update limo

Author: dungscout96

code/plugins/reformat_plugin.py

@@ -21,7 +21,7 @@ def reformat_wiki_pages(filepath, filename, parent, output_file, wiki_input_dir=
 '''.format(filename=filename, parent=parent)
 
     print(f"Reformatting {filename} of {parent}...")
-    if parent in ["nsgportal", "LIMO"]:
+    if parent in ["nsgportal"]:
         pages = []
         titles = []
         # load _Sidebar.md and extract all links from markdown file

plugins/index.md

@@ -5,6 +5,10 @@ has_children: true
 has_toc: true
 nav_order: 7
 ---
+# EEGLAB plugin documentation
+
+Below is a list of plugins that have documentation copied from GitHub. Please note that this is only a small subset of all EEGLAB plugins, as not all plugin documentation is compatible with visualization and search functionalities on the EEGLAB website. The complete list of plugins can be found [here](https://sccn.ucsd.edu/eeglab/plugin_uploader/plugin_list_all.php).
+
 ## Import
 * [EEG-BIDS](/plugins/EEG-BIDS): Imports and export EEG data to the BIDS format
 * [NWB-io](/plugins/NWB-io): Import and export to the NWB format

plugins/limo/01.-Preprocessing.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: Preprocessing
-long_title: Preprocessing
-nav_order: 3
+title: 01.-Preprocessing
+long_title: 01.-Preprocessing
 ---
 # Data for the tutorial
 

plugins/limo/02.-Within-Subject-Categorical-Designs.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: Within Subject Categorical Designs intro
-long_title: Within Subject Categorical Designs intro
-nav_order: 4
+title: 02.-Within-Subject-Categorical-Designs
+long_title: 02.-Within-Subject-Categorical-Designs
 ---
 - [1 way repeated measures ANOVA (Famous, Unfamiliar, Scrambled faces as conditions)](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/wiki/2.-One-way-repeated-measures-ANOVA-(Famous,-Unfamiliar,-Scrambled-faces-as-conditions)))
 - [One way repeated measures ANOVA revised (Famous, Unfamiliar, Scrambled faces as 1st level contrasts)](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/wiki/3.--One-way-repeated-measures-ANOVA-revised-(Famous,-Unfamiliar,-Scrambled-faces-as-1st-level-contrasts))

plugins/limo/03.1.-One-way-repeated-measures-ANOVA-Famous-Unfamiliar-Scrambled-faces-as-conditions.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: One way repeated measures ANOVA (Famous, Unfamiliar, Scrambled faces as conditions)
-long_title: One way repeated measures ANOVA (Famous, Unfamiliar, Scrambled faces as conditions)
-nav_order: 5
+title: 03.1.-One-way-repeated-measures-ANOVA-(Famous,-Unfamiliar,-Scrambled-faces-as-conditions)
+long_title: 03.1.-One-way-repeated-measures-ANOVA-(Famous,-Unfamiliar,-Scrambled-faces-as-conditions)
 ---
 Here we will use the three basic conditions to run a group level ANOVA. LIMO runs a [hierarchical model](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/master/resources/2016_SanDiego_StatisticalanalysisofEEGdata.pdf), first a GLM at the subject level (first level), second a GLM at the group level (second level). Under some assumptions about the data, this is equivalent to running mixed model analysis on all trials for all subjects with subjects as random effects -- but much faster to calculate. 
 

plugins/limo/03.2.--One-way-repeated-measures-ANOVA-revised-Famous-Unfamiliar-Scrambled-faces-as-1st-level-contrasts.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: One way repeated measures ANOVA revised (Famous, Unfamiliar, Scrambled faces as 1st level contrasts)
-long_title: One way repeated measures ANOVA revised (Famous, Unfamiliar, Scrambled faces as 1st level contrasts)
-nav_order: 6
+title: 03.2.--One-way-repeated-measures-ANOVA-revised-(Famous,-Unfamiliar,-Scrambled-faces-as-1st-level-contrasts)
+long_title: 03.2.--One-way-repeated-measures-ANOVA-revised-(Famous,-Unfamiliar,-Scrambled-faces-as-1st-level-contrasts)
 ---
 In the [previous analysis](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/wiki/2.-One-way-repeated-measures-ANOVA-(Famous,-Unfamiliar,-Scrambled-faces-as-conditions)), at the 1st level, we selected ‘face_type’ ([figure 7](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/master/resources/images/7.jpg)) as our variable. By doing so, beta parameters reflect the average height of each face type. We know that there is also a repetition effect – and if one repetition differs a lot more than the others that average can be biased. **It is therefore recommended to always create a full design (all known effects) and pool conditions to create contrasts**. 
 

