Description

Computes group level (between-subjects) statistical analyses on multiple condition types with 1 sample, two-sided statistics for each condition. Includes canonical HRF GLM topoplots and FIR concentration estimates (with significant masking).

Report sections

This pipeline generates a report in HTML format that contains the following:

• Analysis by Channels
• Canonical HRF GLM Analysis (topoplots)
• FIR GLM Analysis (line plots)

References

PARAMETERS

Category: Input

This pipeline expects one file (or a Path Iterator formatted path name) and its path is relative to your Neuroscale account storage bucket. (The easiest way to run it is to go to your Storage page, select the file, click the Task button which will take you to the Create Task page, and then select this pipeline. The filename will be filed in automatically.)

input path

The input data files (in .pkl format) that the pipeline will process. This needs to be a list of files or a wild card expression (e.g. studies/Subject/*.pkl) to indicate multiple files. Each file is the computed session means output from the individual session NIRs block average pipeline. The conditions specified in the 'conditions' list parameter needs to have been previously assigned in the individual session pipeline.

group pkl filename

Filename of concatenated groups (.pkl). If file exists, pipeline will load file, otherwise it will save .pkl file after concatenating groups.

output path

The output root path to save the report to. If empty, report will export to the root path of the input path specified above. Note that if the input path above uses curly braces (e.g. '{}') to create factors from folder names the output path will default to the top folder of the first factor folder (e.g. 'studies/meditation/{Group}/{Task}/{Subject}_*.pkl' will output to 'studies/meditation/'.

Category: Events

conditions

List of named condition(s) that was used in individual session analysis (e.g. 'tap-left').

Category: Style

style figure font

Font to be used throughout the report

Category: Statistics

winsorize

Winsorization is the transformation of statistics by limiting extreme values in the statistical data to reduce the effect of possibly spurious outliers. If true, outlier values will be replaced by the mean when taking means over the individual trials.

mean type

Use mean or median for averaging.

trim proportion

Proportion of data instances to trim (per variable) on the lower and upper tails of the data. If only a single number is provided then it is used for both lower and upper. If no non-zero numbers are provided then no trimming is performed (default). For example, [0.05, 0.05] will remove and replace the least and greatest 5% of extreme data values with the mean.

error type

std: standard deviation

sem: standard error of the mean

var: variance

mad: median absolute difference

CI: confidence interval

CI range

Range for confidence interval if 'CI' selected as error_type. This is a proportion (between 0 and 1) such that 0.95 is the 95% confidence interval.

mcc

Whether to correct for multiple comparisons using False Discovery Rate (FDR).

pval threshold

P-value to determine significance level. Values below specified threshold (e.g. 0.05) will be considered significant for statistics.