Halliday poster

1
HEMISPHERIC DIFFERENCES IN PROCESSING DICHOTIC
SPEECH AND NON-SPEECH I.Yasin and D. V. M.
Bishop Department of Experimental Psychology
University of Oxford
2
INTRODUCTION

Dichotic listening tasks using speech sounds
have often shown a right-ear advantage,
indicating left-hemisphere dominance (Schwartz
and Tallal, 1980). These results are often
explained by the relatively strong contralateral
projections, which may block the processing of
the ipsilateral projections, to the auditory
cortices in the temporal lobes (Bryden, 1988). An
objective measure of brain laterality can be
obtained by using the mismatch negativity (MMN).
The MMN is a cortical brain response reflecting
detection by the brain of a change in the
stimulus. The MMN may be elicited by presenting
an occasional deviant sound in a sequence of
standard sounds. MMNs elicited by non-speech
(Paavilainen et al., 1991) and speech (Shtyrov et
al., 1998) sounds have been shown to be greater
in the right and left hemispheres, respectively,
but this has not always been replicated. The aim
of this study was to investigate the extent of
left and right hemisphere contribution to MMN
strength for both speech and non-speech stimuli
using a dichotic presentation. The speech stimuli
were two consonant-vowel (CV) pairs ba/da and
ta/ka. The non-speech stimuli were a pair of 1.0-
and 1.2-kHz pure tones. 10 normal-hearing
listeners were presented with a repeating
standard to both right and left ears (ba, ta, or
1-kHz tone), with a deviant (da, ka, or 1.2-kHz
tone) presented at a probability of 10 in either
the left or right ear. Since late MMNs to speech
may arise as a consequence of more complex higher
processing prior to categorisation (Korpilahti et
al., 1995), it was predicted that lateralisation
of speech stimuli would be more evident in the
late MMN (350-500 ms after onset) than in the
early MMN (100-200 ms after onset).
3
AIM

• The aim of this study was to investigate the
  extent of left and right hemisphere contribution
  to MMN strength for both speech and non-speech
  stimuli using a dichotic presentation.
• An advantage of dichotic presentation since
  there is ipsilateral suppression of the auditory
  pathways (Cranney et al., 1989
• McNeil et al., 1981), any lateralised response
  may be more pronounced

4
METHOD

• Ten normal-hearing, right-handed listeners
  participated
• The speech stimuli were two consonant-vowel (CV)
  pairs ba/da and ta/ka
• synthesised with the IPOX All Prosodic
  Speech Synthesis Architecture
• The non-speech stimuli were a pair of 1.0- and
  1.2-kHz pure tones with a 352-ms steady-state
  portion and 5-ms on- and off-set ramps
• For all stimuli, the stimulus onset to offset
  durations were 362 ms and the offset to onset
  inter-stimulus interval was 638 ms
• The signal was presented at 75 dB SPL for all
  listeners
• The listeners were presented with a repeating
  standard of each stimulus pair to both right and
  left ears (ba, ta, or 1-kHz tone), with a deviant
  of each pair (da, ka, or 1.2-kHz tone) presented
  at a probability of 10 in either the left or
  right ear
• Cortical responses were measured from an array of
  30 scalp electrodes

5
METHOD The cortical response and MMN
For the purposes of analysis only the responses
from the electrode positions highlighted in blue
will be discussed
Mid line
Right hemisphere
A typical cortical response
Left hemisphere
8
Amplitude (µV)
0
Fz
FT7
FT8
Cz
C4
C3
T7
T8
-8
100
200
CP3
CP4
TP7
TP8
Latency (ms)
_
Standard response
Deviant response
6
RESULTS Cortical response to a deviant da
among standards ba
Left hemisphere
Right hemisphere
Mid line
Averaged cortical responses across the 10
listeners An MMN to both the right and left
deviant at around 100-200 ms Also a late MMN to
both left and right deviants at around 350-500
ms. Particularly noticeable for Fz, Cz, C3, and
C4.
7
RESULTS Cortical response to a deviant ka
among standards ta
Left hemisphere
Right hemisphere
Mid line
Averaged cortical responses across the 10
listeners An MMN to both the right and left
deviant at around 100-200 ms Also a late MMN to
both left and right deviants at around 350-500
ms. Particularly noticeable for Fz, Cz, C3, and
C4.
8
RESULTS Cortical response to a deviant tone of
1.2 kHz among standards tones of 1 kHz
Left hemisphere
Mid line
Right hemisphere
Averaged cortical responses across the 10
listeners MMNs were the greatest for the tone
stimuli An MMN to both the right and left deviant
at around 100-200 ms Also a late MMN to both left
and right deviants at around 350-500 ms.
Particularly noticeable for Fz, Cz, C3, and C4.
9
RESULTS

• MMN (100-200 ms)
• A 3x3x2 ANOVA (right, mid-line, and left
  cortical response regions) x (ba/da, ta/ka and 1
  kHz/1.2 kHz stimuli) x (ear of deviant stimulus
  presentation, right or left) showed no
  significant effect at plt0.05.

