A subcortical correlate of P300 in man

Event-related potentials in visual and auditory target detection tasks were recorded simultaneously from the scalp, somatosensory thalamus and periaqueductal gray in a chronic pain patient with electrodes implanted subcortically for therapeutic purposes. Short latency tactile responses confirmed the location of the thalamic electrodes.Rare auditory stimuli which were detected by the subject were accompanied by a prominent P300 component at the scalp, and by negative activity at the subcortical sites with the same latency as the scalp positivity. This activity was not seen in responses to frequent non-target stimuli and was not dependent on an overt motor response.Similarly, rare visual stimuli generated a scalp P300 and negative activity subcortically; both scalp and subcortical waves had a longer latency than in the auditory experiment. The reaction time was similarly longer to visual targets.These data are inconsistent with a hippocampal generator for P300, but are consistent with a generator in the thalamus or more dorsally located structures.     

Event-related potentials recorded from the scalp and nasopharynx. II. N2, P3 and slow wave

Scalp and nasopharyngeal recordings of the N2, P3 and slow wave components were compared in a target-detection task. The effects of probability, interstimulus interval, intensity, discrimination difficulty, attention, stimulus omission and modality were evaluated. Waves of opposite polarity to the scalp N2 and P3 components were recorded in the nasopharynx. The scalp and nasopharyngeal N2 components showed different patterns of variation across experimental conditions. These findings indicate that there are two different cerebral processes occurring at the latency of the scalp N2. The scalp and nasopharyngeal P3 components consistently covaried across conditions, suggesting a single underlying process. The slow wave was observed only in the scalp recordings.     

Effects of aging on event-related brain potentials (ERPs) in a visual detection task

Event-related brain potentials (ERPs) were recorded from 74 subjects (45 men) between 18 and 82 years of age in a simple visual detection task. On each trial the subject reported the location of a triangular flash of light presented briefly 20° laterally to the left or right visual field or to both fields simultaneously. ERPs to targets exhibited a similar morphology including P1, N1, P2, N2, and P3 components across all age groups. The principal effects of advancing age were (1) a marked reduction in amplitude of the posterior P1 component (75–150 latency) together with an amplitude increase of an anterior positivity at the same latency; (2) an increase in amplitude of the P3 component that was most prominent over frontal scalp areas; and (3) a linear increase in P3 peak latency. These results extend the findings of age-related changes in P3 peak latency and distribution to a non-oddball task in the visual modality and raise the possibility that short-latency ERPs may index changes in visual attention in the elderly.     

Contributions from the auditory nerve to the brain-stem auditory evoked potentials (BAEPs): results of intracranial recording in man

Intraoperative recordings obtained from electrodes placed on the scalp (vertex and earlobe or ear canal) in response to click stimulation were compared with recordings made directly from the auditory nerve in patients undergoing microvascular decompression (MVD) operations to relieve hemifacial spasm (HFS) and disabling positional vertigo (DPV). The results support earlier findings that show that the auditory nerve is the generator of both peak I and peak II in man, and that it is the intracranial portion of the auditory nerve that generates peak II. The results indicate that the second negative peak in the potentials recorded from the earlobe is generated by the auditory nerve where it passes through the porus acusticus into the skull cavity, and that the proximal portion of the intracranial portion of the auditory nerve generates a positive peak in the potentials that are recorded from the vertex. This peak appears with a latency that is slightly longer than that of the second negative peak in the potentials recorded from the earlobe (or ear canal). The second negative peak in the recording from the ear canal and the positive peak in the vertex recording contribute to peak II in the differentially recorded BAEP. Since our results indicate that the difference in the latency of the second negative peak in the recording from the earlobe and that of the positive peak in the vertex recording represents the neural travel time in the intracranial portion of the auditory nerve, this measure may be valuable in the differential diagnosis of eighth nerve disorders such as vascular compression syndrome.     

