Rising tones and rustling noises: metaphors in gestural depictions of sounds. Data

Data for the article. The file ObservationalData.mat contains the following variables: - Data is 4 x 8 x 10 x 3 x 10 matrix. Dimension 1 corresponds to the four annotators. Dimension 2 corresponds to the 8 referent sounds. Dimension 3 corresponds to the 10 imitators. Dimension 4 corresponds to the three response fields. Dimension 5 corresponds to the different possible answers. When the content of the matrix is 0, it means that the annotator has not indicated this answer for this particular sound and response field. When the content of the matrix is 1, this means that the annotator has selected this answer. - Annotators is the list of annotators: 'HS4' 'OH4' 'GL4' 'PS2' - SoundNames is the list of sound names: ‘Upward sweep' 'Downward sweep' 'Stationnary noise' 'Scraping' 'Filling' 'Door' 'Refrigerator' 'Printer' - Imitators are the imitator IDs: '08' '16' '20' '22' '25' '26' '28' '29' '30' '32' - Fields are the field names: 'Main gestural features' 'Main vocal features' 'Direction of the gestures' - NMax are the number of possible answers for each field: 10, 8, 7 Data for the experimental study are stored in two files: GesturalDescriptor.txt and VocalDescriptor.txt Each file is a table, where each row is an observation: one imitator imitating one referent sound. The columns correspond to the imitators, the referent sounds, the factors (granularity, profile, and toneless), and the features.

本文所用数据集说明如下：文件`ObservationalData.mat`包含如下变量：- 数据矩阵（Data）的维度为4×8×10×3×10。其中第1维对应4名标注者（annotators），第2维对应8个基准声源（referent sounds），第3维对应10位模仿者（imitators），第4维对应3个响应字段（response fields），第5维则代表所有可选答案。当矩阵元素取值为0时，表示该标注者未针对该特定声源与响应字段选择此答案；取值为1时，则代表该标注者已选中该答案。- 标注者变量（Annotators）：包含4个条目，分别为'HS4'、'OH4'、'GL4'、'PS2'。- 声源名称变量（SoundNames）：包含8个条目，分别为'上行扫频音'、'下行扫频音'、'稳态噪声'、'刮擦声'、'填充声'、'门声'、'冰箱运行噪声'、'打印机运行声'。- 模仿者ID变量（Imitators）：包含10个条目，分别为'08'、'16'、'20'、'22'、'25'、'26'、'28'、'29'、'30'、'32'。- 响应字段名变量（Fields）：包含3个条目，分别为'主要手势特征'、'主要发声特征'、'手势方向'。- 各字段可选答案数上限（NMax）：依次对应上述3个响应字段，取值分别为10、8、7。本实验的核心实验数据存储于两个文件中：`GesturalDescriptor.txt`与`VocalDescriptor.txt`。两个文件均采用表格格式，其中每一行代表一条观测数据，对应一位模仿者模仿某一基准声源的结果；表格的列依次对应模仿者、基准声源、实验因子（粒度、轮廓与无调式）以及特征参数。