plugins/limo/04.-Summary-statistics-Effects-and-Effect-sizes.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: Summary statistics to measure and report effects and effect sizes
-long_title: Summary statistics to measure and report effects and effect sizes
-nav_order: 7
+title: 04.-Summary-statistics:-Effects-and-Effect-sizes
+long_title: 04.-Summary-statistics:-Effects-and-Effect-sizes
 ---
 # Statistics course plot 
 

plugins/limo/05.-One-sample-t-test-contrasting-Full-Faces-vs-Scrambled-Faces-at-the-subject-level.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: One sample t-test (contrasting Full Faces vs Scrambled Faces at the subject level)
-long_title: One sample t-test (contrasting Full Faces vs Scrambled Faces at the subject level)
-nav_order: 8
+title: 05.-One-sample-t-test-(contrasting-Full-Faces-vs-Scrambled-Faces-at-the-subject-level)
+long_title: 05.-One-sample-t-test-(contrasting-Full-Faces-vs-Scrambled-Faces-at-the-subject-level)
 ---
 Let’s consider again the contrast of interest (Famous+Unfamiliar) Faces vs. Scrambled faces. This can be obtained from the 1-way ANOVA analysis, using a contrast [0.5 -1 0.5] (figures [16](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/master/resources/images/16.jpg) - [17](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/master/resources/images/17.jpg)). This can also be obtained by computing this contrast per subject and performing a one sample t-test on this contrast. Since we have 9 conditions with the full design, the contrast is [0.5 0.5 0.5 -1 -1 -1 0.5 0.5 0.5]. To add one or many contrast, one must create a variable and save this as a file (while we could have a GUI, using a saved variable allows 1. to run many contrasts (each line is a new contrast to run) and 2. to be able to return and check this file a few weeks/months later after the analysis).
 

plugins/limo/06.-Summary-statistics-of-differences.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: Summary statistics of differences
-long_title: Summary statistics of differences
-nav_order: 9
+title: 06.-Summary-statistics-of-differences
+long_title: 06.-Summary-statistics-of-differences
 ---
 The [one sample t-test](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/wiki/5.-One-sample-t-test-(contrasting-Full-Faces-vs-Scrambled-Faces-at-the-subject-level)) was computed on the contrasts faces vs scrambled faces, i.e. on differences. To fully appreciate the effect, we thus have to check differences on contrasts before looking at raw data.
 

plugins/limo/07.-Two-ways-ANOVA-Faces-x-Repetition.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: Two-ways ANOVA (Faces x Repetition)
-long_title: Two-ways ANOVA (Faces x Repetition)
-nav_order: 10
+title: 07.-Two-ways-ANOVA-(Faces-x-Repetition)
+long_title: 07.-Two-ways-ANOVA-(Faces-x-Repetition)
 ---
 Lets’ consider now all 9 conditions: 3 types of faces (familiar, unfamiliar, scrambled) and 3 repetition levels (immediate, small delay, long delay). This is analysed using a repeated measure ANOVA.   
 

plugins/limo/08.-Paired-t-test-Famous-vs-Unfamiliar.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: Paired t-test (Famous vs Unfamiliar)
-long_title: Paired t-test (Famous vs Unfamiliar)
-nav_order: 11
+title: 08.-Paired-t-test-(Famous-vs-Unfamiliar)
+long_title: 08.-Paired-t-test-(Famous-vs-Unfamiliar)
 ---
 Let say you only want to know if famous faces differ from unfamiliar faces – again an ANOVA could be set up test the main effect and using contrasts. Alternatively, if that is the only effect of interest, you can compute a contrast at the subject level and do a paired- t-test on contrasts. 
 