Topography of middle-latency somatosensory evoked potentials following painful laser stimuli and non-painful electrical stimuli

The aim of this study was to compare cerebral evoked potentials following selective activation of Aβ and Aδ fibers. In 15 healthy subjects, Aβ fibers were activated by electrical stimulation of the left radial nerve at the wrist. Aδ fibers were activated by short painful radian heat pulses, applied to the dorsum of the left hand by a CO₂ laser. Evoked potentials were recorded with 15–27 scalp electrodes, evenly distributed over both hemispheres (bandpass 0.5–200 Hz). The laser-evoked potentials exhibited a component with a mean peak latency of 176 msec (N170). Its scalp topography showed a parieto-temporal maximum contralateral to the stimulus side. In contrast, the subsequent vertex negativity (N240), which appeared about 60 msec later, had a symmetrical scalp distribution. Electrically evoked potentials showed a component at 110 msec (N110), that had a topography similar to the laser-evoked N170. The topographies of the N170 and N110 suggest that they may both be generated in the secondary somatosensory cortex. There was no component in the electrically evoked potential that had a comparable interpeak latency to the following vertex potential: for N60 it was longer, for N110 it was shorter. On the other hand, in the laser-evoked potentials no component could be identified the topography of which corresponded to the primary cortical component N20 following electrical stimulation.     

Temporal correspondence of intracranial,cochlear and scalp-recorded human auditory nerve action potentials

Conventional, vertex-ipsilateral ear records (‘A’), as well as 3-channel Lissajous' trajectories (3-CLTs) of auditory brain-stem evoked potentials (ABEPs) were recorded from the scalp simultaneously with tympanic membrane electrocochleograms (‘TME’) and auditory nerve compound action potentials (‘8-AP’) recorded intracranially using a wick electrode on the auditory nerve between the internal auditory meatus and the brain-stem. The recordings were made during surgical procedures exposing the auditory nerve.The peak latency recorded from ‘TME’ corresponded to trajectory amplitude peak ‘a’ of 3-LLT and to peak ‘I’ of the ‘A’ channel ABEP. Peak latency of ‘8-AP’ was slightly longer than the latency of peak ‘II’ of ‘A’ when ‘8-AP’ was recorded from the root entry zone and the same or shorter when recorded from the nerve trunk. ‘8-AP’ peak latency was shorter than trajectory amplitude peak ‘b’ of 3-CLT regardless of where the wick electrode was along the nerve. Peak latencies from all recordings sites clustered into two distinct groups—those that included N1 from ‘TME’, peak ‘I’ of the ‘A’ record and trajectory amplitude peak ‘a’ of 3-CLT, and those that included the negative peak of ‘8-AP’ and trajectory amplitude peak ‘b’ of 3-CLT, as well as peak ‘II’ of the ‘A’ record, when present. In one case, the latency of peak ‘II’ and trajectory amplitude peak ‘b’ was manipulated by changing the conductive properties of the medium surrounding the auditory nerve.These results are consistent with other evidence proposing: (1) the most distal (cochlear) portion of the auditory nerve is the generator of the first ABEP component (‘I’, ‘a’); (2) the proximal auditory nerve is the major contributor to the ‘A’ channel ABEP component ‘II’; (3) in addition to the auditory nerve, more central structures participate in the generation of the 3-CLT ‘b’ component.     

Laser-evoked potentials: exogenous and endogenous components

The aim of this study was to distinguish the exogenous component (related to the physical properties of the stimulus) and the endogenous component (reflecting event-related cognitive processing) of the laser-evoked potential (LEP). Short painful radiant heat pulses generated by a CO₂-laser were applied to the dorsum of the right and left foot. LEPs were recorded with 5 scalp electrodes in the midline versus linked earlobes in 26 healthy subjects. In order to identify the exogenous component, the LEP was recorded during a standardised distraction task (reading a short story). To identify the endogenous component P3 for the LEP, a 2-stimulus oddball paradigm was used (20% probability of targets). When the task of the oddball paradigm consisted of pressing a button, a movement-related long-latency negativity (N1200) was recorded in frontal leads that was absent in a counting task. The LEP of targets, frequent non-targets and during distraction was dominated by a single large positivity. The amplitude of this positivity was task-dependent and increased the more attention the subject payed to the laser stimuli (distraction < neutral < non-target < target). The laser-evoked positivity during distraction had a peak latency of about 400 msec (P400) and a maximum amplitude at the vertex, which was independent of inter-stimulus interval. The P3 following laser stimulation had a significantly later peak at about 570 msec (P570) and a different scalp topography with a parietal maximum. Its amplitude decreased when the interstimulus interval was reduced from 10 to 6 sec. Under neutral instructions, the LEP positivity consisted of a superposition of both the exogenous P400 and the endogenous P570.     

