Journal of the Neurological Sciences, 1978, 37:51-58 © Elsevier/North-Holland Biomedical Press
C O R T I C A L SLOW P O T E N T I A L S A N D CONGENITAL BLINDNESS
51
THE OCCIPITAL EEG IN
JEFFREY L. NOEBELS*, WALTON T. ROTH and BERT S. KOPELL Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, Calif. 94305 and Veterans Administration Hospital, Palo Alto, Calif. 94305 (U.S.A.)
(Received 20 December, 1977) (Accepted 9 January, 1978)
SUMMARY In a group of 7 congenitally blind adults, electroencephalographic occipital alpha rhythms were absent, but slow negative cortical potentials (CNV) were recorded over the visual cortex and were similar to those of normally sighted controls. Frontal and central CNV amplitudes, but not auditory evoked potentials or reaction times, were decreased significantly in the blind. The results confirm the presence of a developmental E E G abnormality following early blindness, and demonstrate the independance of slow potential generation from rhythmic 8-12 c/sec E E G activity in deafferented visual cortex. Non-sensory specific, event-related potentials such as the Contingent Negative Variation may prove to be useful psychophysiological probes of residual cortical function in brain regions which are deprived of primary sensory input.
INTRODUCTION Two questions are central to an understanding of neurologic and psychiatric conditions underlying developmental sensory deficits in man. What residual function persists in the denervated cortex? To what extent are other brain mechanisms altered in response to selective injury of a sensory system? In the visual system, long-term deficits in sensation produce characteristic changes in the electrophysiology of the brain. Unit recordings of abnormal cellular activity within the visual cortex are
This work was supported by NIMH training grant MH-08304-11 (J.L.N.) and the Veterans Administration. * Present address: Department of Neuroscience, Childrens Hospital, 300 Longwood Avenue, Boston, Mass. 02115, U.S.A. - - to whom all correspondance should be addressed. This work was awarded the 1975 student prize at the Annual Meeting of the Western EEG Society, Seattle, Wash.
52 associated with histologically evident synaptic disorganization, resulting from the absence of tonic retinal influence during critical periods of development (Globus and Scheibel 1967; Hubel and Wiesel 1970; Chow 1973). Clinically, alterations of spontaneous electroencephalographic (EEG) activity in human peripheral blindness have been reported since the earliest studies of the Berger rhythm (Berger 1929; Adrian and Matthews 1934). The most repeated finding in early blindness is a paradoxical absence of the occipital alpha rhythm (8-12 c/sec) normally present in sighted persons when the eyes are closed or deprived of patterned stimuli. The implications of alpha wave absence to theories of altered brain function in blindness have led to considerable clinical and theoretical interest in redefining what is electroencephalographically normal for the visually deafferented brain (see Noebels 1977 for references). Recently, the prospect of a cortical prosthesis (Dobelle, Mladejovsky and Girvin 1974) in blind patients underlines the importance of establishing criteria for estimating the degree of residual function by non-invasive procedures prior to surgical intervention. When blindness is complete, the use of direct visual evoked potential recordings is of little value and alternative methods of examining cortical function are required. The Contingent Negative Variation (CNV) is an event-related slow potential sensitive to levels of attention, arousal, and anxiety which arises between the paired presentation of warnmg and imperative stimuli (Walter, Cooper, Aldridge, McCallum and Winter 1964; Tecce 1972). This cerebral slow potential is unrelated to stimulus modality and is distributed widely over the scalp. In the present study, the CNV paradigm was selected in order to assess the regional integrity of slow potential processes in the deafferented human visual cortex, which is otherwise inaccessible to evoked potential analysis. METHODS
Subjects Subjects were selected on the highly restrictive basis of early retinopathic blindness uncomplicated by other developmental abnormalities, recurrent neurological complaints, or concurrent medication. Seven totally blind college students participated in the study with 7 normally sighted, age- and sex-matched controls. Blindness was diagnosed as due to retrolental fibroplasia (RLF) (5 cases), retinoblastoma (1), and uveitis with glaucoma (1) (see Table 1).
Procedures The EEG and CNV were recorded with platinum pin electrodes (Corby, Roth and Kopell 1974) at midline scalp locations (Oz, Pz, Cz, Fz) referred to linked disc electrodes on the earlobes. Previous work in our laboratory (Corby et al. 1974) has shown that subdermal electrodes significantly reduce cephalic skin potential artifact in CNV recordings. Eye movement was monitored with electrodes located at the mid-pupillary line at the right superior and inferior orbital margins. Bioelectric activity was amplified through calibrated Princeton Applied Research amplifiers
53 TABLE 1 SUMMARY OF RETINOPATHIC DEFICIT IN 7 CONGENITALLY BLIND SUBJECTS Residual activity Subject
Sex
Age
os.
od.
