In the literature there has been a prevailing view that the right hemisphere is associated with the perception and expression of negative emotions [8, 21, 83]. Our studies, and review of the literature, suggests that this view point is overly dogmatic, and that negative valence may be left lateralized in a substantial number of individuals [44, 56, 57]. Further, our new data provides evidence that the degree and direction of laterality is an important individual trait, on par with gender, in the degree to which it can account for individual differences in regional brain size, functional brain activity, and psychiatric symptomatology. Although we have found gender to be an important factor in regard to HEV, we feel that a future study with larger numbers of males and females would be helpful to more precisely delineate its level of importance. In this study we did not evaluate functional imaging during the different memory conditions, but we hope to do such a study in the future.
We found that gender alone accounted for about 13% of the variance between individuals in total GMV. In contrast, direction of HEV, gender, and their interaction accounted for up to 60% of the variance. Degree and direction of HEV was also associated with substantial differences in hippocampal and amygdala volumes. Females with the most right-sided HEV tended to have the smallest hippocampal and largest amygdala volumes. (Volumes were not measured in males). Degree of HEV in AEP response accounted for 36% and 54% of the variance in hippocampal and amygdala measures, respectively.
T2 relaxation time was used as indirect measure of resting relative cerebral blood volume, with lower levels of T2 relaxation times correlating with higher blood volumes . Subjects with left negative HEV had lower T2 relaxation times in their left vs. right hemisphere, suggesting greater left-sided rCBV, relative to subjects with right negative HEV.
The right negative HEV group performed significantly better on 3 of 4 memory tests. This difference persisted even when their scores were covaried for intracranial and hippocampal volumes. When direction of hemispheric emotional valence and its interactions were added to regression models examining the effects of gender and childhood maltreatment on 15 administered psychological test parameters, the R2 values for 11 of the 15 models improved, and the effect sizes for the influence of childhood maltreatment were increased on most of these psychological variables.
We presented two methods for determining HEV, probe AEPs and LVFS. They provide highly correlated results, and both appear useful for delineating distinct populations of left vs. right lateralized responders. LVFS, performed in only 15 subjects (versus 28 in the AEP group), correlated with the total volume of corpus callosum, and the right and left hemispheric T2 relaxation times. The changes in GMV and T2 relaxation time by LVFS were in the same direction as those found with the AEP procedure.
Overall, having right negative HEV was associated with enhanced memory, reduced total GMV, smaller hippocampal and larger amygdala volumes (in females), and increased association between exposure to childhood abuse and symptom ratings for dissociation, paranoia and phobic avoidance. In contrast, left negative HEV was associated with diminished memory, increased total GMV, larger hippocampal and smaller amygdala volumes (in females), more left-lateralized hemispheric blood flow, and an increased association between exposure to childhood abuse and symptom ratings for suicidaility, depression, anxiety, hostility, interpersonal sensitivity, and attention deficit. We do not have an explanation for the association between enhanced memory and decreased hippocampal size, but hope that future studies clarify the effects of other factors such as lateralized blood flow and symptom ratings from abuse, and lead to further illumination.
Human biology is highly lateralized, but to varying degrees. Situs inversus, in which there is a mirror image reversal of thoracic and abdominal organs, affects only 0.01% of the population . Between 2–6% of the population appears to have exclusively right-sided, or predominantly right-sided, language lateralization (Loring et al., 1990). Left-handedness occurs in about 12% of adults . We now suggest that left-hemisphere based negative emotional valence occurs in about a third of the population and in an earlier study we reported it to occur in 40% of patients . This view that negative emotional valence is lateralized, but to either right or left hemispheres, stands in marked contrast to earlier theories regarding the specialized role of the right hemispheric in the processing of negative emotions [2–5]. Assuming that negative emotional valence is exclusively right lateralized may result in research findings in which lateralized difference are either diminished or imperceptible. In contrast, recognizing that right or left-sided laterality of negative emotional valence is an important individual difference, may help to resolve discrepancies in the literature that have emerged in studies of the neurobiology and treatment of emotional disorders.
