Research Description
Highlights of Major Accomplishments
- Demonstration by quantified EEG (qEEG) analysis of a disturbance in rapid auditory processing in the left parieto-temporal region (i.e., Wernickeâs area) in reading-impaired children,
- Discovery by qEEG and 3D-source analysis of a defect in function of the parvocellular component of the visual system or the visual word forming system or both in reading-impaired children,
- Demonstration by qEEG spectral and coherence studies of an apparent defect in the accessing of information from the left temporal region in reading-impaired children.
Major Results
2. QEEG studies of infant behavior
In concert with Dr. Deborah Waber of this MRRC, we have been studying qEEG data in learning impaired (LI) and non-learning impaired (non-LI) children ranging in ages from 7 to 11 years. The overall goal of the work is to determine whether there exist fundamental behavioral and neurophysiological abnormalities within this population that could be uncovered by detailed psychological and qEEG evaluations. QEEG data in fact have delineated a number of prominent abnormalities in LI and reading impaired (RI) children.
1.1 Two-tone auditory evoked potential difference in reading impairment. Tallal and others have hypothesized that inability to resolve rapidly changing auditory stimuli is a fundamental defect in children with RI. This defect in the processing of rapidly changing auditory stimuli may make phonemes more difficult to understand since they contain very rapid auditory transitions. We stimulated children with single tones and tone pairs of varying separation. No differences were noted in the resulting auditory evoked potential (EP) to single tones between RI and non-RI children. However, when tone pairs separated by 50 msec were delivered as stimuli, a large region of between-group difference was located overlying the parieto-temporal junction region on the left, in Wernickeâs region. Evidence suggests an absent component in the 2-tone EP for the RI children, a finding which supports a disturbance in resolution of rapidly changing auditory stimuli in RI.
1.2 Color evoked potential difference in reading impaired children. The visual system is believed to comprise at least two data streams, one termed magnocellular (M) and the other parvocellular (P). The M system is most sensitive to large, peripheral, moving stimuli and is relatively insensitive to color and stationary stimuli. In contrast the P system is maximally activated by smaller, central, stationary stimuli and is quite sensitive to color. Work by others has suggested that the M system is abnormal in RI children, causing problems with eye movement that, in turn, impedes reading. More recent work has cast some doubt on this notion. Our approach was to compare RI and non-RI children on the basis of stimuli to larger black and white alternating checks (M stimulus) as compared to smaller red and green alternating checks (P stimulus). As the P system seems most likely to convey information about small stationary stimuli (like letters) we hypothesized that it would be involved in RI. Indeed red-green check visual EP demonstrated much greater RI vs. non-RI group difference than did black-white check EP. Moreover this P system difference was shown by 3D-source analysis to be located at the inferior aspect of the left occipital-temporal junction region. This region is known as the Visual Word Forming System (VWFS) where in adults word blindness results from stroke. It is also very close to the region shown by fMRI as being maximally sensitive to color stimulation. We hypothesize that deficits either in the P visual system or in the VWFS may be partially causative of dyslexia.
1.3 Abnormal cortical connectivity in poor readers during a tapping task. Work by Dr. Peter Wolff of this MRRC has shown that poor tapping performance is associated with poor reading ability, although the causal inter-relationship has remained unclear. We studied a group of children with wide scatter in reading and tapping ability. The task was to tap both fingers alternately (and in various other combinations) in time with a metronome. Then the metronome was stopped and the children asked to continue tapping at the same (remembered) rate. During this task we collected ongoing EEG data. Spectral coherence was gathered among all electrode pairs across all spectral bands. A single coherence factor demonstrated a large difference between the good and poor tappers. The coherence factor involved the fast alpha and slow beta bands and indicated a relative disconnection between the left midtemporal electrode and electrodes located over motor and motor association cortices of both hemispheres, left more than right. We hypothesize that the link between the poor reading and poor tapping results from relative difficulty in accessing information within the left temporal lobe. To tap well, one must access the working memory of the metronome in the left temporal lobe. To read well one must also access information in the left temporal region. The diminished coherence signaled reduced accessibility which could impede both tapping and reading.
2. QEEG studies of infant behavior
In our continuing work with Dr. Heidi Als in this MRRC (see Dr. Alsâ Progress Report), we are capturing and analyzing qEEG data from infants who are premature and/or small for gestational age (SGA). In our past qEEG studies, the findings have supported Dr. Alsâ hypothesis that individualized behaviorally-based intervention in premature infants leads to a more favorable outcome. Our data have demonstrated more "full-term-like" spectral, spectral coherence, and evoked potential features in the intervened-with infants. We are now beginning to evaluate SGA preterm infants, who are at higher risk for later developmental problems than are AGA preterms and full term controls. We are gathering concurrently behavioral, MRI, and qEEG data on SGA and AGA infants shortly after birth and again when they reach full term. One-half of the population will have the comprehensive behaviorally-based intervention.