Onset
Etiology
B.S. C.L. K.V. D.R. B.H. T.P. J.S.
M F M M F M M
21 22 22 23 24 26 27
0 0 0 0 0 0 0
0 0 0 light • light a 0 0
birth 6 months 1 month birth birth 3 years birth
RLF retinoblastoma RLF RLF RLF uveitis, glaucoma RLF
Could detect on/off change of a brilliant light source. (input impedance 100 Megohm, bandpass 0.03-100 Hz), reset to baseline prior to each trial through D C level zeroing circuitry (MacPherson and Kopell 1972), and recorded on magnetic tape. Following electrode placement and system calibration, each participant was confortably seated in a darkened, sound-attenuated enclosure and instructed to perform in a reaction time task. Warning tones ($I = 500 Hz at 85dbSL) delivered through earphones were followed 1.5 sec later by imperative tones ($2 = 800 Hz at 85dbSL) to which a pushbutton response (P) was required within 300 msec in order to avoid a penalty hiss. Over a period of 15 min, 50 trials were repeated at intervals varying randomly from 4-8 sec, and were divided into 2 conditions. For the first 25 trials, sighted subjects were asked to fixate on a small lit star in order to minimize eye movement artifact. For the remainder of the experiment, all subjects were asked to close their eyes. The E O G was absent in all but one blind subject, eliminating the problematic requirement of ocular fixation in the blind. With this one subject, as well as with all sighted controls, any trial containing detectable E O G activity was excluded from the average. Thirty-two consecutive four sec epochs of E E G and EOG, 16 with the eyes open and 16 with eyes closed were averaged on a Fabritek computer for each electrode location. Off-line spectral analysis of the EEG, computed with a D E C PDP12 for each electrode in the form of a Compressed Spectral Array (CSA), allowed qualitative inspection of frequency characteristics across subjects. The CNV amplitude was quantified by measurement from the prestimulus baseline (automatically reset to zero prior to each trial) to a point of maximum negativity (averaged over 0.1 sec) during the $1-$2 interval. Auditory evoked potential (EP) amplitudes in responses to $1 were measured peak to peak. RESULTS
Occipital EEG and C N V amplitudes Raw E E G records and the computed CSAs were examined for occipital alpha wave abundance. Figure 1 shows a representative CSA at each of four electrode locations for a typical blind and sighted subject. Occipital alpha rhythm was found
54 BLIND
Fz Fz
Cz
Pz
Oz Oz
0
4
8
1
16
Hz
Fig. 1. Absence of occipital alpha rhythm in blindness, Compressed Spectral Array (CSA) represents
relative abundance of EEG frequency components (3-16 c/see) recorded at 4 scalp locations from a sighted subject with eyes closed (left, J.F.) and a blind subject (right, B.H.). Each horizontal trace beginning with the lowest in each sequence represents consecutive 4 sec EEG epochs. Data from representative blind subject show absence of 8-12 c/see activity at Oz normally prominent in sighted subjects. to be absent in all blind subjects, and regularly present on eye closure in all sighted controls. In 4 of the blind subjects, prominant 8-12 c/see activity was present with a primarily central distribution. CNV averages consisting of 16 trials from the same subjects are shown in Fig. 2. The CNV amplitude in the blind subject (Fig. 2B) is decreased with respect to the normal subject at the Fz and Cz leads (Fig, 2A), but present and nearly equal at Pz and Oz. Similar CNV amplitudes at Oz (13.5, 10, 4, 6, 6.5, 7.5 #V respectively) were obtained in each of the 6 other blind subjects. Figure 3A illustrates the mean CNV amplitudes of the blind group at Oz, Pz, Cz, and Fz, which represented 101.4 ~ , 73.5 ~ , 59.6 ~ , and 63 ~ respectively, of normally sighted amplitudes. The CNV amplitudes were analyzed using a three factor ANOVA with repeated measures on two factors. The blind group had smaller CNVs than the sighted (P < 0.01) and electrode position significantly affected CNV amplitude in all subjects (P < 0.001). An interaction between blindness and electrode position showed, with subsequent lead by lead t-tests, that the blind produced smaller CNVs at the vertex (P < 0.01) and frontal leads (P < 0.05) than did normally sighted controls. At the vertex the effect was present in 6 of the 7 blind subjects. Occipital and parietal differences were not significant.
BLIND
SIGHTED
I CAL ~ S 1 ÷
i i(
S~p )1
1.5
S1 I
i
4 sec.
S2 !
Fig. 2. Contingent Negative Variation (CNV) recorded from four midline locations in a typical sighted and blind subject (same subjects as Fig. 1). Vertical bars in upper traces (Fz) indicate measurement technique. CNV stimulus paradigm described in Methods is shown below. A calibration pulse of 10/~V initiates each averaging trial.