For example, there is controversy in the literature regarding the efficacy of left-sided rapid transcranial magnetic stimulation (rTMS) for treatment of refractory depression [86–88]. However we found that outcome of rTMS varies greatly depending on the whether the patient has right or left lateralized negative HEV. We predicted that stimulating a hemisphere with a positive HEV would be more likely to be helpful than stimulating one with a negative HEV. As we predicted , 86% percent of those with left negative HEV (based on LVFS response) had a poor outcome to a subsequent course of left prefrontal rTMS versus only 20% of subjects with right negative HEV. The right negative HEV group had a 42% mean reduction in HDRS compared to an 11% reduction in the group with left negative HEV.
We recently completed a replication of our 2002 rTMS study (unpublished observations) at MindCare Centres, British Colombia, Canada, a clinic that specializes in the treatment of depression with rTMS. Data were obtained from 23 depressed patients assessed for HEV laterality who received a 2-week course of left-sided rTMS. The right negative HEV group had a 61% decrease in their depression rating scores versus a 31% decrease in scores in the left negative HEV group. Hence, we predict that the demonstrable efficacy of rTMS could be substantially increased by selecting suitable candidates for left-sided treatment based on laterality of negative HEV, or by adjusting the side of treatment to target the hemisphere with more positive HEV.
This hypothesis may also provide a mechanistic explanation of the findings of Cohen and associates , who conducted a double-blind, controlled trial of high and low frequency rTMS in the treatment of 24 patients with PTSD. They found that high frequency stimulation of the right frontal region produced a more favorable outcome than high frequency stimulation of the left side. This stands in contrast to studies of depression, which favor left-sided rTMS. This difference makes sense as Schiffer  found that among psychotherapy patients with lateralized affective responses to LVFS, that 73% of patients with major depression (n = 15) had a right negative HEV, whereas 71% of patients with PTSD (n = 14) had a left negative HEV. Hence, left-sided rTMS should benefit most patients with depression, while right-sided rTMS should benefit most patient with PTSD. Targeting treatment to the appropriate side for each individual based on LFVS may further enhance outcomes, though this remains to be determined.
These findings may apply to other lateralized treatments such as ECT. There is evidence that psychotropic medications have lateralized effects [49, 51], and several authors have predicted responses to psychotropic medications by measurement of asymmetric brain activation by dichotic listening [47, 52], electroencephalogram [50, 52], fMRI , and PET . LVFS should be explored as a possible method for predicting such outcomes.
These tests for HEV were inspired by observations from split-brain studies that revealed that each hemisphere was capable of supporting independent mentation [1, 38, 39]. Bogen [40, 41] was the first to suggest that these split-brain findings might relate to intact individuals, and his assertion is supported by a number of reports of Wada studies that found, not just affect changes, but dramatic personality changes with the anesthetization of one hemisphere [35–37, 90]. For example, Masia et al reported 4 patients who recalled with severe emotional distress a major trauma such as the decapitation of a friend or an incestuous rape when one hemisphere was anesthetized, but not at baseline nor when the other hemisphere was anesthetized. The side from which the memory was released varied among patients (2 left and 2 right). Ahern et al  described two patients with vivid personality changes. One case went from withdrawn and sullen to affable and social following anaesthetization of the left hemisphere, and in the other went from pleasant and well adjusted to belligerent and abusive when his left hemisphere was anesthetized.
Wittling [91, 92] has reported affect, blood pressure, heart rate, and cortisol changes depending on which side an upsetting film was shown to subjects. Placebo controlled studies have shown that LVFS can induce changes in affect [44, 56, 57] and Schiffer  and Morton  have reported not only affect but also cognitive changes with LVFS. Others [94–97] have reported similar changes following lateralized auditory stimulation in patient populations. In all of these studies, the side that induced negative affects and/or cognitions varied among individuals. Schiffer [42, 43] has suggested that LVFS can be a useful adjunct to psychotherapy.
Considerable evidence indicates that unilateral sensory or motor stimulation, including LVFS, activates the contralateral hemisphere. This evidence includes studies using theta EEG [57, 98], lateral ear temperature , BOLD fMRI [58, 99], and PET . The combination of lateralized psychological and physiological responses, leads Schiffer to hypothesize that LVFS might preferentially activate the contralateral hemisphere  and produce an associated mental state that is consistent with that hemisphere's emotional valence.