Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Removal of ocular artifact from visual evoked response recordings O’Toole, Dennis Michael 1985

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1985_A8 O86.pdf [ 3.56MB ]
Metadata
JSON: 831-1.0096505.json
JSON-LD: 831-1.0096505-ld.json
RDF/XML (Pretty): 831-1.0096505-rdf.xml
RDF/JSON: 831-1.0096505-rdf.json
Turtle: 831-1.0096505-turtle.txt
N-Triples: 831-1.0096505-rdf-ntriples.txt
Original Record: 831-1.0096505-source.json
Full Text
831-1.0096505-fulltext.txt
Citation
831-1.0096505.ris

Full Text

REMOVAL OF OCULAR ARTIFACT FROM VISUAL EVOKED RESPONSE RECORDINGS By DENNIS MICHAEL 0'TOOLE B.A., M c G i l l U n i v e r s i t y , 1980 THESIS SUBMITTED IN PARTIAL FULFILLMENT THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Psychology) We accept t h i s t h e s i s as conforming to the r e q u i r e d standard-THE UNIVERSITY OF BRITISH COLUMBIA August 1985 Dennis M i c h a e l O'Toole, 1985 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It i s understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 i Table of Contents A b s t r a c t i i Acknowledgements i v L i s t of Tables v L i s t of F i g u r e s v i In t r o d u c t i o n 1 H i s t o r i c a l Overview 1 Overview of Eye Movement A r t i f a c t Research 5 Study R a t i o n a l e 18 Method 28 Subjects 28 Recording and Apparatus 28 Procedure 29 Data A n a l y s i s 31 Res u l t s and D i s c u s s i o n 34 Spontaneous B l i n k s 35 B l i n k s i n Response to L i g h t Flashes 42 E v a l u a t i o n of C o r r e c t i o n Procedures • ••• 50 General D i s c u s s i o n and Summary 71 References 77 Appendix I 80 Appendix II . 82 i i A b s t r a c t P o t e n t i a l s generated by the eye cause unwanted a r t i f a c t in V i s u a l Evoked Response (VER) r e c o r d i n g s . These a r t i f a c t s often contaminate the data in a systematic way that can lead to spurious experimental r e s u l t s . Although i t i s widely agreed that o c u l a r a r t i f a c t must be accounted f o r , the methods used to deal with t h i s problem are v a r i e d . The present study compared four methods used to c o n t r o l o c u l a r a r t i f a c t ; b l i n k r e j e c t i o n , eyes c l o s e d , s u b t r a c t i o n and r e g r e s s i o n . Twenty normal, female sub j e c t s were t e s t e d twice w i t h i n the same s e s s i o n . Subjects watched l i g h t f l a s h e s of 4 i n t e n s i t i e s ; 2, 30, 80, and 240 f t lamberts. The l i g h t s were presented at 1 h e r t z , reached maximum brig h t n e s s i n 0.5 msec and l a s t e d for 0.5 sec. During t e s t i n g the VER, and e l e c t r o o c c u l o g r a p h i c (EOG) response generated by a b l i n k , were recorded. In the b l i n k r e j e c t i o n method, any VER epoch that contained b l i n k a r t i f a c t was excluded from the average. The eyes c l o s e d method c o n s i s t e d of having su b j e c t s watch the s t i m u l i through c l o s e d e y e l i d s . The s u b t r a c t i o n method c o r r e c t s b l i n k a r t i f a c t by d i g i t a l l y s u b t r a c t i n g the averaged EOG from the EEG. The p r o p o r t i o n of EOG su b t r a c t e d was determined by the EEG/EOG r a t i o estimated while s u b j e c t s b l i n k e d spontaneously i n a darkened environment. The r e g r e s s i o n method determines what p r o p o r t i o n of EOG i s to be subtra c t e d on the b a s i s of the c o r r e l a t i o n between EOG and EEG w i t h i n VER epochs. Two c o r r e c t i o n , f a c t o r s are c a l c u l a t e d , one to c o r r e c t f o r v e r t i c a l movements and one to c o r r e c t f o r h o r i z o n t a l movements. The b l i n k r e j e c t i o n method was found to be u s e f u l with s u b j e c t s who had 40% or more b l i n k - f r e e epochs, but was an u n r e l i a b l e method f o r the m a j o r i t y of s u b j e c t s . The eyes c l o s e d method was a l s o found to produce poor VER data. The e y e l i d s appear to attenuate the l i g h t reaching the r e t i n a and there may be e y e b a l l movement d e s p i t e having the eyes c l o s e d . Both the s u b t r a c t i o n and r e g r e s s i o n methods s u b s t a n t i a l l y reduced the o c u l a r a r t i f a c t . H o r i z o n t a l eye movements do not appear to be a s i g n i f i c a n t problem over the short i n t e r v a l s of VER r e c o r d i n g because the r e g r e s s i o n method was not super i o r to the s u b t r a c t i o n method i n removing a r t i f a c t . Although the s u b t r a c t i o n and r e g r e s s i o n methods e f f e c t i v e l y reduce o c u l a r a r t i f a c t , both are l e s s e f f e c t i v e at p o s t e r i o r e l e c t r o d e placements. The reason f o r t h i s may be that o c u l a r p o t e n t i a l i s not propagated across the s c a l p i n a l i n e a r f a s h i o n , as of t e n assumed. Using spontaneously generated b l i n k s i n a darkened environment, i t was found that the o c u l a r p o t e n t i a l waveform changes shape as i t moves towards the back of the head. i v Acknowledgements I would l i k e to thank my t h e s i s committee f o r t h e i r guidance and suggestions: Dr. W i l l i a m Iacono ( S u p e r v i s o r ) , Dr. Robert Hare, and Dr. Demetrios Papageorgis. I would a l s o ' l i k e to thank Michael S a t t e r f i e l d f o r h i s programming work and h e l p f u l a d v i c e . V L i s t of Tables page Table 1 Mean Amplitude of Spontaneous B l i n k s 37 Table 2 Mean C o r r e l a t i o n C o e f f i c i e n t s Between EEG and EOG 39 Table 3 Mean Times of Peak Latency 39 Table 4 Spontaneous B l i n k Corrected EEG; Mean RMS De v i a t i o n s 43 Table 5 Range and Mean Number of B l i n k s to L i g h t Flashes . 43 Table 6 Number of B l i n k s ; C o r r e l a t i o n Between T r i a l 1 and 3. . . * 46 Table 7 B l i n k Amplitude; C o r r e l a t i o n Between T r i a l 1 and 3 . 46 Table 8 Mean Bl i n k Amplitude ( T r i a l s 1 and 2) 48 Table 9 Mean Peak L a t e n c i e s of B l i n k s 50 Table 10 Mean Onset L a t e n c i e s of B l i n k s 50 Table 11 P 2 Amplitudes C o r r e l a t e d with EOG Amplitude ( T r i a l 1) 52 Table 12 P 2 Amplitudes C o r r e l a t e d with EOG Amplitude ( T r i a l 2) 52 Table 13 P 2 Amplitudes - T r i a l 1 c o r r e l a t e d with T r i a l 2 54 Table 14 C o r r e l a t i o n s Between Methods ( P 2 Amplitudes averaged across a l l i n t e n s i t i e s ) . . 57 Table 15 C o r r e l a t i o n s Between Methods (P 2 Amplitudes at I n t e n s i t y 3 only) 59 v i L i s t of F i g u r e s page Fig u r e 1 F i g u r e 2 Fig u r e 3 Fi g u r e 4 Fi g u r e 5 Fig u r e 6 Fi g u r e 7 Fi g u r e 8 Spontaneous b l i n k p o t e n t i a l recorded simultaneously at the eyes (EOG) and at three EEG placements; Fz, Cz, and Oz 41 Number of b l i n k s d i s t r i b u t e d at each l i g h t i n t e n s i t y 45 V i s u a l evoked response amplitude by l i g h t i n t e n s i t y , using d i f f e r e n t methods to reduce b l i n k a r t i f a c t 62 V i s u a l evoked response amplitude by l i g h t i n t e n s i t y , recorded at each of three e l e c t r o d e placements 64 V i s u a l evoked response amplitude by e l e c t r o d e placement, average of a l l methods 65 V i s u a l evoked response amplitude by e l e c t r o d e placement using d i f f e r e n t methods to reduce b l i n k a r t i f a c t 66 V i s u a l evoked response amplitude by method of a r t i f a c t c o n t r o l , average of a l l l i g h t i n t e n s i t i e s 68 The e f f e c t s of 4 methods of d e a l i n g with b l i n k a r t i f a c t c o n t r a s t e d with the uncorrected v i s u a l evoked response 69 1 I n t r o d u c t i o n H i s t o r i c a l Overview Since the s t u d i e s of Buchsbaum and Silverman (1968) and S p i l k e r and Callaway (1969), a number of i n v e s t i g a t i o n s have been c a r r i e d out in an area of stimulus i n t e n s i t y c o n t r o l known as augmenting/reducing. Augmenting/reducing r e f e r s to a paradigm in which the magnitude of a s u b j e c t ' s responses to s t i m u l i of varying i n t e n s i t y are examined. In a t y p i c a l experiment, c o r t i c a l responses to l i g h t f l a s h e s of four d i f f e r e n t i n t e n s i t i e s ( v i s u a l evoked responses) are measured. Those subjects who show an increase i n v i s u a l evoked response (VER) amplitude as l i g h t i n t e n s i t y i n c r e a s e s are augmenters, and those showing a decrease i n response amplitude are reducers. Buchsbaum and Silverman (1968) reported a tendency f o r i n d i v i d u a l s who show a reduction of experienced stimulus i n t e n s i t y on P e t r i e ' s K i n e s t h e t i c F i g u r a l A f t e r e f f e c t s (KFA) per c e p t u a l task to show a comparable tendency on a VER procedure. The p e r c e p t u a l task r e q u i r e d a b l i n d f o l d e d subject to manually estimate the width of a c o n t r o l bar using a comparison bar of graduated width. Results showed that s u b j e c t s who tended to underestimate the s i z e of the c o n t r o l bar a l s o showed reduced amplitude of c o r t i c a l responses evoked by l i g h t f l a s h e s of i n c r e a s i n g i n t e n s i t y . S p i l k e r and Callaway (1969) confirmed the f i n d i n g s of Buchsbaum and Silverman (1968) by examining the change in VER 2 amplitude to changes in sine-wave modulated l i g h t using s u b j e c t s who had a l s o completed the KFA task. They found a rank order c o r r e l a t i o n of .66 (p<.0l) between the slope of the curve r e l a t i n g VER amplitude to s t i m u l u s i n t e n s i t y and KFA performance. The greater the underestimation of c o n t r o l bar width, the smaller the VER s l o p e . They concluded that t h i s phenomenon, r e f e r r e d to as " p a r a d o x i c a l d i m i n u i t i o n " by Kamphiusen and van Leeuwen (1968), was r e a l l y an example of "reducing" i n the newly named augmenting/reducing paradigm. Since these f i r s t s t u d i e s , VER research has become i n c r e a s i n g l y r e f i n e d and i t i s now g e n e r a l l y accepted that "augmenters" are i n d i v i d u a l s who show an i n c r e a s e i n amplitude in the e a r l y (0-150 ms) components of the VER and "reducers" are more l o o s e l y d e f i n e d as those who show a slower rate of increase, a l e v e l l i n g o f f , or a decrease in VER amplitude as stimulus i n t e n s i t y i n c r e a s e s . F u r t h e r , i t i s hypothesized (Buchsbaum, 1976) that the d i f f e r e n t p a t t e r n s of response r e f l e c t a c e n t r a l nervous system mechanism f o r modulating sensory input. Using the augmenting/reducing paradigm, i n d i v i d u a l d i f f e r e n c e s have been a s s o c i a t e d with psychopathology. Landau, Buchsbaum, Carpenter, Strauss and Sacks (1975), using a group of 1 9 acute, m e d i c a t i o n - f r e e s c h i z o p h r e n i c p a t i e n t s , showed that these p a t i e n t s produced s m a l l e r VER amplitudes and e i t h e r a decrease or no change i n amplitude at i n c r e a s i n g l i g h t i n t e n s i t i e s . Normal s u b j e c t s and s c h i z o p h r e n i c p a t i e n t s were d i s c r i m i n a t e d with 71% accuracy using VER v a r i a b l e s . A l s o , 3 p a t i e n t s who showed VER reducing tended to have r e l a t i v e l y good premorbid h i s t o r i e s and were found to make greater c l i n i c a l improvement. Asarnow, Cromwell, and Rennick (1978) reported r e s u l t s c o n s i s t e n t with the f i n d i n g s of Landau et a l . (1975). They found that a group of 12 s c h i z o p h r e n i c p a t i e n t s , who were l e s s c h r o n i c and who had a b e t t e r premorbid adjustment than a comparison group of s i m i l a r s i z e , showed a VER reducing tendency. Psychotic d e p r e s s i v e p a t i e n t s were shown to have greater v a r i a b i l i t y and g r e a t e r augmenting tendencies on both VER and KFA scores. T h i s f i n d i n g c orroborates the f i n d i n g of Buchsbaum and Silverman (1968) that there i s an a s s o c i a t i o n between KFA and VER. In a l a t e r study Rappaport, Hopkins, B e l l e z a , and H a l l (1975) found that a group of 123 male sc h i z o p h r e n i c p a t i e n t s (most with a good premorbid h i s t o r y ) had smaller amplitude VERs and greater v a r i a b i l i t y of VER p a t t e r n s that a normal c o n t r o l group. The authors a l s o looked at the e f f e c t s of medication on VER and found no s i g n i f i c a n t e f f e c t on e i t h e r v a r i a b i l i t y or amplitude, suggesting that VER may r e f l e c t a s t a b l e t r a i t that i s u n a f f e c t e d by medication. Other b e h a v i o r a l problems a s s o c i a t e d with i n d i v i d u a l VER d i f f e r e n c e s are s u i c i d e r i s k (Buchsbaum, Haier, & Murphy 1977), a l c o h o l i s m (Martin, Becker, S> B u f f i n g t o n , 1979), chronic pain (von Knorring, Almay, Johansson, & Terenius, 1979) and tobacco smoking (Knott & Venables, 1978). Va r i o u s measures of augmenting/reducing have a l s o been demonstrated to have moderate to high r e t e s t r e l i a b i l i t y . Soskis and Shagass (1974), addressed the question of VER s t a b i l i t y . 4 They showed a short-term (over the same session) r e l i a b i l i t y of 0.92 f o r mean amplitude and 0.66 for slope of VER recorded at the vertex e l e c t r o d e s i t e . T h e i r long term (over a p e r i o d of 16 weeks) r e l i a b i l i t y c o e f f i c i e n t s were not as good, but methodological problems make the two measures d i f f i c u l t to compare. Buchsbaum and Pfefferbaum (1971) found t e s t / r e t e s t r e l i a b i l i t i e s of 0.70 and 0.67, r e s p e c t i v e l y , for normals and p s y c h i a t r i c p a t i e n t s examined at i n t e r v a l s of one to two months. Stark and Norton (1974) reported data that suggested amplitude i s a moderately r e l i a b l e measure of VER but that slope of the VER was c l e a r l y the most r e l i a b l e parameter. In c o n t r a s t to Stark and Norton's (1974) c o n c l u s i o n , Iacono, Gabbay and Lykken (1982) showed that amplitude and slope measures were r e l i a b l e but demonstrated that the slope measure may not be a good d e s c r i p t o r of the a m p l i t u d e / l i g h t i n t e n s i t y f u n c t i o n . In a d d i t i o n to r e t e s t r e l i a b i l i t y , augmenting/reducing measures have been shown to have high h e r i t a b i l i t y . Buchsbaum (1974) showed that VERs were very s i m i l a r between monozygotic twins but showed comparatively l i t t l e s i m i l a r i t y between d i z y g o t i c twins. B u f f i n g t o n , M a r t i n , and Becker (1981) found that VER waveforms and slopes were more s i m i l a r between nontwin r e l a t i v e s than between randomly p a i r e d s u b j e c t s . Taken together the above s t u d i e s i n d i c a t e that the augmenting/reducing paradigm has promise as a r e l i a b l e p h y s i o l o g i c a l measure, but work s t i l l must be done to r e f i n e measurement techniques. The p o s s i b i l i t y of demonstrating a r e l a t i o n s h i p between s t a b l e p h y s i o l o g i c a l responses and 5 b e h a v i o r a l problems j u s t i f i e s f u r t h e r i n v e s t i g a t i o n of augmenting/reducing phenomena, p a r t i c u l a r l y in the area of v i s u a l evoked responses. However, before i n v e s t i g a t i o n of responses to v i s u a l s t i m u l i can proceed, methodological problems in VER r e c o r d i n g must be addressed. A s e r i o u s problem that has not been given s u f f i c i e n t c o n s i d e r a t i o n in augmenting/reducing l i t e r a t u r e i s that of eye b l i n k contamination of the v i s u a l evoked response. The purpose of t h i s study i s to examine various methods for d e a l i n g with b l i n k a r t i f a c t s in VER recordings and to i n v e s t i g a t e the v a l i d i t y of procedures that d i s r e g a r d eye b l i n k contamination. Overview of Eye Movement A r t i f a c t Research E a r l y s t u d i e s of evoked p o t e n t i a l s e i t h e r ignored the e f f e c t s of eye b l i n k s or r e j e c t e d any epoch that contained b l i n k a r t i f a c t . The problems a s s o c i a t e d with both these approaches w i l l be d i s c u s s e d l a t e r . H i l l y a r d and Galambos (1970) were the f i r s t to address the q u a n t i t a t i v e extent of eye movement a r t i f a c t i n evoked p o t e n t i a l r e c o r d i n g s . T h e i r study was concerned with a r t i f a c t p o t e n t i a l s generated by eye r o t a t i o n through the c o r n e o r e t i n a l d i p o l e . The cornea i s e l e c t r i c a l l y p o s i t i v e r e l a t i v e to the r e t i n a , making the e y e b a l l i n e f f e c t a miniature b a t t e r y with a p o s i t i v e and a negative p o l e . The o r i e n t a t i o n of t h i s c o r n e o r e t i n a l p o t e n t i a l s h i f t s as the eyes change p o s i t i o n r e s u l t i n g i n a change of s c a l p p o t e n t i a l . When a person b l i n k s , the e y e l i d