Separate generators with distinct orientations for N20 and P22 somatosensory evoked potentials to finger stimulation?

Sequential spatial maps of scalp potentials, obtained with a 16-channel montage, were used in 12 healthy subjects in order to assess the temporal and spatial distribution of early cortical SEPs to single finger stimulation. It was found that when the contralateral parietal N20 negativity peaks there is a synchronous frontal P20 positivity, supporting the view of a tangentially orientated dipolar generator for this couple of scalp SEPs components. It was not possible to show a distribution of N20 peak on the scalp that would parallel the somatotopic finger representations in area S1; however, the orientation of the putative dipolar source of the N20/P20 complex was found to change according to the finger stimulated. A central P22 component was also constantly obtained without any synchronous negativity on the scalp surface corresponding to the electrode array; a clear somatotopic organisation was found for P22. These features favour the hypothesis that this latter component has a radially orientated generator situated in the prerolandic motor cortex, close to the scalp surface. Because of overlapping between the P20 and P22 components, the determination of P22 onset latency was hazardous in some cases, and spatial mapping was then essential to identify this component. The conclusion that the contralateral parietal N20 and central P22 could be generated by separate dipolar generators with distinct orientations is supported by recent data from combined electrical and magnetic field recording.     

Pain-related and cognitive components of somatosensory evoked potentials following CO2 laser stimulation in man

We recorded cortical potentials evoked by painful CO₂ laser stimulation (pain SEP), employing an oddball paradigm in an effort to demonstrate event-related potentials (ERP) associated with pain. In 12 healthy subjects, frequent (standard) pain stimuli (probability 0.8) were delivered to one side of the dorsum of the left hand while rare (target) pain stimuli (probability 0.2) were delivered to the other side of the same hand. Subjects were instructed to perform either a mental count or button press in response to the target stimuli. Two early components (N2 and P2) of the pain SEP demonstrated a Cz maximal distribution, and showed no difference in latency, amplitude or scalp topography between the oddball conditions or between response tasks. In addition, another positive component (P3) following the P2 was recorded maximally at Pz only in response to the target stimuli with a peak latency of 593 msec for the count task and 560 msec for the button press task. Its scalp topography was the same as that for electric and auditory P3. The longer latency of pain P3 can be explained not only by its slower impulse conduction but also by the effects of task difficulty in the oddball paradigm employing the pain stimulus compared with electric and auditory stimulus paradigms. It is concluded that the P3 for the pain modality is mainly related to a cognitive process and corresponds to the P3 of electric and auditory evoked responses, whereas both N2 and P2 are mainly pain-related components.     

Midlatency auditory evoked responses: Differential effects of sleep in the cat

Middle latency responses (MLRs) in the 10–100 msec latency range, evoked by click stimuli, were studied in 8 adult cats during sleep-wakefulness to determine whether such changes in state were reflected by any MLR component. In particular, we wanted to determine whether the 20–22 msec positivity recorded at the vertex, ‘wave A,’ shown in previous studies to reflect a generator substrate within the ascending reticular formation, was tightly linked to changes in sleep-wakefulness, as reported for single neurons in the ascending reticular activating system. Evoked potentials were collected in 100 trial averages during continuous presentation of 1/sec clicks during initial awake recordings and thereafter during all-night sleep sessions. Continuously recorded EEG, EOG and EMG were scored for wakefulness, slow wave sleep (SWS), and rapid eye movement (REM) sleep during each evoked potential epoch. Recordings were obtained from electrodes implanted at the vertex and overlying the primary auditory cortex referenced to frontal sinus or to neck. In agreement with others, components of the auditory brain-stem response and the 12 msec primary cortical response showed no change in amplitude from wakefulness to either SWS or REM. Only wave A, among the components evaluated in the 1–100 msec range, decreased and disappeared during SWS and dramatically reappeared during REM to an amplitude equal to that during wakefulness. These data lend particular support to a functional relation between wave A and the ascending reticular activating system and suggest that this potential may provide a unique and dynamic probe of tonic brain activity. Moreover, this animal model provides a hypothetical basis for expecting a similar surface recorded potential in the human, a potential which has consequently been discovered.     

