- Email: [email protected]
P50 event-related brain potential amplitude and suppression measurements recorded with subjects seated versus supine
~
!:W~i~]
BRIEF REPORTS
P50 Event-Related Brain Potential Amplitude and Suppression Measurements Recorded with Subjects Seated versus Supine Keith McCallin, Valerie A. Cardenas, and George Fein Key Words: Auditory evoked potentials, auditory P50, electroencephalography, myoclonus, middle latency auditory evoked potentials BIOL PSYCHIATRY 1997;41:902--905
Introduction Auditory P50 event-related brain potentials (ERPs) have been recorded with subjects either seated (Boutros et al 1991; Fein et al 1996; Jerger et al 1992; Kathmann and Engel 1990; Schwarzkopf et al 1993) or supine (Adler et al 1992, 1994; Freedman et al 1987; Nagamoto et al 1989, 1991; Waldo and Freedman 1986). Recordings from supine subjects are based on the conjecture that P50 cannot be reliably recorded from seated subjects due to possible neck muscle artifact (Bickford et al 1964; Robinson and Rudge 1982). Bickford (Bickford et al 1964) elicited myogenic "microreflexes" using 120-dB clicks while loading the neck muscles with weights. These microreflexes were maximal at the inion and occurred as a positive-negative complex at 25-30 msec poststimulus. The myogenic origin was inferred from a reduction in amplitude following reduced loading of the neck muscles in correlation with electromyographic (EMG) recordings, although decreased stimulus intensity also resulted in decreased amplitude. These early studies provide the rationale for some researchers: 1) recording P50 potentials with their subjects supine so that the neck is not supporting the head; and 2) rejecting data that contain a vertex positivity at 30 msec (Adler et al 1990a, 1990b, 1992; Baker et al 1990). More
From the Department of Psychiatry, University of California, San Francisco, California (KM, VAC, GF); and San Francisco Department of Veterans Affairs Medical Center, San Francisco, California (KM, VAC, GF). Address reprint requests to George Fein, PhD, 4150 Clement Street (116R), San Francisco, CA 94121. Received March 19, 1996; revised October 2t, 1996.
© 1997 Society of Biological Psychiatry
recently, a deflection at 30 msec of neural origin has been well documented (Harker et al 1977; Kraus et al 1982; Li6geoisChauvel et al 1994; Mendel and Hosick 1975). This P30 (or Pa) potential is evoked by clicks or tones, is positive at the vertex, approximately 0.5-1 IxV in amplitude (Buchwald et al 1992), has a frontocentral topography (Deiber et al 1988), and can be recorded at high stimulus rates (Erwin and Buchwald 1986). If the 30 msec vertex positivity in P50 paradigm recordings has a myogenic component in addition to the neural P30, this myogenic component could compromise P50 measurement, contributing to the unreliability of P50 amplitude and suppression measurements in seated subjects (Boutros et al 1991; Cardenas et al 1993; Kathmann and Engel 1990; Schwarzkopf et al 1993). In this brief report we compared P50 and P30 recordings in subjects seated versus supine.
Methods
Subjects Six male and 7 female normal subjects, with no history of drug or alcohol abuse or family history of neurologic or psychiatric disorder, were studied. Subjects were between 23 and 34 years of age, were medication free at the time of the study, gave informed consent, and were paid for their participation.
