Interhemispheric communication after section of the forebrain commissures

Can the mind be split? A historical introduction

The idea that the mind might be composed of distinct conscious entities goes back at least to the mid-19th century, and was at first based on the bilateral symmetry of the brain, with each side seemingly a mirror-image replica of the other. This led to early speculation as to whether section of the forebrain commissures might lead to separate, independent consciousnesses. This was not put to the test until the 1960s, first in commissurotomized cats and monkeys, and then in humans who had undergone commissurotomy for the relief of intractable epilepsy. Initial results did indeed suggest independent consciousness in each separated hemisphere, but later findings have also revealed a degree of mental unity, especially in some perceptual functions and in motor control. Some of these findings might be interpreted in terms of subcortical connections or external cross-cuing, and also address questions about the nature of consciousness in a more concrete way.

Music Playing and Interhemispheric Communication: Older Professional Musicians Outperform Age-Matched Non-Musicians in Fingertip Cross-Localization Test

OBJECTIVE

Numerous investigations have documented that age-related changes in the integrity of the corpus callosum are associated with age-related decline in the interhemispheric transfer of information. Conversely, there is accumulating evidence for more efficient white matter organization of the corpus callosum in individuals with extensive musical training. However, the relationship between making music and accuracy in interhemispheric transfer remains poorly explored.

METHODS

To test the hypothesis that musicians show enhanced functional connectivity between the two hemispheres, 65 professional musicians (aged 56-90 years) and 65 age- and sex-matched non-musicians performed the fingertip cross-localization test. In this task, subjects must respond to a tactile stimulus presented to one hand using the ipsilateral (intra-hemispheric test) or contralateral (inter-hemispheric test) hand. Because the transfer of information from one hemisphere to another may imply a loss of accuracy, the value of the difference between the intrahemispheric and interhemispheric tests can be utilized as a reliable measure of the effectiveness of hemispheric interactions.

RESULTS

Older professional musicians show significantly greater accuracy in tactile interhemispheric transfer than non-musicians who suffer from age-related decline.

CONCLUSIONS

Musicians have more efficient interhemispheric communication than age-matched non-musicians. This finding is in keeping with studies showing that individuals with extensive musical training have a larger corpus callosum. The results are discussed in relation to relevant data suggesting that music positively influences aging brain plasticity.

Age-related decline in interhemispheric transfer of tactile information: The fingertip cross-localization task

According to the disconnection hypothesis of cognitive aging, cognitive deficits associated with brain aging could be a result of damage to connective fibres. It has been suggested that the age-related decline in cognitive abilities is accompanied by age-related changes in interhemispheric communication ensured by commissural fibres. This study aimed to contribute to this topic by investigating the effects of aging on the efficiency of interhemispheric transfer of tactile information. A total of 168 right-handed subjects, aged 20-90 years, have been tested using the fingertip cross-localization task: the subject must respond to a tactile stimulus presented to one hand using the ipsilateral (uncrossed condition) or contralateral hand (crossed condition). Because the crossed task requires interhemispheric transfer of information, the value of the difference between the uncrossed and crossed conditions (CUD) can be deemed to be a reliable measure of the efficiency of the interhemispheric interactions. The uncrossed condition was more accurate than the crossed condition for all ages. However, the degree of the CUD was significantly age-dependent. The effectiveness of the interhemispheric transfer of tactile information decreased significantly with age and may indicate the occurrence of age-related changes of the corpus callosum. Considerably, performance appears to decline around the seventh decade of life with the fastest decline in the subsequent decades. The results suggest a relationship between brain aging and the efficiency of the interhemispheric transfer of tactile information. The findings are discussed in relation to the strategic role of white matter integrity in preserving behavioural performances.