plugins/limo/09.1.-Between-subjects’-ANOVAs-with-repeated-factors.md

@@ -3,6 +3,9 @@ layout: default
 parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
+
+title: 09.1.-Between-subjects’-ANOVAs-with-repeated-factors
+long_title: 09.1.-Between-subjects’-ANOVAs-with-repeated-factors
 ---
 After [editing the STUDY for group](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/wiki/Between-Subjects-Categorical-Designs), we have the same Beta files as before, but also txt files split per group which makes file selection easier.  
 

plugins/limo/09.2.-Between-Subjects-Categorical-Designs.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: Between Subjects Categorical Designs
-long_title: Between Subjects Categorical Designs
-nav_order: 13
+title: 09.2.-Between-Subjects-Categorical-Designs
+long_title: 09.2.-Between-Subjects-Categorical-Designs
 ---
 We replicate here the 1-way ANOVA with familiar, unfamiliar and scrambled faces but split the data in two age groups. Of course, we can take the txt files, edit them and save copies for each group – then in LIMO MEEG we simply use these files. Here, instead, we recompute the subjects model adding in the STUDY design our groups, which will consequently save txt files per group (but not change estimates per subjects). Since some subjects have unspecified age – we create three groups based on the median (figure 35).  
 

plugins/limo/10.-Two-sample-t-tests.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: Two sample t-tests
-long_title: Two sample t-tests
-nav_order: 14
+title: 10.-Two-sample-t-tests
+long_title: 10.-Two-sample-t-tests
 ---
 Given the (non-significant) group effect observed in the between subjects’ ANOVA with repeated factor, we can also do a simple two samples t-tests between groups 1 and 2. From the 2nd level GUI ([figure 9](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/master/resources/images/9.jpg)), click on two samples t-test, and select the beta files for group 1 and group 2. Let’s select parameter 1, i.e. famous faces ([figure 38](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/master/resources/images/38.jpg)). Parameter question dialogue is repeated because you could for instance compare groups from 2 different studies, i.e. with different 1st level designs.  
 

plugins/limo/11.-Regression-among-subjects.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: Regression among subjects
-long_title: Regression among subjects
-nav_order: 15
+title: 11.-Regression-among-subjects
+long_title: 11.-Regression-among-subjects
 ---
 In the [between subjects’ ANOVA with repeated factor](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/wiki/Between-subjects%E2%80%99-ANOVAs-with-repeated-factors), we artificially split subjects into young and old subjects. Such post-hoc splitting is not recommended and typically create spurious results. Instead, we could test how much age influences face perception. For this, we will use the contrast faces vs scrambled computed previously in the [one-sample t-test](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/wiki/5.-One-sample-t-test-(contrasting-Full-Faces-vs-Scrambled-Faces-at-the-subject-level)). 
 

plugins/limo/12.-Regression-at-the-trial-level.md

@@ -4,9 +4,8 @@ parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
 
-title: Regression at the trial level
-long_title: Regression at the trial level
-nav_order: 16
+title: 12.-Regression-at-the-trial-level
+long_title: 12.-Regression-at-the-trial-level
 ---
 In previous analyses, the repetition levels were either averaged or used as a categorical variable. Here, we instead used the time between each repetition of the same stimulus – thus for a given subject we have 3 conditions (familiar faces, unfamiliar faces and scrambled faces) and one continuous variable (the distance between the repeat of a stimulus type). 
 

plugins/limo/Between-Subjects-Categorical-Designs.md

@@ -3,6 +3,9 @@ layout: default
 parent: LIMO
 grand_parent: Plugins
 render_with_liquid: false
+
+title: _Sidebar
+long_title: _Sidebar
 ---
 # [Home](https://raw.githubusercontent.com/LIMO-EEG-Toolbox/limo_meeg/wiki)