4110 Words
          1 Table
          1 Figure
          (781) 237-9620

EEG, Bilateral Ear Temperature, and Affect Changes Induced by Lateral Visual Field Stimulation

Fredric Schiffer, M.D.
Carl M. Anderson, Ph.D.
Martin H. Teicher, M.D., Ph.D.

From the Department of Psychiatry, Harvard Medical School, and the Developmental Biopsychiatry Research Program, McLean Hospital, Belmont.

Previous Presentation:  Some data included in this report were presented as a poster at The Annual Meeting of The American Psychiatric Association, May, 1997
Address for Reprint Requests:  Fredric Schiffer, M.D., McLean Hospital, 115 Mill Street, Belmont, MA 02178
Running Title:  EEG, Ear Temperature, and Affect Changes
Key Words:  Lateral visual fields, Cerebral dominance, Electroencephalogram, Ear Canal Temperature, Emotions
Corresponding Author:  Fredric Schiffer, M.D., McLean Hospital, 115 Mill Street, Belmont, MA 02178, (781) 237-9620.
Submitted to: Comprehensive Psychiatry, April 30, 1998.  Resubmitted September 23, 1998; Accepted September 28, 1998
 Fifteen subjects evaluated while wearing in random order 4 pairs of glasses, 2 limiting vision to either the left or right lateral visual field (LVF) and 2 monocular glasses (MG), limiting vision to either eye, manifested significant differences in laterality indices in the predicted directions for theta EEG (p < 0.003) and for ear canal temperature (p < 0.02) between the LVF glasses but not the MG.  There was a significant correlation between lateral shifts in theta EEG and ear canal temperature (r2 = 0.47,  F(1,10) = 8.72, p < 0.015).  The LVF glasses induced an absolute difference in anxiety which was greater (p < 0.05) than that induced by the MG.  LVF glasses appear to induce shifts in affect and hemispheric dominance.

 An interesting literature, beginning in the early 1970's with a paper by Kinsborne () and continuing into the 1990's (), suggests that lateral visual field stimulation can enhance contralateral hemispheric activity.  Schiffer () recently reviewed this literature which includes a number of reports of EEG shifts in response to changes in lateral visual field stimulation (2,,,,) as well as one report of a PET study reporting increased brain activity contralateral to lateralized visual stimulation ().  In an elaborate study by Levick and associates (2), 8 male subjects had their EEGís recorded while they wore contact lenses occluded over all but a lateral portion.  The lenses were rotated so that the subjects could be tested while looking to the left visual field and to the right.  They found statistically significant shifts in theta EEG consistent with brain activation contralateral to the visual stimulation.  They found no significant changes in the alpha spectrum.  In the same report Levick et al (2) described 23 right handed males who had expected enhancements in verbal abilities when vision was allowed to the right side and enhancements in spacial abilities when vision was directed to the left.  They did not find consistent changes in affect except that there was more fatigue reported when subjects looked to the left (right hemisphere).
 In the present study we sought to investigate the effects of lateral visual field stimulation on affect, EEG laterality, and bilateral ear canal temperatures.  Schiffer (3) recently reported his findings in 70 psychotherapy patients of absolute changes in anxiety levels induced by two pairs of lateral visual field glasses, each taped to allow vision out of the left or right lateral field.  Two pairs of glasses allowing monocular vision from either the left or right eye did not induce similar changes.  Most of the patients were found by SCID interviews () to fall within three diagnostic categories: dysthymic disorder (n=20), major depression (n = 21), and posttraumatic stress disorder (PTSD) (n = 18).  The lateral visual field which evoked more anxiety varied with diagnosis.  Those with major depression tended to have more anxiety when looking to the left (right hemisphere), the PTSD group, when looking to the right (left hemisphere), and the dysthymic groupís responses were evenly divided.
 In psychiatric interviews 40 of the 70 patients manifested personality changes such that one lateral view induced a more mature view of the world attended by a greater sense of security and the other view, a less mature perspective attended by a relatively heightened anxiety.  In patients who had such emotional responses, the lateral visual field glasses frequently appeared to be clinically valuable in the patient's ongoing psychotherapy and psychopharmacological management because they allowed the patient to see that his negative perceptions of himself were challenged by his own positive perceptions evoked by looking out the opposite visual field.   The patients were further helped by being able to see with their own perception (out of one lateral field) that they were safer and more valuable than they had believed.  Switching between lateral visual fields enabled patients to begin to communicate within themselves that their self perceptions needed re-evaluation, enabling a less troubled, more mature view of themselves.
 To gather further evidence of these reported effects and their possible mechanisms, we decided to study the lateral visual field glasses and comparison monocular vision glasses in a laboratory setting with subjects previously unknown to the investigators.  We sought to compare EEG and affect responses to the lateral visual field glasses and to the monocular vision glasses using two established measures (2), EEG laterality indices and POMS affect scales.
 We sought also to compare left and right ear canal temperatures with the different sets of glasses.  Our hypothesis was that changes in cerebral blood flow might be reflected in changes in lateral ear temperature differences as suggested by Boyce et al () who found significant correlations in eight year old children between the left minus right ear temperatures and scores on tests of behavior problems.  Their findings in these children were constant with results from a similar investigation of theirs in Rhesus monkeys.  We hypothesized that since the cortex and the brain areas around the ear are both supplied predominately by the internal carotid artery, that increased cortical blood flow might induce a decrease flow around the ipsilateral ear canal.  We sought to determine if bilateral differences in ear temperature might provide a convenient, inexpensive indication of cerebral laterality.

