Skip to content

Latest commit

 

History

History
63 lines (53 loc) · 2.45 KB

README.md

File metadata and controls

63 lines (53 loc) · 2.45 KB

______________________________________________________________

  • Programmer: Farhan Bin Faisal
  • Files: preprocess.ipynb, wordListMaker.m, postWuggy.ipynb, generateConditions.ipynb
  • Date Created: 10 November 2022
  • Meltzer Lab

______________________________________________________________

RESEARCH POSTER

NBack_Poster

______________________________________________________________

EXPERIMENT PARADIGM

The NBAck Training paradigm was programmed in JavaScript (using PsychoJS) and hosted on Pavlovia. A trial run can be accessed here. Please input the following credentials

  • session: 1
  • participantID: 19995

______________________________________________________________

STIMULY GENERATION PIPELINE

1.) preprocess.ipynb

  • Loads word csv file into a pandas dataframe
  • Filters rows for frequency (4.0 Zipfvalue < 5.0)
  • Gets syllables of each word
    • Uses cmu_dict from nltk
  • Filters rows for syllables (1 < syllable < 2)
  • Replaces each word with its corresponding lemma
    • Uses WordNetLemmatizer for this
  • Discards rows not found in cmudict
  • Discards profane words and names
  • Formats dataframe into SOS-compatible input
  • Outputs dataframe as a tab delimited txt file
    • File Named "sos_input.txt"

______________________________________________________________

2.) wordListMaker.m

  • Uses SOS to make 18 lists of 10 words each
  • Lists matched on ZipfValue and syllables
    • Used soft constraints for this
    • Used hard constraints to floor syllable count and frequency
  • Lists can be found in the folder "wordLists"

______________________________________________________________

3. a) wuggy/postSOS.ipynb

  • Generates nonword lists for every wordList
  • nonWordLists outputted to folder nonWordList
  • Prints filenames of files which contain words that could not be converted to a nonWord automatically
    • Need to generate those pseudowords manually

______________________________________________________________

3. b) Wuggy || Generate nonWords manually

  • Download from http://crr.ugent.be/programs-data/wuggy
  • Settings used:
    • Orthographic English
    • Match syllable length
    • Match word length
    • Match transition frequency
    • Match 2 out of 3 segments
  • Manually pass the words that could not be found in the lexicon through Wuggy