# Measuring and Modeling Physical Intrinsic Motivation

arXiv: https://arxiv.org/abs/2305.13452 · HTML: https://mlsanity.com/paper/2305.13452/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 24 references that are in the map.
- First posted on arXiv 22 May 2023.

## Builds on (most cited first, 13 of 13)

- [Curiosity-driven Exploration by Self-supervised Prediction](https://mlsanity.com/paper/1705.05363/) (2017)
- [Exploration by Random Network Distillation](https://mlsanity.com/paper/1810.12894/) (2018)
- [MCVD: Masked Conditional Video Diffusion for Prediction, Generation, and Interpolation](https://mlsanity.com/paper/2205.09853/) (2022)
- [Self-Supervised Exploration via Disagreement](https://mlsanity.com/paper/1906.04161/) (2019)
- [Automated Curriculum Learning for Neural Networks](https://mlsanity.com/paper/1704.03003/) (2017)
- [ThreeDWorld: A Platform for Interactive Multi-Modal Physical Simulation](https://mlsanity.com/paper/2007.04954/) (2020)
- [Learning Particle Dynamics for Manipulating Rigid Bodies, Deformable Objects, and Fluids](https://mlsanity.com/paper/1810.01566/) (2018)
- [Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models](https://mlsanity.com/paper/1507.00814/) (2015)
- [Physion: Evaluating Physical Prediction from Vision in Humans and Machines](https://mlsanity.com/paper/2106.08261/) (2021)
- [Flexible Neural Representation for Physics Prediction](https://mlsanity.com/paper/1806.08047/) (2018)
- [Surprise-Based Intrinsic Motivation for Deep Reinforcement Learning](https://mlsanity.com/paper/1703.01732/) (2017)
- [Learning to Play with Intrinsically-Motivated Self-Aware Agents](https://mlsanity.com/paper/1802.07442/) (2018)
- [Active World Model Learning with Progress Curiosity](https://mlsanity.com/paper/2007.07853/) (2020)

## Abstract (arXiv)

Humans are interactive agents driven to seek out situations with interesting physical dynamics. Here we formalize the functional form of physical intrinsic motivation. We first collect ratings of how interesting humans find a variety of physics scenarios. We then model human interestingness responses by implementing various hypotheses of intrinsic motivation including models that rely on simple scene features to models that depend on forward physics prediction. We find that the single best predictor of human responses is adversarial reward, a model derived from physical prediction loss. We also find that simple scene feature models do not generalize their prediction of human responses across all scenarios. Finally, linearly combining the adversarial model with the number of collisions in a scene leads to the greatest improvement in predictivity of human responses, suggesting humans are driven towards scenarios that result in high information gain and physical activity.