A
B
30• Blind o Sighted 25-
T
20/
joJ
/
' >
%
//
15-
T
I
t0-
5-
i
i
i
i
i
i
~
i
Oz
Pz
Cz
Fz
Oz
Pz
Cz
Fz
Electrode Location
Fig. 3. Mean amplitudes with standard error of the CNV (A), and the auditory EP (N]-P2) to the tone Sl (B) at each of 4 midline recording sites in blind and sighted groups.
56 In a similar analysis of peak to peak N~-P2 auditory EP amplitudes to S~. the means did not differ significantly across groups (Fig. 3B), and only electrode location had a significant effect (P < 0.001). Reaction times to $2 of the two groups did not significantly differ. The blind tended to have shorter reaction times (mean : 192 msec) than the sighted (mean ~- 218 msec). Eye closure had no significant effect on CNV amplitudes, N1-Pz amplitudes, or reaction time in either group. DISCUSSION The principal finding in this study was that early primary deafferentation of the visual cortex does not abolish occipital slow potential production in the adult, despite the absence of occipital alpha rhythm found in each of the blind subjects. A statistical dissociation of CNV amplitudes from background 8-12 c/sec EEG rhythms in normal cortex has been considered previously (Heinneman 1973). However, the conclusion that generation of rhythmic EEG activity and slow potentials depends on separate physiological substrates is clearly supported by the results of the blind subjects, where a congenital lesion, presumably confined to the deeper layers of visual cortex (Chow 1973), disrupted only the alpha rhythm. The presence of an occipital CNV in the blind is anatomically consistant with the electrophysiological evidence that diffuse thalamo-cortical projections mediate cortical slow potential production (Skinner 1971; McCallum, Papakostopoulos, Gombi, Winter, Cooper and Griffith 1973) since fibers in this system terminate primarily in superficial cortical layers where no histological effects of visual deafferentation have been described (Globus and Scheibel 1967; Chow 1973). This interpretation is further supported by the demonstration that patterns of neuronal response generated within different cortical laminae can be selectively abolished (Amassian, Waller and Macy 1964; Calvet, Calvet and Scherrer 1964), and that diffuse destruction of cortex can markedly decrease CNV amplitudes overlying the lesion (McCallum, Walter, Winter, Scotton and Cummins 1970). Thus to the extent to which scalprecorded slow potentials reflect focal cortical activity at the electrode site, the unambiguous presence of' an occipital CNV may prove to be a valid indicator of residual, non-sensory specific function in the visual cortex of the blind.
Extra-occipital changes in the blind Along with the EEG abnormalities attributed to disuse of primary visual pathways in congenital blindness, an additional finding in our subjects with life-long retinal disease was a large decrease of slow potential production over the anterior portion of the scalp to nearly 6 0 ~ of the sighted amplitude, without significant alteration of auditory Nt-Pz EP amplitudes or motor performance. A similar result has been reported in a preliminary study of a subgroup of congenitally blind children (Lairy and Guibal 1969) (but see Cohen 1973). The possibility that the lack of a steady corneoretinal potential might spuriously affect the measured CNV amplitudes is inconsistant with the observed tOl~ography of CNV decrement, which was greater at the vertex than at the frontal location. Fast reaction times in the blind rule out
57 the contribution of psychological variables (such as attention) known to affect both RT and the CNV, even though these processes may be fundamentally independent (Rebert and Tecce 1973). We can only speculate on the significance of this result. High levels of anxiety are known to decrease CNV amplitudes (Low and Swift 1971), and Berger (1935) originally suggested that anxiety may be elevated in the blind. It is possible that habituation to the testing procedure might reduce the difference observed. However, because of the difficulties in estimating the degree of perceived stress during the experimental situation, and the interaction of autonomic (cephalic skin potential) contaminants in the CNV which were controlled for in this study, the contribution of anxiety to frontal CNV differences between blind and sighted subjects remains uncertain. A second problem is that the frontal CNV has been previously distinguished from CNVs recorded at other sites, and may reflect separate psychological variables (Weinberg and Papakostopoulos 1975). Finally, we have no evidence to exclude the possibility that frontal slow potential generation, along with other electrophysiological (Cohen 1970; Kasamatsu 1970; Novikova 1974), morphological (Ryugo, Ryugo, Globus and Killackey 1975), and neurochemical (Bennett, Diamond, Krech and Rosenzweig 1964) processes occurring in areas outside the occipital lobes are sensitive to the normal development of central visual pathways. ACKNOWLEDGEMENTS Provision of CSA software by Dr. Reginald Bickford and Lee Berger is gratefully acknowledged.
