E1 to E3 correspond order-wise to the experiments in the McNamara, Moore, & Conway (2011).
- E1 - Simple word span, complex reading span, and complex reading span with TBR-words as last item in sentence.
- E2 - Long and short versions of reading span. Short corresponds to first complex reading span in E1. Long was used in Copeland and Radvansky.
- E3 - Reading span manipulating phonemic overlap instead of rhyming. (should there also be word span data for this experiment?)
- E4 - Regular OSPAN.
- E7 - (A) Reading span (same as short RSPAN in E1 and E2) with words randomly shuffled within each sentence. (B) Spelled-out OSPAN (no processing component)
- E9 - Pooled word span.
- E10 - Ospan with scrambled, spelled-out words (no verification).
- E11 - Long and short names as distractor items.
Scripts for processing / analysis can be found on github
Each folder has a data sub-folder with the unprocessed experimental files. Where applicable, I exported the e-merge files to tab-seperated .txt files, generally labeled output.txt.
A single script (0\_processing.py) preprocesses these output files, and saves the results as data/scored_all.csv.
This can then be fed to subsequent analyses. So, E-merge => tab-seperated => 0_processing.py => 1_...
Some complex span tasks for these experiments have identical structures / labels. They were preprocessed in the same way:
E2, E4, E6, E7, (E8 to E11?)..
Other experiments had to be preprocessed seperately. See 0_preprocess.py for details.
Complete E-dat files for: E2, E4, E6, E7, E8, E9, E10, E11
Missing in Brooke's data..
- E1 - RT's?
- E3 - Only have recall data
- E5 - Only have recall data