READ ME for ÒDataset supporting the publication Grasping New Material DensitiesÓ 

Dataset DOI: https://doi.org/10.5258/SOTON/D3407
Created on 4th March 2025.

-------------------
GENERAL INFORMATION
-------------------

ReadMe Author: Wendy Adams, University of Southampton.

Date of data collection: September 2018 - September 2024 
Location of data collection:  Ulm University, Germany.


--------------------------
SHARING/ACCESS INFORMATION
-------------------------- 

Licenses/restrictions placed on the data, or limitations of reuse: none. 

If you use this data / analysis code, please cite the following paper:

AUTHORS: Wendy J Adams, Sina Mehraeen & Marc O Ernst
TITLE: Grasping New Material Densities
JOURNAL: Multisensory Research 
DATE: 2025


--------------------
DATA & FILE OVERVIEW
--------------------

This dataset contains:

READ Me
Data.zip
Analysis.zip

Description of methods used for collection/generation of data can be found in the associated publication. 

Data:
Raw data for each experiment are stored in an appropriately named .mat file (e.g. Data_Exp1.mat).  The format of the data for Expts 1, 2a, 2b, 3a, 3b is as follows: Column 1: subject ID, column2: repetition, column 3: object ID, column 4: object orientation, column 5: grasping position (in cm, measured from metal end for bipartite objects). For Expt 4, there is an additional column (6) that indicates which hand was used for grasping.  The data are also provided in a single excel spreadsheet (AllData.xlsx), with the data for each experiment in a separate tab. 

Data were collected by Sina Mehraeen, Daniel Craven and Chandramita Bhattacharya. Data were processed (grasp points extracted from video) by Sina Mehraeen and analysed by Wendy Adams.


Analysis:
All analysis files are in Matlab.  
ExponentialDensity3.m % the main fitting file Ð it fits various possible models to data, i.e. finds the model parameters that maximise the likelihood of the data, as described in the associated manuscript. 
This file calls:
o Define_Vars; % info about stimulus objects, model variables, etc
o Model_setup; % prepares a structure that specifies which model parameters are fitted (vs. fixed values) 
o Get_resultFileName % does what it says on the tin, for current expt and model. 
o Likelihood_PPT3 % calculates the likelihood of single subjectÕs data, given model & fit params, calls:
* getSubFit
o getSubData  % extracts single subjectÕs data
o getExpo % creates exponential fit, given set of parameters
o dens2grasp2 % converts a density ratio to the ideal grasp location (CoM)
o grasp2norm2 % converts a grasping position (in cm) to a normalised grasp location
o setParams.m. % stores all necessary model parameters in a simple structure, given the current best-fitting free parameters, model, expt and data. 

PaperPlots.m. % plots the main results figures
OrientationEffects_PerSub2.m % stats on orientation effects and creates orientation plot (Figure 8)
Stats_tests2.m %% various stats tests done on model parameters

Model_Comparisons.m %% to compare alternative models via F-ratio test

DensityNormGrasp.m %% creates Figure S1 Ð relationship between CoM in cm, CoM in normalised position units and Log Density Ratio