dataset-ErrP-HRI

Dataset: A feasibility study for validating robot actions using EEG-based error-related potentials (ErrP)

This repository contains the dataset accompanying the publication:

Ehrlich, S. K., & Cheng, G. (2019).
A feasibility study for validating robot actions using EEG-based error-related potentials.
International Journal of Social Robotics, 11(2), 271–283.
http://dx.doi.org/10.1007/s12369-018-0501-8

The dataset is designed for research on error-related potentials (ErrP) in human–robot interaction (HRI), where ErrPs are elicited when a human observes robot behavior that is incorrect or unexpected.

---

Repository structure

dataset-ErrP-HRI/
├── cursor/                              # EEG data: "cursor" condition / scenario
│   └── s02_cursor.fdt
│   └── s02_cursor.set
│   └── ...
├── robot/                               # EEG data: "robot" condition / scenario
│   └── s02_robot.fdt
│   └── s02_robot.set
│   └── ...
├── documentation_dataset-ErrP-HRI.pdf    # dataset documentation (recommended)
└── README.md

Study description (brief)

ErrP dataset in HRI observation tasks

Purpose: provide EEG recordings that allow analysis and classification of error-related potentials (ErrP) elicited by observed robot actions.

Design: Participants observed system behavior across different scenarios/conditions. Trials are annotated such that EEG epochs can be assigned to error vs. correct events (and potentially further task phases depending on the dataset documentation).

This dataset supports typical ErrP pipelines such as:

• epoch extraction around error/non-error events
• ERP averaging and statistical analysis
• single-trial classification (e.g., LDA / SVM / Riemannian / CNN baselines)

---

About the dataset

Intended use

This dataset supports research in:

• ErrP detection and modeling
• passive BCI for human–robot interaction
• neuroergonomics (implicit feedback signals)
• evaluation of robust EEG classification under realistic HRI conditions

Modalities and measures (overview)

Depending on the condition folder (cursor/, robot/), the dataset includes:

• EEG recordings
• event markers / labels enabling error vs. correct segmentation
• scenario-dependent task metadata

For exact data formats, EEG configuration, event codes, and label semantics, consult the provided documentation.

• documentation_dataset-ErrP-HRI.pdf

---

Citation

If you use this data please cite below publication.

@article{ehrlich2019feasibility,
  title   = {A feasibility study for validating robot actions using eeg-based error-related potentials},
  author  = {Ehrlich, Stefan K. and Cheng, Gordon},
  journal = {International Journal of Social Robotics},
  volume  = {11},
  number  = {2},
  pages   = {271--283},
  year    = {2019},
  doi     = {10.1007/s12369-018-0501-8}
}