Temporal signatures of prediction error reveal external forces on an embodied robot.
- 1Lund University Cognitive Robotics Lab, Department of Philosophy and Cognitive Science, Lund University, Lund, Sweden.
Abstract
Embodied robots must distinguish self-generated sensory consequences from externally caused perturbations. In predictive processing, such deviation is accounted for as prediction error, but it is commonly treated primarily as a signal to be minimised. Here, we test whether the temporal profile of load-related prediction error provides diagnostic information about the type of physical interaction a robot experiences. We trained and evaluated windowed multilayer perceptrons to predict servo current, used as a proxy for joint load, from proprioceptive, inertial, and servo current history. The training data were recorded during random movements of tilt and pan servo motors of a robotic head. The best model was deployed in the robot in real time and evaluated across free movements and external perturbations, including brief pushes, sustained forces, and obstacle interactions. Prediction error and raw current time series were summarised using features capturing magnitude statistics, temporal structure, and signal shape. A random forest classifier was then used to classify the conditions based on the features. Different interaction conditions produced distinct profiles that supported robust perturbation classification with high cross-validated accuracy. Classification performance for prediction error was equal to the classification of raw current. The results suggest that load-related prediction error is not only a residual of imperfect prediction. Its temporal structure can act as an embodied self-monitoring signal for classifying physical interactions.
Related
- Disentangling the temporal signatures of conditioned pupil dilation: Distinct valence- and prediction-error-related components revealed by mega-analysis
- Embodied Emergence of Temporal Complexity in Finger Tapping: Synaptic and Dopaminergic Control in Cortico-Basal Ganglia Loops
- Ascending Inputs to Inferior Colliculus Subdivisions Reveal Pathway-Specific Hybrid Organization in the External Cortex
- Neural signatures of spontaneous transitions between internal and external thought
- RFind-sc maps single cells onto biological axes defined by external gene signatures
BibTeX
@article{klintefors2026temporal,
title = {{Temporal signatures of prediction error reveal external forces on an embodied robot.}},
author = {Pierre Klintefors and Birger Johansson and Christian Balkenius},
journal = {Frontiers in neurorobotics},
year = {2026},
month = oct,
doi = {10.3389/fnbot.2026.1918275},
url = {https://europepmc.org/article/MED/42953198},
note = {CC BY},
}
RIS
TY - JOUR
TI - Temporal signatures of prediction error reveal external forces on an embodied robot.
AU - Pierre Klintefors
AU - Birger Johansson
AU - Christian Balkenius
PY - 2026
DA - 2026/10/14
JO - Frontiers in neurorobotics
DO - 10.3389/fnbot.2026.1918275
UR - https://europepmc.org/article/MED/42953198
AB - Embodied robots must distinguish self-generated sensory consequences from externally caused perturbations. In predictive processing, such deviation is accounted for as prediction error, but it is commonly treated primarily as a signal to be minimised. Here, we test whether the temporal profile of load-related prediction error provides diagnostic information about the type of physical interaction a robot experiences. We trained and evaluated windowed multilayer perceptrons to predict servo current, used as a proxy for joint load, from proprioceptive, inertial, and servo current history. The training data were recorded during random movements of tilt and pan servo motors of a robotic head. The best model was deployed in the robot in real time and evaluated across free movements and external perturbations, including brief pushes, sustained forces, and obstacle interactions. Prediction error and raw current time series were summarised using features capturing magnitude statistics, temporal structure, and signal shape. A random forest classifier was then used to classify the conditions based on the features. Different interaction conditions produced distinct profiles that supported robust perturbation classification with high cross-validated accuracy. Classification performance for prediction error was equal to the classification of raw current. The results suggest that load-related prediction error is not only a residual of imperfect prediction. Its temporal structure can act as an embodied self-monitoring signal for classifying physical interactions.
SN - 1662-5218
ER -