Data and code associated with manuscript "The effect of interocular contrast differences on the appearance of augmented reality imagery" M. Wang, J. Ding, D.M. Levi and E.A. Cooper ACM Transactions on Applied Perception, in press
The content is organized into 3 folders:
data for both experiments are here (Expt 1 and Expt2_AR subfolders)
- data structure fields:
- reference = whether the reference stimulus was the top one or the bottom one
- adjust = the adjustable one was top or bottom
- stim = columns of stimulus info --col1 is stimulus type (also listed in the .pattern field) --col2 is left eye contrast --col3 is right eye contrast --col4 (Expt 1 only) is surround contrast (average of left and right eye)
- pattern = what stimulus pattern was coded by stim field col1 -- example: for Expt 1, stim col1=1 means grating, col1=5 was bandpassed noise
- resp = responses --col1 is matching response --col2 is exact match response (1 = yes, 2 = no) --col3 to col7 are perceptual effect responses: brightness, contrast, luster, rivalry, depth ---1 = top selected, 2 = bottom selected, 4 = same/neither, 5 = unsure
matching task analysis code for Expt2 only (see other repo for Expt1)
- code for making Figure 8:
- plot_ARmatch_byICR.m --> determine high contrast stim weight for each interocular contrast ratio condition
- plot_ARmatch_byPattern.m --> determine high contrast stim weight for each stimulus pattern condition
- genBino.m --> helper function for the simple weighted combination model
- subfolder: eye dominance --> eye dominance analysis based on matching
- parseByEye.m --> determine eye dominance status for each subject, and save result as 'ARexpt_eyedom.mat'
- ARmatch_weight.m --> determine the weight for dominant eye high contrast trials vs. nondominant eye, and make plots
- subfolder: R stats --> processed data files and R code to run ANOVA/t-test for the matching task, Table 4
scripts for analyzing perceptual question responses for both Expt 1 and 2
- main plot and analysis code
- plot_exactMatch_Expt*.m --> plot Figure 6 and Figure 8 (proportion of exact match)
- doGLME_Match_Expt*.m --> run mixed effect logistic regression model, Table 1, 2, 5, 6
- plot_propEffect_Expt*.m --> plot Figure 7 and Figure 9 (proportion of each perceptual effect)
- doGLME_Effect_Expt*.m --> run logistic regression for different perceptual effects, Table 3, Table 7
- misc. analysis
- check_luster_rivalry.m --> check if dichoptic stim was selected to be lustrous or rivalrous
- check_unsure.m --> check the number of unsure responses
- check_numEffectPerTrial.m --> check effect co-occurence
- check_expt1_vs_2_ttest.m --> compare expt 1 and 2 for the 5 effects (uses expt*_effect.mat saved from plot_propEffect_Expt*.m)