Semantic Networks in the Divided Cerebral Hemispheres

Hemispheric differences in the recognition and manipulation of meaning may be based on distinctions in size, composition, or organization of the right and left semantic networks The present study describes these features of pictorially based semantic networks in 3 subjects with complete forebrain commissurotomy Stimuli were presented for prolonged viewing to the left and right visual hemifields For each trial, the subjects chose from a 20-choice array all pictures that were associated with a target, then indicated the member of each pair of chosen associates that was more closely related to the target The hemispheres' networks were found to be of similar size and composition, but were organized differently The right hemisphere more often produced linear rankings of semantic associates to a target than did the left, and rankings by the two hemispheres were not strongly correlated Hemispheric differences in semantic organization mirror differences in perceptual organization, with the right hemisphere specialized for conventional meaning and the left hemisphere specialized for detecting and processing deviations from standard meaning

Does functional MRI detect activation in white matter? A review of emerging evidence, issues, and future directions

Functional magnetic resonance imaging (fMRI) is a non-invasive technique that allows for visualization of activated brain regions. Until recently, fMRI studies have focused on gray matter. There are two main reasons white matter fMRI remains controversial: (1) the blood oxygen level dependent (BOLD) fMRI signal depends on cerebral blood flow and volume, which are lower in white matter than gray matter and (2) fMRI signal has been associated with post-synaptic potentials (mainly localized in gray matter) as opposed to action potentials (the primary type of neural activity in white matter). Despite these observations, there is no direct evidence against measuring fMRI activation in white matter and reports of fMRI activation in white matter continue to increase. The questions underlying white matter fMRI activation are important. White matter fMRI activation has the potential to greatly expand the breadth of brain connectivity research, as well as improve the assessment and diagnosis of white matter and connectivity disorders. The current review provides an overview of the motivation to investigate white matter fMRI activation, as well as the published evidence of this phenomenon. We speculate on possible neurophysiologic bases of white matter fMRI signals, and discuss potential explanations for why reports of white matter fMRI activation are relatively scarce. We end with a discussion of future basic and clinical research directions in the study of white matter fMRI.

Interhemispheric transfer of spatial and semantic information: electrophysiological evidence

The goal of this study was to cast light on the existence of functional callosal channels for the interhemispheric transfer (IHT) of spatial and semantic information. To do so, we recorded event-related potentials in healthy humans while performing a primed odd-even discrimination task. Targets were visually presented numbers preceded by single-letter primes signaling the probable presentation of an odd or an even number. Primes and targets could appear either in the same or in different visual fields, thus requiring an IHT in the latter case. The P1 and N2 components were influenced by IHT of spatial information only, whereas the later N400 was influenced by IHT of both spatial and semantic information. This was not the case for the P3b, which was modulated by semantic validity only. These results provide novel evidence of the existence of a temporally separated interhemispheric exchange of spatial and semantic information.

Detecting functional magnetic resonance imaging activation in white matter: interhemispheric transfer across the corpus callosum

Background

It is generally believed that activation in functional magnetic resonance imaging (fMRI) is restricted to gray matter. Despite this, a number of studies have reported white matter activation, particularly when the corpus callosum is targeted using interhemispheric transfer tasks. These findings suggest that fMRI signals may not be neatly confined to gray matter tissue. In the current experiment, 4 T fMRI was employed to evaluate whether it is possible to detect white matter activation. We used an interhemispheric transfer task modelled after neurological studies of callosal disconnection. It was hypothesized that white matter activation could be detected using fMRI.

Results

Both group and individual data were considered. At liberal statistical thresholds (p < 0.005, uncorrected), group level activation was detected in the isthmus of the corpus callosum. This region connects the superior parietal cortices, which have been implicated previously in interhemispheric transfer. At the individual level, five of the 24 subjects (21%) had activation clusters that were located primarily within the corpus callosum. Consistent with the group results, the clusters of all five subjects were located in posterior callosal regions. The signal time courses for these clusters were comparable to those observed for task related gray matter activation.

Conclusion

The findings support the idea that, despite the inherent challenges, fMRI activation can be detected in the corpus callosum at the individual level. Future work is needed to determine whether the detection of this activation can be improved by utilizing higher spatial resolution, optimizing acquisition parameters, and analyzing the data with tissue specific models of the hemodynamic response. The ability to detect white matter fMRI activation expands the scope of basic and clinical brain mapping research, and provides a new approach for understanding brain connectivity.