Auditory Stimulation and P50 Recording Methods The auditory click paradigm and the P50 recording methods were as described previously (Fein et al 1996; Jerger et al 1992). 0006-3223/97/$17.00 PII S0006-3223(97)00545-8
Brief Reports
BIOL PSYCHIATRY !997;41:902-905
903
the channel with the maximal P50 amplitude (identified using topographic maps) recorded to the average reference (Fein et al 1994); and 3) singular value decomposition (SVD) on the average referenced waveforms (Cardenas et al 1995). For the peak-picking approaches, P50 or P30 amplitude of the 100-trial average (bandpass filter: 10-50 Hz; see Jerger et al 1992) was chosen using a semiautomatic algorithm, which chose the most positive (or negative) voltage value within specific time intervals. The time windows were 0 - 4 0 and 2 0 - 4 0 msec for the negativities preceding the P30 and P50 peaks, and 20-50 and 40-70 msec for the P30 and P50 peaks. P30 and P50 amplitudes were measured from the peak to the preceding trough. When P30 overlap obscured P50, P50 amplitude was measured to baseline. Conditioning-testing (C-T) ratios were computed as the test amplitude divided by the conditioning amplitude. C-T ratios greater than 2 were truncated to 2 to minimize the effect of outliers on group means (Nagamoto et al 1991). For the SVD method, topographic maps were examined (blind to subject position) by one of us (VAC) to determine whether P50 was more frontally or centrally distributed, then either a frontal (Fz, FCz, Cz, F3, F4, FC3, and FC4) or central (Fz, FCz, Cz, C 1, C2, C3, and C4) montage of seven electrodes was used for analysis. The SVD method was applied to bandpass-filtered
Subject's threshold for detecting stimuli was determined using a methods of limits procedure, and stimuli during the experiments were presented at 55 dB above threshold. A conditioning-testing click paradigm with a 500-msec interclick interval was implemented with a 32-electrode montage. A 10-sec interpair interval was employed to ensure full neuronal recovery between click pairs (Davis et al 1966; Fruhstorfer et al 1970; Zouridakis and Boutros 1992). Seated and supine recordings were obtained with the subjects relaxed and awake. Vertical and horizontal electro-oculographic (EOG) channels were used for eye movement detection, and single trials were rejected if EOG activity exceeded ---60 ixV on either channel. Trials were recorded until 100 eye-movement-free trials were achieved. The supine testing took place on a mattress with the subjects' head lying upon a pillow. The sitting and supine protocols were two of a battery of nine P50 protocols, with approximately a 1-hour interval between the sitting and supine recordings. Supine recordings were always run last, since it was necessary to rearrange the chamber to include a mattress.
P50 Waveform Analysis Three approaches were taken to P50 waveform analysis: 1) peak-picking on the vertex-A1 recordings; 2) peak-picking on
1 Q) I
0 EL
E
0
0 E
0
:=L
¢-
.¢: =L " ~
oo
"o
f
1
0
20
40
60
80
t O0
-1
120
i
i
i
i
0
20
40
60
time in ms
80
100
120
1 O0
120
time in ms
0
E
0
o
o
d5 t--
E ::~
"~ tO c~
._c
EL
t-
"~-
-1
-2
¢0
-2
oo i
0
20
40
i
60
time in ms
80
1 O0
120
0
20
40
60
80
time in ms
Figure 1. Responses for an illustrative subject to the conditioning (left side of figure) and testing (right side of figure) clicks for the seated (top of figure) and supine (bottom of figure) positions. These traces show electrode FCz versus the average reference. The click stimulus occurred at 0 msec. The closed circles mark P30 on the traces, and the open circles mark P50.
904
Brief Reports
BIOL PSYCHIATRY 1997;41:902-905
Table 1. Mean -4- SD for P50 C-T Ratios, P50 Conditioning and Testing Amplitudes, and P30Amplitudes for the Seated and Supine Positions, and the Mean _+ SD for the Difference between the Seated and Supine Positions for All Components Measured (N = 13)
C-T ratio Vertex-A1 Topographic, average referenced SVD P50 conditioning amplitude Vertex-A1 (ixV) Topographic, average referenced (ixV) SVD P50 testing amplitude VertexA1 (txV) Topographic, average referenced (IxV) SVD P30 conditioning amplitude (tzV) Vertex-A1 Topographic, average referenced P30 testing amplitude (p.V) Vertex-A1 Topographic, average referenced
Seated
Supine
Seated - supine
t~2, P
0.64 _+ 0.46 0.61 _+ 0.36 0.37 ± 0.26
0.72 -+ 0.80 0.59 ± 0.41 0.42 -- 0.24
-0.08 -+ 1.02 0.03 _+ 0.34 -0.05 ± 0.24
-0.29, .78 0.31, .76 -0.75, .47
1.62 - 0.90 1.32 ± 0.79 11.59 +-- 5.55
1.29 - 0.41 1.12 ± 0.56 10.33 ± 3.60
0.33 _+ 0.89 0.20 ± 0.56 1.26 +--4.56
1.35, .20 1.28, .22 1.00, .34
1.14 -+- 1.00 0.79 _+ 0.22 3.73 ± 3.01
0.81 ± 0.83 0.59 --- 0.41 4.30 ± 2.87
0.33 ± 1.20 0.20 ± 0.58 -0.57 +_ 2.49
0.98, .34 1.25, .24 -0.82, .43
1.41 _+ 1.25 0.38 _+ 0.37
1.13 ± 0.96 0.60 ± 0.41
0.28 _+ 1.06 -0.22 +_ 0.48
0.95, .36 -1.69,.13
1.09 + 1.19 0.47 + 0.38
0.71 ± 0.62 0.57 ± 0.43
0.38 _+ 1.33 -0.09 -+ 0.53
1.03, .32 -0.63, .54
t values are derived from a matched-pairs t test, and p values are two tailed and uncorrected for multiple comparisons.
averages within a time window chosen by one of us (VAC) to contain ascending and descending aspects of the P50 peak.
