# InFoRM: a unified inverse and forward model for sensorimotor control

## TL;DR

This paper introduces InFoRM, a single neural network that combines inverse and forward models for sensorimotor control, showing better performance and efficiency than traditional separate models.

## Contribution

The novel contribution is a unified neural network model that integrates inverse and forward sensorimotor functions, demonstrating improved performance and adaptability.

## Key findings

- InFoRM outperforms traditional architectures in reproducing cyclic reaching movements under various conditions.
- The unified model requires fewer computational resources than separate inverse and forward models.
- InFoRM can generate motor commands for untrained movement directions, showing adaptability beyond learned patterns.

## Abstract

Sensorimotor control models traditionally consist of two types of internal models: inverse models, which compute the motor commands needed to reach a desired movement goal, and forward models, which predict the resulting sensory feedback. These models are usually considered separate entities, but it is unclear whether such separation exists in the nervous system. Additionally, maintaining separate networks may be more computationally expensive. Therefore, we investigated whether these functions could be executed within a single neural circuit: an inverse-forward-recognition model (InFoRM). We implemented InFoRM using neural networks and compared their ability to reproduce cyclic reaching movements with that of control architectures based on classical, separated inverse and forward models. Desired movement trajectories were represented by recorded three-dimensional kinematics, while efferent (muscle activation) and afferent (muscle length and velocity) signals were obtained through inverse dynamics. Our findings show that InFoRM significantly outperforms control architectures across various conditions, while requiring fewer resources. The network is also able to morph to untrained movement directions, generating motor commands and predicted feedback that had not been learned. These findings demonstrate the computational advantages of integrating inverse and forward processes within a single neural network, suggesting that such unified sensorimotor models may be worthwhile to explore further.