Interhemispheric transfer following callosotomy in humans: Role of the superior colliculus

It is now common knowledge that the total surgical section of the corpus callosum (CC) and of the other forebrain commissures prevents interhemispheric transfer (IT) of a host of mental functions. By contrast, IT of simple sensorimotor functions, although severely delayed, is not abolished, and an important question concerns the pathways subserving this residual IT. To answer this question we assessed visuomotor IT in split-brain patients using the Poffenberger paradigm (PP), that is, a behavioral paradigm in which simple reaction time (RT) to visual stimuli presented to the hemifield ipsilateral to the responding hand is compared to stimuli presented to the contralateral hemifield, a condition requiring an IT. We tested the possibility that the residual IT is mediated by the collicular commissure interconnecting the two sides of the superior colliculus (SC). To this purpose, we used short-wavelength visual stimuli, which in neurophysiological studies in non-human primates have been shown to be undetectable by collicular neurons. We found that, in both totally and partially callosotomised patients, IT was considerably longer with S-cone input than with L-cone input or with achromatic stimuli. This was not the case in healthy participants in whom IT was not affected by color. These data clearly show that the SC plays an important role in IT of sensorimotor information in the absence of the corpus callosum.

Contribution of posterior corpus callosum to the interhemispheric transfer of tactile information

Three total and three partial callosotomy patients underwent neuropsychological testing to evaluate interhemispheric transfer of tactile information. Tactile transfer is required to name objects presented to the left hand, to compare objects held in either hand, and to transfer topological information between hands. Tactile Naming, Same-Different Recognition, and Tactile Finger Localization Tests (intra- and intermanual tasks) were administered as specific tools. Results were compared with previous fMRI data from the same subjects and with the performance of a control group (20 age-matched subjects). Total callosotomy patients performed modestly: mean correct responses were 93% and 30% (right and left hand, respectively) in Tactile Naming; 68% in Same-Different Recognition; 84% and 76% (right and left hand stimulation, respectively) in intermanual Tactile Finger Localization, and 100% in the intramanual task. Partial callosotomy patients achieved 93-100% accuracy: all have an intact splenium, and one, and possibly all, also an intact posterior callosal body. Controls scored 99% in Tactile Naming, both hands, and Same-Different Recognition; 100% in intramanual Tactile Finger Localization; and 96% and 95%, with right and left hand stimulation, respectively, in the intermanual task. Differences between the two callosotomy groups were significant, as were those between total callosotomy patients and controls. The partial callosotomy group scored like the control subjects. Neuropsychological data agree with previous functional findings, further demonstrating that interhemispheric tactile transfer requires posterior corpus callosum integrity.

Posterior corpus callosum and interhemispheric transfer of somatosensory information: an fMRI and neuropsychological study of a partially callosotomized patient

Interhemispheric somatosensory transfer was studied by functional magnetic resonance imaging (fMRI) and neuropsychological tests in a patient who underwent resection of the corpus callosum (CC) for drug-resistant epilepsy in two stages. The first resection involved the anterior half of the body of CC and the second, its posterior half and the splenium. For the fMRI study, the hand was stimulated with a rough sponge. The neuropsychological tests included: Tactile Naming Test (TNT), Same-Different Recognition Test (SDRT), and Tactile Finger Localization Test (intra- and intermanual tasks, TFLT). The patient was studied 1 week before and then 6 months and 1 year after the second surgery. Before this operation, unilateral tactile stimulation of either hand activated contralaterally the first (SI) and second (SII) somatosensory areas and the posterior parietal (PP) cortex, and SII and PP cortex ipsilaterally. All three tests were performed without errors. In both postoperative sessions, somatosensory activation was observed in contralateral SI, SII, and PP cortex, but not in ipsilateral SII and PP cortex. Performance was 100% correct in the TNT for the right hand, but below chance for the left; in the other tests, it was below chance except for TFLT in the intramanual task. This case provides the direct demonstration that activation of SII and PP cortex to stimulation of the ipsilateral hand and normal interhemispheric transfer of tactile information require the integrity of the posterior body of the CC.

Interhemispheric transfer of colour and shape information in the presence and absence of the corpus callosum

Two split-brained subjects, one (L.B.) with full forebrain commissurotomy and one (R.B.) with callosal agenesis, and a group of twenty neurologically intact subjects were tested in three discrimination tasks: a go-no go task, a two-choice task, and a three-choice task. The discriminations were based on colour in Experiment 1, and on shape in Experiment 2. The stimuli were presented in one or other visual field, and the subjects responded with the fingers of one or other hand, allowing the differences in reaction time between crossed and uncrossed responses (CUD) to be calculated. For the normal subjects the CUD tended to diminish with the complexity of the tasks, suggesting that both hemispheres were increasingly involved. Unlike R.B. and the normal controls, who made virtually no errors, L.B. had increasing difficulty as task complexity increased. He was better able to transfer information from the right to the left hemisphere than vice versa, but an analysis of his accuracy under the crossed conditions showed that the amount transferred was always well under one bit. This confirms previous evidence that L.B. has very limited subcortical transfer of either colour or shape.