REFERENCES Adrian, E. D. and Matthews, B. H. C. (1934) The Berger rhythm - - Potential changes from the occipital lobes in man, Brain, 57: 355-385. Amassian, V. E., Waller, H. J. and Macy, J. (1964) Neural mechanism of the primary somatosensory evoked potential, Ann. N.Y. Acad. Sci., 112: 5-32. Bennett, E., Diamond, M., Krech, D. and Rosenzweig, M. (1964) Chemical and anatomical plasticity of brain, Science, 146: 610-619. Berger, H. (1929) Ober das Elektrenkephalogramm des Menschen, Arch. Psychiat. Nervenkr., 87: 527-570. Berger, H. (1935) Ober das Elektrenkephalogramm des Menschen, Arch. Psychiat. Nervenkr., 103: 444--454. Calvet, J., Calvet, M. C. and Scherrer, J., (1964) Etude stratigraphique corticale de i'activit6 EEG spontan6e, Electroenceph. clin. Neurophysiol., 17: 109-125. Chow, K. L. (1973) Neuronal changes in the visual system following visual deprivation. In: Handbook of Sensory Physiology, VoL 7. Sect. 8, Springer-Verlag, New York, pp. 599-630. Cohen, J. (1970) Brain waves and blindness. In Proceedings of the Conference on New Approaches to the Evaluation of Blind Persons, American Foundation for the Blind Research Bulletin, June,
New York, pp. 112-132. Cohen, J. (1973) Developmental aspects of the Contingent Negative Variation, Electroenceph. clin. NeurophysioL, Supplement 33, pp. 133-137. Corby, J., Roth, W. T. and Kopell, B. S. (1974) Prevalence and methods of control of the cephalic skin potential EEG artifact, Psychophysiology, 11 : 350-360.
58 Dobelle, W. H., Mladejovsky, M. G. and Girvin, J. P. (1974) Artificial vision for the blind - - Electrical stimulation of visual cortex offers hope for a functional prosthesis, Science, 183 : 440-443. Globus, A. and Scheibel, A. B. (1967) The effect of visual deprivation on cortical neurons - - A Golgi study, Exp. Neurol., 19: 331-345. Heinneman, U., (1973) Time variability of slow negative potentials and changes in alpha rhythm. Electroenceph. clin. Neurophysiol., Supplement 33, pp. 125-126. Hubel, D. H. and Wiesel, T. N. (1970) The period of susceptibility to the physiological effects of unilateral eye closure in kittens, J. Physiol. (Lond.), 207: 417-436. Kasamatsu, T. (1970) Effects of visual deafferentation on mesencephalic reticular activity, EaTs. NeuroL, 29: 251-267. Lairy, G. C. and Guibal, M. (1969) Donn6es pr61iminaires concernant l'6tude de l'onde d'expectative chez l'enfant d6ficient visuel. In J. Dargent and M. Dongier, (Eds.), Variations Contingentes N~gatives, Universit6 de Li/~ge, pp. 177-190. Low, M. D. and Swift, S. (1971) The Contingent Negative Variation and the "resting" d.c. potential of the human brain, Neuropsychologia, 9: 203-208. McCallum, C., Walter, W. G., Winter, A., Scotton, L. and Cummins, B. (1970) The contingent Negative Variation in cases of known brain lesion, Electroenceph. clin. Neurophysiol., 28: 210. McCallum, W. C., Papakostopoulos, D., Gombi, R., Winter, A. L., Cooper, R. and Grittith, H. B. (1973) Event-related slow potentials in the human brainstem, Nature (Lond.), 242: 465-467. MacPherson, L. and Kopell, B. S. (1972) A zero-setter and voltage reference unit for EEG amplifiers, Psychophysiology, 9: 262-265. Noebels, J. L. (1977) Electroencephalography of Human Blindness, American Foundation for the Blind Research Report, New York. Novikova, L. A. (1974) Blindness and the Electrical Activity oJ the Brain, American Foundation for the Blind Research Series, No. 23, New York. Rebert, C. and Tecce, J. J. (1973) A summary of Contingent Negative Variation and reaction time, Electroenceph. clin. Neurophysiol., Supplement 33, pp. 173-178. Ryugo, D. K., Ryugo, R., Globus, A. and Killackey, H. P. (1975) Increased spine density in auditory cortex following visual or somatic deafferentation, Brain Res., 90: 143-146. Skinner, J. E. (1971) Abolition of a conditioned, surface-negative, cortical potential during cryogenic blockade of the non-specific thalamo-cortical system, Electroenceph. clin. Neurophysiol., 31 : 197-209. Tecce, J. J. (t972) Contingent Negative Variation (CNV) and psychological processes in man, Psychol. Bull., 77:73 -108. Walter, W. G., Cooper, R., Aldridge, V. J., McCallum, W. C. and Winter, A. L. (1964) Contingent Negative Variation - - An electric sign of sensorimotor association and expectancy in the human brain, Nature (Lond.), 203: 380-384. Weinberg, H. and Papakostopoulos, D., (1975) The frontal Contingent Negative Variation - - Its dissimilarity to CNVs recorded from other sites, Electroenceph. clin. Neurophysiol., 39: 21-28.