moving over the surface of the e y e b a l l changes the c o r n e o r e t i n a l p o t e n t i a l . Muscle p o t e n t i a l s generated 6 by eye l i d and cheek movements, and r e f l e x i v e upward r o t a t i o n of the e y e b a l l add to c o r n e o r e t i n a l p o t e n t i a l change and make up the p o t e n t i a l conducted across the s c a l p . In t h e i r 1970 study, H i l l y a r d and Galambos recorded a slow, negative p o t e n t i a l s h i f t , known as contingent negative v a r i a t i o n (CNV), from the s c a l p of 10 normal a d u l t s u b j e c t s . The CNV was generated d u r i n g the preparatory i n t e r v a l between a warning c l i c k and a l e v e r press s i g n a l l e d by a tone. Eye movement p o t e n t i a l s were recorded with e l e c t r o d e s p l a c e d above and below the eye and on the outer c a n t h i . EEG was recorded from the vertex (Cz) as w e l l as a s i t e 4 cm. a n t e r i o r to Cz, using the r i g h t mastoid as a r e f e r e n c e . To e s t a b l i s h the r e l a t i o n s h i p between the amplitude of the EEG eye a r t i f a c t p o t e n t i a l and the amplitude of the EOG, each subject was d i r e c t e d to move h i s or her eyes through v a r i o u s degrees of v i s u a l a r c , both h o r i z o n t a l l y and v e r t i c a l l y . The angles of v i s u a l arc were chosen so that the r e s u l t i n g EOG d e f l e c t i o n s would extend over the range expected during the CNV t r i a l s . The p o t e n t i a l s h i f t s of both EOG and EEG channels were d i g i t i z e d and averaged i n blocks of e i g h t consecutive t r i a l s . The CNV was c h a r a c t e r i z e d as having two components; the " t r u e " CNV of c e r e b r a l o r i g i n and the eye a r t i f a c t p o t e n t i a l (EAP). The EAP amplitudes were c a l c u l a t e d from concurrent r e c o r d i n g s of EOG a f t e r e s t i m a t i n g the r e l a t i o n s h i p between EAP and EOG during the voluntary eye movements through p r e s c r i b e d angles of v i s u a l a r c . R e s u l t s showed that, in the average subject, 23% of the t o t a l CNV c o n s i s t e d of EAP. T h i s estimate was supported by comparing the c o r r e c t e d CNV to " t r u e " CNV recorded with eyes immobilized 7 by f i x a t i o n . Once H i l l y a r d and Galambos (1970) e s t a b l i s h e d the r e l a t i o n s h i p between EOG and EEG a r t i f a c t they c o r r e c t e d i t by s u b t r a c t i n g a constant, subject s p e c i f i c percentage of the EOG from the EEG. Corby and K o p e l l (1972) extended the f i n d i n g s of H i l l y a r d and Galambos (1970) by comparing b l i n k - g e n e r a t e d a r t i f a c t s to v e r t i c a l eye movement a r t i f a c t s i n an e f f o r t to see i f the two a r t i f a c t f i e l d s were s i m i l a r . In t h i s study EOG e l e c t r o d e s were placed above and below the r i g h t eye and EEG was recorded from the vertex vs l i n k e d mastoids. H o r i z o n t a l eye movement was not recorded s i n c e the authors f e l t i t c o n t r i b u t e d minimally to EEG a r t i f a c t at the vertex placement. The s u b j e c t s were asked to move t h e i r eyes through s p e c i f i e d a r c s by f o l l o w i n g two black l i n e s on the w a l l . Whereas H i l l y a r d and Galambos (1970) looked at v i s u a l a r c s up to 15°, Corby and K o p e l l (1972) i n v e s t i g a t e d a r t i f a c t c o n t r i b u t i o n f o r arcs up to 60°. A l s o , the subject was asked to b l i n k "normally, without e x c e s s i v e e f f o r t " when the i n s t r u c t i o n " b l i n k " was heard over an intercom. Concurrent time i n t e r v a l s of EEG and EOG were used to c a l c u l a t e the shape and amplitude of the "t r u e " EEG. The a l g o r i t h m used was: EEG t = EEG 0 - [EOG X A/B] where: EEG t= " t r u e " EEG, f r e e from a r t i f a c t EEG Q= observed EEG A = v i s u a l l y estimated average peak to trough amplitude of EEG r e c o r d d u r i n g v o l u n t a r y eye movements. 8 B = analogous amplitude in EOG Using t h i s a l g o r i t h m the authors demonstrated that c o n t r i b u t i o n of v e r t i c a l eye movements to EEG remained constant across the v i s u a l arcs 10°.to 60°. However, the mean eye b l i n k c o n t r i b u t i o n to vertex EEG of 9.3% was s i g n i f i c a n t l y l e s s that the mean v e r t i c a l eye movement c o n t r i b u t i o n of 13.9%. In l i g h t of t h i s f i n d i n g the authors suggested that b l i n k s and v e r t i c a l eye movements have to be d e a l t with s e p a r a t e l y s i n c e they do not c o n t r i b u t e e q u a l l y to the s c a l p a r t i f a c t . Support for Corby and K o p e l l ' s (1972) f i n d i n g s was provided by G i r t o n and Kamiya (1973). T h e i r approach to removal of eye movement a r t i f a c t s was an o n - l i n e technique that e l e c t r o n i c a l l y s u b t r a c t e d a f r a c t i o n of both h o r i z o n t a l and v e r t i c a l EOG p o t e n t i a l s from the raw EEG s i g n a l . EOG was recorded from e l e c t r o d e s above and below the eye as w e l l as e l e c t r o d e s on the outer c a n t h i . The EEG was d e r i v e d from e l e c t r o d e s at the vertex vs. r i g h t mastoid. Each EOG s i g n a l was a m p l i f i e d by a f a c t o r of 200 and t h i s output was connected to a 10-turn p r e c i s i o n potentiometer. Both of the EOG s i g n a l s and the a m p l i f i e d EEG s i g n a l were connected to a d i f f e r e n t i a l a m p l i f i e r , a l l o w i n g f o r s u b t r a c t i o n of a f r a c t i o n of each EOG s i g n a l from the raw EEG s i g n a l without a l t e r i n g the c o n t r i b u t i o n of the EEG e l e c t r o d e . To determine the potentiometer s e t t i n g s , s u b j e c t s were asked to make a s e r i e s of eye movements v e r t i c a l l y and h o r i z o n t a l l y . The potentiometers were adju s t e d so that upon v i s u a l i n s p e c t i o n there was minimal a r t i f a c t in the EEG t r a c e . The r e s u l t s support the concept that there i s a s u f f i c i e n t l y l i n e a r r e l a t i o n s h i p 9 between the EOG and the vertex-mastoid a r t i f a c t to j u s t i f y using a s u b t r a c t i o n technique. However, as i n the Corby and K o p e l l (1972) study the method d i d not work e q u a l l y w e l l f o r v e r t i c a l eye movements and b l i n k s . Although eye movement a r t i f a c t was removed for v i r t u a l l y a l l s u b j e c t s , b l i n k a r t i f a c t s were only p a r t i a l l y removed i n some s u b j e c t s . T h i s f i n d i n g seemed to r e f l e c t i n t e r i n d i v i d u a l d i f f e r e n c e s s i n c e the b l i n k a r t i f a c t in many su b j e c t s appeared to be completely removed. Roth, Ford, and K o p e l l (1978) used an o f f - l i n e s u b t r a c t i o n technique s i m i l a r in concept to the o n - l i n e technique of G i r t o n and Kamiya and noted the importance of c a l c u l a t i n g a separate c o r r e c t i o n c o e f f i c i e n t for b l i n k s and v e r t i c a l eye movements. However, d e t a i l s of t h e i r methodology were not given and success of a r t i f a c t removal i s not di s c u s s e d because the focus of the paper i s not a r t i f a c t c o r r e c t i o n . In an attempt to attenuate both eye movement and eye b l i n k a r t i f a c t s , Whitton, Lue, and Moldofsky (1978) reported a technique that compensated for eye a r t i f a c t s by a n a l y z i n g the frequency composition of the EEG and EOG s i g n a l . The EEG a s s o c i a t e d with voluntary eye movements, b l i n k s , and i n v o l u n t a r y random eye movements, was recorded from the vertex v s . l i n k e d e a r l o b e s . EOG was simultaneously recorded but p o s i t i o n s of the e l e c t r o d e s were not given. The data were stored on analog tape and d i g i t i z e d o f f - l i n e . The frequency spectrum f o r both the EEG and EOG was determined using the F o u r i e r A n a l y s i s . The EEG and v e r t i c a l EOG sp e c t r a were then l i n e a r l y s c a l e d so that the low frequency peaks were equal in amplitude. This s c a l i n g s t e p 10 r e q u i r e s the assumption that the low frequencies in the EEG are due to eye a r t i f a c t and are t h e r e f o r e e q u i v a l e n t to the EOG low fr e q u e n c i e s . The two sp e c t r a were then s u b t r a c t e d . Because s u b t r a c t i o n of s p e c t r a l frequencies i s the same as f i l t e r i n g the raw EEG s i g n a l , the net e f f e c t was to f i l t e r the EEG of a r t i f a c t s , the p r o p e r t i e s of the f i l t e r being determined, by the EOG. In s p e c t i o n of the c o r r e c t e d EEG spectrum showed no eye movement a r t i f a c t . At the same time the authors a p p l i e d the e l e c t r o n i c s u b t r a c t i o n method of G i r t o n and Kamiya (1973) to attenuate the eye a r t i f a c t s in the EEG of some s u b j e c t s . S p e c t r a l a n a l y s i s of t h i s c o r r e c t e d EEG showed a prominent low frequency peak. The authors concluded that t h i s peak was a remnant of the eye movement a r t i f a c t remaining a f t e r c o r r e c t i o n . The c o n c l u s i o n of the authors in regard to t h e i r c o r r e c t i o n technique was that the dominant peaks of the s c a l p EEG c o u l d be recovered from eye movement contaminated data, and that the. technique i s u s e f u l when power spectra are to be c a l c u l a t e d . A r e s e r v a t i o n the authors make about t h i s method i s that s t a b l e s p e c t r a are r e q u i r e d to obtain c o n s i s t e n t r e s u l t s . For short time s e r i e s ( i n t h i s study 4 seconds was co n s i d e r e d short) r e s o l u t i o n of the high frequency content i s poor, making t h i s an unacceptable method at time i n t e r v a l s l e s s than 4 seconds. Since the d u r a t i o n of a b l i n k i s g e n e r a l l y a matter of m i l l i s e c o n d s t h i s method would not be an acceptable way to c o r r e c t f o r b l i n k a r t i f a c t . R e a l i z i n g the importance of b l i n k a r t i f a c t s in v i s u a l evoked response r e c o r d i n g s , Iacono et a l . (1982) addressed that 11 s p e c i f i c problem as an issue separate from slow eye movements. In a study which used 20 s u b j e c t s , v i s u a l s t i m u l i c o n s i s t i n g of 10 microsecond pulses of l i g h t at four i n t e n s i t i e s were presented i n a pseudo-random order. During s t i m u l i p r e s e n t a t i o n , EEG r e c o r d i n g s were made from above and below the r i g h t eye. For each subject the gain of the EOG was adjusted such that the amplitude of a n a t u r a l l y emitted b l i n k c l o s e l y approximated (by v i s u a l i n s p e c t i o n ) the a r t i f a c t generated in the EEG channel. Both EEG and EOG were recorded o n - l i n e with a PDP-12 computer which a l s o c o n t r o l l e d p r e s e n t a t i o n of the s t i m u l i . Subjects were t e s t e d in t h i s manner twice over a one week p e r i o d . To c o r r e c t the EEG record f o r b l i n k a r t i f a c t s , a p o i n t by p o i n t s u b t r a c t i o n of the averaged EOG was made from the average of the evoked responses. On the b a s i s of t h e i r r e s u l t s the authors concluded that eye b l i n k s exert a strong, and r e t e s t r e l i a b l e , a r t i f a c t on average evoked responses (AER). The b l i n k p o t e n t i a l c o n t r i b u t e d a l a r g e a r t i f a c t to both the shape and the p a t t e r n of AER amplitude across i n t e n s i t i e s of l i g h t . It was found that a b l i n k p o t e n t i a l begins w i t h i n 100 ms and peaks about 150 ms a f t e r stimulus p r e s e n t a t i o n . T h i s means the b l i n k occurs e a r l y enough to enhance the P, component of the AER and, more importantly, i t reduces the magnitude of the N, component of the AER. The net e f f e c t of P, enhancement and the l a r g e N, r e d u c t i o n i s a decrease in the P,- N, d i f f e r e n c e ( P i - N, being the dependent measure o f t e n used in the v i s u a l augmenting/reducing paradigm). Because the amplitude of the b l i n k p o t e n t i a l i n c r e a s e s with l i g h t i n t e n s i t y , and b l i n k s 1 2 d i m i n i s h the P^- N, d i f f e r e n c e , the AER measurements in uncorrected data w i l l be p r o p o r t i o n a l l y smaller as l i g h t i n t e n s i t y i n c r e a s e s . In experimental terms t h i s means that the more a person b l i n k s the more he or she w i l l look l i k e a reducer, r e g a r d l e s s of c o r t i c a l responses. A l l the s t u d i e s mentioned above are based on a conceptual model that can be expressed as f o l l o w s : EEG o b = [j3 x EOG] + EEG af where: EEG o b = EEG observed EEGaf = EEG a r t i f a c t f r e e 0 = the p r o p o r t i o n of EOG p o t e n t i a l t r a n s m i t t e d to the EEG e l e c t r o d e . The value of (5 i n the above time domain s t u d i e s was estimated i n two ways: 1. Corby and K o p e l l (1972) and H i l l y a r d and Galambos (1970) estimated the r a t i o between EOG and EEG amplitude in c a l i b r a t i o n t r i a l s where subjects moved t h e i r eyes in p r e s c r i b e d v i s u a l a r c s . The a r t i f a c t produced in the EEG channel was assumed to c o n t a i n l i t t l e or no evoked c o r t i c a l p o t e n t i a l and t h e r e f o r e to be a measure of the EOG p o t e n t i a l t r a n s m i t t e d to the EEG e l e c t r o d e . The r a t i o c a l c u l a t e d i n these c a l i b r a t i o n t r i a l s was used to c o r r e c t the EEG channel in experimental t r i a l s . 2. G i r t o n and Kamiya (1975) and Iacono et a l . (1982) 1 3 estimated /3 d i r e c t l y by a d j u s t i n g the amplitude of EOG so that s u b t r a c t i o n o n - l i n e or o f f - l i n e l e d to a EEG t r a c e that had no v i s i b l e remnants of a r t i f a c t . A formal e x p r e s s i o n of these approaches shows that they EOG and a r t i f a c t f r e e EEG, |cov (EOG,EEG - 0EOG)| , approaches 0. The same r e s u l t can be accomplished more a c c u r a t e l y without having to d i s p l a y data by c a l c u l a t i n g the r e g r e s s i o n c o e f f i c i e n t of EOG on the observed EEG (V e r g l e r , Gasser, & Mocks 1982) : where e = EEG observed o = EOG S = standard d e v i a t i o n then r e o i s the product-moment c o r r e l a t i o n of EEG observed with EOG, and S e/S 0 i s a c o r r e c t i o n f a c t o r f o r nonequivalent standard d e v i a t i o n s . To s i m p l i f y computation, 0 can be expressed as f o l l o w s : both estimate j3 such that the absolute value of covariance of 0 = r, e o x S e/S o then s i n c e r, e o where S„ = covariance of EEG and EOG. 1 4 In EEG epochs of o c u l a r a r t i f a c t where 0 i s estimated, the a r t i f a c t f r e e EEG i s considered to be a random v a r i a b l e with an expected value of 0. Any unsystematic noise should t h e r e f o r e disappear with averaging, l e a v i n g a good estimate of the EOG p o t e n t i a l t r a n s m i t t e d to the EEG e l e c t r o d e . I t can be shown that j3 as determined by e i t h e r s u b t r a c t i o n methods or by the more formal c o v a r i a n c e approach, provides an estimate f o r the same q u a n t i t y , (see V e r g l e r et a l . , 1982, p. 473). V e r g l e r et a l . (1982) used the above mathematical model to c o r r e c t the EEG t r a c e of 67 s u b j e c t s i n a study of CNV. E i g h t EEG channels (F„,F 3,C„,C 3,C 2,Pz,0 2,0, vs. l i n k e d earlobes) were used to r e c o r d CNV p o t e n t i a l s and EOG was recorded from above and below the, r i g h t eye. Subjects were engaged i n two t a s k s . F i r s t , they were presented with 20 consecutive p a i r s of p i c t u r e s which they had to judge as same or d i f f e r e n t . Secondly, 16 d i f f e r e n t drawings d e p i c t i n g f i s h were to be c l a s s i f i e d by means of a r u l e which the subject had to d i s c o v e r . R e l i a b i l i t y of the c o r r e c t i o n procedure was estimated by determining 0 for both sets of data and then computing, f o r the two t a s k s , a rank order c o r r e l a t i o n of the 0 v a l u e s . V a l i d i t y of the c o r r e c t i o n procedure was evaluated i n the f i r s t task in three ways: 1. c o r r e c t e d data were expected to have face v a l i d i t y , meaning v i s u a l i n s p e c t i o n of the c o r r e c t e d EEG t r a c e should show no EOG a r t i f a c t . 2. the c o r r e c t i o n procedure should s i g n i f i c a n t l y decrease data l o s s that would otherwise occur due to r e j e c t i o n of 1 5 contaminated t r i a l s . Unless the number of contaminated t r i a l s saved i s s u b s t a n t i a l , use of a c o r r e c t i o n procedure may not be worthwhile. 