Results Out of the 78 C-T ratios that resulted (13 subjects × 2 protocols × 3 measurement methods), only three C-T ratios were greater than two, all occurring in vertex-A 1 data. Conditioning P50s were present in all recordings, and P30 occurred in 85% of the conditioning responses and 87% of the testing responses. In five averages, P30 and P50 overlap required that P50 be measured to baseline. Figure 1 presents a nonoverlapping P30 and P50 response. The conditioning and testing P30 deflections are of comparable amplitudes, with classic P50 suppression evident in the testing response amplitude. The comparisons between seated and supine measurements are presented in Table 1. For the P50 measures, these data are presented for all three measurement methods, whereas for P30, only vertex-A1 and topographic measurements are presented, as we found that SVD did not always work welt for P30 estimation, since multiple sources may contribute to the measured waveforms (possibly myogenic and neural sources). There were no statistically significant differences between seated and supine recordings for any P50 or P30 amplitude measure or for the C-T ratio (for amplitude, all two-tailed ps > . 13; for the C-T ratio, all ps > .47). After finding no significant differences in C-T ratios between seated and supine recordings, we computed the C-T ratio test-retest reliability across seated and supine recordings. The reliability of the C-T ratio, estimated using the vertex-A1 recordings, was essentially zero (r = - . 18), but was .55 for SVD and .63 for the topographic peak versus average reference recordings. The standard deviations (SD) for the SVD C-T
measurements were smallest, followed by the topographicaveraged reference measurements, with the vertex-A1 C-T measurements having much larger SD than the other measurement approaches. For the C-T ratio measured by SVD, the SD of the difference score between seated and supine measurements is 0.24. With a sample of 13 subjects and two-tailed p = .05, we have power of .80 to detect a difference in the C-T ratios of 0.19 units using SVD. Thus, we can rule out differences of this magnitude with 80% certainty. Moreover, for SVD, the mean C-T ratios were slightly larger for the supine recordings, so there is no indication in this study of even a trend for reduced suppression in seated recordings. Regarding the P30 recordings, we also have no indication of recording position effects on P30 incidence or amplitude; however, given the very low signal-to-noise ratio of the P30 measurements, we only have .80 power to rule out a 100% change in P30 amplitude as a function of recording position.
Discussion We found no difference between the P50 conditioning or testing amplitudes or the P50 C-T ratios recorded in the sitting versus supine positions, regardless of the P50 measurement method used. We also did not find any difference between the seated versus supine recordings in either the incidence or amplitude of the P30 potential. This suggests that the 30-msec vertex positivity present on P50 recordings is either largely neural in origin, or that neck muscle artifact is present at comparable levels in both sitting and supine recordings. In either case, the results indicate that subject position does not have a large effect on P50
Brief Reports
amplitude or suppression measures. We note that these results do not necessarily generalize to all P50 recordings. It is possible that given other recording parameters (e.g., much louder stimuli) a difference in myogenic activity may exist, and it may affect P30 and P50 measures; however, given the results reported here, the burden of evidence for requiring supine recordings or elimination of trials with 30-msec positivities to insure against myogenic
BIOL PSYCHIATRY 1997;41:902-905
905
contamination of P50 recordings lies with those who suggest these constitute a problem. This work was supported by General Medical Research funds from the Department of Veterans Affairs, the DVA Research Career Scientist program (Dr. Fein), and NIDA grant R01DA08365.