C O R T I C A L SLOW P O T E N T I A L S A N D CONGENITAL BLINDNESS
51
THE OCCIPITAL EEG IN
JEFFREY L. NOEBELS*, WALTON T. ROTH and BERT S. KOPELL Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, Calif. 94305 and Veterans Administration Hospital, Palo Alto, Calif. 94305 (U.S.A.)
(Received 20 December, 1977) (Accepted 9 January, 1978)
SUMMARY In a group of 7 congenitally blind adults, electroencephalographic occipital alpha rhythms were absent, but slow negative cortical potentials (CNV) were recorded over the visual cortex and were similar to those of normally sighted controls. Frontal and central CNV amplitudes, but not auditory evoked potentials or reaction times, were decreased significantly in the blind. The results confirm the presence of a developmental E E G abnormality following early blindness, and demonstrate the independance of slow potential generation from rhythmic 8-12 c/sec E E G activity in deafferented visual cortex. Non-sensory specific, event-related potentials such as the Contingent Negative Variation may prove to be useful psychophysiological probes of residual cortical function in brain regions which are deprived of primary sensory input.
INTRODUCTION Two questions are central to an understanding of neurologic and psychiatric conditions underlying developmental sensory deficits in man. What residual function persists in the denervated cortex? To what extent are other brain mechanisms altered in response to selective injury of a sensory system? In the visual system, long-term deficits in sensation produce characteristic changes in the electrophysiology of the brain. Unit recordings of abnormal cellular activity within the visual cortex are
This work was supported by NIMH training grant MH-08304-11 (J.L.N.) and the Veterans Administration. * Present address: Department of Neuroscience, Childrens Hospital, 300 Longwood Avenue, Boston, Mass. 02115, U.S.A. - - to whom all correspondance should be addressed. This work was awarded the 1975 student prize at the Annual Meeting of the Western EEG Society, Seattle, Wash.
52 associated with histologically evident synaptic disorganization, resulting from the absence of tonic retinal influence during critical periods of development (Globus and Scheibel 1967; Hubel and Wiesel 1970; Chow 1973). Clinically, alterations of spontaneous electroencephalographic (EEG) activity in human peripheral blindness have been reported since the earliest studies of the Berger rhythm (Berger 1929; Adrian and Matthews 1934). The most repeated finding in early blindness is a paradoxical absence of the occipital alpha rhythm (8-12 c/sec) normally present in sighted persons when the eyes are closed or deprived of patterned stimuli. The implications of alpha wave absence to theories of altered brain function in blindness have led to considerable clinical and theoretical interest in redefining what is electroencephalographically normal for the visually deafferented brain (see Noebels 1977 for references). Recently, the prospect of a cortical prosthesis (Dobelle, Mladejovsky and Girvin 1974) in blind patients underlines the importance of establishing criteria for estimating the degree of residual function by non-invasive procedures prior to surgical intervention. When blindness is complete, the use of direct visual evoked potential recordings is of little value and alternative methods of examining cortical function are required. The Contingent Negative Variation (CNV) is an event-related slow potential sensitive to levels of attention, arousal, and anxiety which arises between the paired presentation of warnmg and imperative stimuli (Walter, Cooper, Aldridge, McCallum and Winter 1964; Tecce 1972). This cerebral slow potential is unrelated to stimulus modality and is distributed widely over the scalp. In the present study, the CNV paradigm was selected in order to assess the regional integrity of slow potential processes in the deafferented human visual cortex, which is otherwise inaccessible to evoked potential analysis. METHODS
Subjects Subjects were selected on the highly restrictive basis of early retinopathic blindness uncomplicated by other developmental abnormalities, recurrent neurological complaints, or concurrent medication. Seven totally blind college students participated in the study with 7 normally sighted, age- and sex-matched controls. Blindness was diagnosed as due to retrolental fibroplasia (RLF) (5 cases), retinoblastoma (1), and uveitis with glaucoma (1) (see Table 1).
Procedures The EEG and CNV were recorded with platinum pin electrodes (Corby, Roth and Kopell 1974) at midline scalp locations (Oz, Pz, Cz, Fz) referred to linked disc electrodes on the earlobes. Previous work in our laboratory (Corby et al. 1974) has shown that subdermal electrodes significantly reduce cephalic skin potential artifact in CNV recordings. Eye movement was monitored with electrodes located at the mid-pupillary line at the right superior and inferior orbital margins. Bioelectric activity was amplified through calibrated Princeton Applied Research amplifiers
53 TABLE 1 SUMMARY OF RETINOPATHIC DEFICIT IN 7 CONGENITALLY BLIND SUBJECTS Residual activity Subject
Sex
Age
os.
od.
Onset
Etiology
B.S. C.L. K.V. D.R. B.H. T.P. J.S.
