ICN_related

Thesis_Ailce TRFs Progress and scripts

(DONE) Predictors

# Scripts
All(gammatone, onset, CFG-Lexical): Thesis/Alice_gammatone_word-predictors.py

(DONE) Acoustic (envelope, word onset)

# TRF Scripts
  Natives: /Thesis/TRFs_produce-estimate_Natives.py
  ESLs: /Thesis/TRFs_produce-estimate_ESLs.py

(DONE) Word-Models (NGram, CFG, Fractality)

# CFG & N-gram TRFs (Produced and analyzed)
  Natives: /Thesis/TRFs_produce-estimate_Natives.py
  ESLs: /Thesis/TRFs_produce-estimate_ESLs.py

# Fractality Model script (unfinished)
  /Thesis/Importance_Fractality.py

# box count result (preview ver.)
  /Thesis/boxed_count_results_2.csv

(DONE) Lexical (Content v.s. Function words)

# Lexical & non-lexical TRFs (Produced and analyzed)
  Natives: /Thesis/TRFs_produce-estimate_Natives.py
  ESLs: /Thesis/TRFs_produce-estimate_ESLs.py

Instant Frequency (From envelope by HHT_EMD) (Almost, not yet divided into 12 parts)

Produced by EMD, working on how to trun into the 12 part of the data.
# Predictor: IF
  Alice_IF_EMD.ipynb
# IF TRFs
  Natives: /Thesis/??
  ESLs: /Thesis/??

Foundamental Frequency (F0) (Working on it, Matlab)

figures (Statistics/Analyses)

# envelope & onset
  Natives:/Thesis/Alice_Natives_analysis.ipynb
  ESLs:/Thesis/Alice_ESLs_analysis.ipynb

(DONE) Experiment (for ESLs)

Text Preprocessing

12 Qs: /Thesis/12Qs_preprocess.py
Corrected 12 Qs text: /Thesis/Alice_questionsLIST.json

Procedure

# MacbookPro & iMac 2017
  /Thesis/Alice_SoundPlay.py

# Win10 & NeuroScan (in NCU)
  /Thesis/Alice_SoundPlay_windows_EEGver.py

# Win8 & MEG 160 system(in Academia _Sinica)
  /Thesis/Alice_SoundPlay_windows_MEGver.py

---

Thesis_LTTC TRFs Progress (0% Starting with Predictor tables_)

Predictor tables (Partial, but not orgaanized and saved into csv)

(DONE) word (split)

(DONE) Segment (per tape)

Ongoing Onset & Offset

(DONE) Order (accumalate sequence)

LogFreq & LogFreq_Prev & LogFreq_Next NOPE

SndPower NOPE

Length (Offset - Onset) NOPE

(DONE) Position (by sentence)

(DONE) Sentence (by tape)

Ongoing IsLexical (1 & 0 == Booleen)

Almost NGram

CFG NOPE

Fractality NOPE

Acoustic (envelope, word onset) (NOPE)

Word-Models (NGram, CFG, Fractality) (NOPE)

Lexical (Content v.s. Function words) (NOPE)

Instant Frequency (From envelope by HHT_EMD) (NOPE)

Foundamental Frequency (F0) (NOPE)

figures (Statistics/Analyses) (NOPE)

THE PROGRESS OF THE THESIS

Table of Contents(ALL Content needs to be rewrite!!!)

• (Zero) 2 parts_ Chinese Abstract---
• (Zero) 2 parts_ English Abstract---
• (Zero) 2 parts_ Aknoledgements--
• (Done) Content Table---
• (Last to do) List of Tables---
• (Last to do) List of Figures---

1 Introduction

• 1.1 (Need to revise + add content) 3 parts_1&3 modified by the PI Bilingualism and second language learning
• 1.2 (Need to revise + add content) 3 parts_1/2/3 modified by the PI Speech comprehension
• 1.3 (Zero) 4 parts_ Auditory features (envelope, word onset, F0, instantaneous frequency_IF)
• 1.4 (Zero) 3 parts_ Linguistic features (Lexicality, semantic surprisal, syntactic surprisal_POS tags)
• 1.5 (Zero) 2 parts_ Language model (Ngram, CFG)
  • 1.5.1 (Zero) 3 parts_ N-gram model
  • 1.5.2 (Zero) 3 parts_ Context Free G model
• 1.6 (Need to add content) 3 parts_ Temporal Response Function (TRF) Analysis
• 1.7 (Need to add content) 3 parts_Aims of this study

2 Experiment 1

• 2.1 (Zero) 2 parts_ Intro
• 2.2 (1st Draft) Methods and Materials
  • 2.2.1 (1st Draft_need to check on Natives)Participants
  • 2.2.2 (1st Draft_need revised on REFs) Stimuli
  • 2.2.3 (1st Draft) Procedure (Same as Brennan & Hal, 2019 + Vocabulary Size Test)
  • 2.2.4 (1st Draft_need revised on ESL-EEG) EEG recording
  • 2.2.5 (1st Draft)_need to check refernces descriptions EEG pre-processing
• 2.3 Results (TRF)
  • 2.3.1 1st_Env & onsetEnv//F0 & IF not included in here currently Auditory response functions (Env / onsetEnv)
  • 2.3.2 (Zero) need description Word class Categories: Function word and content word
  • 2.3.2 (Zero) need description Language model: Ngram & CFG
• 2.4 (Zero) 2 parts_ Summary

3 Experiment 2_LTTC

• 3.1 (Zero) 2 parts_summarize it from LTTCIntro
• 3.2 Methods and Materials
  • 3.2.1 (1st Draft) Participants
  • 3.2.2 (1st Draft) Stimuli
  • 3.2.3 (1st Draft) Procedure (Same as LTTC)
  • 3.2.4 (1st Draft) MEG recording
  • 3.2.5 (Need to revise) MEG pre-processing
• 3.3 (1st Draft) Results (TRFs)
  • 3.3.1 (1st Draft) Behavioral results (LDT)
  • 3.3.2 (1st Draft) Sensor level of CD (ERP)
• 3.4 (1st Draft) Summary

4 General Discussion and Conclusion

• 4.1 (Zero) Summary

• 4.2 (Zero) Differences in Natives & ESLs

• 4.3 (Zero) Surprisal representations not that useful?

• 4.4 (Zero) CFG not sig. in both groups (would materials influence the CFG? story主題性較發散, cfg對於主題性預測有關？)

• 4.5 (Zero) (Universal Grammar?)

• 4.6 (Zero) Application TRF analysis

• 4.7 (Zero) Limitation and the future direction

• ### References

• ### Appendixes