 Fifteen right handed () college students, 18 to 24 years of age, comprised of 4 males and 11 females, were recruited by advertisement and paid to participate in another study looking into possible shifts in hemispheric dominance while subjects (abuse victims and controls) recalled a neutral and a traumatic memory.  This study was a more elaborate replication of an earlier study from our laboratory ().  All of the subjects recruited for the abuse study were invited to also participate in the present study.  Subjects were not paid for their participation in this study.  No subject declined the open and unpressured invitation to participate in the present study.  Nine of the subjects had no history of childhood abuse according to an Abbreviated Trauma Questionnaire developed by our laboratory based on the Life Events Questionnaire (), but 6 reported a history of some degree of childhood abuse.  All 9 subjects without an abuse history were asymptomatic and without an Axis I diagnosis by a structured psychiatric interview.  Among the 6 with a history of abuse, three had a current Axis I diagnosis.   One had a depressive disorder and an anxiety disorder, both "not otherwise specified."  The second had an anxiety disorder, not otherwise specified, and the third had a major depressive disorder and a mild posttraumatic stress disorder.  All subjects had a GAF score above 75 and none was taking psychotropic medications.  None had a history of head injury or birth trauma; all were in excellent physical health by history.  All subjects gave written informed consent.
 Each subject arrived at the laboratory for testing and had EEG electrodes applied in the standard 10/20 system ().  The electrodes were referenced to linked mastoids and all impedances were less than 5 K ohms.  The subjects were each randomly offered one of four pairs of taped glasses.  Two pairs of glasses were made by covering safety glasses with white adhesive tape over one side and 50% of the medial aspect of the other.  Each of these two pairs of lateral visual field glasses was taped so that it permitted vision to only either the left or the right lateral visual field.  The two other pairs of glasses were similar safety goggles, taped completely over one side, but only over the bottom one fourth of the other side.  These glasses allowed for monocular vision which has been shown to cause some hemispheric lateralization (,).  The tape on the bottom of the unoccluded lens gave the monocular glasses a  more complex appearance.
 After the first pair of glasses was worn for 2 minutes, an EEG was then recorded for 2 minutes with the subject looking straight ahead with his or her eyes open.  Following the EEG recording, an experimenter measured the subject's left then right ear temperatures with an "Ototemp LTX" infrared ear thermometer, manufactured by Exergen Corporation, Watertown, Massachusetts.  In the differential mode, the device has a clinical accuracy of ± 0.01 degrees Centigrade and a response time of approximately 0.1 second.  After the ear canal temperatures of both ears were obtained, one of the experimenters (CA) verbally asked the subject to rate his or her present feelings for each of eight affects from an abbreviated POMS scale (), from none to extreme on a 5 point scale.  The eight affects measured were: anxiety, tension, anger, sadness, hopelessness, panic, nervousness, and guilt.  Following the POMS measurements, the first pair of glasses were removed, and the subject was allowed to rest for 2 minutes.  Then the next pair were placed on the subject.  The identical procedure was then followed for each of the 4 randomly presented pairs of glasses.
 Following the data collection, all EEG's were inspected, and all observed artifacts were removed.  EEG's for 3 subjects (an unabused male and female and an abused female) upon review contained persistent artifacts, and were eliminated from our EEG data analysis.  In 3 other subjects (2 females, one abused, and 1 abused male) because of time constraints, only the lateral visual field glasses were tested.  For each subject, for each condition, at least 90 seconds of artifacted EEG's were obtained for spectral analysis.  We obtained data for the theta (4-7 Hz) and the alpha (8-13 Hz) power bands.  An asymmetry index (L-R/L+R) x 100 was calculated for four pairs of left-right leads (F3-F4, F7-F8, T3-T4, and T5-T6) from the frontal and temporal areas, and their mean was used as the asymmetry index for each condition.  Increased alpha and theta bands are both associated with decreased brain activity (2,).  An increase in the asymmetry index would result from increased left relative to right theta or alpha EEG activity which would indicate increased right relative to left brain activity.  We predicted that the experimental goggles would activate the contralateral hemisphere relative to the ipsilateral.  Thus we anticipated that the left visual field goggles would have a highter laterality index than the right visual field goggles.  The asymmetry indices for the left and for the right lateral visual field glasses were compared by an repeated measures ANOVA (the results of which are identical to a paired t-test) as were those for the left and right monocular glasses.
 We calculated the average POMS scores for all subjects in each condition.  We anticipated based on earlier results (3) that the side on which greater negative affect would occur with the different experimental goggles would vary among subjects.  We found this to be true for the present population as well.   Because of this observation, we decided to compare also the absolute differences in affect between the two lateral visual field glasses with the absolute differences between the two monocular glasses.  This evaluation allowed for a comparison of the magnitude of the differences in affect between types of goggles without regard to the lateral direction of the individual responses.  Affect scores are reported as means ± standard deviations.  Absolute values were not used in the EEG or ear temperature analyses.
 The differences in ear canal temperatures during each pair of lateral visual field glasses and each pair of monocular glasses were calculated and compared by repeated measures ANOVA.  Since we hypothesized that increased hemispheric cortical blood flow would induce a decrease blood flow around the ipsilateral ear canal (since the cortex and the brain areas around the ear are both supplied predominately by the internal carotid artery), we expected the left ear temperatures to be relatively higher than the right when the subject wore the glasses allowing vision to the left visual field compared to when he wore the right visual field goggles.