Neural generators of the somatosensory evoked potentials: Recording from the cuneate nucleus in man and monkeys

The neural generators of the somatosensory evoked potentials (SEPs) elicited by electrical stimulation of the median nerve were studied in man and in rhesus monkeys. Recordings from the cuneate nucleus were compared to the far-field potentials recorded from electrodes placed on the scalp. It was found that the shape of the response from the surface of the human cuneate nucleus to stimulation of the median nerve is similar to that of the response recorded more caudally in the dorsal column, i.e., an initially small positivity followed by a negative wave that is in turn followed by a slow positive wave. The beginning of the negative wave coincides in time with the N14 peak in the SEP recorded from the scalp, and its latency is 13 msec. The response from the cuneate nucleus in the rhesus monkey has a similar shape and its negative peak appears with the same latency as the positive peak in the vertex response that has a latency of 4.5 msec; the peak negativity has a latency of about 6 msec and thus coincides with P6.2 in the vertex recording. Depth recordings from the cuneate nucleus and antidromic stimulation of the dorsal column fibers in the monkey provide evidence that the early components of the response from the surface of the cuneate nucleus are generated by the dorsal column fibers that terminate in the nucleus.The results support the hypothesis that the P14 peak in the human SEP is generated by the termination of the dorsal column fibers and that the cuneate nucleus itself contributes little to the far-field potentials.     

Scalp topography and dipolar source modelling of potentials evoked by CO2 laser stimulation of the hand

CO₂ laser evoked potentials to hand stimulation recorded using a scalp 19-channel montage in 11 normal subjects consistently showed early N1/P1 dipolar field distribution peaking at a mean latency of 159 ms. The N1 negativity was distributed in the temporoparietal region contralateral to stimulation and the P1 positivity in the frontal region. The N1/P1 response was followed by 3 distinct components: (1) N2a reaching its maximal amplitude at the vertex and ipsilaterally to the stimulated hand, (2) N2b mostly distributed in the frontal region, and (3) P2 with a mid-central topography. Brain electrical source analysis showed that this sequence was explained, with a residual variance below 5%, by a model including two dipoles in the upper bank of the Sylvian fissure of each hemisphere, a frontal dipole close to the midline, and two anterior medial temporal dipoles, thus suggesting a sequential activation of the two second somatosensory areas, anterior cingulate gyrus and the amygdalar nuclei or the hippocampal formations, respectively. This model fitted well with the scalp field topography of grand average responses to stimulation of left and right hand obtained across all subjects as well as when applied to individual data. Our findings suggest that the second somatosensory area contralateral to the stimulation is the first involved in the building of pain-related responses, followed by ipsilateral second somatosensory area and limbic areas receiving noxious inputs from the periphery.     

Median and tibial nerve somatosensory evoked potentials: middle-latency components from the vicinity of the secondary somatosensory cortex in humans

The topography of the middle-latency N110 after radial nerve stimulation suggested a generator in SII. To support this hypothesis, we have tried to identify a homologous component in the tibial nerve SEP (somatosensory evoked potential). Evoked potentials following tibial nerve stimulation (motor+sensory threshold) were recorded with 29 electrodes (bandpass 0.5–500 Hz, sampling rate 1000 Hz). For comparison, the median nerve was stimulated at the wrist. Components were identified as peaks in the global field power (GFP). Map series were generated around GFP peaks and amplitudes were measured from electrodes near map maxima. With median nerve stimulation, we recorded a negativity with a maximum in temporal electrode positions and 106±12 ms peak latency (mean±SD), comparable to the N110 following radial nerve stimulation. After tibial nerve stimulation the latency of a component with the same topography was 131±11 ms (N130). Both N110 and N130 were present ipsi- as well as contralaterally. Amplitudes were significantly higher on the contralateral than the ipsilateral scalp for both median (3.1±2.4 μV vs. 1.7±1.6 μV) and tibial nerve (1.9±1.2 μV vs. 0.6+1 μV). The topography of the N130 can be explained by a generator in the vicinity of SII. The latency difference between median and tibial nerve stimulation is related to the longer conduction distance (cf. N20 and P40). The smaller ipsilateral N130 is consistent with the bilateral body representation in SII.     