References Adler LE, Gerhardt GA, Franks R, et al (1990a): Sensory physiology and catecholamines in schizophrenia and mania. Psychiatry Res 31:297-309. Adler LE, Waldo MC, Tatcher A, Cawthra E, Baker N, Freedman R (1990b): Lack of relationship of auditory gating defects to negative symptoms in schizophrenia. Schizophr Res 3:131-138. Adler LE, Hoffer LJ, Griffith J, Waldo MC, Freedman R (1992): Normalization by nicotine of deficient auditory sensory gating in the relatives of schizophrenics. Biol Psychiatry 32:607. Adler LE, Hoffer L, Nagamoto HT, Waldo MC, Kisley MA, Griffith JM (1994): Yohimbine impairs P50 auditory sensory gating in normal subjects. Neuropsychopharmacology 10: 249 -257. Baker NJ, Staunton M, Adler LE, et al (1990): Sensory gating deficits in psychiatric inpatients: Relation to catecholamine metabolites in different diagnostic groups. Biol Psychiatry 27:519-528. Bickford R, Jacobson J, Thane D, Cody, R (1964): Nature of average evoked potentials to sound and other stimuli in man. Ann NY Acad Sci 112:204-223. Boutros NN, Overall J, Zouridakis B (1991): Test-retest reliability of the P50 mid-latency auditory evoked response. Psychiatry Res 29:181-192. Buchwald JS, Erwin R, Van Lancker D, Guthrie D, Schwafel J, Tanguay P (1992): Midlatency auditory evoked responses: P 1 abnormalities in adult autistic subjects. Electroencephalogr Clin Neurophysiol 84:164-171. Cardenas VA, Gerson J, Fein G (1993): The reliability of P50 suppression as measured by the conditioning/testing ratio is vastly improved by dipole modeling. Biol Psychiatry 33:335-344. Cardenas VA, Yingling CD, Jewett D, Fein G (1995): A Multichannel model-free method for estimation of eventrelated potential amplitudes and its comparison to dipole source localization. J Med Eng Technol 19:88-98. Davis H, Mast T, Yoshie M, Zerlin S (1966): The slow response of the human cortex to auditory stimuli: Recovery process. Electroencephalogr Clin Neurophysiol 21:105-113. Deiber MP, Ibafiez V, Fischer C, Perrin F, Maugui6re F (1988): Sequential mapping favours the hypothesis of distinct generators for Na and Pa middle latency auditory evoked potentials. Electroencephalogr Clin Neurophysiol 71:187-197. Erwin RJ, Buchwald JS (1986): Midlatency auditory evoked responses: Differential recovery cycle characteristics. Electroencephalogr Clin Neurophysiol 64:417-423. Fein G, Biggins C, VanDyke C (1994): The auditory P50 response is normal in Alzheimer's disease when measured via a paired click paradigm. Electroencephalogr Clin Neurophysiol 92:536-545.
ein G, Biggins C, MacKay S (1996): Cocaine abusers have reduced auditory P50 amplitude and suppression compared to both normal controls and alcoholics. Biol Psychiatry 39:955-965. Freedman R, Adler LE, Waldo M (1987): Gating of the auditory evoked potential in children and adults. Psychophysiology 24:223-227. Fruhstorfer H, Soveri P, Jarvilehto T (1970): Short-term habituation of auditory evoked response in man. Electroencephalogr Clin Neurophysiol 28:153-161. Harker LA, Hogich E, Voots R, Mendel M (1977): Influence of succinylcholine on middle component auditory evoked potentials. Arch Otolaryngol 103:133-137. Jerger K, Biggins C, Fein G (1992): P50 suppression is not affected by attentional manipulation. Biol Psychiatry 31:365377. Kathmann N, Engel RR (1990): Sensory gating in normals and schizophrenics: A failure to find strong P50 suppression in normals. Biol Psychiatry 27:1216-1226. Kraus N, Ozdamar C), Hier D, Stein L (1982): Auditory middle latency responses (MLRs) in patients with cortical lesions. Electroencephalogr Clin Neurophysiol 54:275-287. Litgeois-Chauvel C, Musolino A, Badier JM, Marquis P, Chauvel P (1994): Evoked potentials recorded from the auditory cortex in man: Evaluation and topography of the middle latency components. Electroencephalogr Clin Neurophysiol 92:204-214. Mendel MI, Hosick EC (1975): Effects of secobarbital on the early components of the auditory evoked potentials. Rev Laryngol 96:178-184. Nagamoto HT, Adler LE, Waldo MC, Freedman R (1989): Sensory gating in schizophrenics and normal controls: Effects of changing stimulation interval. Biol Psychiatry 25:549-561. Nagamoto HT, Adler LE, Waldo MC, Griffith J, Freedman R (1991): Gating of auditory response in schizophrenics and normal controls. Effects of recording site and stimulation interval on the P50 wave. Schizophr Res 4:31-40. Robinson K, Rudge P (1982): Centrally generated evoked potentials. In Halliday AM (ed), Evoked Potentials in Clinical Testing. London: Churchill Livingstone, pp 368-372. Schwarzkopf SB, Lamberti JS, Smith DA (1993): Concurrent assessment of acoustic startle and auditory P50 evoked potential measures of sensory inhibition. Biol Psychiatry 33:815-828. Waldo MC, Freedman R (1986): Gating of auditory evoked responses in normal college students. Psychiatry Res 19:223239. Zouridakis G, Boutros NN (1992): Stimulus parameter effects on the P50 evoked response. Biol Psychiato' 32:839-841.