Role of the corpus callosum in the somatosensory activation of the ipsilateral cerebral cortex: an fMRI study of callosotomized patients

To verify whether the activation of the posterior parietal and parietal opercular cortices to tactile stimulation of the ipsilateral hand is mediated by the corpus callosum, a functional magnetic resonance imaging (fMRI, 1.0 tesla) study was performed in 12 control and 12 callosotomized subjects (three with total and nine with partial resection). Eleven patients were also submitted to the tactile naming test. In all subjects, unilateral tactile stimulation provoked a signal increase temporally correlated with the stimulus in three cortical regions of the contralateral hemisphere. One corresponded to the first somatosensory area, the second was in the posterior parietal cortex, and the third in the parietal opercular cortex. In controls, activation was also observed in the ipsilateral posterior parietal and parietal opercular cortices, in regions anatomically corresponding to those activated contralaterally. In callosotomized subjects, activation in the ipsilateral hemisphere was observed only in two patients with splenium and posterior body intact. These two patients and another four with the entire splenium and variable portions of the posterior body unsectioned named objects explored with the right and left hand without errors. This ability was impaired in the other patients. The present physiological and anatomical data indicate that in humans activation of the posterior parietal and parietal opercular cortices in the hemisphere ipsilateral to the stimulated hand is mediated by the corpus callosum, and that the commissural fibres involved probably cross the midline in the posterior third of its body.

Bilateral field advantage and evoked potential interhemispheric transmission in commissurotomy and callosal agenesis

The role of the corpus callosum versus other cerebral commissures in the interhemispheric integration of visual information was studied in four individuals with complete agenesis of the corpus callosum, two individuals with partial agenesis, one total commissurotomy patient, and normal individuals. Evoked potential (EP) indices of interhemispheric transmission of visual sensory responses were observed during matching of unilateral and bilateral visual field letters and patterns. Neither the commissurotomy nor any of the acallosal patients had ipsilateral hemisphere visual EPs (P1 and N1), demonstrating that the posterior callosum is necessary for interhemispheric transmission of these components of visual evoked potentials. While the commissurotomy patient could not compare bilaterally presented letters, the anterior commissure of the acallosal patients appeared to be sufficient for interhemispheric comparison of single letters. However, bilateral comparison of more complex visual patterns resulted in considerable difficulty for complete agenesis patients, while comparison of patterns was more nearly normal when anterior callosal fibers were present (partial agenesis).

Same/different judgements about the direction and colour of apparent-motion stimuli after commissurotomy

One subject with complete forebrain commissurotomy (L.B.), another with posterior callosotomy (D.K.) and 6 normal controls were tested for their ability to make same/different judgements about the direction (Experiment 1) and colour (Experiment 2) of apparent-motion stimuli. In the first experiment, all of the subjects were able to judge whether the direction of apparent motion of two stimulus sequences was the same or different, regardless of whether the two sequences were presented unilaterally or bilaterally. In the second experiment, however, L.B. was unable to judge whether the colours of two lights comprising an apparent motion sequence were the same or different when the two lights were presented bilaterally although he was able to do so when both lights were in the same visual field. D.K. was able to make the discrimination on both bilateral and unilateral presentations but he performed more poorly in the bilateral condition than in either visual field. The control subjects performed consistently in all three locations. Given that same/different judgement tasks are a rigorous test of interhemispheric integration, the results indicate that the ability to discriminate apparent motion across the retinal midline is intact in split-brained subjects. Furthermore, the results support a dissociation in the interhemispheric integration of the direction and colour of apparent-motion stimuli after commissurotomy.

A dissociation in naming digits and colors following commissurotomy

Pairs of red and green digits were flashed in the opposite visual fields of L.B., a man with complete forebrain commissurotomy. He was highly accurate in naming both digits, the green ones alone, or the red ones alone, regardless of field. In contrast, he was relatively poor at naming the colors of the digits in the left visual field, and in comparing colors between fields. This dissociation suggests either (a) that information about shape was tagged by color and transferred from right to left hemisphere, but information about color itself was not transferred, or, more likely (b) that L.B.'s right hemisphere has acquired the ability to overtly name highly familiar stimuli, such as digits, but not colors.

