home index by author index by subject

Poster P2

Friday stream 1-4 Session 09.50 - 11.10 Length 25 minutes

A Study of Compensated Adult Dyslexics Using Whole-Head Magnetoencephalography (MEG): Preliminary Findings

Alison Fisher, Ian Holliday and Ian Richards

Neurosciences Research Institute, Aston University. fisherae@aston.ac.uk

Abstract

MEG's temporal resolution allows for the tracking of rapid and dynamic changes in cortical activity and promotes it as the technique of choice for examining complex language-related disorders such as dyslexia. Psychophysical and magnetically recorded neurophysiological measures of the performance of dyslexics on sensory and reading related tasks will be reported. More specifically, groups of compensated adult dyslexics and matched controls, with full psychometric profiles, will be compared to investigate the neural correlates of a possible temporal processing deficit in dyslexia. Theoretical implications will be drawn and future directions for MEG research into dyslexia discussed.

Introduction

It has recently been suggested that low-level auditory deficits may play a causal role in developmental dyslexia. More specifically, dyslexics may be impaired in their ability to process brief and rapidly presented auditory stimuli. Significantly weaker magnetically recorded responses to such stimuli have been reported in a group of dyslexic adults [1]. Data presented here are part of an ongoing study examining low-level auditory deficits in dyslexia and focus on the effect of stimulus timing.

Method

Participants

? 4 dyslexics (2 males, mean age-24y9m, range 19y5m-29y11m) and 4 controls (2 males, mean age-25y10m, range 23y10m-29y9m). ? Students with FSIQ>90 and normal threshold hearing levels. ? Dyslexics selected on the basis of previous diagnosis and family history of literacy difficulties. ? See Table1 for psychometric profiles.

Table1: Psychometric Data for Dyslexics and Controls

Mean Standardised Scores

WAIS-III = Wechsler Adult Intelligence Scale, 3rd Edition; PhAB = Phonological Assessment Battery; WRAT3 = Wide Range Adult Achievement Test, 3rd Revision; WORD = Wechsler Objective Reading Dimensions

*Unpaired t-tests show significant differences between groups (p<0.05). FSIQ was not significantly different between groups.

Stimuli, Task and Recordings

• Stimuli demonstrated in Figure1.
• Tone discrimination (same or different); responses recorded on a separate occasion.
• MEG responses recorded using a 151-channel OMEGA Whole-Head System (CTF Inc.). 200 epochs of 2s (500msec pre-trigger) recorded at a sample rate of 1250Hz. Epochs triggered to first stimulus of each tone pair. 2s duration between epochs. Averages computed and band pass filtered between 1-70Hz.

Figure1: Tone Pair Stimuli

Data Analysis

• Two bilateral equivalent current dipoles were fitted to explain the recorded M100 response to a single tone in a pair, using a spherical head model. Only results for spatio-temporal fits with a Monte Carlo Volume <5cm3 were reported. Peak latency and amplitude of the fitted dipoles were noted for the M100 component to both tones.
• Coregistered Magnetic Resonance Images and MEG data in two control participants demonstrated that dipoles were located within the left and right auditory cortices.

Results

Figure2: A plot of the time course of fitted dipoles for a control participant in the 500msec ISI condition. The first tone is presented at 0msec, the second at 500msec. M100 responses are demonstrated by the amplitude peaks of the dipoles, approximately 100msec after each tone.

Figure3: Fitted dipoles overlaid on a control participant's MRI. Mean Percentage of Correct Discrimination Responses

Figure4: A mixed design ANOVA found a main effect for ISI (p<0.05) but not group.

Mean Amplitude of Left Dipole (nAm)

Figure5: Unpaired t-tests show significant differences in amplitude of response between groups to both tone1 and tone2 in the 500msec ISI condition (p<0.05). No significant differences were found in amplitude of response between groups in the 200msec ISI condition.

Mean Latency of Left Dipole (msec)

Figure6: No significant differences were found in latency of response between groups

Mean Percent Difference in Amplitude of Response to Successive Tones

Figure7: The amplitude of response to the second tone was subtracted from the amplitude of response to the first tone, and this difference was expressed as a percentage of the first tone. A positive percent difference indicates a reduction in the amplitude of response to the second tone, while a negative percent difference indicates an increase in the amplitude of response to the second tone.

Mean Difference in Latency Across Hemispheres (msec)

Figure8: A mixed ANOVA found a significant 3-way [Group*ISI*Tone] interaction (p<.02). Main effects for the three factors were not significant. Unpaired t-tests show significant differences between groups in response to the first tone in the 200msec ISI condition.

Conclusions

• Increasing the rate at which auditory stimuli are presented did not have a detrimental effect on the performance of dyslexics relative to controls on a tone-pair discrimination task.
• Trends within magnetically recorded data suggest impaired neural responses to low-level auditory stimuli in dyslexics. However, the timing of stimuli did not account for the differences observed between the participant groups.

References

1) Nagarajan, S., Mahncke, H., Salz, T., Tallal, P., Roberts, T. and Merzenich, M. M. (1999) Cortical Auditory Signal Processing in Poor Readers. Proc. Nat. Acad. Sci. USA, 96: 6483-6488.