!:W~i~]
BRIEF REPORTS
P50 Event-Related Brain Potential Amplitude and Suppression Measurements Recorded with Subjects Seated versus Supine Keith McCallin, Valerie A. Cardenas, and George Fein Key Words: Auditory evoked potentials, auditory P50, electroencephalography, myoclonus, middle latency auditory evoked potentials BIOL PSYCHIATRY 1997;41:902--905
Introduction Auditory P50 event-related brain potentials (ERPs) have been recorded with subjects either seated (Boutros et al 1991; Fein et al 1996; Jerger et al 1992; Kathmann and Engel 1990; Schwarzkopf et al 1993) or supine (Adler et al 1992, 1994; Freedman et al 1987; Nagamoto et al 1989, 1991; Waldo and Freedman 1986). Recordings from supine subjects are based on the conjecture that P50 cannot be reliably recorded from seated subjects due to possible neck muscle artifact (Bickford et al 1964; Robinson and Rudge 1982). Bickford (Bickford et al 1964) elicited myogenic "microreflexes" using 120-dB clicks while loading the neck muscles with weights. These microreflexes were maximal at the inion and occurred as a positive-negative complex at 25-30 msec poststimulus. The myogenic origin was inferred from a reduction in amplitude following reduced loading of the neck muscles in correlation with electromyographic (EMG) recordings, although decreased stimulus intensity also resulted in decreased amplitude. These early studies provide the rationale for some researchers: 1) recording P50 potentials with their subjects supine so that the neck is not supporting the head; and 2) rejecting data that contain a vertex positivity at 30 msec (Adler et al 1990a, 1990b, 1992; Baker et al 1990). More
From the Department of Psychiatry, University of California, San Francisco, California (KM, VAC, GF); and San Francisco Department of Veterans Affairs Medical Center, San Francisco, California (KM, VAC, GF). Address reprint requests to George Fein, PhD, 4150 Clement Street (116R), San Francisco, CA 94121. Received March 19, 1996; revised October 2t, 1996.
© 1997 Society of Biological Psychiatry
recently, a deflection at 30 msec of neural origin has been well documented (Harker et al 1977; Kraus et al 1982; Li6geoisChauvel et al 1994; Mendel and Hosick 1975). This P30 (or Pa) potential is evoked by clicks or tones, is positive at the vertex, approximately 0.5-1 IxV in amplitude (Buchwald et al 1992), has a frontocentral topography (Deiber et al 1988), and can be recorded at high stimulus rates (Erwin and Buchwald 1986). If the 30 msec vertex positivity in P50 paradigm recordings has a myogenic component in addition to the neural P30, this myogenic component could compromise P50 measurement, contributing to the unreliability of P50 amplitude and suppression measurements in seated subjects (Boutros et al 1991; Cardenas et al 1993; Kathmann and Engel 1990; Schwarzkopf et al 1993). In this brief report we compared P50 and P30 recordings in subjects seated versus supine.
Methods
Subjects Six male and 7 female normal subjects, with no history of drug or alcohol abuse or family history of neurologic or psychiatric disorder, were studied. Subjects were between 23 and 34 years of age, were medication free at the time of the study, gave informed consent, and were paid for their participation.
