Data from: Rethinking predictive suppression: expert pianists do not show greater tactile suppression than musical novices
Data files
Jun 02, 2026 version files 23.10 KB
-
Data_ExpertTactileSuppression.csv
9.21 KB
-
DisciplineSpecificMetadata.json
8.64 KB
-
README.md
5.25 KB
Abstract
Expert performance is often attributed to automatized movement, where reliance upon predictive motor programs is increased, and upon sensory feedback decreased. However, few studies have examined how expert performance interacts with tactile feedback. A contradiction exists in current literature. On one hand, more predictable, external tactile feedback associated with movement (e.g., through previous experience) is perceived less (i.e., tactile suppression). Experts, whose experience presumably leads to stronger predictions, should therefore show greater tactile suppression than novices during their trained performance. On the other hand, reductions in tactile inputs impair fine motor performance: opposite to what expertise should yield. To better understand this paradox around expert performance and tactile perception, we examined tactile suppression and fine motor performance in expert pianists and musical novices during piano key presses. We measured thresholds for perceiving tactile, electrical stimuli of variable intensities at rest and right before a piano key press (i.e., during movement planning). Tactile suppression was determined as an increased threshold at movement planning compared to rest. Experts did not show greater tactile suppression compared to novices. Fine motor performance was measured as participants’ accuracy at matching a key press velocity target. Experts showed greater accuracy in matching a slow velocity target than novices. There was a trend towards less tactile suppression relating to better fine motor performance across both groups. Together, these findings suggest that 1) less tactile suppression may be beneficial to fine motor performance, and 2) performance-benefitting inputs may not be subjected to predictive suppression.
Dataset DOI: 10.5061/dryad.c59zw3rqz
GENERAL INFORMATION
Author Information:
Corresponding Author: Julia Tom (University of Toronto): j.tom@utoronto.ca
Co-authors: Joseph X. Manzone (University of Toronto), Joyce L. Chen (University of Toronto)
Funding Information:
Natural Sciences and Engineering Council of Canada Discovery Grant (to J.L. Chen).
SHARING/ACCESS INFORMATION
Recommended citation for this dataset: Tom, J., Manzone, J. X., & Chen, J. L. (2026). Data from: Rethinking predictive suppression: expert pianists do not show greater tactile suppression than musical novices.
DATA & FILE OVERVIEW
Data_ExpertTactileSuppression.csv: Main experimental trial data, including tactile thresholds in different movement conditions, tactile suppression in different movement conditions, and fine motor performance.
METHODOLOGICAL INFORMATION
Methods used for data collection: See Methods section of the associated manuscript for full experimental design.
Human Subjects & Privacy Compliance: All participants provided explicit consent for their de-identified data to be published. Direct personal identifiers (such as names or participant IDs) were removed, and indirect identifiers were aggregated to ensure anonymity.
Files and variables
File: Data_ExpertTactileSuppression.csv
Variables
- P#: Unique numerical identifier assigned to each participant (e.g., 1, 2, 3...).
- Group: Expertise group assignment: 1 == Expert; 2 == Novice
- CE.30: slow velocity accuracy (Musical Instrument Digital Interface: MIDI)
- CE.70: fast velocity accuracy (MIDI)
- avg_peak_velocity: mean peak velocity (millimeters/second: mm/s)
- ve_peak_velocity: variable error (consistency) peak velocity (mm/s)
- avg_onset: mean movement onset timing (second: s)
- ve_onset: variable error (consistency) movement onset timing (s)
- Zero: zero value
- Baseline1: tactile perceptual threshold from Baseline 1 trials (pre-test rest; milliamperes: 'mA')
- planning: tactile perceptual threshold from movement planning trials (mA)
- execution: tactile perceptual threshold from movement execution block (mA)
- Baseline2: tactile perceptual threshold from Baseline 2 trials (post-test rest; mA)
- P-B1: planning suppression (planning - Baseline1: mA)
- E-B1: execution suppression (execution - Baseline1: mA)
- Age: age on date of testing (years)
- Sex: female == 1; male == 2
- Catch B1: count of false positive catch trials during Baseline 1 block
- Catch PE: count of false positive catch trials during joint planning and execution block
- Catch B2: count of false positive catch trials during Baseline 2 block
Code/software
Required Software for Viewing Data
The primary dataset is provided in a standard, open format (.csv) to ensure long-term accessibility. It can be viewed using:
- Spreadsheet Software: Microsoft Excel, Google Sheets, or LibreOffice Calc.
- Text/Code Editors: VS Code, Notepad++, or TextEdit (for raw text inspection).
Statistical Analysis Environment
All data processing, statistical analysis, and data visualizations were executed using the following environment:
- Software: RStudio (Version 2024.04.2.764)
Loaded R Packages
The data analysis was performed using the following specific R packages (installed via CRAN):
- Data Manipulation:
dplyr,tidyr - Data Visualization:
ggplot2,patchwork - T-Tests & Comparisons:
rstatix(t-test calculations) andeffsize(computing Cohen's d) - ANOVA Effect Sizes:
effectsize(calculate partial eta squared (η²p) from ANOVA outputs) - Bayesian Statistics:
BayesFactor(calculate Bayes factors) - General Resampling:
boot(bootstrapping) - Correlation Analyses:
psychandbroom(Fisher's z correlation calculations and tidying outputs)
Workflow & Relationship of Files
- Data Input: The primary tabular data file (
Data_ExpertTactileSuppression.csv) contains the de-identified raw metrics used for the study. - Analysis Execution: The data was processed and analyzed using RStudio (Version 2024.04.2.764). The R packages listed above were utilized to run the primary statistical tests and generate the figures presented in the manuscript directly from this
.csvfile. Please note that this script is not included here.
Access information
Other publicly accessible locations of the data:
- None. This is the primary and exclusive repository for this dataset.
Data was derived from the following sources:
- Not applicable. This dataset consists entirely of original data collected and produced by the authors for the associated study.
Human subjects data
All participants provided explicit consent for their de-identified data to be published in the public domain. To ensure anonymity, all direct personal identifiers (e.g., names, contact information) were completely removed from the dataset, and indirect identifiers were aggregated into broader categories where necessary.
