|
During RNB, compared to a period of BAW, there was a significantly higher peak amplitude of the Na wave on the right side. This is the negative peak preceding the positive Pa wave (25–32 ms). Similarly, the peak amplitude of the Nb wave (the first maximum negative peak following the Pa wave) was significantly higher on the right side during RNB compared to the preceding period as well the period of BAW. There were no changes found on either side during BAW. There was no correlation between changes during the session and pre-session nostril patency. Regarding the origins of the MLAEPs, there are two schools of thought which claimed the neural and the muscular origin of MLAEPs, and have been equivocal. Ruhm and colleagues [15] termed the smaller potentials as cochleoneurogenic and the bigger ones as vestibulomyogenic. Davis [16] also differentiated these responses according to their myogenic and neurogenic origin. The myogenic part which originated from posterior auricular and neck muscles was called the sonomotor reflex with an onset at 12 ms. In the present study it is also important to note that the auricular reference electrodes may have picked up myogenic activity. However, ear lobe recording has been shown to have less muscle artefacts compared to recording from the mastoid [17], which was the reason for selecting auricular reference recording. The neurogenic part according to Davis [16] originated at the medial geniculate body (thalamus), primary cortical areas and the immediate adjoining secondary areas. Ruhm and colleagues [15] were among the first to establish the neurogenic origin of auditory middle latency responses by recordings from the brain surface during craniectomies . Currently the neural generators of the different compo- nents are as follows: the Na wave has been postulated to be due to activity at the mesencephalic or diencephalic level [18], the Pa wave corresponds to the activity at the superi-or temporal gyrus [19] and the generator of the Nb wave is relatively localised in the dorso-posterior-medial part of the Heschl’s gyrus, i.e., the primary auditory cortex [8]. In summary, the MLAEPs have a myogenic component that can be minimised by fully relaxing the subject [20]. In the present study the amplitudes of the Na wave and the Nb wave increased during yoga breathing. While the earli- er component (i.e., the Na wave) may have been modified by myogenic activity, the chance of muscle artefacts con- tributing to the Nb wave amplitude is unlikely. Also, in pranayamas involving nostril manipulation, voluntary occlusion of the nostrils is traditionally done with the right hand [13]. Hence the tonic muscular contraction of the right arm could have influenced (if at all) the activity over the left hemisphere in the form of movement-related activ- ity. This was not seen in the present study. Apart from the myogenic contribution to MLAEPs, a report on the variation of the middle latency evoked poten- tials with the physical characteristics of the stimuli, showed the Na and Pa waves are the most consistent and reliable waves evoked [21–23]. P0 and Nb are the next most reliable. An increase in the amplitude of an evoked potential component has been interpreted as being indicative of effective activation of the underlying neural generator [24]. The Na wave has been postulated to be due to activity at the mesencephalic or diencephalic level [18]. Intracerebral recording in man has shown that the neural generator of the Nb wave is relatively localised in the dorso-posterior-medial part of the Heschl’s gyrus, i.e., the primary auditory cortex [8]. As described above, the Na wave is believed to correspond to the mesencephalic-dien- cephalic (thalamic) level, possibly the medial geniculate body, while the Nb wave is relatively localised in the Heschl’s gyrus. Changes at the cortical level during the yoga breathing practice studied here could be expected to be lateralised, as was described in other studies on uninostril breathing [2, 5]. In attempting to understand why sub-cortical changes were also lateralised (i.e., the Na wave), a possible explanation is that the descending corticofugal control mechanisms could be expected to exert significant influences on the processing of information at the brainstem and thalamic levels [25]. As it has been recognised that the corticofugal descending inputs to the medial geniculate body are stronger ipsilaterally than contralaterally, these descending pathways may explain the lateralised effect at these subcortical levels [26]. The Pa component has been shown to be correlated with activity at the level of the superior temporal gyrus [19]. If the uninostril yoga breathing were to produce a lateralised effect, it would be likely to have had a lateralised effect at this level. In the present study the Pa wave amplitude showed a higher value on the right side during RNB as compared to the changes on the left side, but this differ- ence was not significant. A possible explanation is as follows. When measuring the peak amplitudes of individual