Auditory Stimulation and P50 Recording Methods The auditory click paradigm and the P50 recording methods were as described previously (Fein et al 1996; Jerger et al 1992). 0006-3223/97/$17.00 PII S0006-3223(97)00545-8
Brief Reports
BIOL PSYCHIATRY !997;41:902-905
903
the channel with the maximal P50 amplitude (identified using topographic maps) recorded to the average reference (Fein et al 1994); and 3) singular value decomposition (SVD) on the average referenced waveforms (Cardenas et al 1995). For the peak-picking approaches, P50 or P30 amplitude of the 100-trial average (bandpass filter: 10-50 Hz; see Jerger et al 1992) was chosen using a semiautomatic algorithm, which chose the most positive (or negative) voltage value within specific time intervals. The time windows were 0 - 4 0 and 2 0 - 4 0 msec for the negativities preceding the P30 and P50 peaks, and 20-50 and 40-70 msec for the P30 and P50 peaks. P30 and P50 amplitudes were measured from the peak to the preceding trough. When P30 overlap obscured P50, P50 amplitude was measured to baseline. Conditioning-testing (C-T) ratios were computed as the test amplitude divided by the conditioning amplitude. C-T ratios greater than 2 were truncated to 2 to minimize the effect of outliers on group means (Nagamoto et al 1991). For the SVD method, topographic maps were examined (blind to subject position) by one of us (VAC) to determine whether P50 was more frontally or centrally distributed, then either a frontal (Fz, FCz, Cz, F3, F4, FC3, and FC4) or central (Fz, FCz, Cz, C 1, C2, C3, and C4) montage of seven electrodes was used for analysis. The SVD method was applied to bandpass-filtered
Subject's threshold for detecting stimuli was determined using a methods of limits procedure, and stimuli during the experiments were presented at 55 dB above threshold. A conditioning-testing click paradigm with a 500-msec interclick interval was implemented with a 32-electrode montage. A 10-sec interpair interval was employed to ensure full neuronal recovery between click pairs (Davis et al 1966; Fruhstorfer et al 1970; Zouridakis and Boutros 1992). Seated and supine recordings were obtained with the subjects relaxed and awake. Vertical and horizontal electro-oculographic (EOG) channels were used for eye movement detection, and single trials were rejected if EOG activity exceeded ---60 ixV on either channel. Trials were recorded until 100 eye-movement-free trials were achieved. The supine testing took place on a mattress with the subjects' head lying upon a pillow. The sitting and supine protocols were two of a battery of nine P50 protocols, with approximately a 1-hour interval between the sitting and supine recordings. Supine recordings were always run last, since it was necessary to rearrange the chamber to include a mattress.
P50 Waveform Analysis Three approaches were taken to P50 waveform analysis: 1) peak-picking on the vertex-A1 recordings; 2) peak-picking on
1 Q) I
0 EL
E
0
0 E
0
:=L
¢-
.¢: =L " ~
oo
"o
f
1
0
20
40
60
80
t O0
-1
120
i
i
i
i
0
20
40
60
time in ms
80
100
120
1 O0
120
time in ms
0
E
0
o
o
d5 t--
E ::~
"~ tO c~
._c
EL
t-
"~-
-1
-2
¢0
-2
oo i
0
20
40
i
60
time in ms
80
1 O0
120
0
20
40
60
80
time in ms
Figure 1. Responses for an illustrative subject to the conditioning (left side of figure) and testing (right side of figure) clicks for the seated (top of figure) and supine (bottom of figure) positions. These traces show electrode FCz versus the average reference. The click stimulus occurred at 0 msec. The closed circles mark P30 on the traces, and the open circles mark P50.
904
Brief Reports
BIOL PSYCHIATRY 1997;41:902-905
Table 1. Mean -4- SD for P50 C-T Ratios, P50 Conditioning and Testing Amplitudes, and P30Amplitudes for the Seated and Supine Positions, and the Mean _+ SD for the Difference between the Seated and Supine Positions for All Components Measured (N = 13)
C-T ratio Vertex-A1 Topographic, average referenced SVD P50 conditioning amplitude Vertex-A1 (ixV) Topographic, average referenced (ixV) SVD P50 testing amplitude VertexA1 (txV) Topographic, average referenced (IxV) SVD P30 conditioning amplitude (tzV) Vertex-A1 Topographic, average referenced P30 testing amplitude (p.V) Vertex-A1 Topographic, average referenced
Seated
Supine
Seated - supine
t~2, P
0.64 _+ 0.46 0.61 _+ 0.36 0.37 ± 0.26
0.72 -+ 0.80 0.59 ± 0.41 0.42 -- 0.24
-0.08 -+ 1.02 0.03 _+ 0.34 -0.05 ± 0.24
-0.29, .78 0.31, .76 -0.75, .47
1.62 - 0.90 1.32 ± 0.79 11.59 +-- 5.55
1.29 - 0.41 1.12 ± 0.56 10.33 ± 3.60
0.33 _+ 0.89 0.20 ± 0.56 1.26 +--4.56
1.35, .20 1.28, .22 1.00, .34
1.14 -+- 1.00 0.79 _+ 0.22 3.73 ± 3.01
0.81 ± 0.83 0.59 --- 0.41 4.30 ± 2.87
0.33 ± 1.20 0.20 ± 0.58 -0.57 +_ 2.49
0.98, .34 1.25, .24 -0.82, .43
1.41 _+ 1.25 0.38 _+ 0.37
1.13 ± 0.96 0.60 ± 0.41
0.28 _+ 1.06 -0.22 +_ 0.48
0.95, .36 -1.69,.13
1.09 + 1.19 0.47 + 0.38
0.71 ± 0.62 0.57 ± 0.43
0.38 _+ 1.33 -0.09 -+ 0.53
1.03, .32 -0.63, .54
t values are derived from a matched-pairs t test, and p values are two tailed and uncorrected for multiple comparisons.
averages within a time window chosen by one of us (VAC) to contain ascending and descending aspects of the P50 peak.
