EEG-MEG

Data collected from normal hearing younger adults (18-30) and from normal hearing and hearing impaired older adults (>= 60) to study age-related deficits in the representation of speech in noise. EEG data were collected with Biosemi system by using the ABR module from one electrode placed in Cz and reference with respect to the left and right ear lobes. The data uploaded are the raw ones in bdf format. BDF files can be opened by using matlab scripts that can be found in toolboxes such as EEGLab or that can be found directly from Biosemi's website. Each mat file contains information about sampling frequency, channels, triggers, etc. Each participants was tested in 9 conditions: Quiet, +3 dB, 0 dB, -3 dB and -6 dB with English (H) and Dutch (L) speakers used as background noise. Files were named based on the SNR and background speaker used. For instance, S01_M3_H, means Subject 01, SNR = -3 dB and English speaker in the background, while S01_P3_L, means Subject 01, SNR = +3 dB and Dutch speaker in the background. The "Q" denomination is used for the condition in quiet. Subjects id# from 1 to 17 are younger adults, subjects with id# from 21 to 35 are normal hearing older adults and subjects named S01_HL.....S17_HL are older adults with hearing loss.

MEG data were collected from 157 sensors. Each mat file contains information about the data, such as sampling frequency. A 3D matrix is used to store the 3 repetitions recorded from each conditions. Each participants was tested in 10 conditions: Quiet, +3 dB, 0 dB, -3 dB and -6 dB with English (H) and Dutch (L) speakers used as background noise. Two conditions in quiet were played. Files were named based on the SNR and background speaker used, in the same way as for the EEG data.

Auditory stimuli were also uploaded. The stimuli have been pre-processed as described in our publications "Evidence of degraded representation of speech in noise, in the aging midbrain and cortex" and "Effect of informational content of noise on speech representation in the aging midbrain and cortex". The envelope needs to be extracted.

本数据集采集自听力正常的青年成年人（18~30岁）以及听力正常与听力受损的老年成年人（≥60岁），旨在研究噪声环境下言语表征的年龄相关缺陷。

采用Biosemi系统的听觉脑干反应（Auditory Brainstem Response, ABR）模块采集脑电图（Electroencephalogram, EEG）数据，电极放置于Cz点，参考电极置于左右耳垂。上传数据为原始bdf格式文件，可通过EEGLab等工具包中的Matlab脚本，或直接从Biosemi官网获取的脚本打开。每个Mat文件包含采样频率、通道数、触发信号等相关信息。

每名被试完成9种测试条件：安静环境，以及信噪比（Signal-to-Noise Ratio, SNR）为+3 dB、0 dB、-3 dB、-6 dB的英语（H）、荷兰语（L）背景噪声环境。文件命名规则基于信噪比与背景讲话者类型，例如S01_M3_H代表被试01、信噪比为-3 dB、背景为英语讲话者；S01_P3_L代表被试01、信噪比为+3 dB、背景为荷兰语讲话者。安静环境条件使用“Q”标识。被试编号1至17为青年成年人，编号21至35为听力正常的老年成年人，编号以S01_HL至S17_HL命名的为听力受损老年成年人。

脑磁图（Magnetoencephalogram, MEG）数据采集自157个传感器，每个Mat文件包含数据相关信息（如采样频率），采用三维矩阵存储每种测试条件下的3次重复记录数据。每名被试完成10种测试条件：2次安静环境测试，以及信噪比为+3 dB、0 dB、-3 dB、-6 dB的英语（H）、荷兰语（L）背景噪声环境。文件命名规则与EEG数据一致。

本数据集同时上传了听觉刺激材料。刺激材料已按照本团队已发表论文《Evidence of degraded representation of speech in noise, in the aging midbrain and cortex》与《Effect of informational content of noise on speech representation in the aging midbrain and cortex》中描述的流程完成预处理，需提取其包络信息。