3. the product-moment c o r r e l a t i o n between c o r r e c t e d CNV records and the EOG should be low; that i s , there should be l i t t l e s i m i l a r i t y in the two r e c o r d i n g s . In examining r e l i a b i l i t y , V e r g l e r et a l . (1982) found rank order c o r r e l a t i o n s ranging from 0.11 to 0.58 across the two t a s k s . Although the /3 values were moderately s t a b l e they were not i n agreement with those found i n other s t u d i e s . The Cz 0 value (0.098) found i n t h e i r study was s i m i l a r to the value (0.093) found by Corby and K o p e l l (1972) for b l i n k s , but was below the estimates found i n other s t u d i e s f o r v e r t i c a l eye movements (0.14, Corby & K o p e l l , 1972; 0.10 - 0.15, G i r t o n & Kamiya, 1973). The f a c t that there were d i f f e r e n c e s between /3 values at Cz i s not s u r p r i s i n g i n l i g h t of the Corby and K o p e l l (1972) evidence that b l i n k s and v e r t i c a l eye movements d i f f e r e n t i a l l y contaminate the EEG. V e r g l e r et a l . (1982) estimated a s i n g l e 0 value f o r both b l i n k and v e r t i c a l eye movements and compromised t h e i r c a l c u l a t i o n s by f a i l i n g to d i s c r i m i n a t e between the two. In terms of data l o s s , V e r g l e r et a l . (1982) found a s i g n i f i c a n t savings of 53% - 65% of contaminated epochs that would otherwise have been r e j e c t e d . The i n t e r p r e t a t i o n of product-moment c o r r e l a t i o n s between the c o r r e c t e d EEG and the EOG was not c l e a r c u t . For the 8 e l e c t r o d e leads, 5 s i g n i f i c a n t 1 6 c o r r e l a t i o n s were found. However, 3 of these 5 were in the p o s t e r i o r p o s i t i o n s (P 2,0 2,0,) and were neg a t i v e . The 3 c o r r e l a t i o n s that were not s i g n i f i c a n t , i n d i c a t i n g s u c c e s s f u l a t t e n u a t i o n of EOG a r t i f a c t , were a n t e r i o r ( F 4 , C 3 and C 2 ) . The face v a l i d i t y of the procedure was a l s o u n c l e a r . The authors presented a graphic i l l u s t r a t i o n of one f r e q u e n t l y b l i n k i n g subject and l i m i t e d these data to four s e l e c t e d e l e c t r o d e placements. Because they d i d not comment on the data of other s u b j e c t s , the question of face v a l i d i t y remains unanswered. Gratton,Coles, and Donchin (1983) d e s c r i b e a technique that avoids two problems evident i n the V e r g l e r et a l . (1982) study. F i r s t , they c a l c u l a t e d c o r r e c t i o n f a c t o r s s e p a r a t e l y for b l i n k s and eye movements; and secondly, they computed these c o r r e c t i o n f a c t o r s a f t e r event r e l a t e d a c t i v i t y had been removed from both EEG and EOG records. If the c o r t i c a l evoked p o t e n t i a l appears in both the EEG and EOG records, then the e f f e c t of the EOG s i g n a l at EEG e l e c t r o d e s i t e s would be overestimated during the c o r r e c t i o n procedure. To remove event r e l a t e d a c t i v i t y , the authors began by c a l c u l a t i n g a raw average c o n s i s t i n g of the mean of a l l epochs in a given experimental c o n d i t i o n . T h i s raw average was c a l c u l a t e d f o r each EEG e l e c t r o d e s i t e as w e l l as the EOG s i t e . These raw averages were then subtracted, epoch by epoch, from the a p p r o p r i a t e raw EEG and EOG data. The authors contend that t h i s s u b t r a c t i o n step g i v e s an estimate of the e l e c t r i c a l a c t i v i t y that i s not stimulus r e l a t e d , that i s , e l e c t r i c a l a c t i v i t y other than o c u l a r p o t e n t i a l which occurs at the EOG and 1 7 EEG e l e c t r o d e s i t e s . T h i s s u b t r a t i o n step avoids the problem that EEG a c t i v i t y which may be pi c k e d up at the EOG e l e c t r o d e s i t e i s included i n the estimate of oc u l a r p o t e n t i a l . At t h i s p o i n t i n the c a l c u l a t i o n the EEG data are con s i d e r e d to be a dependent v a r i a b l e and the EOG data serve as the independent v a r i a b l e . The EOG data are regressed on the EEG data and a r e g r e s s i o n c o e f f i c i e n t i s c a l c u l a t e d . The r e g r e s s i o n c o e f f i c i e n t i s then used to d e r i v e the c o r r e c t i o n f a c t o r s that determine how much EOG i s subtrac t e d from the EEG. The r e g r e s s i o n c o e f f i c i e n t (r) i s m u l t i p l i e d by the r a t i o of EEG standard d e v i a t i o n over EOG standard d e v i a t i o n . S r x — = c o r r e c t i o n f a c t o r Seog T h i s i s a s c a l i n g step that a d j u s t s f o r amplitude, d i f f e r e n c e s because EOG p o t e n t i a l s are g e n e r a l l y much greater i n magnitude at the eyes than they are at the EEG e l e c t r o d e s i t e s . The r e s u l t i n g c o r r e c t i o n f a c t o r d i c t a t e s what p r o p o r t i o n of the EOG s i g n a l i s to be sub t r a c t e d from the EEG s i g n a l . C o r r e c t i o n f a c t o r s are c a l c u l a t e d s e p a r a t e l y f o r b l i n k and fo r eye movement data. B l i n k s are detected, and the data within the b l i n k epoch are used to d e r i v e a c o r r e c t i o n f a c t o r as d e s c r i b e d above. T h i s c o r r e c t i o n f a c t o r i s used to c o r r e c t the EEG data p o i n t s , i n each epoch, which f a l l between the onset and o f f s e t of b l i n k s . The premise i s that the tr a n s m i s s i o n of the b l i n k p o t e n t i a l i s q u a l i t a t i v e l y d i f f e r e n t from the transmission 1 8 of eye movement p o t e n t i a l and must be d e a l t with s e p a r a t e l y . To account for any eye movement p o t e n t i a l , a separate c o r r e c t i o n f a c t o r i s c a l c u l a t e d , i n the above manner, based on data a f t e r removal of b l i n k p o t e n t i a l . T h i s second c o r r e c t i o n f a c t o r i s a p p l i e d to a l l epochs of raw data a f t e r b l i n k c o r r e c t i o n . Once the s c a l e d EOG i s s u b t r a c t e d from each epoch the data are averaged to y i e l d the c o r r e c t e d VER. It i s important to note that although the c o r r e c t i o n f a c t o r s are computed using only e l e c t r i c a l a c t i v i t y that i s not event r e l a t e d , the a c t u a l c o r r e c t i o n i s made on raw data that include event r e l a t e d p o t e n t i a l s . Study R a t i o n a l e The problems c o n f r o n t i n g b l i n k c o r r e c t i o n procedures and s t u d i e s can be c l a s s i f i e d broadly in four ways: 1 . E l i m i n a t i o n of contaminated t r i a l s , the commonly used method, of d e a l i n g with b l i n k s , may not always be an acceptable procedure from e i t h e r a t h e o r e t i c a l or p r a c t i c a l p o i n t of view. 2 . Most c o r r e c t i o n techniques have been used to attenuate slow eye movements, not b l i n k s . 3 . In most s u b t r a c t i o n techniques l i t t l e c o n s i d e r a t i o n i s given to the p o s s i b i l i t y that v o l u n t a r y b l i n k s , r o u t i n e l y used in c a l i b r a t i o n , may have an evoked c e r e b r a l component that would add to the b l i n k p o t e n t i a l picked up at EEG e l e c t r o d e s . 4 . Studies that have looked at a r t i f a c t removal have t y p i c a l l y l e f t v a l i d i t y and r e l i a b i l i t y of t h e i r procedures open 19 to q u e s t i o n . 1. Problems with E l i m i n a t i o n of Blink-Contaminated T r i a l s . The poverty of b l i n k c o r r e c t i o n procedures in the l i t e r a t u r e i s due mainly to the f a c t that the issue has been l a r g e l y ignored. As pointed out by Iacono et a l . (1982), in very few s t u d i e s of v i s u a l evoked p o t e n t i a l s to l i g h t f l a s h e s was eye movement monitored. Those who do monitor eye movement r o u t i n e l y r e j e c t any epochs that are contaminated by movement a r t i f a c t s . T h i s procedure of epoch r e j e c t i o n has been questioned r e c e n t l y on the grounds that b l i n k i n g i t s e l f may be a means of modulating sensory input and may be a p h y s i o l o g i c a l expression of c e r t a i n p a t h o l o g i c a l c o n d i t i o n s (Iacono et a l . , 1982). Ponder and Kennedy (1927), i n t h e i r c l a s s i c study of b l i n k r a t e s , noted that one of the most pronounced f e a t u r e s of p o s t e n c e p h a l i t i c Parkinson's disease i s a n e a r l y t o t a l absence of b l i n k i n g . As well they noted that b l i n k i n g , while i t tends to be s t a b l e over time in a given i n d i v i d u a l , can be temporarily increased by anger, anxiety, and other emotional s t a t e s . More r e c e n t l y , Stevens (1978) demonstrated an abnormality of b l i n k r a t e and b l i n k r e f l e x in m e d i c a t i o n - f r e e s c h i z o p h r e n i c p a t i e n t s . Stevens (1978) a l s o p o i n t s out that increased b l i n k i n g i s a common si g n of G i l l e s de l a Tourette syndrome, s c h i z o p h r e n i a , and other p s y c h i a t r i c d i s o r d e r s that respond w e l l to dopamine-b l o c k i n g n e u r o l e p t i c s . Karson, Freed, Kleinman, Bigelow, and Wyatt (1981) support Stevens' (1978) hypothesis that b l i n k r a t e i s modulated in some way by c e n t r a l dopamine a c t i v i t y . To f u r t h e r t e s t the r e l a t i o n s h i p between b l i n k r a t e and monoamine 20 a c t i v i t y they examined the e f f e c t s of n e u r o l e p t i c medication on s c h i z o p h r e n i c p a t i e n t s . T h e i r c o n c l u s i o n was that n e u r o l e p t i c s s i g n i f i c a n t l y reduced b l i n k r a t e s i n these p a t i e n t s but they co u l d not say whether t h i s e f f e c t was due d i r e c t l y to changes in dopamine metabolism or to r e d u c t i o n s i n psychopathology. Although the neurophysiology of b l i n k i n g i s not well understood, i t i s c l e a r that an i n d i v i d u a l ' s r a t e i s a f f e c t e d by f a c t o r s which include emotional and p h y s i o l o g i c a l s t a t e s . Given that b l i n k i n g may be d i r e c t l y r e l a t e d to c o g n i t i v e processes i t would not be prudent to ignore the importance of b l i n k s in evoked p o t e n t i a l r e c o r d i n g s . By r e j e c t i n g contaminated t r i a l s and keeping only t r i a l s that are b l i n k f r e e , a s e l e c t i o n bias of unknown importance i s introduced that could make experimental r e s u l t s m i s l e a d i n g . Apart from the t h e o r e t i c a l reasons for not r e j e c t i n g contaminated t r i a l s , there i s a l s o the p r a c t i c a l reason that a subject p o o l , e s p e c i a l l y one c o n t a i n i n g n e u r o l o g i c or p s y c h i a t r i c p a t i e n t s , may not support the data l o s s (Vergler et a l . , 1982). The matter of whether or not t r i a l r e j e c t i o n i s an acceptable means of d e a l i n g with b l i n k a r t i f a c t in the VER paradigm must be adressed. 2. L i m i t a t i o n s of CNV C o r r e c t i o n Techniques. I n t e r e s t in eye movement c o r r e c t i o n techniques o r i g i n a t e d with CNV s t u d i e s . In a t y p i c a l CNV experiment a subject i s given an a u d i t o r y or v i s u a l s t i m u l u s . T h i s stimulus a l e r t s the subject to a second stimulus that r e q u i r e s a response. During the i n t e r s t i m u l u s i n t e r v a l a slowly developing, negative waveform, of i n t e r e s t to p s y c h o p h y s i o l o g i s t s , i s generated. However, 21 during t h i s i n t e r s t i m u l u s p e r i o d , p o t e n t i a l s from the corneo-r e t i n a l d i p o l e due to e y e b a l l r o t a t i o n are a l s o generated. These p o t e n t i a l s are subsequently t r a n s m i t t e d across the s c a l p to the EEG e l e c t r o d e s . It was i n the i n t e r e s t of e l i m i n a t i n g t h i s slow eye movement a r t i f a c t d u r i n g the i n t e r s t i m u l u s i n t e r v a l that c o r r e c t i o n techniques were developed. In CNV s t u d i e s b l i n k s t y p i c a l l y do not pose the problem they do in s t u d i e s which evoke responses with l i g h t f l a s h e s . Since CNV p o t e n t i a l s tend to develop slowly and can be sustained f o r as long as 20 seconds (Martin & Venables, 1980), b l i n k a r t i f a c t s do not d i s t o r t r e s u l t s i n a systematic way. Often, CNV re s e a r c h e r s concerned with b l i n k a r t i f a c t s would apply the same r e g r e s s i o n c o e f f i c i e n t (or manual adjustment of EOG amplitude) to b l i n k and slow movement a r t i f a c t s . Since Corby and K o p e l l (1972) have shown that b l i n k s and slow- eye movement p o t e n t i a l s are d i f f e r e n t i a l l y t r a n s m i t t e d across the s c a l p , any c o r r e c t i o n procedure a p p l i e d i n d i s c r i m i n a t e l y to both types of a r t i f a c t may give l e s s than optimal r e s u l t s . It i s necessary to determine how eye movement c o r r e c t i o n techniques, o r i g i n a t i n g with CNV s t u d i e s , w i l l perform i n VER stu d i e s where b l i n k s are the major contaminant of VER r e c o r d i n g s . 3. Evoked Responses and Voluntary B l i n k s . A l l c o r r e c t i o n procedures, whether using o n - l i n e analog devices or o f f - l i n e c o r r e c t i o n formulas, r e q u i r e c a l i b r a t i o n data. When c a l i b r a t i n g f o r b l i n k s these data have been generated by having the subject b l i n k on command (Corby & K o p e l l , 1972) or by using random b l i n k s ( V e r g l e r et a l . , 1982) or both (Iacono et 22 a l . , 1982). There are problems i n v o l v e d in using e i t h e r c a l i b r a t i o n method, whether asking the subject to b l i n k or using random b l i n k s . In instances when the subject i s given the command " b l i n k " the experimenter i s i n f a c t generating an a u d i t o r y evoked response in a d d i t i o n to the b l i n k p o t e n t i a l . Thus the r e s u l t i n g EEG record would c o n t a i n more than j u s t a b l i n k p o t e n t i a l and c o r r e c t i o n c o e f f i c i e n t s estimated on these data would be i n a c c u r a t e . On the other hand, using random b l i n k s avoids the a u d i t o r y evoked response but can have i t s own drawbacks. Armington (1981) reported that b l i n k s b r i e f l y i n t e r r u p t l i g h t f a l l i n g on the r e t i n a and in doing so evoke a c e r e b r a l response s i m i l a r to that evoked by a l i g h t f l a s h . He p o s t u l a t e d that b l i n k i n g may be one mechanism, c o e x i s t i n g with others such as head movement and eye movement, that produces changes i n r e t i n a l s t i m u l a t i o n when viewing a steady scene. These changes in s t i m u l a t i o n serve to prevent r e c e p t o r s from becoming adapted to the stimulus being viewed. The i m p l i c a t i o n of t h i s f i n d i n g for re c o r d i n g v i s u a l evoked p o t e n t i a l s i s that a s u b j e c t ' s b l i n k should not be allowed to cause a change in l i g h t i n t e n s i t y f a l l i n g on the r e t i n a . The simple way to avoid t h i s problem i s . to have the subject i n a darkened environment throughout c a l i b r a t i o n and experimental t r i a l s . Iacono et a l . (1982) used such a darkened environment; however, they a l s o used gross c o r r e c t i o n procedures. These r e l a t i v e l y crude c o r r e c t i o n procedures which i n v o l v e d v i s u a l , o n - l i n e e s t i m a t i o n of the amount of EOG p o t e n t i a l s u b t r a c t e d from the EEG, may have 23 l i m i t e d the degree of p r e c i s i o n a f f o r d e d by using a darkened environment. Although V e r g l e r et a l . (1982) used spontaneously generated eye movement and b l i n k p o t e n t i a l s , they made no mention of a darkened environment. 