Results Out of the 78 C-T ratios that resulted (13 subjects × 2 protocols × 3 measurement methods), only three C-T ratios were greater than two, all occurring in vertex-A 1 data. Conditioning P50s were present in all recordings, and P30 occurred in 85% of the conditioning responses and 87% of the testing responses. In five averages, P30 and P50 overlap required that P50 be measured to baseline. Figure 1 presents a nonoverlapping P30 and P50 response. The conditioning and testing P30 deflections are of comparable amplitudes, with classic P50 suppression evident in the testing response amplitude. The comparisons between seated and supine measurements are presented in Table 1. For the P50 measures, these data are presented for all three measurement methods, whereas for P30, only vertex-A1 and topographic measurements are presented, as we found that SVD did not always work welt for P30 estimation, since multiple sources may contribute to the measured waveforms (possibly myogenic and neural sources). There were no statistically significant differences between seated and supine recordings for any P50 or P30 amplitude measure or for the C-T ratio (for amplitude, all two-tailed ps > . 13; for the C-T ratio, all ps > .47). After finding no significant differences in C-T ratios between seated and supine recordings, we computed the C-T ratio test-retest reliability across seated and supine recordings. The reliability of the C-T ratio, estimated using the vertex-A1 recordings, was essentially zero (r = - . 18), but was .55 for SVD and .63 for the topographic peak versus average reference recordings. The standard deviations (SD) for the SVD C-T
measurements were smallest, followed by the topographicaveraged reference measurements, with the vertex-A1 C-T measurements having much larger SD than the other measurement approaches. For the C-T ratio measured by SVD, the SD of the difference score between seated and supine measurements is 0.24. With a sample of 13 subjects and two-tailed p = .05, we have power of .80 to detect a difference in the C-T ratios of 0.19 units using SVD. Thus, we can rule out differences of this magnitude with 80% certainty. Moreover, for SVD, the mean C-T ratios were slightly larger for the supine recordings, so there is no indication in this study of even a trend for reduced suppression in seated recordings. Regarding the P30 recordings, we also have no indication of recording position effects on P30 incidence or amplitude; however, given the very low signal-to-noise ratio of the P30 measurements, we only have .80 power to rule out a 100% change in P30 amplitude as a function of recording position.
Discussion We found no difference between the P50 conditioning or testing amplitudes or the P50 C-T ratios recorded in the sitting versus supine positions, regardless of the P50 measurement method used. We also did not find any difference between the seated versus supine recordings in either the incidence or amplitude of the P30 potential. This suggests that the 30-msec vertex positivity present on P50 recordings is either largely neural in origin, or that neck muscle artifact is present at comparable levels in both sitting and supine recordings. In either case, the results indicate that subject position does not have a large effect on P50
Brief Reports
amplitude or suppression measures. We note that these results do not necessarily generalize to all P50 recordings. It is possible that given other recording parameters (e.g., much louder stimuli) a difference in myogenic activity may exist, and it may affect P30 and P50 measures; however, given the results reported here, the burden of evidence for requiring supine recordings or elimination of trials with 30-msec positivities to insure against myogenic
BIOL PSYCHIATRY 1997;41:902-905
905
contamination of P50 recordings lies with those who suggest these constitute a problem. This work was supported by General Medical Research funds from the Department of Veterans Affairs, the DVA Research Career Scientist program (Dr. Fein), and NIDA grant R01DA08365.