M F M M F M M
21 22 22 23 24 26 27
0 0 0 0 0 0 0
0 0 0 light • light a 0 0
birth 6 months 1 month birth birth 3 years birth
RLF retinoblastoma RLF RLF RLF uveitis, glaucoma RLF
Could detect on/off change of a brilliant light source. (input impedance 100 Megohm, bandpass 0.03-100 Hz), reset to baseline prior to each trial through D C level zeroing circuitry (MacPherson and Kopell 1972), and recorded on magnetic tape. Following electrode placement and system calibration, each participant was confortably seated in a darkened, sound-attenuated enclosure and instructed to perform in a reaction time task. Warning tones ($I = 500 Hz at 85dbSL) delivered through earphones were followed 1.5 sec later by imperative tones ($2 = 800 Hz at 85dbSL) to which a pushbutton response (P) was required within 300 msec in order to avoid a penalty hiss. Over a period of 15 min, 50 trials were repeated at intervals varying randomly from 4-8 sec, and were divided into 2 conditions. For the first 25 trials, sighted subjects were asked to fixate on a small lit star in order to minimize eye movement artifact. For the remainder of the experiment, all subjects were asked to close their eyes. The E O G was absent in all but one blind subject, eliminating the problematic requirement of ocular fixation in the blind. With this one subject, as well as with all sighted controls, any trial containing detectable E O G activity was excluded from the average. Thirty-two consecutive four sec epochs of E E G and EOG, 16 with the eyes open and 16 with eyes closed were averaged on a Fabritek computer for each electrode location. Off-line spectral analysis of the EEG, computed with a D E C PDP12 for each electrode in the form of a Compressed Spectral Array (CSA), allowed qualitative inspection of frequency characteristics across subjects. The CNV amplitude was quantified by measurement from the prestimulus baseline (automatically reset to zero prior to each trial) to a point of maximum negativity (averaged over 0.1 sec) during the $1-$2 interval. Auditory evoked potential (EP) amplitudes in responses to $1 were measured peak to peak. RESULTS
Occipital EEG and C N V amplitudes Raw E E G records and the computed CSAs were examined for occipital alpha wave abundance. Figure 1 shows a representative CSA at each of four electrode locations for a typical blind and sighted subject. Occipital alpha rhythm was found
54 BLIND
Fz Fz
Cz
Pz
Oz Oz
0
4
8
1
16
Hz
Fig. 1. Absence of occipital alpha rhythm in blindness, Compressed Spectral Array (CSA) represents
relative abundance of EEG frequency components (3-16 c/see) recorded at 4 scalp locations from a sighted subject with eyes closed (left, J.F.) and a blind subject (right, B.H.). Each horizontal trace beginning with the lowest in each sequence represents consecutive 4 sec EEG epochs. Data from representative blind subject show absence of 8-12 c/see activity at Oz normally prominent in sighted subjects. to be absent in all blind subjects, and regularly present on eye closure in all sighted controls. In 4 of the blind subjects, prominant 8-12 c/see activity was present with a primarily central distribution. CNV averages consisting of 16 trials from the same subjects are shown in Fig. 2. The CNV amplitude in the blind subject (Fig. 2B) is decreased with respect to the normal subject at the Fz and Cz leads (Fig, 2A), but present and nearly equal at Pz and Oz. Similar CNV amplitudes at Oz (13.5, 10, 4, 6, 6.5, 7.5 #V respectively) were obtained in each of the 6 other blind subjects. Figure 3A illustrates the mean CNV amplitudes of the blind group at Oz, Pz, Cz, and Fz, which represented 101.4 ~ , 73.5 ~ , 59.6 ~ , and 63 ~ respectively, of normally sighted amplitudes. The CNV amplitudes were analyzed using a three factor ANOVA with repeated measures on two factors. The blind group had smaller CNVs than the sighted (P < 0.01) and electrode position significantly affected CNV amplitude in all subjects (P < 0.001). An interaction between blindness and electrode position showed, with subsequent lead by lead t-tests, that the blind produced smaller CNVs at the vertex (P < 0.01) and frontal leads (P < 0.05) than did normally sighted controls. At the vertex the effect was present in 6 of the 7 blind subjects. Occipital and parietal differences were not significant.
BLIND
SIGHTED
I CAL ~ S 1 ÷
i i(
S~p )1
1.5
S1 I
i
4 sec.
S2 !
Fig. 2. Contingent Negative Variation (CNV) recorded from four midline locations in a typical sighted and blind subject (same subjects as Fig. 1). Vertical bars in upper traces (Fz) indicate measurement technique. CNV stimulus paradigm described in Methods is shown below. A calibration pulse of 10/~V initiates each averaging trial.
