Selective auditory attention in CI users - EEG data

<h2><span>Experiment</span></h2> <p><span>We conducted a selective auditory attention experiment,&nbsp;</span><span>where stimuli were presented in the free room over two </span><span>loudspeakers separated by</span> <span>60&deg;. </span><span>Before each trial, participants received instructions directing&nbsp;their attention to a specific audiobook. This guidance was&nbsp;provided visually on a screen, featuring a symbol indicating the selected audiobook&rsquo;s direction. </span><span>Initially,&nbsp;we conducted eight trials in a single-speaker scenario, where&nbsp;only one audiobook was presented from alternating sides. Each&nbsp;trial lasted approximately two minutes. After every two trials,&nbsp;the presented story changed. In twelve subsequent trials, we implemented a competing-speaker paradigm, where two stories were presented simultaneously. However, the distractor story started 10 s later, affording participants time to discern the target speaker. We further&nbsp;randomized the organization by starting with the block of s2&nbsp;stimuli instead of s1 for every second participant.&nbsp;</span></p> <p>&nbsp;</p> <h2><span>EEG Recording</span></h2> <p><span>We collected EEG data using an actiCHamp System (BrainProducts GmbH, Germany) equipped with 32 electrodes. For&nbsp;each participant between two and four electrodes were removed due to their proximity to the CI magnet and sound&nbsp;processor. The sampling rate was set at 1 kHz, and an online&nbsp;low-pass filter with a cutoff frequency of 280 Hz was implemented.&nbsp;Prior to the experiment, electrode impedances were maintained below 20 kΩ. We monitored the impedances through<br>both the single-speaker and competing-speaker scenarios, and,&nbsp;if needed, applied additional conductive gel to ensure that the&nbsp;impedances remained below the threshold of 20 kΩ.&nbsp;For synchronizing the audio and the EEG recording, we&nbsp;used an audio splitter and recorded the presented audio as two&nbsp;auxiliary channel over the EEG recorder with two StimTraks&nbsp;(BrainProducts GmbH, Germany) as adapter. We performed an&nbsp;offline correlation analysis between the recorded audio signal&nbsp;and delayed versions of the clean stimuli. The delay with&nbsp;highest value in Pearson&rsquo;s r is used to align the respective&nbsp;stimuli with the EEG recording. Additionally, we sent onse</span></p> <h2><span>Dataset</span></h2> <p><span>We provide the data in hdf5 format. </span></p> <p><span>It includes:</span></p> <ul> <li><span>The EEG recording (raw and ICA-cleaned (read the paper for the method))</span></li> <li><span>Stimuli and precalculated features (speech envelope and onset envelope)</span></li> </ul> <p><span>For organization of the file and an example of how to read the data using python, see the <a href="../api/records/10980117/draft/files/hdf5_dataset_info.txt/content" target="_blank" rel="noopener noreferrer">hdf5_dataset_info.txt</a> file.</span></p>