References Adler LE, Gerhardt GA, Franks R, et al (1990a): Sensory physiology and catecholamines in schizophrenia and mania. Psychiatry Res 31:297-309. Adler LE, Waldo MC, Tatcher A, Cawthra E, Baker N, Freedman R (1990b): Lack of relationship of auditory gating defects to negative symptoms in schizophrenia. Schizophr Res 3:131-138. Adler LE, Hoffer LJ, Griffith J, Waldo MC, Freedman R (1992): Normalization by nicotine of deficient auditory sensory gating in the relatives of schizophrenics. Biol Psychiatry 32:607. Adler LE, Hoffer L, Nagamoto HT, Waldo MC, Kisley MA, Griffith JM (1994): Yohimbine impairs P50 auditory sensory gating in normal subjects. Neuropsychopharmacology 10: 249 -257. Baker NJ, Staunton M, Adler LE, et al (1990): Sensory gating deficits in psychiatric inpatients: Relation to catecholamine metabolites in different diagnostic groups. Biol Psychiatry 27:519-528. Bickford R, Jacobson J, Thane D, Cody, R (1964): Nature of average evoked potentials to sound and other stimuli in man. Ann NY Acad Sci 112:204-223. Boutros NN, Overall J, Zouridakis B (1991): Test-retest reliability of the P50 mid-latency auditory evoked response. Psychiatry Res 29:181-192. Buchwald JS, Erwin R, Van Lancker D, Guthrie D, Schwafel J, Tanguay P (1992): Midlatency auditory evoked responses: P 1 abnormalities in adult autistic subjects. Electroencephalogr Clin Neurophysiol 84:164-171. Cardenas VA, Gerson J, Fein G (1993): The reliability of P50 suppression as measured by the conditioning/testing ratio is vastly improved by dipole modeling. Biol Psychiatry 33:335-344. Cardenas VA, Yingling CD, Jewett D, Fein G (1995): A Multichannel model-free method for estimation of eventrelated potential amplitudes and its comparison to dipole source localization. J Med Eng Technol 19:88-98. Davis H, Mast T, Yoshie M, Zerlin S (1966): The slow response of the human cortex to auditory stimuli: Recovery process. Electroencephalogr Clin Neurophysiol 21:105-113. Deiber MP, Ibafiez V, Fischer C, Perrin F, Maugui6re F (1988): Sequential mapping favours the hypothesis of distinct generators for Na and Pa middle latency auditory evoked potentials. Electroencephalogr Clin Neurophysiol 71:187-197. Erwin RJ, Buchwald JS (1986): Midlatency auditory evoked responses: Differential recovery cycle characteristics. Electroencephalogr Clin Neurophysiol 64:417-423. Fein G, Biggins C, VanDyke C (1994): The auditory P50 response is normal in Alzheimer's disease when measured via a paired click paradigm. Electroencephalogr Clin Neurophysiol 92:536-545.
ein G, Biggins C, MacKay S (1996): Cocaine abusers have reduced auditory P50 amplitude and suppression compared to both normal controls and alcoholics. Biol Psychiatry 39:955-965. Freedman R, Adler LE, Waldo M (1987): Gating of the auditory evoked potential in children and adults. Psychophysiology 24:223-227. Fruhstorfer H, Soveri P, Jarvilehto T (1970): Short-term habituation of auditory evoked response in man. Electroencephalogr Clin Neurophysiol 28:153-161. Harker LA, Hogich E, Voots R, Mendel M (1977): Influence of succinylcholine on middle component auditory evoked potentials. Arch Otolaryngol 103:133-137. Jerger K, Biggins C, Fein G (1992): P50 suppression is not affected by attentional manipulation. Biol Psychiatry 31:365377. Kathmann N, Engel RR (1990): Sensory gating in normals and schizophrenics: A failure to find strong P50 suppression in normals. Biol Psychiatry 27:1216-1226. Kraus N, Ozdamar C), Hier D, Stein L (1982): Auditory middle latency responses (MLRs) in patients with cortical lesions. Electroencephalogr Clin Neurophysiol 54:275-287. Litgeois-Chauvel C, Musolino A, Badier JM, Marquis P, Chauvel P (1994): Evoked potentials recorded from the auditory cortex in man: Evaluation and topography of the middle latency components. Electroencephalogr Clin Neurophysiol 92:204-214. Mendel MI, Hosick EC (1975): Effects of secobarbital on the early components of the auditory evoked potentials. Rev Laryngol 96:178-184. Nagamoto HT, Adler LE, Waldo MC, Freedman R (1989): Sensory gating in schizophrenics and normal controls: Effects of changing stimulation interval. Biol Psychiatry 25:549-561. Nagamoto HT, Adler LE, Waldo MC, Griffith J, Freedman R (1991): Gating of auditory response in schizophrenics and normal controls. Effects of recording site and stimulation interval on the P50 wave. Schizophr Res 4:31-40. Robinson K, Rudge P (1982): Centrally generated evoked potentials. In Halliday AM (ed), Evoked Potentials in Clinical Testing. London: Churchill Livingstone, pp 368-372. Schwarzkopf SB, Lamberti JS, Smith DA (1993): Concurrent assessment of acoustic startle and auditory P50 evoked potential measures of sensory inhibition. Biol Psychiatry 33:815-828. Waldo MC, Freedman R (1986): Gating of auditory evoked responses in normal college students. Psychiatry Res 19:223239. Zouridakis G, Boutros NN (1992): Stimulus parameter effects on the P50 evoked response. Biol Psychiato' 32:839-841.