Processing lexical semantic and syntactic information in first and second language: evidence form German and Russian

1
Processing lexical semantic and syntactic
information in first and second language
evidence form German and Russian

• S. Ruschemeyer, C. Fiebach, V. Kemp, and A.
  Frederici.
• Presented by A. Gusicova for Psych 526

2
Purpose

• Neural networks comparison between two
  typologically different languages ( semantic and
  syntactic) German and Russian
• Native (German) and non-native (German L2 Russian
  L1) semantic and syntactic processing

3
Experiment 1

• Participants
• 18 native German speakers (8 men mean age 25
  age range 23-30)
• 7 native Russian speakers (3 men mean age 30.5
  range 23-32)
• Ai Leen Gender proportions

4
Materials

• Scott Russian materials caused ELAN only for
  dlya.
• Ai Leen non word after dlya.

5
Procedure

• German
• 2 lists 240 items 48 per condition
• Repetitions of the same participle separated by
  at least 20 trials
• No more than three consecutive sentences belonged
  to the same conditionNo more then 4 consecutive
  trials contained either correct or incorrect
  sentences
• 8 s each ( 0, 400, 800, 1200ms)

• Russian
• 200 items 40 per condition
• Repetitions of a participle and null-event never
  occurred
• No more than 3 consecutive sentences belonged to
  the same condition
• No more then 4 consecutive trials contained
  either correct or incorrect sentences
• 10 s each ( 0, 500, 1, 000, and 1,500ms)

6
fMRI

• 8 axial slices 5mm thickness, 2 mm interslice
  distance, 64x64 data matrix 3mmx3mm inplane
  resolution 3T Bruker Medspec 30/100 system.
• T1-weighted modified driven equilibrium Fourier
  transform images nonslice-selective inversion
  pulse followed by a single excitation of each
  slice.

7
Results

• Note Look at the accuracy rates for Russian
  native speakers on the correct condition

8
(No Transcript)
9
Results cont

• Bonnie what does it mean to have decreased
  activation as a result of anomaly (posterior
  cingulate and inferior precuneus regions) ?

10
Results cont
11
Results Cont
12
Experiment 2

• Participants
• 18 native speakers of German ( same as in Exp 1)
• 14 Russian native speakers learners of German
• Note no independent measure of proficiency was
  reported

13
Results
14
Results
15
Results cont
16
conclusions

• Neural networks engaged in native speakers of
  typologically different languages process
  semantic and syntactic information are similar.
• Non-native speakers rely on different areas of
  neural networks then native speakers while
  processing semantic and syntactic information.