4. Outcome Measures: V a l i d i t y and r e l i a b i l i t y . In order to evaluate r e l i a b i l i t y and v a l i d i t y of c o r r e c t i o n procedures we must have a s u i t a b l e standard or c r i t e r i o n against which the a r t i f a c t c o r r e c t e d EEG t r a c e s can be compared. In past s t u d i e s , outcome e v a l u a t i o n t y p i c a l l y r e l i e d on v i s u a l i n s p e c t i o n of the c o r r e c t e d EEG rec o r d to see i f there was any r e s i d u a l eye b l i n k a r t i f a c t . T h i s approach i s not o b j e c t i v e and provides only a crude e s t i m a t i o n of b l i n k removal. V e r g l e r et a l . (1982) employed one o b j e c t i v e outcome measure when they c o r r e l a t e d the c o r r e c t e d EEG r e c o r d with the EOG r e c o r d . On the assumption that the EEG and EOG p o t e n t i a l s are orthogonal, they reasoned that the p r o b a b i l i t y of the two traces being the same should not exceed chance. The c o r r e l a t i o n c o e f f i c i e n t of a s u c c e s s f u l l y c o r r e c t e d EEG t r a c e and i t s complimentary EOG t r a c e should be i n s i g n i f i c a n t and near 0. The authors found i n s i g n i f i c a n t c o r r e l a t i o n s at only 3 of 8 e l e c t r o d e s i t e s using t h i s procedure. This poor r e s u l t i s not s u r p r i s i n g s i n c e the authors recorded over long response epochs (7.5 sec) a l l o w i n g the p o s s i b i l i t y of systematic contamination by slow eye movement, yet only one c o r r e l a t i o n c o e f f i c i e n t was c a l c u l a t e d f o r both v e r t i c a l eye movements and b l i n k s . As di s c u s s e d e a r l i e r , f a i l u r e to account for the d i f f e r e n t i a l e f f e c t of slow eye movement can r e s u l t in poor a r t i f a c t 24 a t t e n u a t i o n . Gratton et a l . (1983) u t i l i z e d two t e s t s of v a l i d i t y in a s s e s s i n g t h e i r r e g r e s s i o n approach to a r t i f a c t c o r r e c t i o n . The f i r s t v a l i d i t y check i n v o l v e d comparing the c o r r e c t e d averaged evoked response (AER) with what the authors c a l l e d the " t r u e " AER. T h i s " t r u e " AER c o n s i s t e d of an average of the epochs during which EOG a c t i v i t y d i d not exceed a v a r i a n c e c r i t e r i o n (nonblink epochs). The authors themselves p o i n t out a problem in using t h i s procedure, s t a t i n g that t h i s " t r u e " AER may be computed on too small a number of epochs to e l i m i n a t e random noise from the average. Such noise would reduce the accuracy of the " t r u e " AER e s t i m a t i o n . The other t e s t used to support the v a l i d i t y of t h e i r c o r r e c t i o n procedure was a comparison between c o r r e c t e d AERs from t r i a l s with d i f f e r e n t degrees of a r t i f a c t . For each s u b j e c t , two sets of data were d e r i v e d , one i n which EOG v a r i a n c e was greater than a c r i t e r i o n and one set i n which EOG was l e s s than the c r i t e r i o n . I t was found that the d i f f e r e n c e between t r i a l s contaminated by v a r y i n g amounts of a r t i f a c t decreased a f t e r c o r r e c t i o n , suggesting that the c o r r e c t i o n procedure was removing much of the unwanted a r t i f a c t var i a n c e . Another approach I c a l l the "spontaneous b l i n k " method may serve as a more d i r e c t means of e v a l u a t i n g c o r r e c t i o n procedures. During c a l i b r a t i o n t r i a l s s u b j e c t s are allowed to b l i n k spontaneously as they s i t comfortably in a darkened room, producing eye movement p o t e n t i a l s in both the EOG and EEG channels. In t h i s s i t u a t i o n the EEG t r a c e would represent the 25 p o t e n t i a l t r a n s m i t t e d from the eye as we l l as random c e r e b r a l a c t i v i t y . Averaging of suc c e s s i v e spontaneous b l i n k epochs i n the EEG channel would e l i m i n a t e random EEG a c t i v i t y and y i e l d a good r e p r e s e n t a t i o n of the b l i n k a r t i f a c t waveform at the EEG e l e c t r o d e placement. Averaging of the EOG channel would s i m i l a r l y y i e l d an estimate of the b l i n k p o t e n t i a l at the EOG e l e c t r o d e placements. By s u b t r a c t i n g the c o r r e c t p r o p o r t i o n of averaged EOG from the corresponding averaged EEG b l i n k epochs, most of the recorded p o t e n t i a l w i l l be e l i m i n a t e d l e a v i n g a t r a c e resembling that of averaged random ( r e s t i n g ) EEG. This p r o v i d e s a standard against which we can assess the success of a c o r r e c t i o n procedure. The more s u c c e s s f u l a method i s i n removing the b l i n k s the more c l o s e l y the c o r r e c t e d record w i l l approximate a s t r a i g h t l i n e . E i t h e r o v e r - c o r r e c t i o n or under-c o r r e c t i o n w i l l cause a t e l l t a l e d e v i a t i o n from l i n e a r i t y . The purpose of the present study i s to address each of the iss u e s d i s c u s s e d above. Before a thorough a n a l y s i s of these i s s u e s can be made, an understanding of the c h a r a c t e r i s t i c s of b l i n k s i s r e q u i r e d . Parameters such as b l i n k onset, peak l a t e n c y , and b l i n k r e l i a b i l i t y w i l l t h e r e f o r e be i n v e s t i g a t e d . Once the b l i n k c h a r a c t e r i s t i c s and t h e i r r e l a t i o n to the VER are determined, i n v e s t i g a t i o n of b l i n k a r t i f a c t i n the EEG can proceed. The b e n e f i t s and disadvantages of s e v e r a l approaches to d e a l i n g with b l i n k a r t i f a c t w i l l be looked at and an attempt made to determine which approach i s most a p p l i c a b l e to the VER paradigm. Twenty subjects w i l l begin by s i t t i n g i n a completely darkened room for 5 minutes where they are allowed to b l i n k 26 f r e e l y . T h i s i s the spontaneous b l i n k t r i a l and i s intended to produce an EEG record of b l i n k p o t e n t i a l that i s f r e e of any c e r e b r a l evoked p o t e n t i a l . The spontaneous b l i n k data w i l l be used to provide a v a l i d i t y check of two c o r r e c t i o n procedures ( r e g r e s s i o n and s u b t r a c t i o n ) and as a means of e s t i m a t i n g a c o r r e c t i o n f a c t o r . The s u b j e c t s w i l l then be presented with 3 t r i a l s of l i g h t f l a s h e s . Each t r i a l c o n s i s t s of 64 l i g h t f l a s h e s at each of 4 i n t e n s i t i e s f o r a t o t a l of 256 f l a s h e s . During T r i a l 1, s u b j e c t s w i l l keep t h e i r eyes open and watch the center of the screen. During T r i a l 2, the s u b j e c t s w i l l be asked to keep t h e i r eyes c l o s e d and watch the l i g h t s through c l o s e d eye l i d s . The eyes c l o s e d data w i l l be used to compare the e f f i c a c y of t h i s method with other approaches to b l i n k a r t i f a c t c o n t r o l . T r i a l 3 i s a r e p l i c a t i o n of T r i a l 1 where subjects again w i l l watch the l i g h t s with t h e i r eyes open. B l i n k a r t i f a c t i n the data from T r i a l s 1 and 3 w i l l be d e a l t with using three d i f f e r e n t approaches. The r e g r e s s i o n approach of Gratton et a l . (1983) w i l l be used as w e l l as an o f f l i n e v a r i a t i o n of the s u b t r a c t i o n method d e s c r i b e d e a r l i e r . The s u b t r a c t i o n method that w i l l be employed here uses a s c a l i n g f a c t o r d e r i v e d from the spontaneous b l i n k t r i a l data. T h i s s c a l i n g f a c t o r w i l l be used to determine what p r o p o r t i o n of EOG i s s u b t r a c t e d from the EEG. A t h i r d approach to d e a l i n g with b l i n k s w i l l be the b l i n k r e j e c t i o n method. This method i n c l u d e s i n the data average only those epochs which are not contaminated with a b l i n k . F i n a l l y , the raw data, c o n t a i n i n g both b l i n k -contaminated and b l i n k - f r e e t r i a l s , w i l l be averaged and 27 r e f e r r e d to as uncorrected data. The uncorrected data w i l l be used fo r comparison when es t i m a t i n g the v a l i d i t y of the c o r r e c t i o n procedures. 28 Method Subjects Twenty female s u b j e c t s were r e c r u i t e d from undergraduate psychology courses at the U n i v e r s i t y of B r i t i s h Columbia to p a r t i c i p a t e in a p s y c h o p h y s i o l o g i c a l experiment. The mean age of the women who p a r t i c i p a t e d in the study was 24.3 years (SD.=3.0 yrs) with a range of 19 to 45 years.. A l l v o l u n t e e r s were Caucasian. Recording and Apparatus Four l i g h t f l a s h i n t e n s i t i e s ; 2, 30, 80, and 240 f t lamberts were used. These w i l l be r e f e r r e d to as i n t e n s i t i e s 1, 2, 3, and 4, r e s p e c t i v e l y , in the f o l l o w i n g t e x t . The f l a s h e s were generated by 8 f l a s h t u b e s mounted behind a 46 x 33 centimeter, semitranslucent screen which d i s t r i b u t e d the l i g h t evenly. The l i g h t s were presented at 1 h e r t z , reached maximum br i g h t n e s s in 0.5 m i l l i s e c o n d s and l a s t e d f o r 0.5 seconds. The EEG was recorded with g o l d - p l a t e d Grass e l e c t r o d e s f i x e d to the s c a l p at three s i t e s ; Fz and Cz (10-20 I n t e r n a t i o n a l System) and Oz. Linked earlobes were used as r e f e r e n c e . Resistances were kept below 5 kohms by brushing the skin with an abra s i v e paste before a t t a c h i n g the e l e c t r o d e s . In order to monitor b l i n k i n g , the v e r t i c a l EOG was a l s o recorded. Ag-AgCl e l e c t r o d e s were attached above and below the r i g h t eye 29 a f t e r vigorous massage of the area with a b r a s i v e paste. Subjects were grounded with an e l e c t r o d e placed on the r i g h t arm. A l l data were recorded on a Beckman Type R612 polygraph with time c o n s t a n t s for the EEG and EOG channels set at 1.0 second. The 50% bandwidth f o r high frequency f i l t e r i n g was set at 30 h e r t z . A marker channel i n d i c a t i n g the onset and i n t e n s i t y of each l i g h t f l a s h was a l s o used. For the purposes of r e c o r d i n g spontaneous b l i n k s , the- marker channel was replaced by an EOG channel with a time constant set at .03 seconds. T h i s EOG channel served as a "marker" of the onset of a b l i n k and was used to d e f i n e the onset of a spontaneous b l i n k epoch for the purposes of d i g i t i z a t i o n . Recordings were s t o r e d on magnetic tape and d i g i t i z e d o f f - l i n e u s ing a V e t t e r Model A FM r e c o r d e r . Procedure L i g h t f l a s h e s were presented to s u b j e c t s in a pseudorandom order with 64 p r e s e n t a t i o n s at each i n t e n s i t y . The 256 f l a s h e s of l i g h t were presented in one block. Before each block of l i g h t s the s u b j e c t s were given prerecorded i n s t r u c t i o n s e x p l a i n i n g the experimental procedure (see Appendix I ) . Each block of p r e s e n t a t i o n s was c o n s i d e r e d to be one t r i a l and w i l l be r e f e r r e d to as a t r i a l i n t h i s study. In t o t a l there were1 three l i g h t t r i a l s (eyes open, T r i a l 1; eyes c l o s e d , T r i a l 2; and eyes open, T r i a l 3) and one t r i a l i n v o l v i n g no l i g h t s (spontaneous b l i n k t r i a l ) . S ubjects were asked to remove g l a s s e s or c o n t a c t l e n s e s before the experimental procedures began. The s u b j e c t s sat in a high backed c h a i r with a small p i l l o w 30 placed behind the neck to a i d in keeping t h e i r head s t a t i o n a r y . A l l s u b j e c t s were asked to watch the center of the screen and to avoid moving t h e i r head or eyes d u r i n g stimulus p r e s e n t a t i o n (see i n s t r u c t i o n s Appendix I ) . Stimulus p r e s e n t a t i o n was c o n t r o l l e d by a microprocessor. Each subject began with the spontaneous b l i n k t r i a l . Subjects sat in a t o t a l l y darkened room for 5 minutes while EOG and EEG were recorded. Subjects were i n s t r u c t e d to keep t h e i r face pointed towards the center of the screen and not to c l o s e t h e i r eyes while the room was dark. T h i s procedure provided an opportunity to r e c o r d spontaneous b l i n k s ( b l i n k s that occurred without a command to b l i n k ) i n a darkened room such that the eye c l o s u r e a s s o c i a t e d with b l i n k i n g would not r e s u l t i n a change i n l i g h t f a l l i n g on the r e t i n a and evoke an " o f f " response in the EEG channel ( E f r o n , 1964). Foll o w i n g t h i s dark p e r i o d the subject was i n s t r u c t e d to watch the center of the screen while the f i r s t block of 256 l i g h t f l a s h e s was presented ( l i g h t t r i a l 1). No i n s t r u c t i o n s regarding b l i n k i n g during l i g h t f l a s h e s were given. For the second block of 256 l i g h t f l a s h e s ( l i g h t t r i a l 2) subjects were asked to keep t h e i r eyes c l o s e d while keeping t h e i r heads s t i l l and faces pointed toward the center of the screen. The t h i r d block of 256 l i g h t f l a s h e s ( l i g h t t r i a l 3) was presented in the same manner as the f i r s t t r i a l , with s u b j e c t s i n s t r u c t e d to open t h e i r eyes and watch the center of the screen. 31 Data A n a l y s i s EEG and EOG were sampled at a r a t e of 250 h e r t z over a 700 m i l l i s e c o n d epoch f o l l o w i n g stimulus onset. A z e r o - v o l t a g e b a s e l i n e was estimated by averaging the 200 m i l l i s e c o n d period, of EEG o c c u r r i n g j u s t before l i g h t s timulus onset. A l l EEG amplitudes were computed with r e f e r e n c e to t h i s b a s e l i n e . Component peaks of the VER were d e f i n e d as f o l l o w s ; P, - the maximum amplitude from 60-112 m i l l i s e c o n d s , N, - the minimum amplitude from 80-152 m i l l i s e c o n d s , and P 2 - the maximum amplitude from 160-254 m i l l i s e c o n d s . B l i n k d e t e c t i o n was done a u t o m a t i c a l l y by a computer algorithm. Any epoch in which a b l i n k was detected i n the 200 ms b a s e l i n e p e r i o d before stimulus onset or in the 700, ms p e r i o d a f t e r stimulus onset was d e f i n e d as a blink-contaminated epoch. The purpose of d e t e c t i n g b l i n k s i n the prestimulus p e r i o d was to e l i m i n a t e them from the b a s e l i n e average, thereby p r o v i d i n g a more accurate e s t i m a t i o n of the z e r o - v o l t a g e b a s e l i n e . The number and latency of b l i n k s o c c u r r i n g a f t e r stimulus onset were a l s o determined. A b l i n k was d e f i n e d as any r i s e i n the EOG that had a minimum amplitude of 50 m i c r o v o l t s but d i d not exceed an amplitude of 1.0 m i l l i v o l t , had a minimum dura t i o n of 50 m i l l i s e c o n d s , and a maximum d u r a t i o n of 600 m i l l i s e c o n d s . Latency to b l i n k onset was d e f i n e d as the time when the b l i n k was at 10 percent of i t s peak amplitude. Two c o r r e c t i o n methods were used in t h i s study; r e g r e s s i o n and s u b t r a c t i o n . The r e g r e s s i o n method was that of Gratton, Coles, and Donchin (1982) d e s c r i b e d i n d e t a i l e a r l i e r i n the 32 i n t r o d u c t i o n . In b r i e f t h i s approach uses a c o r r e l a t i o n c o e f f i c i e n t c a l c u l a t e d by r e g r e s s i n g the EOG amplitude measures on the EEG measures fo r a l l b l i n k contaminated epochs... T h i s c o e f f i c i e n t i s used to c a l c u l a t e a s c a l i n g f a c t o r that determines the a p p r o p r i a t e p r o p o r t i o n of EOG to be s u b t r a c t e d , in a po i n t by point f a s h i o n , from a l l b l i n k contaminated epochs of the EEG. T h i s same procedure i s repeated using a l l epochs of the b l i n k c o r r e c t e d data and a second c o e f f i c i e n t i s c a l c u l a t e d to c o r r e c t f o r nonblink, eye movement a r t i f a c t . T h i s second c o r r e c t i o n f a c t o r i s used i n the same manner as the f i r s t but i s a p p l i e d to a l l epochs i n c l u d i n g the p r e v i o u s l y c o r r e c t e d b l i n k contaminated epochs. The second c o r r e c t i o n f a c t o r i s expected to remove any EEG a r t i f a c t that has not p r e v i o u s l y been d e f i n e d as b l i n k a r t i f a c t . These twice c o r r e c t e d data are then averaged y i e l d i n g the f i n a l VER. The same p r i n c i p l e used in the r e g r e s s i o n method, that of s u b t r a c t i n g a p r o p o r t i o n of EOG amplitude from the EEG, i s employed i n the s u b t r a c t i o n method but the process of e s t i m a t i n g the c o r r e c t i o n f a c t o r i s s i m p l i f i e d somewhat. The s u b t r a c t i o n method used here i s s i m i l a r i n concept to the s u b t r a c t i o n procedures used p r e v i o u s l y ( G i r t o n & Kamiya, 1973; Iacono et a l . , 1972) but employs a more s o p h i s t i c a t e d means of e s t i m a t i n g the amount of EOG to be s u b t r a c t e d from the EEG. In t h i s study a r a t i o of EEG to EOG was c a l c u l a t e d by using the spontaneous b l i n k data. To detect b l i n k s d u r i n g the spontaneous b l i n k t r i a l , an EOG channel with a time constant set at .03 seconds was used. T h i s EOG channel s i g n a l e d the onset of a b l i n k . The other EOG 33 channel and the three EEG channels were then sampled at a rate of 250 h e r t z over a 200 m i l l i s e c o n d p e r i o d preceding b l i n k onset and a 500 m i l l i s e c o n d p e r i o d f o l l o w i n g b l i n k onset. B l i n k s were averaged in each channel to y i e l d the spontaneous b l i n k data. I t was assumed that there was no s i g n i f i c a n t evoked p o t e n t i a l i n v o l v e d i n b l i n k i n g d u r i n g the spontaneous b l i n k t r i a l . An average of the EEG at each e l e c t r o d e placement s h o u l d , t h e r e f o r e , give an estimate of the o c u l a r a r t i f a c t t r a n s m i t t e d to the EEG e l e c t r o d e that i s uncontaminated by c e r e b r a l p o t e n t i a l . A r a t i o of averaged EEG to averaged EOG i s c a l c u l a t e d at each e l e c t r o d e placement. This r a t i o i s then used as a s c a l i n g , f a c t o r to determine the p r o p o r t i o n of EOG to be s u b t r a c t e d , point by p o i n t , from the EEG d~.ta. The b l i n k r e j e c t i o n method of d e a l i n g with o c u l a r a r t i f a c t simply r e j e c t s any a r t i f a c t contaminated epochs. Included in the average are only those epochs which do not c o n t a i n a b l i n k as d e f i n e d above. The nonblink average i s then considered to be r e p r e s e n t a t i v e of the the " t r u e " VER s i n c e i t i s uncontaminated by b l i n k a r t i f a c t . T h i s method c o n t r a s t s with the eyes c l o s e d approach where subjects keep t h e i r head s t a t i o n a r y and watches the l i g h t f l a s h e s through c l o s e d e y e l i d s . The aim of t h i s method i s to a v o i d b l i n k i n g by having the e y e l i d s c l o s e d throughout the experimental procedure. Data r e f e r r e d to as "uncorrected data" i n the f o l l o w i n g text are raw data, obtained from the eyes-open c o n d i t i o n , that have not had o c u l a r a r t i f a c t removed in any manner. 