EEG theta and alpha power:
 The differences between the laterality indices for the theta bands between the left and the right lateral visual field glasses was highly significant in the expected direction by repeated measures ANOVA (F (1,11) = 15.17, p < 0.003).  In 11 of the 12 subjects the asymmetry index, as predicted, was larger during the left than the right lateral visual field glasses.  The remaining subject showed essentially no difference in the theta EEG asymmetry indices between the two lateralized glasses.  In contrast, there was no significant difference in laterality index with right versus left monocular glasses (F (1,8) = 0.60, ns).  Table 1 shows the theta laterality indices for the left and right lateral visual field glasses and the left and right monocular glasses.  To be sure that the differences between the experimental and control glasses were not due to there being fewer controls, we repeated our ANOVA on the experimental glasses eliminating the 3 subjects who did not have control trials.  For these 9 subjects the repeated measures ANOVA (F1,8) = 10.94, p = 0.01, remained significant.  Figure 1 shows BEAM scans for theta (4 - 7 Hz) power during the two lateralized conditions in a typical subject.  There was no significant interaction with sex or history of abuse with theta EEG asymmetry with the different glasses.
 For the alpha band, the differences between the EEG laterality indices for two lateral visual field glasses (F (1,11) = 1.89, ns) or for the two monocular glasses (F (1,8) = 0.75, ns) did not achieve significance.

Ear Temperature measurements:
 The lateral visual field glasses appeared to produce significant differential effects on ear canal temperature.  As predicted, with the left lateral visual field glasses, the left ear canal was 0.13 oC warmer than the right, while with the right sided glasses, the left ear canal was 0.08 oC cooler (F (1,14) = 7.46, p < 0.02).  This analysis for the monocular glasses did not approach significance (F (1,11) = 0.978, p = ns).
 There was a significant Pearson's correlation between the differences in the left  minus right ear temperatures during the left and right lateral visual field glasses and the differences in the laterality indices for theta EEG during the same conditions (r2 = 0.46, F(1,10) = 8.57, p < 0.016).  The same factors had a more significant correlation when fit to a polynomial of two degrees (r2 = 0.67, F (2,9) = 9.20, p < 0.007).  The individual factors in these correlations were both in the expected direction.  There were no significant correlations between any ear temperature differences and any EEG laterality indices for the monocular glasses.

