MorseEEG-ATP

MorseEEG-ATP: a multitask EEG dataset for exploring auditory temporal processing with Morse code

===================================

Introduction
------------
MorseEEG-ATP is a publicly available multitask electroencephalography (EEG) dataset for studying auditory temporal processing and temporal pattern discrimination using Morse code (MC) stimuli. Auditory temporal processing refers to the ability of the auditory system to perceive, encode, and integrate the temporal structure of sounds. The dataset was designed to address the relative lack of open resources targeting the intermediate stage between low-level acoustic encoding and higher-level language or cognitive processing, namely the extraction and discrimination of stable temporal patterns from discrete acoustic elements governed by explicit temporal rules. MC provides a controlled and parameterizable stimulus model for this purpose because it consists of short and long acoustic elements combined according to fixed rules, has a clear and stable temporal structure, and depends relatively little on lexical or semantic knowledge in untrained listeners.

Dataset Content
---------------
The dataset contains synchronously recorded EEG and behavioral data from 32 healthy adult participants (14 females). Data were collected with a 62-channel EEG system at a sampling rate of 600 Hz and are organized in an EEG-BIDS compatible structure (BIDS v1.10.1). The release includes:

1. Raw EEG recordings in BrainVision format (`.vhdr`, `.vmrk`, `.eeg`) together with BIDS sidecar files.
2. Minimally preprocessed EEG data under `derivatives/preprocessed/` for reproducible downstream analysis.
3. Trial-aligned event annotations, including event onsets, trigger values, block labels, speed conditions, and stimulus file references.
4. Behavioral measures including accuracy and reaction time, together with participant-level summary information.
5. Companion preprocessing and benchmark analysis scripts.

Experimental Design
-------------------
- **Tasks**: The experiment includes two auditory discrimination tasks that share the same stimulus set and trial structure but differ in decision rules.
- **Dot-count task (`task-dotcount`)**: Participants report the number of dots contained in the presented Morse code character.
- **First-last match task (`task-firstlast`)**: Participants judge whether the first and last elements of the presented Morse code character are identical.
- **Presentation-rate conditions**: Each task includes three presentation-rate conditions defined by Morse character transmission speed: 15, 30, and 40 characters per minute (CPM). These conditions support comparisons across different temporal-rate demands.
- **Rationale**: By jointly manipulating task rules and presentation rate, the dataset supports comparative analyses of auditory temporal processing, temporal pattern discrimination, and listening ability assessment across processing demands and speed conditions.
- **Trial procedure**: Each trial consists of a fixation period, stimulus presentation, a response window, and a jittered inter-trial interval.

Data Acquisition
----------------
- **EEG System**: g.tec medical engineering GmbH g.HIamp.
- **Sampling Rate**: 600 Hz.
- **Online Filtering**: 0.1-100 Hz bandpass filter and a 48-52 Hz notch filter.
- **Effective Channels**: 62 scalp electrodes arranged according to the International 10-20 system.
- **Reference Electrode**: right earlobe.
- **Stimulus and behavior software**: Stimuli were presented and behavioral responses were collected using E-Prime 3.0.

Research Value
--------------
The value of MorseEEG-ATP lies in three main aspects. First, at the level of experimental paradigm, the dataset is built on the discrete acoustic units and explicit temporal rules of MC, while systematically integrating task demands and presentation-rate manipulations within a unified framework. This provides a reproducible and parameterizable paradigm for studying auditory temporal pattern discrimination based on explicit temporal rules. Second, at the level of assessment methodology, the dataset combines multitask conditions, cross-rate manipulations, trial-aligned EEG recordings, and behavioral measures, supporting comparisons of neural and behavioral performance across different processing demands and speed conditions, as well as the development and validation of methods for listening ability assessment. Third, as a research resource, the dataset provides raw and preprocessed data, event annotations, behavioral measures, and analysis scripts in a standardized format, thereby supporting studies of rhythm perception, related individual differences, EEG analysis methods, neural decoding models, and auditory brain-computer interface approaches under cross-condition generalization settings.

Technical Validation Highlights
-------------------------------
According to the accompanying paper, technical validation is provided through behavioral performance, event-related potential responses, and listening-related measures, supporting the quality and usability of the dataset.

Data Quality Notes
------------------
- **Data corruption**:
- Channel 56 data for participant `sub-13` are corrupted.
- Channel 56 data for participant `sub-14` are corrupted.
- Channel 43 data for participant `sub-15` are corrupted.
- In the minimally preprocessed data, these corrupted channels have been interpolated using neighboring electrodes.

Usage Recommendations
---------------------
- **Citation**: Please cite the associated Scientific Data article when using this dataset.
- **Preprocessing**: The raw data preserve complete trigger markers. The companion preprocessing pipeline can be used directly or adapted for related studies.
- **Analysis**: The derivative `.mat` files are organized for efficient comparison across blocks and presentation-rate conditions.
- **Applications**: The dataset primarily supports studies of auditory temporal processing, temporal pattern discrimination, cross-condition EEG analysis, and listening-related individual differences. It may also be useful for benchmarking EEG analysis and neural decoding methods.

Summary
-------
MorseEEG-ATP provides a standardized and shareable EEG resource for studying auditory temporal processing with controlled Morse code stimuli. It bridges a gap between low-level acoustic encoding datasets and higher-level language datasets by targeting the extraction and discrimination of temporal patterns governed by explicit rules. The resource supports research on temporal pattern discrimination, individual differences, and listening ability assessment, while also providing a useful benchmark for EEG analysis and decoding methods.