34 R e s u l t s and D i s c u s s i o n In t h i s s e c t i o n d i s c u s s i o n of the r e s u l t s w i l l begin with an a n a l y s i s of spontaneous b l i n k data. T h i s w i l l include an a n a l y s i s of how the b l i n k p o t e n t i a l i s propagated across the s c a l p as well: as an e s t i m a t i o n of how w e l l the r e g r e s s i o n and s u b t r a c t i o n techniques e l i m i n a t e spontaneous b l i n k p o t e n t i a l from the EEG. Next, b l i n k i n g in response to the l i g h t f l a s h e s w i l l be i n v e s t i g a t e d . Onset and peak l a t e n c y along with amplitude changes of the- b l i n k at the four l i g h t i n t e n s i t i e s w i l l be examined. The r e l i a b i l i t y of v a r i o u s b l i n k parameters, i n c l u d i n g the number of b l i n k s , from T r i a l s 1 and 3 w i l l be estimated. E v a l u a t i o n of c o r r e c t i o n procedures w i l l i n clude an a n a l y s i s of how the b l i n k i n f l u e n c e s v a r i o u s p o r t i o n s of the VER. C o r r e l a t i o n of EOG with c o r r e c t e d and uncorrected EEG and c o r r e l a t i o n s between i n d i v i d u a l c o r r e c t i o n approaches w i l l a l s o be c a l c u l a t e d . Although the p r o b a b i l i t y values for i n d i v i d u a l c o r r e l a t i o n s w i l l not be presented, the reader should be aware that a l i b e r a l estimate of s i g n i f i c a n c e i s as f o l l o w s : a c o e f f i c i e n t of .4400 or g r e a t e r represents p<.05 and a c o e f f i c i e n t of .5600 or g r e a t e r represents p<.0l. These estimates are based on o n e - t a i l e d t e s t s of s i g n i f i c a n c e because only p o s i t i v e c o r r e l a t i o n s are expected. R e l i a b i l i t y of methods between T r i a l s 1 and 3 w i l l be determined. An a n a l y s i s of v a r i a n c e (ANOVA) w i l l be c a r r i e d out using data from the r e g r e s s i o n and s u b t r a c t i o n procedures, the eyes c l o s e d 35 c o n d i t i o n , and the uncorrected VERs from the eyes-open con d i t i o n . In the course of t h i s study a l a r g e number of c o r r e l a t i o n s were c a l c u l a t e d . Some c o r r e l a t i o n c o e f f i c i e n t s w i l l be reported in which the same two v a r i a b l e s are c o r r e l a t e d f o r each of 3 e l e c t r o d e s i t e s , 4 l i g h t i n t e n s i t i e s , and 2 t r i a l s , i n c r e a s i n g the p o s s i b i l i t y of Type I e r r o r . Rather than a d j u s t i n g the p r o b a b i l i t y values to accomodate repeated estimates of the same phenomenon, and r i s k i n g Type II e r r o r , c o r r e l a t i o n s w i l l be presented in a manner such that r e l a t i o n s h i p s across a l l c o r r e l a t i o n s can be seen. The i n t e n t i o n of t h i s study i s to look at p a t t e r n s in the c o r r e l a t i o n matrix, t h e r e f o r e i t would be m i s l e a d i n g to focus on the s t a t i s t i c a l s i g n i f i c a n c e of each c o r r e l a t i o n as though i t represented an independent t e s t of a h y p o t h e s i s . To c o n t r o l f o r the number of f a l s e p o s i t i v e s i n the repeated measures ANOVAs f o l l o w i n g , the Greenhouse-Geisser (1959) c o r r e c t i o n was used. This c o r r e c t i o n procedure a d j u s t s the degrees of freedom of the F - t e s t based on the homogeneity of both the v a r i a n c e s and covariances of the repeated measures. When there i s reason to question the homogeneity assumptions a more c o n s e r v a t i v e F - t e s t i s used. The degrees of freedom are a d j u s t e d by a f a c t o r (e) that r e f l e c t s the degree of h e t e r o g e n e i t y of the repeated measures variances and c o v a r i a n c e s . Spontaneous B l i n k s 36 The r e c o r d i n g of spontaneous b l i n k s in t h i s study served a dual purpose. These data provided a means to i n v e s t i g a t e the c h a r a c t e r i s t i c s of b l i n k s as they occurred and were propagated across the s c a l p , and a l s o served as a means to evaluate the v a l i d i t y of both the s u b t r a c t i o n and r e g r e s s i o n methods of a r t i f a c t c o r r e c t i o n . As the b l i n k p o t e n t i a l t r a v e l s from i t s source at the eye across the s c a l p , t i s s u e r e s i s t a n c e should r e s u l t i n a decreased waveform amplitude. T h i s proved to be the case, as amplitude measures were attenuated at each of the three e l e c t r o d e placements [F(3,57)=117.02, e=.3394, p<.000l], (see Table l ) . Since the p o t e n t i a l recorded at the EOG diminishes as i t t r a v e l s a crosg the s c a l p i t i s reasonable to assume that i t s c o n t r i b u t i o n to the EEG record w i l l vary with e l e c t r o d e l o c a t i o n . If t h i s assumption i s c o r r e c t , then any c o r r e c t i o n procedure that s u b t r a c t s the same EOG s i g n a l at each e l e c t r o d e placement i s l i k e l y to y i e l d i n a c c u r a t e estimates of the VER. The most accurate approach would be one that attempts to estimate a c o r r e c t i o n f a c t o r for each e l e c t r o d e s i t e and s c a l e s the EOG s i g n a l by that f a c t o r before s u b t r a c t i o n . Both the r e g r e s s i o n and s u b t r a c t i o n c o r r e c t i o n procedures s c a l e the EOG in t h i s manner. Although the b l i n k amplitude i s decreased p o s t e r i o r l y the q uestion remains as to whether or not the form of the wave changes. Any change in waveform would, of course, have i m p l i c a t i o n s for those c o r r e c t i o n procedures that estimate EEG amplitude on the b a s i s of EOG. Data here i n d i c a t e that the 37 Table 1 Mean Amplitude of Spontaneous B l i n k s ( u v o l t s ) EOG Fz Cz Oz 217.63 54.36 21 .86 12.88 38 o c u l a r p o t e n t i a l may change in form as i t moves back across the s c a l p . C o r r e l a t i o n c o e f f i c i e n t s of spontaneous b l i n k EOG regressed on the corresponding EEG at each e l e c t r o d e placement were c a l c u l a t e d and are presented in Table 2. A one-way repeated measures ANOVA was performed using Z-score t r a n s f o r m a t i o n s of the c o r r e l a t i o n c o e f f i c i e n t s as the dependent v a r i a b l e . The transformations were done s i n c e a l l the c o r r e l a t i o n s are greater than 0, making the d i s t r i b u t i o n between -1 and 1 nonnormal. The ANOVA shows a s i g n i f i c a n t d i f f e r e n c e [F(2,38) = 111.68 , e=.6580, p< .0001] across e l e c t r o d e placements, d e c r e a s i n g i n magnitude p o s t e r i o r l y . T h i s i n d i c a t e s that the s i g n a l recorded at the EEG s i t e becomes i n c r e a s i n g l y d i f f e r e n t from the EOG s i g n a l as the two e l e c t r o d e s i t e s move f u r t h e r a p a r t . The manner i n which b l i n k p o t e n t i a l changes at va r i o u s EEG placements i s not c l e a r . One p o s s i b l e e x p l a n a t i o n of the decreasing c o r r e l a t i o n s i s that the p h y s i c a l c h a r a c t e r i s t i c s of the s c a l p and s k u l l act as a f i l t e r and change the form of the EOG s i g n a l as i t i s conducted to the back of the head. This would r e s u l t in the same p a t t e r n of de c r e a s i n g c o r r e l a t i o n s but would have i m p l i c a t i o n s for procedures that u t i l i z e EOG information to c a l c u l a t e c o r r e c t i o n f a c t o r s . I f the ocu l a r p o t e n t i a l does change shape, b l i n k removal would become l e s s e f f i c i e n t at the p o s t e r i o r s i t e s because the c o r r e l a t i o n c o e f f i c i e n t would r e f l e c t an underestimate of the o c u l a r p o t e n t i a l and too l i t t l e EOG would be s u b t r a c t e d . The e f f e c t of ocu l a r p o t e n t i a l on EEG v a r i e s , and the waveform i t s e l f may be a l t e r e d as i t moves acr o s s the sc a l p , but 39 Table 2 Mean C o r r e l a t i o n C o e f f i c i e n t s Between EEG and EOG - Average of N=20 Subjects Fz C z Oz .89459 .67655 .48193 Table 3 Mean Times of Peak Latency (ms) N= 1 2 EOG Fz Cz Oz 78.33 78.33 80.40 78.66 40 there i s no reason to expect a s i g n i f i c a n t change in the speed of t r a n s m i s s i o n . A n a l y s i s of the l a t e n c y of peak amplitudes at each e l e c t r o d e placement (Table 3) shows that there are no s i g n i f i c a n t d i f f e r e n c e s i n t r a n s m i s s i o n time [F(3,33)=1.07, e=.4697, p=.3424]. In the i n t e r e s t of making the most accurate estimate of the peak of the b l i n k , f o r the above a n a l y s i s , only those s u b j e c t s with strong, c l e a r l y d e f i n e d b l i n k s were used. In t h i s case, the 12 s u b j e c t s with an average EOG of greater than 200 m i c r o v o l t s were used. The above analyses suggest that the b l i n k p o t e n t i a l i s t r a n s m i t t e d e s s e n t i a l l y i n s t a n t a n e o u s l y across the s c a l p but appears to change in form as i t moves from f r o n t to back (see F i g u r e 1). I f the b l i n k p o t e n t i a l does indeed change form p o s t e r i o r l y , the r e g r e s s i o n technique, which r e l i e s h e a v i l y on the form of EOG in the c a l c u l a t i o n of a c o r r e c t i o n c o e f f i c i e n t , might be somewhat l e s s accurate than the s u b t r a c t i o n technique which r e l i e s on EOG amplitude rather than form. Both approaches, however, depend on the EOG form as a b a s i s f o r the p o i n t by point s u b t r a c t i o n of the c o r r e c t i o n f a c t o r , so both would be a f f e c t e d by any change in the waveform. As a t e s t of which c o r r e c t i o n procedure performed b e t t e r , the EEG epochs c o n t a i n i n g spontaneous b l i n k s were c o r r e c t e d using both the r e g r e s s i o n and the s u b t r a c t i o n techniques. The e x p e c t a t i o n i s that the b e t t e r procedure w i l l produce an EEG t r a c e that most c l o s e l y approximates a s t r a i g h t l i n e . Since the spontaneous b l i n k s are assumed to be f r e e of any evoked p o t e n t i a l , once the b l i n k i s removed a l l other EEG a c t i v i t y should c a n c e l with averaging, thus l e a v i n g a s t a i g h t l i n e or 41 F i g u r e 1. S p o n t a n e o u s b l i n k p o t e n t i a l r e c o r d e d s i m u l t a n e o u s l y a t t h e e y e s (EOG) and a t t h r e e EEG p l a c e m e n t s ; F z , Cz, and Oz. The r e c o r d i s t h a t o f one r e p r e s e n t a t i v e s u b j e c t d u r i n g t h e S p o n t a n e o u s B l i n k T r i a l . 42 f l a t r e c o r d . The more s u c c e s s f u l the c o r r e c t i o n procedure the more c l o s e l y the EEG should approach a s t r a i g h t l i n e . As a measure of c o r r e c t e d EEG " f l a t n e s s " , the root mean square (RMS) d e v i a t i o n of each epoch was c a l c u l a t e d . The magnitude of the RMS i s i n c r e a s e d by e i t h e r over- or u n d e r - c o r r e c t i o n and thus i t can be c o n s i d e r e d an absolute measure of b l i n k removal, a l a r g e RMS i n d i c a t i n g poor c o r r e c t i o n of the EEG. A 3 ( e l e c t r o d e s i t e s ) x 2 ( c o r r e c t i o n methods) repeated measures ANOVA was performed on the RMS d e v i a t i o n s of spontaneous b l i n k EEG a f t e r c o r r e c t i o n by each technique (Table 4). A s i g n i f i c a n t d i f f e r e n c e between c o r r e c t i o n techniques was found [F(1,19)=11.64, p<.01] with the s u b t r a c t i o n method producing smaller RMS values at a l l placements. There was a l s o a s i g n i f i c a n t d i f f e r e n c e , independent of method, between RMS d e v i a t i o n at the three placements [F(2,3 8 ) = 5.62, e = . 8 l 8 5 , p< . 0 5 ] with RMS d e v i a t i o n i n c r e a s i n g i n magnitude p o s t e r i o r l y . A l a r g e RMS i n d i c a t e s poor removal of b l i n k a r t i f a c t . The above r e s u l t s suggest that both methods are a f f e c t e d by the changing b l i n k waveform at the p o s t e r i o r e l e c t r o d e s i t e s . B l i n k s i n Response to L i g h t Flashes The spontaneous b l i n k t r i a l p r o vided i n f o r m a t i o n about b l i n k s under non-test s i t u a t i o n s but i t i s a l s o important to understand how b l i n k i n g occurs in r e l a t i o n to the l i g h t f l a s h and how the o c u l a r p o t e n t i a l a f f e c t s the VER. For the purpose of the f o l l o w i n g a n a l y s i s , only b l i n k s that occurred 80 m i l l i s e c o n d s or l a t e r a f t e r the l i g h t f l a s h were i n c l u d e d i n the Table 4 Spontaneous B l i n k Corrected EEG Mean RMS Dev i a t i o n s (jxvolts) - Average of N=20 Subjects Fz Cz Oz Regression 2.99520 3.37984 3.83525 Method Subtract ion 2.53419 2.84679 3.33189 Method Table 5 Range and Mean Number of B l i n k s to L i g h t Flashes I n t e n s i t y T r i a l B l i n k s 1 2 3 4 Low 2 2 1 3 25 1 High 30 50 60 62 Mean 11.0 23.6 44. 1 48.9 Low 0 2 7 1 1 3 High 32 45 63 62 Mean 7.5 18.3 37.0 42.6 44 s u b j e c t ' s average. Since b l i n k s sometimes occur before the l i g h t f l a s h , e s p e c i a l l y at low i n t e n s i t i e s , the 80 ms onset allowed us to i n c l u d e in c a l c u l a t i o n s only those b l i n k s that were a response to the l i g h t f l a s h . The number of b l i n k s at each i n t e n s i t y v a r i e d c o n s i d e r a b l y as can be seen by the mean and range values of Table 5. Since b l i n k i n g i s a r e f l e x i v e r e a c t i o n to b r i g h t l i g h t , an i n c r e a s e i n b l i n k frequency would be expected as l i g h t i n t e n s i t y i n c r e a s e s . A 2-way Repeated Measures ANOVA showed the number of b l i n k s to increase s i g n i f i c a n t l y with l i g h t i n t e n s i t y for both t r i a l s [F(3,57)=109.23, e=.6323, p<.000l]. There was a l s o found to be a s i g n i f i c a n t d i f f e r e n c e [F(1,19)=8.54, p<.0l] between t r i a l s , with mean number of b l i n k s decreasing in T r i a l 3. There was no s i g n i f i c a n t i n t e r a c t i o n between i n t e n s i t y of l i g h t f l a s h and t r i a l . The d i s t r i b u t i o n of b l i n k s at each i n t e n s i t y can be seen in F i g u r e 2. For t h i s F i g u r e , the number of b l i n k s at each l i g h t i n t e n s i t y i s an average over both t r i a l s to s i m p l i f y the p r e s e n t a t i o n of the data. The most common frequency of b l i n k s at the lower two i n t e n s i t i e s of l i g h t i s in the 0-20 b l i n k range whereas at l i g h t i n t e n s i t i e s 3 and 4 the common frequency i s in the 41-64 b l i n k range. These d i f f e r e n c e s are r e f l e c t e d i n c o r r e l a t i o n s between T r i a l 1 and T r i a l 3. Although b l i n k frequency i s r e l i a b l e across t r i a l s (Table 6), b l i n k amplitude i s a r e l i a b l e measure only at l i g h t i n t e n s i t i e s 3 and 4 (Table 7). A glance at Table 7 shows how u n r e l i a b l e b l i n k amplitude can be when measured at l i g h t i n t e n s i t i e s 1 and 2. T h i s i s 6 0 . 0 Intensity i Intensity 2 Intensity 3 Intensity 4 Figure 2. Number of blinks distributed at each light intensity, recorded in EOG. Blinks were 46 Table 6 Number of B l i n k s C o r r e l a t i o n Between T r i a l 1 and 3 I n t e n s i t y 1 2 3 4 .6604 .7844 .6518 .5187 P<.01 P<.001 p<. 01 p<.05 Table 7 B l i n k Amplitude C o r r e l a t i o n Between T r i a l 1 and 3 I n t e n s i t y 1 2 3 4 .3397 .3563 .5944 .6502 p=.167 p=.123 p<. 