A
B
30• Blind o Sighted 25-
T
20/
joJ
/
' >
%
//
15-
T
I
t0-
5-
i
i
i
i
i
i
~
i
Oz
Pz
Cz
Fz
Oz
Pz
Cz
Fz
Electrode Location
Fig. 3. Mean amplitudes with standard error of the CNV (A), and the auditory EP (N]-P2) to the tone Sl (B) at each of 4 midline recording sites in blind and sighted groups.
56 In a similar analysis of peak to peak N~-P2 auditory EP amplitudes to S~. the means did not differ significantly across groups (Fig. 3B), and only electrode location had a significant effect (P < 0.001). Reaction times to $2 of the two groups did not significantly differ. The blind tended to have shorter reaction times (mean : 192 msec) than the sighted (mean ~- 218 msec). Eye closure had no significant effect on CNV amplitudes, N1-Pz amplitudes, or reaction time in either group. DISCUSSION The principal finding in this study was that early primary deafferentation of the visual cortex does not abolish occipital slow potential production in the adult, despite the absence of occipital alpha rhythm found in each of the blind subjects. A statistical dissociation of CNV amplitudes from background 8-12 c/sec EEG rhythms in normal cortex has been considered previously (Heinneman 1973). However, the conclusion that generation of rhythmic EEG activity and slow potentials depends on separate physiological substrates is clearly supported by the results of the blind subjects, where a congenital lesion, presumably confined to the deeper layers of visual cortex (Chow 1973), disrupted only the alpha rhythm. The presence of an occipital CNV in the blind is anatomically consistant with the electrophysiological evidence that diffuse thalamo-cortical projections mediate cortical slow potential production (Skinner 1971; McCallum, Papakostopoulos, Gombi, Winter, Cooper and Griffith 1973) since fibers in this system terminate primarily in superficial cortical layers where no histological effects of visual deafferentation have been described (Globus and Scheibel 1967; Chow 1973). This interpretation is further supported by the demonstration that patterns of neuronal response generated within different cortical laminae can be selectively abolished (Amassian, Waller and Macy 1964; Calvet, Calvet and Scherrer 1964), and that diffuse destruction of cortex can markedly decrease CNV amplitudes overlying the lesion (McCallum, Walter, Winter, Scotton and Cummins 1970). Thus to the extent to which scalprecorded slow potentials reflect focal cortical activity at the electrode site, the unambiguous presence of' an occipital CNV may prove to be a valid indicator of residual, non-sensory specific function in the visual cortex of the blind.
Extra-occipital changes in the blind Along with the EEG abnormalities attributed to disuse of primary visual pathways in congenital blindness, an additional finding in our subjects with life-long retinal disease was a large decrease of slow potential production over the anterior portion of the scalp to nearly 6 0 ~ of the sighted amplitude, without significant alteration of auditory Nt-Pz EP amplitudes or motor performance. A similar result has been reported in a preliminary study of a subgroup of congenitally blind children (Lairy and Guibal 1969) (but see Cohen 1973). The possibility that the lack of a steady corneoretinal potential might spuriously affect the measured CNV amplitudes is inconsistant with the observed tOl~ography of CNV decrement, which was greater at the vertex than at the frontal location. Fast reaction times in the blind rule out
57 the contribution of psychological variables (such as attention) known to affect both RT and the CNV, even though these processes may be fundamentally independent (Rebert and Tecce 1973). We can only speculate on the significance of this result. High levels of anxiety are known to decrease CNV amplitudes (Low and Swift 1971), and Berger (1935) originally suggested that anxiety may be elevated in the blind. It is possible that habituation to the testing procedure might reduce the difference observed. However, because of the difficulties in estimating the degree of perceived stress during the experimental situation, and the interaction of autonomic (cephalic skin potential) contaminants in the CNV which were controlled for in this study, the contribution of anxiety to frontal CNV differences between blind and sighted subjects remains uncertain. A second problem is that the frontal CNV has been previously distinguished from CNVs recorded at other sites, and may reflect separate psychological variables (Weinberg and Papakostopoulos 1975). Finally, we have no evidence to exclude the possibility that frontal slow potential generation, along with other electrophysiological (Cohen 1970; Kasamatsu 1970; Novikova 1974), morphological (Ryugo, Ryugo, Globus and Killackey 1975), and neurochemical (Bennett, Diamond, Krech and Rosenzweig 1964) processes occurring in areas outside the occipital lobes are sensitive to the normal development of central visual pathways. ACKNOWLEDGEMENTS Provision of CSA software by Dr. Reginald Bickford and Lee Berger is gratefully acknowledged.
