Early Time Course Hemisphere Differences in Phonological

1
Early Time Course Hemisphere Differences in
Phonological Orthographic ProcessesLaura K.
Halderman1, Christine Chiarello1 Natalie
Kacinik21University of California, Riverside
2University of California, Davis
Introduction
Predictions
Simple Masking Experiment

• Left Hemisphere
• LH will show facilitation for orthographic
  condition at both SOAs
• LH will show facilitation for phonological
  condition at both SOAs
• Right Hemisphere
• RH will show facilitation for orthographic
  condition at both SOAs
• RH will not show facilitation for phonological
  condition
• However, if phonological processes are slower in
  the RH, facilitation may occur in the 50 ms SOA

• Chiarello (2003) has proposed that the cerebral
  hemispheres use different processing styles for
  linguistic information
• LH - processes information rapidly, quickly
  moving to a deeper level, discarding initial,
  more shallow representations
• RH - processes information more slowly,
  maintaining shallow representations of the
  information, even if no longer relevant
• Form-Specific vs Abstract Form (Marsolek et al.,
  1992)
• Word Length effects in RH (Lavidor Ellis, 2002)
• This model did not specifically consider
  phonological processes in reading
• Research on split-brain patients has revealed
  little evidence for Grapheme-to-Phoneme
  Conversion processes in RH (Baynes Eliassen,
  1998 Zaidel, 1985)
• GPC processes in the LH have been attributed to
  speech mechanisms being located in the same
  hemisphere
• Normal population studies have revealed mixed
  results
• Phonology
• No evidence for phonology in RH (Khateb et al.,
  2000 Crossman Polich, 1988 Rayman Zaidel,
  1991)
• Used metalinguistic tasks - Rhyme Judgment
• Evidence for bilateral phonological processes
  (Chiarello et al., 1999 Coney, 2002 Weekes,
  1999)
• Orthography
• Greater orthographic processes in the RH
  (Crossman Polich, 1988 Marsolek, 1992)
• Orthographic facilitation in RH, but no
  phonological facilitation (Lavidor Ellis, 2003)

• A second experiment was conducted using a similar
  paradigm to compare the effects of the pseudoword
  mask to a condition with no pseudoword mask
• All methods were repeated except no pseudoword
  appeared between the target and final pattern
  mask
• Only the 30 ms SOA was examined
• Participants - 48 Native English speaking,
  Right-handed undergraduates

Results
Backward Masking Methods
No pseudoword condition compared to pseudoword
condition - 30 ms

• Participants
• 96 Native English speaking undergraduates
• Right-handed
• Stimuli Presentation
• 150 Experimental Trials
• Pseudowords matched for orthographic regularity
  using Bigram frequencies (Mayzner Tresselt,
  1965)
• Forced Choice items matched for frequency
  (Francis Kucera, 1982)
• 30 ms 50 ms SOAs
• Each component of the stimuli was presented for
  30 ms or 50 ms

Lateralized Backward Masking

• Significant RVF advantage
• NoMask condition interacts with OP and OP-
  conditions
• No interaction between NoMask conditon and O-P-
  condition

Results
30 ms SOA - RT - Visual Field X Mask Type
Conclusions

• LH
• OP lt OP-
• OP- lt O-P-
• RH
• OP OP-
• OP- lt O-P-

• When no pseudoword mask is presented, the RVF
  advantage is equal to the Unrelated pseudoword
  mask condition
• However, when a pseudoword mask sharing relevant
  orthographic or phonological information is
  present, the hemispheres differentially use the
  overlapping information to facilitate target
  recognition
• The LH is benefited most by overlapping phonology
  at the shortest SOA, while the RH is benefited
  most by orthography
• RH does not show access to phonology, but it does
  demonstrate greater orthographic facilitation
  than the LH at this early time course, an effect
  similar to Lavidor and Ellis (2003)
• These results suggest the LH is capable of
  generating a phonological representation of words
  very quickly, perhaps automatically
• The RH, on the other hand, is capable of
  generating an orthographic representation very
  rapidly which stands in contrast to general
  findings of slower, RH processes (Chiarello,
  2003)

• Presentation of target beings initial processing
  which is interrupted by the pseudoword mask
• Similar masks reinstate some of the initial
  decoded properties of the target and facilitates
  identification
• Pattern mask restricts perception to the initial
  processed properties of the target and mask
• Participants are unaware of the nonword mask

50 ms SOA - RT - Visual Field X Mask Type

• LH
• OP OP-
• OP- lt O-P-
• RH
• OP OP-
• OP- lt O-P-

Mask Type Conditions
OP OP- O-P-
boat boat boat
BOTE BOTS NENE
This research was conducted under the support of
the National Science Foundation grant
BCS-0079456, granted to the second author.