• MMN (350-500 ms)
• A 3x3x2 ANOVA (right, mid-brain, and left
  cortical regions) x (ba/da, ta/ka and 1 kHz/1.2
  kHz stimuli) x (ear of deviant stimulus
  presentation, right or left) showed that there
  was a significant interaction between cortical
  response region and the ear of deviant stimulus
  presentation F(2,18)12.41 plt0.001. On further
  analysis using one-sample t-tests the late MMN
  was significantly larger then zero for a right
  ear deviant eliciting a right hemisphere response
  t(9)5.397 plt0.001.

10
CONCLUSIONS

• MMN (100-200 ms)
• Hemispheric asymmetry is thought to depend
  on whether the auditory stimuli are perceived as
  speech (processed predominantly in the left
  hemisphere) or non speech (processed
  predominantly in the right hemisphere) (Rinne et
  al., 1999 Hertrich et al., 2002). The presented
  findings do not support lateralisation for either
  the speech or non-speech stimuli. A similar
  finding was reported by Shtyrov et al. (2000)
  the mismatch negativity revealed no hemispheric
  differences in processing of speech sounds in a
  dichotic listening task. Our data are more in
  line with some recent evidence from functional
  neuroimaging suggesting parallel streams of
  auditory processing with complex interactions
  within, as well as across, hemispheres (Scott et
  al., 2003) may not lead to clear evidence of
  early lateralisation of speech.
• MMN (350-500 ms)
• It was predicted in the introduction that
  lateralisation of speech stimuli would be more
  evident in the late MMN (350-500 ms after onset)
  than in the early MMN (100-200 ms after onset).
  The presented findings show that there is no
  evidence of a significantly greater MMN for the
  speech stimuli as compared to the non-speech
  stimuli. However, regardless of stimulus type,
  there is a greater MMN elicited in the right
  hemisphere (as compared to the left hemisphere)
  by a deviant sound presented to the right ear (as
  compared to a deviant presented to the left ear).
  These findings are similar to those reported by
  Hertrich et al. (2002). Hertrich et al. also
  reported a significantly larger mismatch response
  from the right hemisphere generated by a deviant
  sound presented to the right ear in a dichotic
  listening task. One explanation for these results
  may be that the right hemisphere is activated by
  the switching of attention to the deviant sound
  (Fiez et al., 1995 Schulte-Korne et al., 2001)
  which is then engaged in the processing of
  non-linguistic aspects of the stimuli.

11
REFERENCES

• Bryden, M. (1988). An overview of the dichotic
  listening procedure and its relation to cerebral
  organization, in Handbook of dichotic
  listening theory, methods and research, edited
  by K. Hugdahl (John Wiley Sons, New York), pp.
  1-43.
• Fiez, J. A., Raichle, M. E., Miezin, F. M.,
  Petersen, S. E., Tallal, P., and Katz, W. F.
  (1995) PET studies of auditory and phonological
  processing effects of stimulus characteristics
  and task demands, J. Cognititive Neuroscience 7,
  357-375.
• Hertrich, I., Mathiak, K., Lutzenberger, W., and
  Ackermann, H. (2002). Hemispheric lateralization
  of the processing of consonant-vowel syllables
  (formant transitions) effects of stimulus
  characteristics and attentional demands on evoked
  magnetic fields, Neuropsychologia 40, 1902-1917.
• Korpilahti, P., Lang, H., and Aaltonen, O.
  (1995). Is there a late latency mismatch
  negativity (MMN) component? Electroencephalogr.
  Clin. Neurophysiol. 95-96
• Paavilainen, P., Alho, K., Reinikainen, K., Sams,
  M., and Näätänen, R. (1991). Electroencephalogr.
  Clin. Neurophysiol. 78, 466-479.
• Rinne, T., Alho, K., Alku, P., Holi, M.,
  Sinkkonen, J., and Virtanen, J. (1999). Analysis
  of speech sounds is left hemisphere predominant
  at 100-150 ms after sound onset., Neuroreport
  10, 113-117.
• Schwartz, J., and Tallal, P. (1980). Science 207,
  1380-1381.
• Scott, S. K., and Johnsrude, I. S. (2003). The
  neuroanatomical and functional organization of
  speech perception, Trends in Neuroscience
  26,100-107.
• Schulte-korne, G., Deimel, W., Bartling, J., and
  Remschmidt, H. (2001). Speech perception deficit
  in dyslexic adults as measured by mismatch
  negativity (MMN), Int. J. Psychophysiol. 40,
  77-87.
• Shtyrov, Y., Kujala, T., Lyytinen, H., Kujala,
  J., Ilmoniemi, R. J., and Näätänen, R. (2000).
  Lateralization of speech processing in the brain
  as indicated by mismatch negativity and dichotic
  listening, Brain and cognition 43, 392-398.
• Shtyrov, Y., Kujala, T., Ahveninen, J.,
  Tervaniemi, M., Alku, P., Ilmoniemi, R. J., and
  Näätänen, R. (1998). Background acoustic noise
  and the hemispheric lateralization of speech
  processing in the human brain magnetic mismatch
  negativity study, Neuroscience Letters 251,
  141-144.
• The research was supported by the Wellcome Trust