01 P<.01 47 understandable i n view of the b l i n k d i s t r i b u t i o n data. Amplitude measures are determined by averaging only those epochs in which a b l i n k occurred in response to a l i g h t f l a s h . Since the low i n t e n s i t y b l i n k averages c o n t a i n s i g n i f i c a n t l y fewer b l i n k s than those of high i n t e n s i t i e s ( i n some cases only 2 or 3 b l i n k s ) i t i s not s u r p r i s i n g that they are l e s s r e l i a b l e . Because normal s u b j e c t s tend to be augmenters, an in c r e a s e in response amplitude with increased l i g h t i n t e n s i t y should be seen. T h i s amplitude i n c r e a s e , being a r e l i a b l e phenomenon, should not change from T r i a l 1 to T r i a l 3. A 2-way repeated measures ANOVA showed the peak amplitudes to in c r e a s e across i n t e n s i t i e s [F(3,59)=7.89, e=.63l5, p<.01] but showed no d i f f e r e n c e between T r i a l 1 and T r i a l 3 [F(3,51)=0.71, e=.5535, p=.476]. Along with the data i n Table 5, t h i s means the number of b l i n k s decreases s i g n i f i c a n t l y , but uniformly for a l l s u b j e c t s , between T r i a l 1 and T r i a l 3 while the magnitude of those b l i n k s remains the same (Table 8 ) . Important b l i n k dimensions, i n terms of how the VER w i l l be a f f e c t e d , are the la t e n c y of onset and la t e n c y of peak amplitude. In order to understand which p o r t i o n of the VER w i l l be most a f f e c t e d by the b l i n k p o t e n t i a l , the mean latency of peak amplitude must be estimated. These values f o r each l i g h t i n t e n s i t y are presented i n Table 9. A 2-way repeated measures ANOVA demonstrated a s i g n i f i c a n t d i f f e r e n c e [F(3,51) = 1 4.30 , e=.5280, p<.000l] between peak l a t e n c i e s across i n t e n s i t i e s but no s i g n i f i c a n t d i f f e r e n c e between t r i a l s [F(1,17)=3.48, p=.080]. As demonstrated e a r l i e r , measures at the lower l i g h t i n t e n s i t i e s 48 Table 8 Mean Bl i n k Amplitude ( y v o l t s ) I n t e n s i t y T r i a l 1 2 3 4 1 145.29 107.34 147.21 170.85 3 117.31 109 .81 134.61 152.69 49 must be i n t e r p r e t e d c a u t i o u s l y due to infrequent and u n r e l i a b l y e l i c i t e d b l i n k s . Latency to b l i n k onset was d e f i n e d i n t h i s study as the time when the b l i n k was at 10 percent of i t s peak amplitude. For a b l i n k with a peak amplitude of 300 m i c r o v o l t s , onset would be considered that point i n time when the b l i n k measured 30 m i c r o v o l t s . Measured in t h i s f a s h i o n the mean b l i n k onset times are presented i n Table 10. A 2-way repeated measures Anova showed that there was no s i g n i f i c a n t d i f f e r e n c e between mean onset l a t e n c i e s in T r i a l 1 and those of T r i a l 3 [F(1,17)=0.32, P=.5811]. There was, however, a s i g n i f i c a n t d i f f e r e n c e between onset l a t e n c i e s across l i g h t i n t e n s i t i e s [F(3,51)=11.70, e=.3772, p<.000l]. I n t e n s i t i e s 3 and 4, being the b r i g h t e s t l i g h t f l a s h e s , are the c o n d i t i o n s when the subj e c t s b l i n k most often and with gr e a t e s t magnitude. I t i s during these two c o n d i t i o n s that b l i n k p o t e n t i a l i s most troublesome as an a r t i f a c t in EEG re c o r d s . As can be seen in Tables 9 and 10, the b l i n k at i n t e n s i t i e s 3 and 4 has an onset in the range of 107 to 118 m i l l i s e c o n d s and a peak lat e n c y of 228 to 235 m i l l i s e c o n d s . On the ba s i s of t h i s information i t appears that the b l i n k p o t e n t i a l w i l l have l i t t l e or no a p p r e c i a b l e e f f e c t on the P, component (60-112 ms) of the VER; p o s s i b l y some e f f e c t on the N, component (80-150 ms); and the strongest e f f e c t on the P 2 component (160-250 ms). Ev a l u a t i o n of C o r r e c t i o n Procedures From the a n a l y s i s of b l i n k c h a r a c t e r i s t i c s d u r i n g l i g h t Table 9 Mean Peak L a t e n c i e s of B l i n k s (ms) I n t e n s i t y T r i a l 1 2 3 4 1 428 310 228 236 3 374 265 229 235 Table 10 Mean Onset L a t e n c i e s of B l i n k s (ms) T r i a l I n t e n s i t y 2 3 209 175 1 22 1 1 5 1 07 1 1 4 1 1 3 1 18 51 f l a s h e s presented in the previous s e c t i o n , i t appears that the P 2 wave i s the p o r t i o n of the VER most a f f e c t e d by ocular p o t e n t i a l s . To assess how b l i n k s i n f l u e n c e v a r i o u s p o r t i o n s of the VER, the EOG b l i n k amplitude, averaged at each l i g h t i n t e n s i t y , was c o r r e l a t e d with P 1 f N,, and P 2 amplitudes at the same i n t e n s i t i e s . T h i s was done f o r each e l e c t r o d e placement. Small c o r r e l a t i o n s were found between b l i n k amplitude and both P, and N, amplitudes f o r a l l three methods (see appendix I I , Tables 1-2) while l a r g e c o r r e l a t i o n s were found between P 2 and b l i n k amplitude only f o r the uncorrected data (see the lower boxes i n Tables 11 and 12). Tables 11 and 12 present only the c o r r e l a t i o n s of b l i n k amplitude with P 2 amplitude at l i g h t i n t e n s i t i e s 3 and 4 because s u b j e c t s blinked, so r a r e l y at the lower two i n t e n s i t i e s (see Table 5 and Figure 2) that a r e l i a b l e b l i n k average c o u l d not be c a l c u l a t e d . T h i s problem was compounded i n the data where b l i n k contaminated epochs were removed ( b l i n k r e j e c t i o n method) because there were too few nonblink epochs at the highest two i n t e n s i t i e s ( Table 5 and Fi g u r e 2) to produce a r e l i a b l e average. Consequently, the b l i n k r e j e c t i o n data were excluded e n t i r e l y from Tables 11 and 12. The uncorrected data c o r r e l a t e s i g n i f i c a n t l y with EOG amplitude at only the Fz placement during T r i a l 1 but s i g n i f i c a n t c o r r e l a t i o n s are found at a l l three placements during T r i a l 3. This suggests that b l i n k i n g during t r i a l 3 has a greater e f f e c t at p o s t e r i o r s i t e s than i t does in T r i a l 1 . The data i n Tables 11 and 12, as well as those' in Appendix 52 Table 11 P 2 Amplitudes C o r r e l a t e d with EOG Amplitude ( T r i a l 1) Method Sub Reg UnCor E l e c t r o d e Placement Fz Cz Oz Fz Cz Oz Fz Cz Oz I n t e n s i t y 3 4 .0051 .0327 .0125 -.0354 -.0159 .0668 .0538 .1413 .0412 .0815 -.0482 .0667 .6061 .6388 .3987 .3266 .2428 . 1355 Table 12 P 2 Amplitudes C o r r e l a t e d with EOG Amplitude ( T r i a l 3) Method E l e c t r o d e Placement I n t e n s i t y 3 Sub Reg UnCor Fz Cz Oz Fz Cz Oz Fz Cz Oz .3996 .31 56 .2724 .4373 .0627 .0156 .1812 .3836 .3058 .1839 .21 52 .1213 .6730 .4831 .4561 .6089 .4844 .4707 Note. Sub=Subtraction, Reg=Regression, Uncor=Uncorrected 53 I I , i n d i c a t e that b l i n k amplitude c o r r e l a t e s s u b s t a n t i a l l y with only the P 2 p o r t i o n of the VER. T h i s suggests that b l i n k p o t e n t i a l d i d not a p p r e c i a b l y a f f e c t e i t h e r P, or N,. It i s expected that the o c u l a r p o t e n t i a l would account f o r a l a r g e p o r t i o n of the v a r i a n c e i n the EEG, e s p e c i a l l y at Fz, such that EOG amplitude would c o r r e l a t e s i g n i f i c a n t l y with that p o r t i o n of the VER that c o n t a i n s a s u b s t a n t i a l amount of b l i n k p o t e n t i a l . Tables 11 and 12 show that both the s u b t r a c t i o n and the r e g r e s s i o n methods have e f f e c t i v e l y removed the b l i n k a r t i f a c t , as the c o r r e l a t i o n s are small at a l l placements for both t r i a l s . In summary, the above data demonstrate, f o r t h i s group of s u b j e c t s , that b l i n k p o t e n t i a l a f f e c t s the P 2 component of the VER but does not a l t e r the P, or N, components. Both the r e g r e s s i o n and s u b t r a c t i o n techniques appear to e f f e c t i v e l y remove the b l i n k a r t i f a c t , showing no c o n s i s t e n t c o r r e l a t i o n with b l i n k amplitude at any e l e c t r o d e placement. Because the b l i n k p o t e n t i a l i s a f f e c t i n g P 2, i t i s t h i s component that w i l l be used in the f o l l o w i n g e v a l u a t i o n s of c o r r e c t i o n methods. To assess the r e t e s t r e l i a b i l i t y of each of the three approaches ( s u b t r a c t i o n , r e g r e s s i o n , and b l i n k r e j e c t i o n ) , as w e l l as the r e l i a b i l i t y of uncorrected data, P 2 amplitudes of s u b j e c t ' s VERs during T r i a l 1 were c o r r e l a t e d with those of T r i a l 3 (see Table 13). C o r r e l a t i o n s between t r i a l s of the s u b t r a c t i o n method above are of the same magnitude as those repo r t e d f o r P^N, measures, over one week, by Iacono, Gabbay and Lykken (1982, Table 1). They used an o n l i n e gain adjustment of the EOG to make b l i n k amplitude approximately the same s i z e 54 Table 13 P 2 Amplitudes - T r i a l 1 c o r r e l a t e d with T r i a l 3 E l e c t r o d e Placement Method 1 I n t e n s i t y 2 3 4 Mean Fz Sub Reg B Rej UnCor .5358 .5787 .6983 .6451 .8899 .7506 .5904 .7423 .8448 .8170 -.1463 .6734 .7269 .5945 .7522 .7241 .7617 .6926 .5965 .6973 Cz Sub Reg B Rej UnCor .6461 .6050 .4729 .5874 .7105 .6959 .5969 .8038 .8700 .81 37 -.0436 .8606 .8296 .7339 .4533 .821 5 .7693 .7161 .4436 .7756 Oz Sub Reg B Rej UnCor .7297 .71 58 .621 1 .7265 .6458 .6818 .61 55 .6603 .8817 .7888 .0727 .8270 .8520 .6904 . 1 654 .8678 .7830 .7204 .1912 .7747 Note. Sub=Subtraction, Reg=Regression, B Rej=Blink R e j e c t i o n , Uncor U n c o r r e c t e d 55 as the b l i n k a r t i f a c t i n the EEG while the subject was asked to b l i n k i n a t o t a l l y darkened chamber. They then s u b t r a c t e d the averaged EOG from the VER o f f l i n e . The modified s u b t r a c t i o n method employed in the present study i s shown to be s l i g h t l y more r e l i a b l e than the r e g r e s s i o n technique at a l l e l e c t r o d e placements (see mean i n t e n s i t y values, Table 13). C o r r e l a t i o n means were computed by f i r s t squaring each c o e f f i c i e n t , c a l c u l a t i n g the mean of the squared values and then taking the square root of t h i s mean. Ins p e c t i o n of the b l i n k r e j e c t i o n data in Table 13 shows that r e l i a b i l i t y i s moderate for a l l e l e c t r o d e placements at the lower two l i g h t i n t e n s i t i e s but r e l i a b i l i t y decreases d r a m a t i c a l l y at the higher two i n t e n s i t i e s . T h i s i s c o n s i s t e n t with data presented in Table 5 and F i g u r e 6, supporting the argument that too few nonblink epochs are a v a i l a b l e at the higher i n t e n s i t i e s to produce a r e l i a b l e VER estimate. The uncorrected data show moderate to high r e l i a b i l i t y at a l l i n t e n s i t i e s . At i n t e n s i t i e s 1 and 2 the c o r r e l a t i o n s l i k e l y r e f l e c t the VER r e l i a b i l i t y since there i s minimal b l i n k i n g to i n t e r f e r e with EEG r e c o r d i n g . In c o n t r a s t , the c o r r e l a t i o n s remain high at i n t e n s i t i e s 3 and 4 due to the good r e l i a b l i l i t y of the b l i n k s themselves at these i n t e n s i t i e s (see Table 7). As a r e s u l t the uncorrected data, although appearing to r e l i a b l y r e f l e c t the VER at a l l e l e c t r o d e placements and l i g h t i n t e n s i t i e s , may be d i f f e r e n t i a l l y r e f l e c t i n g the r e l i a b i l i t y of two parameters, VER and b l i n k i n g . As a means of comparing these procedures with each other, a 56 c o r r e l a t i o n matrix was c o n s t r u c t e d (see Table 14) in which a l l methods, i n c l u d i n g eyes c l o s e d and b l i n k r e j e c t i o n , were c o r r e l a t e d with each other. For the purposes of t h i s matrix, the P 2 data were averaged over both t r i a l s of the s u b t r a c t i o n , r e g r e s s i o n , b l i n k r e j e c t i o n , and uncorrected c o n d i t i o n s , and c o r r e l a t e d with data from the s i n g l e eyes c l o s e d t r i a l . In l i g h t of the moderate to high c o r r e l a t i o n s found between T r i a l s 1 and 3 t h i s was f e l t to be a v a l i d way to compare the above procedures with data from the s i n g l e eyes c l o s e d t r i a l . Table 14 shows that the r e g r e s s i o n and s u b t r a c t i o n method have the highest c o r r e l a t i o n with each other at a l l three e l e c t r o d e placements, i n d i c a t i n g more commonality between these two methods than any of the o t h e r s . At f i r s t glance the b l i n k r e j e c t i o n c o n d i t i o n shows a s u r p r i s i n g l y high c o r r e l a t i o n with both the r e g r e s s i o n and s u b t r a c t i o n methods but on i n s p e c t i o n t h i s c o r r e l a t i o n can be seen as the r e s u l t of very strong c o r r e l a t i o n s at the lower i n t e n s i t i e s and weaker c o r r e l a t i o n s at the two high l i g h t i n t e n s i t i e s (see appendix I I , Tables 3 - 6 ) . T h i s i s c o n s i s t e n t with the argument made e a r l i e r that the b l i n k r e j e c t i o n c o n d i t i o n i s u n r e l i a b l e at the higher l i g h t i n t e n s i t i e s . The uncorrected c o n d i t i o n a l s o c o r r e l a t e s moderately with the s u b t r a c t i o n and r e g r e s s i o n methods. I n s p e c t i o n of Table 14 shows that c o r r e l a t i o n s between unco r r e c t e d data and the s u b t r a c t i o n or r e g r e s s i o n data are lowest at Fz and highest at Oz. T h i s i s expected since the b l i n k p o t e n t i a l in the uncorrected data would have i t s g r e a t e s t e f f e c t at Fz, thus reducing the c o r r e l a t i o n . The reverse i s true at Oz. 57 Table 14 C o r r e l a t i o n s Between Methods (P 2 Amplitudes averaged across i n t e n s i t i e s ) 1 Method Method E l e c t r o d e Placement Sub Reg B Rej UnCor EC Sub 1 .0 Reg Fz Cz Oz Mean .8675 .8403 .9057 .8716 1 .0 B Rej Fz Cz Oz .61 46 .8037 .8227 .6431 .7445-.7976 1 .0 Mean .7528 .7312 UnCor Fz Cz Oz .4671 .7717 .8752 .4458 .6375 .761 1 .7066 .621 0 .81 05 1 .0 Mean .7256 .6283 .71 69 Eyes Closed Fz Cz Oz .3574 .4090 .3560 .4265 .491 0 .4227 . 1876 .3365 .2724 .2069 .3557 .3204 1 .0 Mean .3749 .4478 .2724 .301 1 Note. Sub=Subtraction, Reg=Regression, B Rej=Blink R e j e c t i o n , Uncor=Uncorrected, EC=Eyes Closed. 1. For a l l but the eyes c l o s e d method, data were averaged across T r i a l 1 and T r i a l 3. 58 Here the b l i n k has r e l a t i v e l y l i t t l e e f f e c t on P 2 and the VER i s l a r g e s t because the Oz e l e c t r o d e i s over primary v i s u a l c o r t e x . T h i s r e s u l t s i n a higher c o r r e l a t i o n between uncorrected data and the two c o r r e c t i o n methods. The eyes c l o s e d data show weak c o r r e l a t i o n s with data of a l l the other methods. This suggests that the VERs (or at l e a s t the P 2 peaks) produced by an eyes c l o s e d approach have l i t t l e in common with the VERs of other procedures. When data from a l l s u b j e c t s i n t h i s study are i n c l u d e d in the a n a l y s i s of the b l i n k r e j e c t i o n method i t appears to produce l e s s r e l i a b l e VER data that those of the r e g r e s s i o n and s u b t r a c t i o n methods. However, i f only those subjects with r e l a t i v e l y few b l i r k contaminated epochs are examined, the b l i n k r e j e c t i o n method produces s u b s t a n t i a l l y b e t t e r r e s u l t s . Table 1 5 shows the c o e f f i c i e n t s of the 5 procedures c o r r e l a t e d with each other when only s u b j e c t s who had 40% or more b l i n k f r e e epochs at l i g h t i n t e n s i t y 3 were i n c l u d e d . The b l i n k r e j e c t i o n method in t h i s case c o r r e l a t e s h i g h l y with both the r e g r e s s i o n and s u b t r a c t i o n methods. When these r e s u l t s are compared to those at l i g h t i n t e n s i t y 3 with data from a l l s u b j e c t s included (see Appendix I I , Table 5 ) a marked improvement, e s p e c i a l l y at the Fz e l e c t r o d e placement, i s seen. These r e s u l t s suggest that i f enough nonblink epochs are produced by a subject, r e j e c t i n g b l i n k contaminated epochs can y i e l d data s i m i l a r to that of the r e g r e s s i o n and s u b t r a c t i o n methods. With the VER procedures used i n t h i s study the problem a r i s e s that many sub j e c t s do not produce a s u f f i c i e n t number of nonblink epochs to make t h i s 59 Table 15 C o r r e l a t i o n s Between Methods ( P 2 Amplitudes at I n t e n s i t y 3 Only, N=9) Method Method E l e c t r o d e Placement Sub Reg B Rej UnCor EC Sub 1 .0 Reg Fz Cz Oz Mean .9055 .9375 .9455 .9296 1 .0 B Rej Fz Cz Oz .8031 .9727 .9736 .7345 .9641 .9236 1 .0 Mean • 8 9 1 9 .8797 UnCor Fz Cz Oz .5272 .9483 .9658 .3079 .81 74 .8468 .8363 .9201 .9632 1 .0 Mean .8386 .7023 .9080 Eyes Closed Fz Cz Oz .4689 .4959 .4604 .4864 . 4443 .6036 .4105 .4462 . .4915 .3082 .4659 .3912 1 .0 Mean .4753 .51 58 .4506 .3937 Note. Sub=Subtraction, Reg=Regression, B Rej=Blink R e j e c t i o n , Uncor=Uncorrected, EC=Eyes C l o s e d . 