Split decisions: problems in the interpretation of results from commissurotomized subjects

It has been widely accepted that there is little if any interhemispheric transfer of perceptual information in subjects following section of the forebrain commissures, although there may be diffuse transfer of emotional and connotative information, and an ability to direct spatial attention between hemispheres. However Sergent has recently shown that split-brained subjects can make rapid and accurate perceptual judgments based on visual stimuli presented simultaneously to the two visual fields. These include judgments of alignment, judgments of relative quantity, and higher-order judgments such as lexical decisions about letter strings straddling the visual fields. However it is premature to conclude that these tasks are accomplished by subcortical transfer of perceptual information; at least some of them can be explained in terms of strategies based on information available to just one hemisphere, or on the transfer of rudimentary information, perhaps accomplished through cross-cueing.

Binocularity and excitability loss in visual cortex cells of corpus callosum transected kittens and cats

The contribution of the corpus callosum to binocularity of visual cortex cells and to their responsiveness was studied in cats. Electrophysiological recordings of the responses of single cells to visual stimulation was performed in the callosal projection zone, visual cortex area 17-18 boundary in callosotomized cats. Callosotomy was carried out by transection of the visual segment of the corpus callosum in 6-7-week-old kittens and in acute and chronic adult cats (postoperative recovery time: 11 days-39 months). While in our normal cats the common proportion of binocularly driven cells (79.8%) was found (66.3% in the sham controls), a remarkable diminution (29.7%) was found in the callosotomized kittens, in the acute (39.7%) and in the chronic (50.6%)-operated cats. We have also found a change in the amount of binocularity as function of postoperative recovery time. While the proportion of binocular cells was conceivable (60.7%) in the short- and intermediate-term callosotomized cats (postoperative time: 0.3-5.5 months), it was diminished (36.9%) in the long-term (6.5-39 months) chronic cats. As to the responsiveness level, it was found that visual responsive cells constituted 88% of the cells in the normal and 80.3% in the sham controls. In comparison, they constituted 69.2% in the acute, 54.4% in the chronic and 52.8% in the callosotomized kittens. Furthermore, callosal transection had produced a symmetric effect in the two hemispheres, regarding binocularity and responsiveness. It has been thus concluded that the corpus callosum is essential for the mediation of binocular functions between the two hemispheres; in addition, cortical excitability has been also found to depend on callosal integrity.

Judgements about numerosity by a commissurotomized subject

A commissurotomized subject, L.B., was shown asterisks flashed at random locations, up to four in each field, and attempted either to compare the numbers in the two fields or to report the total number. The main results were: (a) Report was more accurate with unilateral than with bilateral presentation, suggesting that the difficulty integrating across fields was partly attentional; (b) in integrating across fields, attention was focused on one field, with only crude 'one-or-many' information from the other; (c) in cross-field comparisons, the focus was on the LVF, but in reporting the number it was on the RVF when report was oral or right-handed, and on the LVF when report was left-handed; (d) cross-field comparisons were improved when the locations were mirrored across the midline.

The MR equivalents of cerebral hemispheric disconnection: a telencephalic commissuropathy

Since the original observations of cerebral disconnection in experimental conditions with animal models, the surgical division of the cerebral commissures for control of seizure activity has led to the description of a human hemispheric disconnection syndrome. More recently, MR has revealed a spectrum of conditions of the forebrain responsible for the natural development and acquired occurrence of this unique commissural syndrome. Acutely, nonspecific expressions of disconnection may be observed; however, there are no explicit behavioral manifestations of this syndrome in the chronic state. The signs of stabilized hemispheric disconnection are only elicitable via sophisticated, neuropsychological testing performed by experienced technicians. Included in the present series of callosal pathology were individuals with agenesis, surgical section, infarction, trauma, glioma, and Marchiafava-Bignami disease. On specialized examination, the surgical "split-brain" patients illustrated the classic syndrome of a halt of interhemispheric transfer of information, a block of nondominant hemisphere access to language functions and an inhibition of dominant hemisphere access to superior visuospatial capacity. The nonsurgical subjects with developmental callosal agenesis and acquired pathologic processes involving the callosum revealed a varied, nonspecific reduction in cognitive function most probably related to associated extracallosal hemispheric pathology. No external abnormalities in behavior which could be attributed specifically to the callosal findings were identified chronically, although some progressive diseases suggested the presence of a subacute symptomatic phase of the disconnection syndrome. Any condition which causes the division, destructive insult, or primary ontogenic nondecussation of commissural axons will produce an interference in interhemispheric, interneuronal communication. This is manifested by individual degrees of covert interhemispheric data transfer arrest and of reduced bidirectional dominant-nondominant hemispheric access and telencephalic integrative capacity, consistent with a cerebral commissuropathy. This study reveals the MR equivalents of the hemispheric disconnection syndrome which occur in many nonspecific pathologic conditions of the cerebrum, but which result in few if any overt behavioral aberrations in the stabilized state.

