This OSF page is a repository for data and analysis in the manuscript:

Mitchell, Harris, Benstock & Ales (2020). The reliability of pseudoneglect is task dependent. Neuropsychologia

DOWNLOAD

The code 'analyzeData.m' uses MATLAB to download data and run all analysis scripts

Alternatively you can:

Download and unzip raw Data from OSF (https://osf.io/5fksw/) - make sure data is saved in folder called 'Data' within repository

Download all code directly from github (https://github.com/aleslab/MitchellEtAlPseudoneglect) - place code in the same folder as 'Data' (NOT inside 'Data')

INSTRUCTIONS FOR RUNNING THE EXPERIMENT STIMULUS

To run the experiment you will need:

• MATLAB (R2017a and above)
• The latest version of psychtoolbox (http://psychtoolbox.org/)
• To run Line bisection task you will also need a touchscreen with a Linux OS and a stylus

** Tactile rod bisection is not run through MATLAB and requires a tactile rod platform and rods **

Open 'stimulus_code'

To run landmark task:

lm.m 

To run line-bisection task:

mlb.m

Output files saved in same folder:

Landmark: LM_date_time.mat Line bisection: MLB_date_time.mat All data-files are saved in individual session folders in each participant folder alongside initial analyses

INSTRUCTIONS FOR DATA ANALYSIS

analyzeData.m will run all analysis code and save out analysis outputs for each participant into each participants folder, and save group analyses into the "Analysis" folder.

To run data analysis you will need:

• MATLAB
• Palamedes toolbox to run psychophysical analyses (http://www.palamedestoolbox.org/download.html)

Load matlab navigate to the directory where the code was downloaded into and run the analyze data script.

analyzeData.m

NOTE: The ICC may not work if you have installed an a toolbox that overrides matlab default statistics function If you have errors calculating ICC check that: finv betapdf and betainv functions are from matlab statistics toolbox and not a 3rd-party toolbox This can be done using:*

which('finv')

All individual analysis files are saved so group analysis should run straight away - but individual analysis scripts are available.

Individual participant analyses

The analyzeData script runs each of the following scripts. This section describes each of the individual scripts run by the main function. Each of these scripts will be saving analysis outputs into the folders described below.

Code: folder/reliabilityanalysis_code

To analyse visual data for each participant run:

pp_analysisVisual.m 

THEN

pp_psychometricVisual.m

To analyse tactile data for each participant run:

pp_analysisTactile.m

Data will end up here: folder/Data/PP/Analysis

../Visual

File: PP_visualanalysisStart.mat <- data before psychometric fitting Results from the landmark task: lm

average percentage of right side responses alongside raw data Results from the line bisection task: mlb average mean bisection error alongside raw data

File: PP_visualanalysis.mat <- data after psychometric fitting No change to mlb but lm now includes PSE data for each session and overall

../Tactile

File: PP_tactileanalysis.mat Results: trb

average mean bisection error alongside raw data

Group level analyses

To run analysis run:

grp_analysisReliability.m

Data will end up here: folder/Analysis

Output: ReliabilityAnalysis.mat

Produces a variety of different plots Cronbach's alpha and t-test results Covariance matrices (not presented in pub) Key structure: 'results' used to calculate alpha and make plots

Organisation of structures (by column):

Results.plotting.modalities:

(1) observers

(2) bisection error total mean

(3) bisection error landmark

(4) bisection error manual line bisection

(5) bisection error tactile rod bisection

(6) bisection error standard deviation all tasks

(7) SEM all tasks

(8) observers organised by bias

Results.plotting.sessions.(task): <- one for each lm/mlb/trb

(1) observers

(2) bisection error task mean

(3) bisection error session 1

(4) bisection error session 2

(5) bisection error session 3

(6) bisection error session 4

(7) standard deviation task

(8) SEM task

(9) observers organised by bias

ALPHA RESULTS:

Reliability across session:

Landmark: results.sessions.lm Line bisection: results.sessions.mlb Tactile rod: results.sessions.trb Reliability across modalitiy:

results.modalities.all T-tests:

Landmark: results.modalities.lmT Line bisection: results.modalities.mlbT Tactile rod: results.modalities.trbT To produce figures found in text run: publicationPlots.m Plots will end up here: folder/Analysis/Plots

This will produce 'Plots' folder within 'Analysis'

Create and save all plots presented in publication

Hand used analysis:

grp_analysis_handUsed.m

Link contrast analysis:

pp_psychometricContrast.m
grp_analysis_LMlineContrast.m