POMS scores:
 Between the two  lateral visual field glasses and between the two monocular glasses there were no significant differences between the POMS ratings for the mean of all 8 affects investigated nor for any of the affects individually, nor were there any significant differences between the lateral visual field glasses and the monocular glasses on these affect scores.  Because it was found in previous research (3) that responses to the lateral visual field glasses can be lateralized to different sides depending on individual's diagnostic category, we decided as in previous work (3), to compare the absolute differences between the left and right lateral visual field glasses with the absolute differences for the two monocular glasses for each of the 8 affects and the mean of all eight.  For the mean of all 8 affects the absolute difference for the lateral visual field glasses was 0.50 ± 0.61 and that for the monocular glasses was 0.28 ± 0.30, (Wilcoxon  Signed-Rank Test, p = 0.164, Signed-Rank = 10.500, N = 12).  Among the 8 individual affects measured, the absolute differences were significant between the lateral visual field glasses and the monocular glasses only for "anxiety."   For "anxiety" the absolute difference between the two lateral visual field glasses was 0.90 ± 0.85, and that between the two monocular glasses was 0.29 ± 0.45 (Wilcoxon Signed-Rank Test, p = 0.031, Signed-Rank = 10.500, N =12).  Previous research (3) had also shown this difference for anxiety measures to be significant.


 Kinsborne (1) was the first to suggest that one might affect cerebral dominance by altering lateral visual field stimulation.  Dimond and associates () used a contact lens occluded on one side to show films to either the medial or the lateral aspect of the retinal and found more intense affect when the films were shown to the lateral aspect of the right retina (right hemisphere).  Whittling and associates () more recently have found that with a complex system of eye tracking and computer screen masking they could effectively show movies to either the left or the right lateral visual field.  They too found different emotional responses depending upon the field to which the film was shown.  Levick et al (2) found that pairs of contact lenses occluded on one side allowing vision to either the left or the right lateral visual field could evoke significant shifts in EEG theta in the mean of a frontal and a temporal lead.  Davidson et al (4), Schweinberger et al (5), and others (6,7) have also found EEG or evoked potential shifts in response to changes in lateral visual field stimulation.  Greenberg his associates (8) found that lateral visual field stimulation could be correlated with shifts in striatal laterality determined by PET scanning.
 Schiffer (3) demonstrated clinically significant changes in anxiety levels in response to lateral visual field stimulation.  In the present study we sought to confirm his findings and to test the hypothesis that such affect changes might be the result of shifts in hemispheric dominance.  We did find a significant change in anxiety, but not other affects, in response to the lateral visual field glasses in these college students not in therapy, and these results are in accord with earlier findings in psychotherapy patients (3).  In the present study, the mean change in anxiety was 0.9 ± 0.85 on 5 point scale from none to extreme.  This change was clinically as well as statistically significant.
 The significant shift in theta EEG laterality in the expected directions with the lateral visual field glasses but not the monocular glasses is consistent with earlier work on EEG and lateral visual field stimulation (2), and supports the hypothesis that the affect changes in response to the lateral visual field glasses are related to shifts in hemispheric activity.  As Levickís group (2), we did not find statistically significant shifts in the alpha spectrum although the alpha laterality indices did a go in the expected directions.  We do not have an explanation for why changes in theta but not alpha spectrums were significant, but the fact that two independent trials found the same result is noteworthy.
 The changes in ear canal temperature occurring between the two lateral visual field glasses but not the two monocular glasses suggests that lateral visual field stimulation but not monocular stimulation was able to effect directional changes in ear canal temperature.  One hypothesis needing further testing is that an increase in hemispheric cortical blood flow should cause an ipsilateral decrease due to shunting in the area around the ear, since both areas are supported by the same internal carotid artery.  The decreased blood flow in the area around the ear should be reflected in a decrease in ear canal temperature with increased hemispheric activity.  This hypothesis is strengthened by the significant correlation between ear canal temperature lateral differences and EEG lateral shifts.  Overall, these results support the suggestion by Boyce et al (10) that bilateral ear canal temperatures may offer an inexpensive, easily applied method of approximating hemispheric activity, and they encourage further study of this possibility.  We are beginning a study of any possible relationships between cerebral blood flow as measured by fMRI and bilateral ear canal temperatures.
 The fact that two independent measures of hemispheric activation (theta EEG and differential ear canal temperatures) changed significantly in the predicted directions supports our hypothesis that the lateral visual field glasses can influence the balance of hemispheric activation.