60 approach u s e f u l . At l i g h t i n t e n s i t y 3 only 9 of 20 subj e c t s met the 40% b l i n k free epochs c r i t e r i o n , and at i n t e n s i t y 4 only 5 of 20 s u b j e c t s f e l l in t h i s category. To t h i s point i n the d i s c u s s i o n , data have been presented i n d i c a t i n g that the s u b t r a c t i o n technique may produce s t a t i s t i c a l l y more v a l i d VERs than those of the r e g r e s s i o n technique (Table 4), but both techniques markedly reduce the b l i n k a r t i f a c t (Tables 11 and 12) and both techniques are h i g h l y r e l i a b l e (Table 13). Aside from t h e o r e t i c a l arguments against using the eyes c l o s e d and b l i n k r e j e c t i o n methods (Iacono, 1982), data presented here suggest that the b l i n k r e j e c t i o n approach produces u n r e l i a b l e r e s u l t s at the higher l i g h t i n t e n s i t i e s . The b l i n k r e j e c t i o n method appears to y i e l d good r e s u l t s i f subjects produce a s u f f i c i e n t number of b l i n k f r e e epochs, but most s u b j e c t s i n t h i s study b l i n k e d too often at i n t e n s i t i e s 3 and 4 to make t h i s method u s e f u l . The eyes c l o s e d c o n d i t i o n does not appear to y i e l d VERs that are s i m i l a r to the other methods at any l i g h t i n t e n s i t y or e l e c t r o d e p o s i t i o n . To evaluate whether these d i f f e r e n c e s in methods, deduced from the above c o r r e l a t i o n a l data, t r a n s l a t e i n t o meaningful d i f f e r e n c e s at the experimental l e v e l , an a n a l y s i s of va r i a n c e was c a r r i e d out. A repeated measures 3-way ANOVA was performed using uncorrected, s u b t r a c t i o n , r e g r e s s i o n , and eyes c l o s e d data (4 methods x 3 e l e c t r o d e placements x 4 i n t e n s i t i e s ) . The dependent measure was P 2 amplitude scores averaged over two t r i a l s of the s u b t r a c t i o n , r e g r e s s i o n and uncorrected methods along with the 61 s i n g l e t r i a l eyes c l o s e d data. R e s u l t s of t h i s ANOVA r e v e a l that P 2 amplitude i n c r e a s e s s i g n i f i c a n t l y with l i g h t i n t e n s i t y [F(3,57)=83.99, e=.6282, p<.000l]. There was a l s o a s i g n i f i c a n t intensity-by-method i n t e r a c t i o n [F(9,171)=53.72, e=.3607, P<.0001] as d i s p l a y e d in Figure 3. The mean of the eyes c l o s e d c o n d i t i o n was found to be s i g n i f i c a n t l y higher at i n t e n s i t y 4 (Newman-Keuls comparison; Q=9.26, CV 0 1=3.64) and i n t e n s i t y 3 (Q=7.24, CV 0 1=3.64) than both s u b t r a c t i o n and r e g r e s s i o n methods and s i g n i f i c a n t l y lower at i n t e n s i t y 1 (Q=5.99, CV 0 1=3.64). One explanation f o r t h i s i n t e r a c t i o n i s that c l o s e d e y e l i d s attenuate the l i g h t f a l l i n g on the r e t i n a and i n turn attenuate the VER at i n t e n s i t y 1. At i n t e n s i t y 4 ocular p o t e n t i a l , due to r e f l e x i v e r o t a t i o n of the e y e b a l l , contaminates the VER making i t l a r g e r than i t should be. Iacono et a l . (1982) reported s i m i l a r eye movements i n t h e i r s u b j e c t s during p i l o t t e s t i n g with the eyes c l o s e d technique. The uncorrected means are s i g n i f i c a n t l y greater than those of the other methods at a l l l i g h t i n t e n s i t i e s . No s i g n i f i c a n t d i f f e r e n c e s were found between the means of the s u b t r a c t i o n and r e g r e s s i o n methods at any i n t e n s i t y . An i n t e n s i t y - b y - e l e c t r o d e i n t e r a c t i o n was a l s o found [F(6,114)=20.32, e=.4985, p<.000l]. F i g u r e 4 shows that the mean of Fz i s s i g n i f i c a n t l y lower at i n t e n s i t y 1 (Q=3.93, CV o l=3.70) than the mean of e i t h e r Cz or Oz. However, at i n t e n s i t y 4 the mean of Fz i s s i g n i f i c a n t l y l a r g e r than that of Cz (Q=15.26, CV o 1=3.70) but not s i g n i f i a n t l y l a r g e r than that of Oz (Q=1.34, CV O 5=2.80). T h i s i n t e r a c t i o n i s l i k e l y due to the c o n t r i b u t i o n 6 2 2 3 Intensity" Figure 3 . Visual evoked response amplitude by light intensity, using different methods to reduce blink artifact. 63 of the uncorrected data and the l a r g e e f f e c t b l i n k s have at the Fz p o s i t i o n as l i g h t i n t e n s i t y i n c r e a s e s . At lower l i g h t i n t e n s i t i e s , where b l i n k s are not an important f a c t o r , the VER i s n a t u r a l l y smaller at the Fz placement (Hassett, 1978; p.119) because f r o n t a l cortex i s l e s s r e a c t i v e to v i s u a l s t i m u l i than o c c i p i t a l or p a r i e t a l c o r t e x . At higher i n t e n s i t i e s , however, the b l i n k p o t e n t i a l becomes the main c o n t r i b u t o r to the P 2 peak and i n f l a t e s the Fz mean. The Fz mean i s not s i g n i f i c a n t l y l a r g e r than that of Oz at i n t e n s i t y 4 because while the b l i n k p o t e n t i a l has l e s s i n f l u e n c e at Oz, the VER there i s g r e a t e s t (Hassett, 1978; p. 120). The e l e c t r o d e main e f f e c t [F(2,38)=8.78, e=.6334, p<.00l] more g r a p h i c a l l y i l l u s t r a t e s the above phenomenon (Figure 5). Due to a l a r g e i n f l u e n c e of the b l i n k p o t e n t i a l at Fz t h i s mean i s l a r g e r than that of Cz, but the Oz mean, due. to a strong VER i n f l u e n c e , i s the l a r g e s t of the t h r e e . F i g u r e 6 i l l u s t r a t e s the i n t e r a c t i o n of method and e l e c t r o d e placement [F(6,114)=62.30, e=.2476, p<.000l] and shows how the two c o r r e c t i o n procedures decrease the e f f e c t of b l i n k p o t e n t i a l at the Fz placement. The mean of the eyes c l o s e d c o n d i t i o n at Fz, being s i g n i f i c a n t l y l a r g e r than those of the two c o r r e c t i o n methods (Q=11.85, CV o 1=3.70), supports the argument made e a r l i e r (see F i g u r e 3) that d e s p i t e having t h e i r eyes c l o s e d subjects s t i l l produce o c u l a r p o t e n t i a l i n response to b r i g h t l i g h t s , probably as a r e s u l t of e y e b a l l r o t a t i o n . I t i s at the Fz placement that we would expect to see the greatest e f f e c t of t h i s problem. A f u r t h e r problem of the eyes c l o s e d 6 4 -I 1 4 (_ 1 2 3 4 I n t e n s i t y F i g u r e 4. V i s u a l evoked response amplitude by l i g h t i n t e n s i t y , recorded a t each of three e l e c t r o d e placements. P2 average of 4 methods. 65 E l e c t r o d e Placement F i g u r e 5. V i s u a l evoked response amplitude by e l e c t r o d e placement, average of a l l methods. 66 2 5 . 0 #)- Uncorrected A \ - Eyes closed Qi— Subtraction 0 - Regression 2 0 . 0 • • 1 5 . 0 w •P o > 3 0) T3 3 •p a, CN a*. 1 0 . 0 - • 5 . 0 Fz Cz Oz Electrode Placement Figure 6 . Visual evoked response amplitude by electrode placement using different methods to reduce blink artifact. 67 c o n d i t i o n i s i l l u s t r a t e d at the Oz placement, i n F i g u r e 6, where the mean i s s i g n i f i c a n t l y smaller than those of the two c o r r e c t i o n methods (Q=3.14, CV 0 5=2.80). T h i s small mean suggests that at the p o s t e r i o r s i t e , where o c u l a r p o t e n t i a l has l i t t l e e f f e c t , the e y e l i d attenuates the amount of l i g h t reaching the r e t i n a and thus produces a small VER. There i s no s i g n i f i c a n t d i f f e r e n c e found between the s u b t r a c t i o n and r e g r e s s i o n methods at any electode placement. Not s u r p r i s i n g l y , the 3-way ANOVA a l s o showed a s i g n i f i c a n t method e f f e c t [F(3,57)=55.70, e=.6334, p<.0001] which i s i l l u s t r a t e d i n Fi g u r e 7. As expected the mean of the uncorrected data, due to i n f l a t i o n by the b l i n k a r t i f a c t , i s s i g n i f i c a n t l y l a r g e r than those of the other three methods (Q=13.08, CV O i=0.78) but the eyes c l o s e d mean, although l a r g e r than the two c o r r e c t i o n methods was not s i g n i f i c a n t l y so (Q=2.41, CV 0 1=2.84). Looking back to Fig u r e 3 we can see that a s i g n i f i c a n t l y lower mean at i n t e n s i t y 1 and a s i g n i f i c a n t l y l a r g e r mean at i n t e n s i t y 4 has l e d to a mean, when averaged across i n t e n s i t i e s , that i s not s i g n i f i c a n t l y d i f f e r e n t from those of the r e g r e s s i o n and s u b t r a c t i o n methods. In a d d i t i o n to the above main e f f e c t s and two-way i n t e r a c t i o n s , a s i g n i f i c a n t three way method-by-electrode-by-i n t e n s i t y i n t e r a c t i o n was found [F(18,342)=52.34, e=.2047, p< .0001], I n t e r p r e t a t i o n of t h i s i n t e r a c t i o n i s very d i f f i c u l t and i t does not seem l i k e l y these r e s u l t s would add s u b s t a n t i a l l y to the present a n a l y s i s . The e f f e c t s of a l l four methods ( s u b t r a c t i o n , r e g r e s s i o n , b l i n k r e j e c t i o n , and eyes c l o s e d ) on 68 Regression S u b t r a c t i o n Eyes Uncorrected Closed Method F i g u r e 7 . V i s u a l evoked response amplitude by method of a r t i f a c t c o n t r o l , average of a l l l i g h t i n t e n s i t i e s . 69 VER U n c o r r e c t e d S u b t r a c t i o n R e g r e s s i o n EOG 20 — U n c o r r e c t e d — Eyes C l o s e d B l i n k Rej. 1 00 ms F i g u r e 8 . The e f f e c t s o f 4 methods o f d e a l i n g w i t h b l i n k a r t i f a c t a r e c o n t r a s t e d w i t h t h e v i s u a l evoked r e s p o n s e . The VER and v e r t i c a l EOG a r e from one s u b j e c t i n r e s p o n s e t o l i g h t i n t e n s i t y 2 d u r i n g T r i a l 1. R e c o r d i n g s were made a t F z . 70 the uncorrected data are i l l u s t r a t e d i n f i g u r e 8. The VERs and EOG are from one r e p r e s e n t a t i v e s u b j e c t and were recorded d u r i n g T r i a l 1 . To summarize, the r e s u l t s of the ANOVA i n d i c a t e that i n terms of method-main e f f e c t s , the uncorrected data are s i g n i f i c a n t l y d i f f e r e n t from the data of the other three methods: eyes c l o s e d , r e g r e s s i o n , and s u b t r a c t i o n . Further a n a l y s i s shows that the eyes c l o s e d method a l s o d i f f e r s s i g n i f i c a n t l y from the two c o r r e c t i o n procedures, y i e l d i n g smaller means at i n t e n s i t y 1 and l a r g e r means at i n t e n s i t i e s 3 and 4. This suggests d i f f i c u l t i e s with the eyes c l o s e d procedure. The method-by-electrode i n t e r a c t i o n f u r t h e r i n d i c a t e s the problem with the eyes c l o s e d method, showing i t to produce l a r g e r means at Fz, l i k e l y due to oc u l a r p o t e n t i a l , and lower means at Oz, l i k e l y due to a t t e n u a t i o n of l i g h t s t i m u l i by the e y e l i d s . The e l e c t r o d e main e f f e c t confirms a l a r g e mean at Oz, c o n s i s t e n t with e a r l i e r evidence that P2 i s l a r g e s t over the o c c i p i t a l r e g i o n . The e l e c t r o d e e f f e c t a l s o r e v e a l s , not unexpectedly, a l a r g e mean at Fz where the b l i n k a r t i f a c t has i t s g r e a t e s t e f f e c t . There were no s i g n i f i c a n t d i f f e r e n c e s between the r e g r e s s i o n and s u b t r a c t i o n procedures in the method main e f f e c t or any of the method i n t e r a c t i o n s . 71 General D i s c u s s i o n and Summary Iacono et a l . (1982) argued that the eye b l i n k e x e r t s a s u f f i c i e n t i n f l u e n c e on VER measurements to warrant monitoring of such a r t i f a c t s d u r i n g evoked p o t e n t i a l r e c o r d i n g . R e s u l t s here are c o n s i s t e n t with t h i s p o s i t i o n i n that data uncorrected f o r eye movements were s i g n i f i c a n t l y d i f f e r e n t from those that had such a r t i f a c t removed. In c o n t r a s t to the Iacono et a l . (1982) f i n d i n g that the primary e f f e c t of a b l i n k was on the N, component (80-150 ms), i n t h i s study b l i n k s had a pronounced i n f l u e n c e i n the 160-254 ms range, i n c r e a s i n g the P 2 p o r t i o n of the VER. Our r e s u l t s are c o n s i s t e n t with those of Soskis and Shagass (1974) who concluded that b l i n k i n g d i d not i n f l u e n c e the VER during the f i r s t 150 ms. The d i f f e r e n c e s in b l i n k l a t e n c i e s among the above s t u d i e s may be a r e s u l t of d i f f e r e n c e s i n p h o t o s t i m u l a t o r s used to present the l i g h t f l a s h e s . Iacono et a l . (1982) used a Grass model PS-2 photostimulator commonly employed in e a r l y augmenting/reducing s t u d i e s . T h i s o l d e r type photostimulator probably e l i c i t e d s h o r t e r l a t e n c y b l i n k s because i t generated b r i g h t e r l i g h t f l a s h e s than models used in other s t u d i e s , i n c l u d i n g t h i s one. Given that b l i n k s exert a systematic and s i g n i f i c a n t e f f e c t on the VER, they must be d e a l t with i n some fa s h i o n before the VER data can be c o n s i d e r e d v a l i d . The most commonly used approach i s .to omit EEG epochs that c o n t a i n a b l i n k a r t i f a c t (Iacono et a l . , 1982). There are two reasons, t h e o r e t i c a l and p r a c t i c a l , why t h i s r e j e c t i o n approach i s u n d e s i r a b l e . From a 72 t h e o r e t i c a l standpoint i t i s not c l e a r that EEG epochs where no b l i n k s have occurred are e q u i v a l e n t to those epochs where b l i n k s have occu r r e d . If the two epochs are not e q u i v a l e n t , EEG averages c a l c u l a t e d with nonblink t r i a l s would produce an innaccurate VER. T h e o r e t i c a l arguments a s i d e , the data l o s s i n c u r r e d by the b l i n k r e j e c t i o n technique p r o h i b i t s i t s use i n the evoked p o t e n t i a l procedures u t i l i z e d i n t h i s and s i m i l a r s t u d i e s . As was demonstrated here, the data l o s s at the highest two l i g h t i n t e n s i t i e s i s great enough to make VER measures h i g h l y u n r e l i a b l e and s i g n i f i c a n t l y d i f f e r e n t in magnitude from those obtained from the two c o r r e c t i o n procedures. Having sub j e c t s f i x a t e t h e i r gaze, or by some other means avoid eye movements, i s not a s u i t a b l e a l t e r n a t i v e because c e r t a i n c l a s s e s of s u b j e c t s such as c h i l d r e n and p s y c h i a t r i c or n e u r o l o g i c a l p a t i e n t s cannot or w i l l not cooperate (Gratton et a l . , 1982). One a l t e r n a t i v e to the b l i n k r e j e c t i o n method i s to present the s t i m u l i through c l o s e d eye l i d s . T h i s procedure a l s o has drawbacks since eye l i d t r a n s l u c e n c y and t h i c k n e s s , and s k i n pigmentation no doubt a f f e c t the stimulus i n t e n s i t y at the r e t i n a . Attenuation of the VER to l i g h t i n t e n s i t y 1 was seen i n the eyes c l o s e d c o n d i t i o n , but t h i s e f f e c t was not apparent at the higher i n t e n s i t i e s because eye movement a r t i f a c t seemed to play a more important r o l e as stimulus b r i g h t n e s s inc r e a s e d . If eye movement occurs behind c l o s e d l i d s , any advantage the eyes c l o s e d procedure might have i s e l i m i n a t e d . At low stimulus i n t e n s i t y , where b l i n k s are not a s i g n i f i c a n t problem, c l o s e d eye l i d s appears to d i m i n i s h the r e t i n a l i l l u m i n a t i o n thereby 73 decreasing P 2 of the VER, and at high s t i m u l u s i n t e n s i t y a systematic eye movement a r t i f a c t may i n f l a t e the P 2 component. The eyes c l o s e d approach does not appear to be an adequate method of d e a l i n g with the problem of b l i n k a r t i f a c t . Since r e j e c t i o n of b l i n k epochs and the eyes c l o s e d approach are not s u i t a b l e methods f o r d e a l i n g with b l i n k s , c o r r e c t i o n of contaminated data appears to be a necessary procedure. C o r r e c t i o n methods have t y p i c a l l y f a l l e n i n t o two c a t e g o r i e s ; those that s u b t r a c t a p o r t i o n of the EOG from the EEG based on a v i s u a l e s t i m a t i o n ( u s u a l l y o n l i n e ) of what p o r t i o n of raw EEG can be a t t r i b u t e d to o c u l a r a c t i v i t y (Corby & K o p e l l , 1972; G i r t o n & Kamiya, 1973; Iacono et a l . , 1982) or those that use a more s o p h i s t i c a t e d r e g r e s s i o n a n a l y s i s to determine a c o r r e c t i o n f a c t o r (Verleger et a l . , 1982; Gratton et a l . , 1983). Both these approaches assume t h a t the s c a l p p o t e n t i a l i s a l i n e a r summation of b r a i n and ocular p o t e n t i a l . T h i s means s u b t r a c t i o n of the ocular p o t e n t i a l from the EEG record should leave a " c l e a n " c e r e b r a l p o t e n t i a l . These techniques, then, depend on both the assumption of l i n e a r i t y and the accuracy of the c o r r e c t i o n f a c t o r e s t i m a t i o n for t