REFERENCES Adrian, E. D. and Matthews, B. H. C. (1934) The Berger rhythm - - Potential changes from the occipital lobes in man, Brain, 57: 355-385. Amassian, V. E., Waller, H. J. and Macy, J. (1964) Neural mechanism of the primary somatosensory evoked potential, Ann. N.Y. Acad. Sci., 112: 5-32. Bennett, E., Diamond, M., Krech, D. and Rosenzweig, M. (1964) Chemical and anatomical plasticity of brain, Science, 146: 610-619. Berger, H. (1929) Ober das Elektrenkephalogramm des Menschen, Arch. Psychiat. Nervenkr., 87: 527-570. Berger, H. (1935) Ober das Elektrenkephalogramm des Menschen, Arch. Psychiat. Nervenkr., 103: 444--454. Calvet, J., Calvet, M. C. and Scherrer, J., (1964) Etude stratigraphique corticale de i'activit6 EEG spontan6e, Electroenceph. clin. Neurophysiol., 17: 109-125. Chow, K. L. (1973) Neuronal changes in the visual system following visual deprivation. In: Handbook of Sensory Physiology, VoL 7. Sect. 8, Springer-Verlag, New York, pp. 599-630. Cohen, J. (1970) Brain waves and blindness. In Proceedings of the Conference on New Approaches to the Evaluation of Blind Persons, American Foundation for the Blind Research Bulletin, June,
New York, pp. 112-132. Cohen, J. (1973) Developmental aspects of the Contingent Negative Variation, Electroenceph. clin. NeurophysioL, Supplement 33, pp. 133-137. Corby, J., Roth, W. T. and Kopell, B. S. (1974) Prevalence and methods of control of the cephalic skin potential EEG artifact, Psychophysiology, 11 : 350-360.
58 Dobelle, W. H., Mladejovsky, M. G. and Girvin, J. P. (1974) Artificial vision for the blind - - Electrical stimulation of visual cortex offers hope for a functional prosthesis, Science, 183 : 440-443. Globus, A. and Scheibel, A. B. (1967) The effect of visual deprivation on cortical neurons - - A Golgi study, Exp. Neurol., 19: 331-345. Heinneman, U., (1973) Time variability of slow negative potentials and changes in alpha rhythm. Electroenceph. clin. Neurophysiol., Supplement 33, pp. 125-126. Hubel, D. H. and Wiesel, T. N. (1970) The period of susceptibility to the physiological effects of unilateral eye closure in kittens, J. Physiol. (Lond.), 207: 417-436. Kasamatsu, T. (1970) Effects of visual deafferentation on mesencephalic reticular activity, EaTs. NeuroL, 29: 251-267. Lairy, G. C. and Guibal, M. (1969) Donn6es pr61iminaires concernant l'6tude de l'onde d'expectative chez l'enfant d6ficient visuel. In J. Dargent and M. Dongier, (Eds.), Variations Contingentes N~gatives, Universit6 de Li/~ge, pp. 177-190. Low, M. D. and Swift, S. (1971) The Contingent Negative Variation and the "resting" d.c. potential of the human brain, Neuropsychologia, 9: 203-208. McCallum, C., Walter, W. G., Winter, A., Scotton, L. and Cummins, B. (1970) The contingent Negative Variation in cases of known brain lesion, Electroenceph. clin. Neurophysiol., 28: 210. McCallum, W. C., Papakostopoulos, D., Gombi, R., Winter, A. L., Cooper, R. and Grittith, H. B. (1973) Event-related slow potentials in the human brainstem, Nature (Lond.), 242: 465-467. MacPherson, L. and Kopell, B. S. (1972) A zero-setter and voltage reference unit for EEG amplifiers, Psychophysiology, 9: 262-265. Noebels, J. L. (1977) Electroencephalography of Human Blindness, American Foundation for the Blind Research Report, New York. Novikova, L. A. (1974) Blindness and the Electrical Activity oJ the Brain, American Foundation for the Blind Research Series, No. 23, New York. Rebert, C. and Tecce, J. J. (1973) A summary of Contingent Negative Variation and reaction time, Electroenceph. clin. Neurophysiol., Supplement 33, pp. 173-178. Ryugo, D. K., Ryugo, R., Globus, A. and Killackey, H. P. (1975) Increased spine density in auditory cortex following visual or somatic deafferentation, Brain Res., 90: 143-146. Skinner, J. E. (1971) Abolition of a conditioned, surface-negative, cortical potential during cryogenic blockade of the non-specific thalamo-cortical system, Electroenceph. clin. Neurophysiol., 31 : 197-209. Tecce, J. J. (t972) Contingent Negative Variation (CNV) and psychological processes in man, Psychol. Bull., 77:73 -108. Walter, W. G., Cooper, R., Aldridge, V. J., McCallum, W. C. and Winter, A. L. (1964) Contingent Negative Variation - - An electric sign of sensorimotor association and expectancy in the human brain, Nature (Lond.), 203: 380-384. Weinberg, H. and Papakostopoulos, D., (1975) The frontal Contingent Negative Variation - - Its dissimilarity to CNVs recorded from other sites, Electroenceph. clin. Neurophysiol., 39: 21-28.