Pain-related somatosensory evoked potentials following CO2 laser stimulation in man

0ain-related somatosensory evoked potentials (SEPs) following CO₂ laser stimulation were analyzed in normal volunteers. Low power and long wavelength CO₂ laser stimuli to the hand induced a sharp pain which was associated with a large positive component, P320, recorded over the scalp. Amplitude decreased and latency increased with reduction in stimulus intensity and subjective pain feeling. P320 was maximal at the vertex but was distributed widely over the scalp. There were no topographic differences between left- and right-hand stimulation, or between hand and chest stimulation. Lidocaine injection to produce anesthetic nerve block resulted in loss of P320, but the potential was relatively preserved during ischemic nerve block. No potential corresponding to P320 could be recorded following electrical or mechanical tactile stimulation.We consider P320 to be generated by impulses arising from pain stimuli and ascending through Aδ fibers. We propose the thalamus as a generator source from considering its scalp topography, but pain-specific cognition or perception may also be involved in generating this potential.     

Comparison in man of short latency averaged evoked potentials recorded in thalamic and scalp hand zones of representation

Recordings were performed in the thalamus of 13 patients suffering from either abnormal movements or intractable pain, with the aim of delimiting the region to be destroyed or stimulated in order to diminish the syndrome. In 11 of these patients averaged evoked potentials were recorded simultaneously from the scalp and specific thalamus (VP) hand area levels following median nerve stimulation. These recordings were done during the operation or afterwards when an electrode was left in place for a program of stimulation.The latencies of onsets and peaks on the scalp ‘P15’ were compared with those of the VP wave; a clear correspondence was found. Moreover, when increased stimulation was used, both waves began to develop in parallel. Thus in the contralateral ‘P15’ a component exists due to the field produced by the thalamic response. To explain the presence of an ipsilateral scalp ‘P15’ wave, we propose that a second wave having the same latency and a slightly shorter peak exists on the scalp due to a field produced by a brain-stem response. This double origin of ‘P15’ is also shown by the different changes which the ipsilateral and contralateral waves present during changes in alertness.The scalp ‘N18–N20’ is also composed of at least 2 components. The first peak appears on the scalp with a latency shorter than that of the negativity which develops in the thalamus. The N wave, moreover, increases in latency with rapid stimulus repetition. We propose with others that ‘N18’ is a cortical event reflecting the arrival of the thalamo-cortical volley. The second component, ‘N20,’ has a peak latency closely correlated to that of the thalamic negativity. This component was present alone in ‘N’ when rapid stimulation (> 4/sec) was used, which did not change the thalamic response. It must be a field produced by the thalamic negativity.     

Comparison of scalp distribution of short latency somatosensory evoked potentials (SSEPs) to stimulation of different nerves in the lower extremity

SSEPs to stimulation of the CPN at the knee and PTN, PN and SN at the ankle were recorded from 15 cephalic sites and compared in 8 normal subjects. The configuration, amplitude, peak latency and distribution of P27, N35 (CPN) and P37, N45 (PTN, PN and SN) were analyzed. The configuration and distribution of SSEPs to stimulation of the 3 nerves at the ankle were similar across subjects. Both P37 and N45 were greatest in amplitude at the vertex and at recording sites ipsilateral to the side of stimulation. At contralateral sites either negative (N37) or negative, positive, negative potentials were recorded. The peak latency of N37 was the same or slightly less than that of P37. CPN-SSEPs were lower in amplitude and their configuration and scalp distribution showed much greater intersubject variability. This suggests that complex mechanisms which variably interact with one another are reflected in scalp SSEPs to CPN stimulation at the knee. The larger amplitude plus the minimal intersubject variability in morphology and topography of PTN-SSEPs indicate that this nerve is the most suitable for routine clinical use.     

Mismatch negativity of ERP in cross-modal attention

Event-related potentials were measured in 12 healthy youth subjects aged 19-22 using the paradigm "cross-modal and delayed response" which is able to improve unattended purity and to avoid the effect of task target on the deviant components of ERP. The experiment included two conditions: (i) Attend visual modality, ignore auditory modality; (ii) attend auditory modality, ignore visual modality. The stimuli under the two conditions were the same. The difference wave was obtained by subtracting ERPs of the standard stimuli from that of the deviant stim-uli. The present results showed that mismatch negativity (MMN), N2b and P3 components can be produced in the auditory and visual modalities under attention condition. However, only MMN was observed in the two modalities un-der inattention condition. Auditory and visual MMN have some features in common: their largest MMN wave peaks were distributed respectively over their primary sensory projection areas of the scalp under attention condition, but over front     

Pain-related somatosensory evoked potentials following CO2 laser stimulation of foot in man