h e i r v a l i d i t y . An assumption that has often been made when es t i m a t i n g the c o r r e c t i o n f a c t o r i s that the f i e l d s generated by vol u n t a r y b l i n k s during c a l i b r a t i o n are the same as those generated i n response to l i g h t s during experimental t r i a l s . T h i s assumption may not be v a l i d , i n which case the r e g r e s s i o n technique, using b l i n k s obtained under the same c o n d i t i o n s to which they are 74 a p p l i e d , should be s u p e r i o r to the s u b t r a c t i o n technique. However, using spontaneous b l i n k s obtained i n a darkened environment to estimate the c o r r e c t i o n f a c t o r , we found that there was no s i g n i f i c a n t d i f f e r e n c e between the VERs of the s u b t r a c t i o n method and those of the r e g r e s s i o n approach. In a d d i t i o n , the s u b t r a c t i o n method eye b l i n k s and eye movements were not d e a l t with s e p a r a t e l y . Although there i s evidence that b l i n k s and eye movements c o n t r i b u t e d i f f e r e n t i a l l y to the EEG re c o r d ( H i l l y a r d & Galambos, 1970), t h i s appears to be of l i t t l e consequence i n the short time i n t e r v a l s i n v o l v e d i n the VER procedure employed here. Separate c o r r e c t i o n f a c t o r s were not c a l c u l a t e d f o r the s u b t r a c t i o n method, yet i t produced VER r e s u l t s e q u i v a l e n t to those of the r e g r e s s i o n method. The problem of eye movements appears to be of more importance i n CNV s t u d i e s where long response i n t e r v a l s are i n v o l v e d . Since eye movement a r t i f a c t s do not appear to c o n t r i b u t e s i g n i f i c a n t l y to the VER, the r e g r e s s i o n method, which c a l c u l a t e s a c o r r e c t i o n c o e f f i c i e n t f o r both b l i n k and eye movement p o t e n t i a l , l o s e s that advantage over the s u b t r a c t i o n approach. In f a c t , when spontaneous b l i n k data are c o r r e c t e d with both methods, the s u b t r a c t i o n technique i s found to be s t a t i s t i c a l l y s u p e r i o r to the r e g r e s s i o n technique. This d i f f e r e n c e d i d not prove to be s i g n i f i c a n t i n the " c l i n i c a l " sense because both methods were e q u i v a l e n t when compared to unco r r e c t e d or eyes c l o s e d data. The r e l i a b i l i t y of both methods was found to be c o n s i s t e n t with that reported by Iacono et a l . (1982), but again the s u b t r a c t i o n method proved to be s l i g h t l y 75 s u p e r i o r at a l l e l e c t r o d e placements. In terms of e s t i m a t i n g the c o r r e c t i o n f a c t o r , i t appears that the s u b t r a c t i o n approach has a s l i g h t edge. The reason fo r t h i s i s not c l e a r but a p o s s i b l e e x p l a n a t i o n may l i e i n the assumption of l i n e a r i t y that i s c e n t r a l to a l l these c o r r e c t i o n procedures. Overton and Shagass (1969) found d e v i a t i o n s from l i n e a r i t y f o r v e r t i c a l eye movements although they f e l t these d e v i a t i o n s were not s u f f i c i e n t to a f f e c t the c o r r e c t i o n a l g o r i t h m . Data here suggest that the assumption of l i n e a r i t y i s s i g n i f i c a n t l y v i o l a t e d by b l i n k p o t e n t i a l s i n c e both c o r r e c t i o n procedures decrease in e f f i c a c y as e l e c t r o d e placements move f u r t h e r from the EOG s i t e . C o r r e l a t i o n of the EOG with EEG at a l l three e l e c t r o d e s i t e s shows that c o r r e l a t i o n s decrease s i g n i f i c a n t l y at the center and back of the head, and t h i s decrease i s not due to simple a t t e n u a t i o n of waveform amplitude. If amplitude decreased but the waveform remained the same, EOG would s t i l l c o r r e l a t e h i g h l y with EEG. The s u b t r a c t i o n technique, while a f f e c t e d , appears to be more robust than the r e g r e s s i o n method in the face of n o n l i n e a r b l i n k propagation. The r e g r e s s i o n technique r e l i e s h e a v i l y on the form of the EOG to c a l c u l a t e a c o r r e c t i o n c o e f f i c i e n t while the s u b t r a c t i o n technique uses only amplitude. T h i s d i f f e r e n c e between the two techniques may e x p l a i n the r e s u l t s seen above. In c o n c l u s i o n , data presented above suggest t h a t , f o r the s u b j e c t s i n t h i s study, b l i n k a r t i f a c t has s u b s t a n t i a l l y a f f e c t e d only the P 2 p o r t i o n of the VER. Although r e j e c t i o n of b l i n k epochs and the eyes c l o s e d procedure are shown to be 76 u n s u i t a b l e f o r d e a l i n g with b l i n k a r t i f a c t , the two c o r r e c t i o n procedures employed s i g n i f i c a n t l y reduced b l i n k contamination. While the c o r r e c t i o n methods produced VERs that were not s i g n i f i c a n t l y d i f f e r e n t in comparison to the other procedures, an o b j e c t i v e t e s t of v a l i d i t y showed the s u b t r a c t i o n method to be s t a t i s t i c a l l y more e f f e c t i v e and somewhat more r e l i a b l e . Both methods were l e s s e f f e c t i v e i n removing b l i n k a r t i f a c t at the p o s t e r i o r e l e c t r o d e p o s i t i o n s . The reason f o r t h i s may be that b l i n k p o t e n t i a l i s not propagated i n a l i n e a r fashion as o f t e n assumed. As w e l l , eye movement a r t i f a c t s , other than b l i n k s , that can cause systematic a r t i f a c t i n CNV recordings do not appear to be a s i g n i f i c a n t problem over the short i n t e r v a l s of VER r e c o r d i n g . 77 References Armington, J.C. (1981). V i s u a l l y evoked c o r t i c a l p o t e n t i a l s accompanying b l i n k s . I n v e s t i g a t i o n s in Ophthalmology and the  V i s u a l Sciences, 20, 691-695. Asarnow, R.F., Cromwell,R.L., & Rennick, P.M. (1978). C o g n i t i v e and evoked response measures of information p r o c e s s i n g in s c h i z o p h r e n i c s with and without a family h i s t o r y of schizophrenia.. J o u r n a l of Nervous and Mental Disease, 166, 719— 730. Buchsbaum, M., & Silverman, J . (1968). Stimulus i n t e n s i t y c o n t r o l and the c o r t i c a l evoked response. Psychosomatic Medicine, 30, 12-22. Buchsbaum, M.S., (1974). Average evoked response and stimulus i n t e n s i t y in i d e n t i c a l and f r a t e r n a l twins. P h y s i o l o g i c a l  Psychology, 2, 365-370. Buchsbaum, M.S., (1976). S e l f - r e g u l a t i o n of stimulus i n t e n s i t y : Augmenting/reducing and the average evoked response. In G. Schwartz & D. Shapiro, (Eds.), Consciousness and S e l f R e g u l a t i o n . New York: Plenum. Buchsbaum, M.S., Haier, R.J., & Murphy, D.L. (1977). S u i c i d e attempts, p l a t e l e t monoamine oxidase and the average evoked response. Acta. P s y c h i a t r a Scandinavia, 56, 69-79. B u f f i n g t o n , V., Martin, D.C, & Becker, J . (1981). VER s i m i l a r i t y between a l c o h o l i c probands and t h e i r f i r s t degree r e l a t i v e s . Psychophysiology, 5, 529-533. Corby, J . C , & K o p e l l , B.S. (1972). D i f f e r e n t i a l c o n t r i b u t i o n s of b l i n k and v e r t i c a l eye movements as a r t i f a c t s i n EEG r e c o r d i n g s . Psychophysiology, 9, 640-644. Ef r o n , R. (1964). A r t i f i c i a l s y n t h e s i s of evoked responses to l i g h t f l a s h . Annals of the New York Academy of Science. 122, 292-304. Gatton, G., Coles, M., & Donchin, E. (1983). A new method f o r o f f -l i n e removal of o c u l a r a r t i f a c t . Electroencephlography and  C l i n i c a l Neurophysiology, 44, 735-741. Gershon, E.S., & Buchsbaum, M.S. (1977). A genetic study of average evoked responses augmentation/reduction i n a f f e c t i v e d i s o r d e r s . In C. Shagass, S. Gershon, & A. F r i e d h o f f (Eds.), Psychopathology and B r a i n D y s f u n c t i o n . New York: Raven P r e s s . G i r t o n , D.G., & Kamiya, J . (1973). A simple o n - l i n e technique f o r removing eye movement a r t i f a c t s from the EEG. Electroencephlography and C l i n i c a l Neurophysiology, 34, 212-216. Greenhouse, S.W. & G e i s s e r , S. (1959). On methods in the a n a l y s i s 78 of p r o f i l e data. Psychometrika. 24, 95-112. Hassett, J . (1978). A Primer of Psychophysioloqy. San F r a n c i s c o : W.H. Freeman and Company. H i l l y a r d , S.A., & Galambos, R. (1970). Eye movement a r t i f a c t i n the CNV. Electroencephloqraphy and C l i n i c a l Neurophysiology, 28, 173-182. Iacono, W.G.., Gabbay, F.H., & Lykken, D.T. (1982). Measuring the average evoked response to l i g h t f l a s h e s : The c o n t r i b u t i o n of ey e - b l i n k a r t i f a c t to augmenting/reducing. B i o l o g i c a l  P s y c h i a t r y , 17, 897-911. Kamphiusen, H.A.C., & van Leeuwen, W.S. (1968) Response to s i n u s o i d a l l y modulated l i g h t as a i d in the study of symmetry and asymmetry of homologous areas of the head. Electroencephloqraphy  and C l i n i c a l Neurophysiology, 24, 489. Karson, C , Freed, W.J., Kleinman, J.E., Bigelow, L.B., & Wyatt, R.J. (1981). N e u r o l e p t i c s decrease b l i n k i n g i n s c h i z o p h r e n i c s u b j e c t s . B i o l o g i c a l P s y c h i a t r y , 16, 679-682. Knott, V.J., & Venables, P.H. (1978). Stimulus i n t e n s i t y c o n t r o l and the c o r t i c a l evoked response in smokers and non-smokers. Psychophysioloqy, 15, 186-192. Landau, S.G., Buchsbaum, M.S., Carpenter, W., S t r a u s s , J . , & Sacks, M. (1975). Schizophrenia and stimulus i n t e n s i t y c o n t r o l . A rchives of General P s y c h i a t r y , 32, 1239-1245. Martin, B.C., Becker, J . , & B u f f i n g t o n , V. (1979). An evoked p o t e n t i a l study of endogenous a f f e c t i v e d i s o r d e r s i n a l c o h o l i c s . In H. Beglester (Ed.), Evoked Brain P o t e n t i a l s and Behavior, (pp. 401-417). New York: Plenum. O'Connor, K.P. (1980). A p p l i c a t i o n of the c o n t i n g e n t negative v a r i a t i o n i n psy'chophysiology. In I. M a r t i n , & P. Venables, (Eds.), Techniques in Psychophysioloqy, New York: John Wiley and Sons. Rappaport, M., Hopkins, H.K. H a l l , K., B e l l e z a , T., & H a l l , R.A. (1975). Schizophrenia and evoked p o t e n t i a l s : Maximum amplitude, frequency of peaks, v a r i a b i l i t y , and phenothiazine e f f e c t s . Psychophysioloqy, 12, 196-207. Roth, W.T., Ford, J.M., & K o p e l l , B.S. (1978). Long-latency evoked p o t e n t i a l s and r e a c t i o n time. Psychophysioloqy, 15, 17-23. Sos k i s , D.A., & Shagass, C. (1974). Evoked p o t e n t i a l t e s t s of augmenting-reducing. Psychophysioloqy, 1 1 , 175-190. S p i l k e r , B., & Callaway, E. (1969). "Augmenting" and "Reducing" i n averaged v i s u a l evoked responses to sine wave l i g h t . 79 Psychophysiology, 6, 49-57. Stevens, J.R. (1978) Eye b l i n k and s c h i z o p h r e n i a : Psychosis or t a r d i v e dyskinesia.. American J o u r n a l of P s y c h i a t r y , 135, 223-227. V e r g l e r , R. , Gasser, T., Mocks, J . , (1982). C o r r e c t i o n of EOG a r t i f a c t s i n e v e n t - r e l a t e d p o t e n t i a l s of the EEG; Aspects of r e l i a b i l i t y and v a l i d i t y . Psychophysiology, 19, 472-480. von Knorring, L., Almay, B.G.L., Johansson, F. , & Terenius, L. (1979). Endorphins in CSF of c h r o n i c pain p a t i e n t s , in r e l a t i o n to augmenting/reducing response i n v i s u a l averaged evoked response. Neuropsychobiology, 5, 322-326. Whitton, J.L., Lue, F., & Moldofsky, H.E. (1978). A s p e c t r a l method for removing eye movement a r t i f a c t s from the EEG. Electroencephlography and C l i n i c a l Neurophysiology, 44, 735-741. 80 Appendix I I n s t r u c t i o n s to Subjects before the Dark P e r i o d The procedure f o r t h i s experiment i s s t r a i g h t f o r w a r d . You w i l l begin by s i t t i n g comfortably with the room darkened f o r a few minutes. During t h i s time i t i s important that you do not c l o s e your eyes and that you keep your head and face as s t i l l as p o s s i b l e . Following the dark p e r i o d you w i l l be r e q u i r e d to watch the screen in f r o n t of you as some f l a s h e s of l i g h t are presented. The l i g h t s may seem b r i g h t at times but they are in no way harmful to your eyes. I t i s a l s o important that you keep your head and face as s t i l l as p o s s i b l e during these l i g h t t r i a l s . A l r i g h t , we are about to s t a r t the dark p e r i o d , make you s e l f comfortable, keep your eyes open and s i t as s t i l l as you can. Here we go. I n s t r u c t i o n s to Subjects before T r i a l 1 Now you are about to watch a s e r i e s of f l a s h i n g l i g h t s . I would l i k e you to watch the center of the screen and as before keep your head and face s t i l l . Here we go. I n s t r u c t i o n s to Subjects before T r i a l 2 For the next t r i a l I would l i k e you to watch the l i g h t f l a s h e s through c l o s e d eye l i d s . T h i s means keeping your eyes c l o s e d throughout the s e r i e s of l i g h t s while your face i s pointed towards the center of the screen. Make y o u r s e l f comfortable. Here we go. I n s t r u c t i o n s to Subjects before T r i a l 3 T h i s l a s t t r i a l w i l l i n v o l v e you lo o k i n g at the l i g h t f l a s h e s with your eyes open ag a i n . As before watch the center the screen, keep your head and your face s t i l l and keep your eyes open. Here we go. 82 Appendix II Table A1 P, Amplitudes C o r r e l a t e d with EOG Amplitude ( C o e f f i c i e n t s averaged over T r i a l s 1 and 3) Method E l e c t r o d e Placement I n t e n s i t y 3 Sub Reg UnCor Fz Cz Oz Fz Cz Oz Fz Cz Oz -.1140 .1740 .1018 .2088 .2324 -.0602 .1764 .2066 .0784 .1704 .2754 .0356 -. 1367 -.0921 -.0408 .0424 .2002 -. 1424 Note. Sub=Subtraction, Reg=Regression, Uncor=Uncorrected Table A2 N! Amplitudes C o r r e l a t e d with EOG Amplitude ( C o e f f i c i e n t s averaged over T r i a l s 1 and 3) I n t e n s i t y Method E l e c t r o d e 3 4 Placement . . Sub Fz Cz Oz -.2845 -.1378 .0780 -.3318 -.2837 r.0831 Fz -.2944 -.3931 Reg Cz -. 1 036 -.2117 Oz .071 3 -.0472 UnCor Fz Cz Oz -.3164 -.1841 -. 1 109 -.2390 -.2099 -.1042 Note. Sub=Subtraction, Reg=Regression, Uncor=Uncorrected 3 84 Table A3 C o r r e l a t i o n s Between Methods (P 2 Amplitudes at I n t e n s i t y 1 Only, N=20) 1 Method Method E l e c t r o d e Placement Sub Reg B Rej UnCor EC Sub 1 .0 Reg Fz Cz Oz .9582 .9866 .9942 1 .0 B Rej Fz Cz Oz .8338 .9086 .9638 .921 4 .9208 .961 1 1 .0-UnCor Fz Cz Oz .6872 .8970 .9827 .7768 .9022 .9739 .8753 .9514 .9782 1.0 Eyes Closed Fz Cz Oz .0540 .3110 .2975 .0038 .2833 .3056 -.0488 .2347 .2194 -.1960 .2543 .2664 1 .0 Note. Sub=Subtraction, Reg=Regression, B Rej=Blink R e j e c t i o n , Uncor=Uncorrected, EC=Eyes Closed 1. For a l l but eyes c l o s e d method, data were averaged across T r i a l 1 and T r i a l 2. Table A4 C o r r e l a t i o n s Between Methods (P 2 Amplitudes at I n t e n s i t y 2 Only, N=20)|B1 Method Method E l e c t r o d e Placement Sub Reg B Rej UnCor EC Sub 1 .0 Reg Fz Cz Oz .9198 .921 5 .9242 1.0 B Rej Fz Cz Oz .7740 .9477 .941 8 .8221 .9254 .8947 1 .0 UnCor Fz-Cz Oz .51 64 .8239 .8831 .3877 .6340 .6978 .6869 .8181 .8424 1 .0 Eyes Closed Fz Cz Oz .3960 .4673 .3701 .3759 .4204 .4426 .2717 .4406 .4040 .261 5 .3825 .2853 1 .0 Note. Sub=Subtraction, Reg=Regression, B Rej=Blink R e j e c t i o n , Uncor=Uncorrected, EC=Eyes Closed 1.. For a l l but eyes c l o s e d method, data were averaged across T r i a l 1 and T r i a l 2. 86 Table A5 C o r r e l a t i o n s Between Methods (P 2 Amplitudes at I n t e n s i t y 3 Only, N=20) 1 Method Method E l e c t r o d e Placement Sub Reg B Rej UnCor EC Sub 1 .0 Reg Fz Cz Oz .7530 ,7249 .8830 1 .0 B Rej Fz Cz Oz .341 3 .8006 .7690 .31 37 .4960 .6269 1 .0 UnCor Fz Cz Oz .2873 .7058 .8088 .1207 .4602 .6937 . 1 1 53 .7499 .8028 1 .0 Eyes Closed Fz Cz Oz .3318 .3358 .3892 .3698 .5080 .5080 -.0880 . 1 082 .1788 . 1 971 .3465 .321 5 1 .0 Note. Sub=Subtraction, Reg=Regression, B Rej=Blink R e j e c t i o n , Uncor=Uncorrected, EC=Eyes Closed 1. For a l l but eyes c l o s e d method, data were averaged across T r i a l 1 and T r i a l 2. 87 Table A6 C o r r e l a t i o n s Between Methods (P 2 Amplitudes at I n t e n s i t y 4 Only, N=20) 1 Method Method E l e c t r o d e Placement Sub Reg B Rej UnCor EC Sub 1 .0 Reg Fz Cz Oz .8240 .6904 .8121 1 .0 B Rej Fz Cz Oz .31 70 .4682 .5480 . 1 771 .51 67 .6538 1 .0 UnCor Fz Cz Oz .2265 .6332 .81 49 . 1 632 .4453 .6327 .8637 .5894 .5632 1 .0 Eyes Closed Fz Cz Oz .6258 .491 3 .3607 .6704 .6705 .4094 .2385 .4383 .2313 .1603 ,41 86 .3938 1 .0 Note. Sub=Subtraction, Reg=Regression, B Rej=Blink R e j e c t i o n , Uncor=Uncorrected, EC=Eyes Closed 1 . For a l l but eyes c l o s e d method, data were averaged across T r i a l 1 and T r i a l 2. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096505/manifest

Comment

Related Items