Table 1.
Differences in Laterality Indices between Left and Right trials, Means ± Standard Deviations.

      theta  alpha
Lateral visual field glasses, N = 12
  5.60 ± 5.48
  2.52 ±  6.51
Monocular Visual Field Glasses, N = 9
  1.70 ± 6.60
  1.85 ± 6.41

Legend for Figure 1

Beam scans derived from theta (4-7Hz) EEG power during the left visual field glasses and during the right visual field glasses in a typical subject.  Darker areas represent decreased theta, suggesting increased brain activity.  With the left visual field glasses the asymmetry index [(L-R)/(L+R)x100] is larger than that during the right visual field glasses suggesting, relative to the right visual field glasses, greater right to left hemispheric activation with the left visual field glasses and relatively greater left to right hemispheric activation with the right visual field glasses.  ROI indicates the regions of interest.

This work was sponsored in part by NIMH Grant RO1 MH-53636 to MHT.  We acknowledge the assistance of Ann Polcari, R.N., M.C.S and Carol Glod, R.N., C.S., Ph.D. in the recruitment and diagnostic assessment of the subjects.

1.Kinsbourne M: Lateral input may shift activation balance in the integrated brain. American Psychologist. 1983;38:228-229
2. Levick SE, Lorig T, Welxler, Gur RE, Gur RC, Schwartz GE: Asymmetrical visual deprivation: a technique to differentially influence lateral hemispheric function. Perceptual and Motor Skills 1993;76:1363-1382
3.Schiffer F: Affect Changes Observed with Right Versus Left Lateral Visual Field Stimulation in Psychotherapy Patients: Possible Physiological, Psychological, and Therapeutic Implications. Comprehensive Psychiatry, 1997;38:289-295
4.Davidson RJ, Schaffer CE, Saron C: Effects of lateralized presentations of faces on self-reports of emotion and EEG asymmetry in depressed and non-depressed subjects. Psychophysiology 1985;22:353364
5.Schweinberger SR, Sommer W, Stiller RM: Event-related potentials and models of performance asymmetries in face and word recognition. Neuropsychologia 1994;32:175-191
6.Lavine RA, Jenkins RL: Hemispheric asymmetry in processing visual half-field pattern-reversal stimuli assessed by reaction time and evoked potentials. Intern J Neurosci 1989;44:197-204
7.Tressoldi PE, Cusumano S: Visual evoked potentials related to behavioral asymmetries during foveal attention in the two extrapersonal hemispaces.  Brain Cogn 1992;18:125-137
8.Greenberg JH, Reivich M, Alavi A: Metabolic mapping of functional activity in human subjects with the [18F]fluorodeoxyglucose technique.  Science 1981;212:678-680
9.First MB,Spitzer RL, Gibbon M, Williams JBW: Structured Clinical Interview for DSM-IV Axis I Disorders - Patient Edition (SCID-I/P, Version 2.0)
10.Boyce TW, Higley JD, Jemerin JJ, Champoux M, Suomi SJ: Tympanic temperature asymmetry and stress in rhesus macaques and children. Archives Pediatrics and Adolescent Med 1996;150:518-523
11.Oldfield RC:  The assessment and analysis of handedness:  The Edinburgh Inventory.  Neuropsychologia 1971; 9: 97-113
12. Schiffer F, Teicher MH, Papanicolaou AC: Evoked potential evidence for right brain activity during the recall of traumatic memories. J Neuropsychiatry Clin Neurosci 1995;7:169-175
13. Tennant C, Andrews G: A scale to measure the stress of life events. Aust N Z J Psychiatry 1976;10:27-32
14. Jasper HH: The ten-twenty electrode system of the international federation. Electroenceph Clin Neurophysiol 1958;10:371-375
15. Weisz J, Szilagyi N, Lang E, Adam G: The influence of monocular viewing on heart period variability. Intern J Psychophysiology 1992;12:11-18
16. Bross M, Milton J: Effect of monocular occlusion on rotary pursuit performance. Perceptual and Motor Skills 1987;65:796-798
17. Lorr M, McNair DM, Droppleman LF:  Manual:  profile of mood states. San Diego, Educational and Industrial Testing Service, 1971
18. Schacter DL: EEG theta waves and psychological phenomena: a review and analysis. Biological Psychiatry 1977;5:47-82
19. Dimond SJ, Farrington L, Johnson P: Differing emotional response from right and left hemispheres. Nature 1976;261:690-692
20.Wittling W, Roschmann R: Emotion-related hemisphere asymmetry: Subjective emotional responses to laterally presented films. Cortex 1993;29:431-448