Schizophrenia, corpus callosum, and interhemispheric communication: a review

The literature suggests that some schizophrenic patients may have a dysfunction in the transfer of information between the two cerebral hemispheres via the corpus callosum (CC). The presence of an abnormally thickened CC in some schizophrenic patients and an abnormally thin CC in others suggests a possible anatomic basis for abnormal organization of cognitive functions in these patients resulting from either increased or decreased communication between the two cerebral hemispheres. The findings of the anatomic studies have suggested a connection between a thickened CC and both early onset and negative symptom patterns in schizophrenic patients. Similarly, onset of symptoms later in life and positive symptom patterns have been associated with a thinner CC. Behavioral and electrophysiological findings appear to support the idea that the same subgroups may be associated with opposite patterns of effective transfer of information between the cerebral hemispheres. Careful analysis of the transfer of information between the cerebral hemispheres for selected groups of schizophrenic patients and normal controls could provide important information about cerebral organization and possible mechanisms operative in schizophrenia.

Negative evidence of difference between right- and left-handers in interhemispheric transfer of information

The present study has been designed to investigate the relationship between handedness and callosal function. Based on the previous finding of a greater anatomical connection between the hemispheres in left-handers than in right-handers and in order to test the possibility of greater functional communication, the efficiency of interhemispheric transfer of information has been measured using a finger localization task. Comparison of two groups of male right-handed and left-handed subjects shows no difference in the efficiency of interhemispheric transfer of information.

Dichotic listening in commissurotomized and hemispherectomized subjects

Three commissurotomized and two left-hemispherectomized subjects were tested on spoken report of sequences of three dichotic pairs of digits. With instruction to report only one digit from each pair, there was an overall advantage to the ear contralateral to the hemisphere mediating speech, but report of ipsilateral-ear digits ranged from 40 to 100%. In commissurotomized subjects, the more extreme ipsilateral suppression under instructions to report all digits may be due to failure to gain access to unattended information stored in the right hemisphere, rather than to suppression of the ipsilateral pathway. However one commissurotmized patient did appear to have access to right-hemisphere items, the result either of subcortical transfer or of external cross-cueing. The hemispherectomized subjects seemed able to store both attended and unattended information in the same hemisphere.

Subcortical transfer of cognitive information in subjects with complete forebrain commissurotomy

Three complete commissurotomy subjects were tested for the ability to integrate cognitive information presented for a prolonged duration to opposite visual hemi-fields. A sample stimulus was projected to one hemi-field and a three-choice array to the other, with one choice related to the sample on an abstract or concrete basis. All subjects performed the non-verbal matching task at well above chance level, with scores comparable to those attained when the task was performed completely within one or the other hemisphere. Transfer was equally successful in the two directions, though the pathway originating in the right and terminating in the left hemisphere may be more sensitive to affective components of the stimulus. The information relayed subcortically is neither verbal nor imagic in nature, but appears to correspond to contextual or connotative associates of the stimulus. The results are discussed in terms of a cognitive system common to the two hemispheres and involving associational networks but not lateralized functions such as language and complex visuospatial processes.

Interhemispheric communication in a case of total visuo-verbal disconnection

Interhemispheric communication strategies as noted in split-brain subjects may serve a compensatory function in patients with total visuo-verbal disconnection. The results of a case support this expectation. Identification of visual material reflected the perceptual and linguistic limitations of the right hemisphere, and naming was sometimes mediated by the rehearsal of alternatives. Implications for disconnection theory are discussed.