Homework 9

Due: Mar 30 by 11:59 pm eastern standard time

Submit a single file called name_09.py to Brightspace/OWL where name is replaced with your last name, e.g. gribble_09.py

Fictitious EEG Dataset

All of the sample code include code to generate the raw data can be found in this GitHub repository: eeg_study.

The description of the fictitious study is here: eeg_study.

Processing raw data and generating a summary table

I have written example code to read the raw data and generate processed data… and I have also written example code to go from the processed data to a summary table called summary_table.csv containing 1 row for each participant, where each column codes a different variable.

The columns of the summary table code the participant number, the group (control vs learning), the CP3 peak-to-peak amplitude pre and post, the SNAP peak-to-peak amplitude pre and post, and the time to complete the 3 behavioural sequences pre and post.

Questions

Answer the following questions by generating figures to illustrate the data and by performing statistical analyses.

  1. Is there a difference in the peak to peak CP3 amplitude pre vs post for the control group?
  2. Repeat Q1 for the learning group.
  3. Is there a difference in peak to peak SNAP amplitude pre vs post for the control group?
  4. Repeat Q3 for the learning group.
  5. Is the pre-to-post change in CP3 different for the learning vs control groups?
  6. Repeat Q5 for the SNAP electrode.
  7. Is there a correlation between the pre-to-post change in CP3 amplitude and pre-to-post change in the sequence time (average over the three sequence measures)?

Generate figures as needed to illustrate the data, and choose statistical analyses as appropriate.

Submit a single .py file containing code that produces all figures and all statistical analyses. Your code should save each figure as a .pdf file. You don’t necessarily need one figure per question, you can generate a smaller number of figures that illustrate the data as you wish. Your code should output a series of sentences, one per question, that report the statistical results, e.g. “There was no reliable difference between ABC and DEF, t(x)=y.xyz, p<.abc”. Your code can depend upon the file summary_table.csv being in the same directory as the code. The TA should be able to run your code with a single command python3 name_09.py (your last name replaces name).