Reality

Goal-related feedback guides motor exploration and redundancy resolution in human motor skill acquisition

The Sequential-Weight Illusion.

Object Visibility, Not Energy Expenditure, Accounts For Spatial Biases in Human Grasp Selection.