Since our previous study of pain somatosensory evoked potentials (SEPs) following CO₂ laser stimulation of the hand dorsum could not clarify whether the early cortical component NI was generated from the primary somatosensory cortex (SI) or the secondary somatosensory cortex (SII) or both, the scalp topography of SEPs following CO₂ laser stimulation of the foot dorsum was studied in 10 normal subjects and was compared with that of the hand pain SEPs and the conventional SEPs following electrical stimulation of the posterior tibial nerve recorded in 8 and 6 of the 10 subjects, respectively. Three components (N1, N2 and P2) were recorded for both foot and hand pain SEPs. N1 of the foot pain SEPs was maximal at the midline electrodes (Cz or CPz) in all data where that potential was recognized, but the potential field distribution was variable among subjects and even between two sides within the same subject. N1 of the hand pain SEPs was maximal at the contralateral central or midtemporal electrode. The scalp distribution of N2 and P2, however, was not different between the foot and hand pain SEPs. The mean peak latency of N1 following stimulation of foot and hand was found to be 191 msec and 150 msec, respectively, but there was no significant difference in the interpeak latency of Nl-N2 between foot and hand stimulation. It is therefore concluded that NI of the foot pain SEPs is generated mainly from the foot area of SI. The variable scalp distribution of the N7 component of the foot pain SEPs is likely due to an anatomical variability among subjects and even between sides.     

A method of calculating spinal cord transit time from potentials evoked by tibial nerve stimulation in normal subjects and in patients with spinal cord disease

Somatosensory potentials were evoked by stimulation of the tibial nerve at the ankle and recorded over the spine and scalp in 16 normal subjects and 26 patients with known or suspected spinal cord disease, with the aim of developing a method of measuring spinal sensory conduction velocity using a tolerable number of stimuli, applied unilaterally to alert subjects.In normal subjects N21 was consistently recorded overL1 vertebra and in most subjects a complex, N27/N29/P33, was recorded over the cervical spine referred to the vertex. Constant latencies at different spinal levels and, in one subject, comparison with the latency of the ascending volley indicate that the complex was not derived from the spinal cord but from more rostral structures, and therefore only transit time, rather than velocity, could be measured.In patients with clinically definite multiple sclerrosis, even with minimal clinical signs, the N27/N29/P33 complex was always abnormal. Abnormalities in this and other forms of spinal cord disease were commonly absence or distortion of the complex, prolonged transit time being rare. The clinical value of the method is limited by the very low amplitude of the responses.     

Direct recording of somatosensory evoked potentials in the vicinity of the dorsal column nuclei in man: their generator mechanisms and contribution to the scalp far-field potentials

Somatosensory evoked potentials (SEPs) in the vicinity of the dorsal column nuclei in response to electrical stimulation of the median nerve (MN) and posterior tibial nerve (PTN) were studied by analyzing the wave forms, topographical distribution, effects of higher rates of stimulation and correlation with components of the scalp-recorded SEPs. Recordings were done on 4 patients with spasmodic torticollis during neurosurgical operations for microvascular decompression of the eleventh nerve. The dorsal column SEPs to MN stimulation (MN-SEPs) were characterized by a major negative wave (N1; 13 msec in mean latency), preceded by a small positivity (P1) and followed by a large positive wave (P2). Similar wave forms (P1′-N1′-P2′) were obtained with stimulation of PTN (PTN-SEPs), with a mean latency of N1′ being 28 msec. Maximal potentials of MN-SEPs and PTN-SEPs were located in the vicinity of the ipsilateral cuneate and gracile nuclei, respectively, at a level slightly caudal to the nuclei. The latencies of P1 and N1 increased progressively at more rostral cervical cord segments and medulla, but that of P2 did not. A higher rate of stimulation (16 Hz) caused no effects on P1 and N1, while it markedly attenuated the P2 component. These findings suggest that P1 and N1 of MN-SEPs, as well as P1′ and N1′ of PTN-SEPs, are generated by the dorsal column fibers, and P2 and P2′ are possibly of postsynaptic origin in the respective dorsal column nuclei.The peak latency of N1 recorded on the cuneate nucleus was identical with the scalp-recorded far-field potential of P13–14 in all patients, while no scalp components were found which corresponded to P2. These findings support the previous assumption that the scalp-recorded P13–14 is generated by the presynaptic activities of the dorsal column fibers at their terminals in the cuneate nucleus.