components of evoked potentials, an inherent difficulty is that an increase in a positive component would influence the amplitude of a subsequent negative component and vice versa [27]. In the present case, the Pa wave is a positive component occurring between two negative waves i.e., the Na and the Nb waves, both of which showed an increase in amplitude. Hence the increase in the amplitudes of the preceding Na and succeeding Nb negative waves may have influenced the amplitude of the Pa wave, which is in between them. A usual way of overcoming this problem is to measure the peak-to-peak amplitude. However, in the present study this method was not chosen as the changes in neural generators corresponding to specific waves individually were intended to be studied. A previous report on cytoarchitechtonic data suggests that the primary auditory cortices of right and left hemispheres are similarly organised [28]. This may explain why under baseline conditions auditory evoked potentials with neural generators in the right and left primary auditory cortices have similar latency and amplitude characteristics [8]. Hence the asymmetric increase in amplitudes of MLAEP components during RNB appears to be related to the breathing practice, rather than asymmetry in the under- lying neural generators. Previous studies on EEG activity of right and left hemispheres related to spontaneous shifts in nostril dominance have suggested that there was an increase in the EEG amplitude over the contralateral hemisphere [2]. Similarly, contralateral hemispheric activation based on EEG activi- ty was also demonstrated during forced uninostril breath- ing [5]. In these studies increased EEG amplitudes were considered indicators of increased mental activity in that hemisphere. Performance in hemisphere-specific tasks immediately after forced uninostril breathing showed that after forced left nostril breathing there was improved performance in spatial tasks, considered as right hemisphere functions. As verbal tasks were performed better after forced RNB, it was stated that contralateral hemisphere activation occurred with forced uninostril breathing [29]. Previous studies described a contralateral hemisphere function enhancing effect based on the EEG, whereas in the present study changes in MLAEPs suggested that enhancement occurred ipsilaterally. Other studies on the performance in hemisphere-specific tasks related to both spontaneous shifts in nostril dominance and forced uni- nostril breathing reported improved performance in tasks specific to the contralateral hemisphere [3, 29]. In contrast, yoga breathing through right, left and alternate nostrils as well as BAW with no nostril manipulation, improved per- formance in a right hemisphere task (spatial memory) with no lateralised effect following yoga [30]. In order to deter- mine whether the trend of an increase in amplitudes of evoked potential components ipsilateral to the nostrilthrough which breathing is practiced occurred irrespective of the nostril, it is necessary to refer to unpublished data of a study also on 14 male volunteers (group average age±SD, 26.86±5.04 years) who were evaluated in ‘left nostril yoga breathing (LNB)’ and in ‘BAW sessions’. There were no significant changes during the left nostril yoga breathing and the peak amplitudes of the Na and Nb waves are mentioned here. For the Na wave on the left side in the LNB session: 1.12±1.45 µV (pre), 1.45±1.72 µV (during), 1.19±1.67 µV (post); and in the BAW session: 0.67±0.87 µV (pre), 0.62±0.45 µV (during), 0.74±0.68 µV (post). For the Na wave on the right side in the LNB ses- sion: 1.22±2.21 µV (pre), 1.03±0.78 µV (during), 1.08±1.45 µV (post); and in the BAW session: 1.37±1.11 µV (pre), 1.14±1.42 µV (during), 1.14±0.79 µV (post). For the Nb wave on the left side in the LNB session: 0.64±0.68 µV (pre), 0.65±0.45 µV (during), 0.52±0.56 µV (post); and in the BAW session: 0.51±0.54 µV (pre), 0.45±0.38 µV (during), 0.38±0.17 µV (post). For the Nb wave on the right side in the LNB session: 0.76±0.81 µV (pre), 0.60±0.51 µV (during), 0.64±0.66 (post); and in the BAW session 0.84±0.53 µV (pre), 0.53±0.47 µV (during), 0.30±0.25 µV (post). Spontaneous shifts in nostril patency and forced uni- nostril breathing have similar effects (i.e., contralateral enhancement of hemispheric activity). In the present study, the higher peak amplitudes of two MLAEP components on the right side suggested that RNB brought about effective activation of underlying diencephalic and primary auditory cortical generators on the right side. Certain psychiatric disorders are known to be associated with selective disruption of the function of a specific hemisphere. Uninostril breathing practices have potential use in conditions like this. For example, left forced uninostril breathing was tried with success in obsessive compulsive disorder, which was described by Shannahoff- Khalsa and Beckett (1996) as a disorder of the right hemisphere [31]. This makes it desirable for the effects of these practices to be understood in normal volunteers.
|
