Release of PsPM 6.0

[teddychao]

Updates included in PsPM 6.0 are listed as following.

Documentation

The manual and developers' guide have been updated in PsPM 6.0.

New features

• PsPM now support a develop mode, which allows
  • Minimised terminal output
  • Automagical parameter selection for testing
  • Always overwrite files (issue: Test mode for PsPM #302, solution: 364).
• pspm_extract_segments
  • now allows an option z-score for normalisation (solution: New option to normalise data in pspm_extract_segments #332).
• pspm_find_valid_fixations
  • now allows processing preprocessed left/right/combined pupil data (issue: Integrate pspm_find_valid_fixations with pspm_pupil_pp #275, solution: Fix issue 275 #290)
• pspm_gaze_pp
  • now allows preprocessing gaze signals (issue: Add a function for preprocessing gaze signals #310, solution: Fix issue 310 - pspm_gaze_pp #311).
• pspm_overwrite
  • now handles the behaviour of overwrite if not defined by the user, and is used where applicable (issues: Command line message in pspm_import missing #374, data_editor not displaying missing epochs when being run from command line #375, solution: Fix issue 375 - pspm_data_editor #376).
• pspm_time2index
  • now processes adjustable conversion from time to index globally (issue: New function for converting time in second to data points #326, solution: Fix issue 326 #327, Update time2indx #388).
• pspm_update_chantype
  • now allows generalised behaviour of updating channel types (issue: Add a function for preprocessing gaze signals #310, solution: Fix issue 310 - pspm_gaze_pp #311).

Bug fix

• General
  • A bug which may lead to failure of pspm path searching has been fixed (issue: File seperator issue #353, solution: Fix issue 353 - pspm_path_test #354).
  • A bug which may display figures in the GUI has been fixed (issue: Data Display "Semi automatic mode" in Pre-processing of Heart data #248, solution: Fix issue 248 - main GUI #345).
• pspm_bf_spsrf_box
  • now uses correct parameter settings (issue: Incorrect parameter setting in GLM for scanpath-speed #298, solution: 299).
• pspm_convert_ppu2hb
  • terminates if only one pulse is detected (issue: pspm_convert_ppu2hb crashes in edge case #381, solution: Fix issue 381 - pspm_convert_ppu2hb #382).
• pspm_data_editor
  • now able to show figures when import a datafile correctly (issue: pspm_data_editor is not working #358, solution: Fix Issue 358 - pspm_data_editor #359).
• pspm_dcm
  • now handles missing epochs that start at 0 correctly (issue: pspm_dcm cannot handle missing epochs that start at 0 (or very close to zer0). #324, solution: Fix issue 324 - pspm_time2index #334).
  • now processes variables properly to make length consistent (issue: pspm_dcm throw error of inconsistent matrix when processing missing_epochs #361, solution: Fix issue 361 - pspm_dcm #363).
• pspm_display
  • now display plots with correct x-axis ranges (issue: Bug in pspm_display (heart beat channel) #291, solution: Fix issue 291 #292)
• pspm_glm
  • now applies correct variable settings (issue: Incorrect variable setting in glm #314, solution: Fix issue 314 #315).
  • now produces results correctly when facing many short missing epochs (issue: pspm_glm produce no results when facing many short missing epochs #338, solution: Fix issue 338 - pspm_glm #339).
• pspm_interpolate
  • now returns results correctly when applying forced extrapolation (issue: Forced extrapolation in pspm_interpolate returns no results #317, solution: Fix issue 317 - pspm_interpolate #337).
• pspm_resp_pp
  • now allows replace as the channel action properly (issue: Conflicting options in pspm_resp_pp? #380, solution: Fix issue 380 - pspm_resp_pp #383).
• pspm_scr_pp
  • now uses inconsistent variables when delivering missing epochs (issue: scr_pp uses inconsistent variables when delivering missing epochs #300, solution: Fix issue 300 #301).
  • now writes data out properly when epochs are removed (issue: pspm_scr_pp may not write out data successfully #335, solution: Fix issue 335 - pspm_scr_pp #336).
  • now properly handles data if the first channel is not scr (issue: Bug in writing channel (pspm_scr_pp) #351, solution: Fix issue 351 - pspm_scr_pp #352).
• pspm_sf
  • now refer to model file correctly (issue: Review and fix bugs in pspm_sf #370, solution: Fix issue 370 - pspm_sf #371).
• pspm_split_sessions
  • now processes missing files with an epoch starting at 0s properly (issue: pspm_split_sessions crashes when a missing file contains an epoch starting at 0 s #323, solution: Fix issue 323 #333).
  • now properly catches edge cases (solutions: small fixes #386, refactor time2index #389).
• ValidSamplesModel
  • now properly throw warnings if histogram counts are zeros (issue: pupil combined processing error #355, solution: Fix issue 355 - zero index issue in ValidSamplesModel.m #365).

Improvements

• General
  • now supports loading ppg data (issue: Add support for channel 'ppu' #391, solution: Fix issue 391 - import ppu issue #392).
  • now displays improved terminal output text (issue: Warning messages from command line output in pspm_import #387, solution: Fix issue 387 - unsupported characters #390).
• pspm_con1
  • now assesses the statistics in the arguments for throwing warnings when detecting invalid values (issue: Incorrect assigned NaN #395, solution: Fix issue 395 - pspm_con1 #398).
• pspm_data_editor
  • now displays the x and y axis label and text according to the input data (issue: Data editor to show missing epochs in time order #224, solution: Fix issue 224 - pspm_data_editor #394).
  • now sorts the epoch list according to the start data point.
• pspm_get_marker
  • now detect and update the field flank when applicable (issue: pspm find marker events: flank setting #308, solution: Fix issue 308 - missing flank #343).
• pspm_import
  • now detect and update the field flank when applicable (issue: pspm find marker events: flank setting #308, solution: Fix issue 308 - missing flank #343).
• pspm_load_data
  • now checks whether the input has an empty channel (issue: Allow empty marker channels #212, solution: Fix issue 212 - pspm_write_channel #344).
  • now autofills some variables (issue: pspm find marker events: flank setting #308, solution: Fix issue 308 - missing flank #343).
• pspm_pupil_pp
  • now calls pspm_load_data to load single channels (issue: pspm find marker events: flank setting #308, solution: Fix issue 308 - missing flank #343).
• pspm_rev_glm
  • now display normalised data for visualisation purpose only (issue: plot of design matrix of pupil size glm with luminance as nuisance file #213, solution: Fix issue 213 - design matrix #342, Re-fix issue 213 - plot matrix #347).
• pspm_write_channel
  • now checks whether the input has an empty channel (issue: Allow empty marker channels #212, solution: Fix issue 212 - pspm_write_channel #344).

GUI

• "Edit defaults" in the GUI is now working properly (issue: Feature "Edit Defaults" does not work #289, solution: Fix issue 289 - edit default error #348).
• GUI typeface is now generalised across the software (issue: Set UI parameters to be globally accessible #293, solution: Fix Issue 293 #294).

For Developers

• General
  • now uses .editorconfig to control the indent (solution: Add the editorconfig file #328).
  • now uses UTF-8, LF, and MATLAB's soft tab indent as the standard code formatting style (issue: PsPM initialisation #130, solution: Fix issue 295 #296, Fix issue 130 #369).
  • now uses test data from the repository PsPM-data (issue: Migrating to a private test data repository #340, solution: Fix issue 340 - test data remapping #341).
• pspm_get_text_test
  • A bug, which leaves residual test files after code testing, has been fixed (issue: get_text_test forgets to delete test files after running #312, solution: Fix issue 312 #313).