Releases: LearningToTalk/L2TWordLists
Support for oddball NonWordRep files from TimePoint3
This update adds support for original, uncorrected version of the NonWordRep experiment from TimePoint3 as well as the shortened, supplemental form of the NonWordRep experiment. Because we are combining the Eprime files from these experiment administrations into a single Eprime file, the package internals are updated to handle files where there are multiple sets of familiarization and testing trials.
- Added new WorldBet corrections for items in the other NonWordRep TP3 experiment.
- WorldBet gloss for "ruhgloke" from TimePoint3 updated to match TimePoint2.
- The trial-numbering procedure now supports Eprime files with multiple sets of familiarization and testing trials.
NonWordRep support completed
Most of the functionality from the original R script DataAnalysis/RScripts/makeL2TwordLists/L2T-NWR-MakeWordLists.R
should be supported by this package.
Core functionality, for both RealWordRep and NonWordRep experiments, include:
- Reading and parsing Eprime text files from experiments
- Correcting wrong item names or abbreviations in the Eprime text files
- Looking up WordList information for a set of trials
- Saving read-only
.txt
files - Wrapping the above functionality into a single shortcut function:
create_rwr_wordlist_file()
andcreate_nwr_wordlist_file()
Enhancements from previous release:
- Experiment session date and animal avatar are now extracted from the header of the Eprime file
- Code shared by both experiments has been factored out, so the same internal code is used to read Eprime files and save generated WordLists. Most of the experiment-specific code now resides in the
lookup_*_wordlist()
functions. - Streamlined the code used to test the package
Limitations:
Versions of the NWR TP3 experiment that the original script did not support are not supported by the package.
RealWordRep support completed
All of the functionality from the original R script DataAnalysis/RScripts/makeL2TwordLists/L2T-RWR-MakeWordLists.R
should be supported by this package.
Core functionality includes:
- Reading and parsing Eprime text files from RealWordRep experiments
- Correcting wrong abbreviations in the Eprime text files
- Looking up WordList information for a set of trials, including the many different versions of the experiment from TimePoint3
- Saving read-only
.txt
files - Wrapping the above functionality into a single shortcut function
create_rwr_wordlist_file()
Enhancements include:
- Determining the TimePoint from the Eprime file. This is no longer a parameter that needs to be set manually.
- Extensive unit-tests based on the output of the original R script
- A comprehensive README file
- Using the rprime package for parsing Eprime files
- Combining the WordLists and trial information by means of a table join, instead of row-by-row look-up.