Inadequacies of self-report data for exclusion criteria detection in marihuana research: an empirical case for multi-method direct examination screening.

Stringent exclusion criteria in drug abuse research are necessary to protect against methodological confounds compromising the interpretation of findings. However, reliance on self-report screening may fail to detect important exclusion variables. We compared three levels of exclusion criteria screening in a study of neurophysiological/neurocognitive sequelae of chronic marihuana use in normals. LEVEL 1 (self-report) consisted of telephone pre-screening. LEVEL 2 (also self-report) involved in-depth personal interviews. LEVEL 3 consisted of several direct examination assessments including a medical/psychiatric examination by a board certified psychiatrist, eight weeks of twice per week urine drug screens, an EEG exam and eight hours of neuropsychological testing. Results indicated that 39.0% of subjects passing self-report screening had significant exclusion criteria findings that were only detected through LEVEL 3 direct examination procedures. Of all subjects found to have exclusion criteria after being provisionally accepted following LEVEL 1 telephone pre-screening, 55.7% were detected only through more rigorous LEVEL 3 direct examination screening methods.

Topographic quantitative EEG sequelae of chronic marihuana use: a replication using medically and psychiatrically screened normal subjects.

In two previous studies it was reported that chronic marihuana (THC) use was associated with unique quantitative EEG features which were present in the non-intoxicated state. THC users, as contrasted with controls, had significant elevations of Absolute Power, Relative Power, and Coherence of alpha activity over the bilateral frontal cortex. Furthermore, a quantitative EEG discriminant function analyses permitted a 95% correct user versus non-user classification. However, because all of the THC users and 58% of the non-user controls were psychiatric inpatients, diagnostic and medication effects, if any, were uncontrolled. In the present study the same quantitative EEG methods were used to study daily THC users and non-user controls who underwent a rigorous screening process to insure that they were medically and psychiatrically healthy. The results of previous studies were replicated and an additional EEG correlate of chronic THC exposure (reduced alpha frequency) was identified.

Reduced P50 auditory gating response in psychiatrically normal chronic marihuana users: a pilot study.

BACKGROUND: Neurophysiological studies of marihuana (THC) often contain uncontrolled confounds [psychiatric diagnoses, polydrug use, central nervous system (CNS)-relevant injury, etc.] that can alter electrophysiological measures. This P50 sensory gating report is part of a larger neurophysiological and neurocognitive investigation of chronic THC exposure using rigorously screened medically and psychiatrically normal individuals without concurrent use of non-THC substances. METHODS: Following medical and psychiatric screening, including serial urine drug screens, technically adequate P50 paired auditory recovery tests were obtained on 19 chronic THC users and 14 control subjects. Fifty pairs of 80-dB auditory clicks (1 pair per 10 sec, 500-msec interclick separation) were delivered through earphones. The sensory gating measure was the ratio between the P50 amplitudes at the vertex elicited by the conditioning (first) and test (second) click. RESULTS: THC subjects had significantly higher sensory gating ratios (i.e., reduced suppression) than did control subjects. Among THC users, sensory gating ratios did not correlate with duration or frequency of THC use, although subjects with ratios above 40 had nearly twice the number of "joint-years" of THC exposure than did those with lower ratios. CONCLUSIONS: Reduced P50 suppression in the sensory gating paradigm may be a possible neurophysiological CNS sequela of long-term cumulative exposure to THC.

Possible EEG sequelae of very long duration marihuana use: pilot findings from topographic quantitative EEG analyses of subjects with 15 to 24 years of cumulative daily exposure to THC.

In previous work we demonstrated and replicated a significant association between increased absolute and relative power and interhemispheric coherence of EEG alpha activity over the bilateral frontal-central cortex ("alpha hyperfrontality") in daily marihuana users as contrasted with nonusers. In this report we focused our analyses on subjects who reported smoking marihuana on a daily basis for 15 to 24 consecutive years. Compared to nonuser controls and subjects who had used marihuana on a daily basis for shorter periods of time, subjects with excessively long cumulative exposures to THC were found to have significantly elevated absolute power of theta activity over bilateral frontal-central cortex, as well as significantly increased interhemispheric coherence of theta activity across central and posterior regions. Concurrent reaction time studies conducted in our laboratory suggest that very long duration cumulative marihuana exposure might be associated with slowed cognitive processing.

Early and middle latency evoked potentials in medically and psychiatrically normal daily marihuana users: a paucity of significant findings.

The use of evoked potentials to study CNS effects of marihuana (THC) have produced inconsistent findings. Our previous pilot studies suggested that auditory P300 latencies and amplitudes, auditory P50 and somatosensory P30 amplitudes and brainstem auditory evoked potential latencies were altered in THC users. Because these findings were flawed by uncontrolled psychiatric diagnostic and medication variables, we undertook a controlled investigation of screened medically and psychiatrically normal THC users and controls. When age effects were controlled, THC related alterations of brain stem and both auditory and visual P300 responses could not be seen. This report extends our analyses to other auditory, somatosensory and visual evoked potentials. With the possible exception of an elevated auditory P50 amplitude, significant evoked potential correlates to daily THC use were not seen when normals were studied and age effects controlled.

Auditory and visual P300 event related potentials are not altered in medically and psychiatrically normal chronic marihuana users.

Attempts to use Event Related Potentials, particularly the cognitive or P300 evoked potential, as measures of CNS effects of THC use have been infrequent and have produced inconsistent results. We published a pilot study in which psychiatric patient THC users had significantly prolonged auditory P300 latencies and reduced amplitudes as contrasted with non-users. Because psychiatric diagnoses and medication effects could not be controlled, we repeated the study with medically and psychiatrically normal subjects selected with extremely stringent exclusion criteria and screening procedures. P300 latency differences between THC users and controls were not detected. Using all subjects, THC users displayed reduced auditory and visual P300 amplitudes. However, when age differences between THC users and controls were removed, all significant P300 amplitude differences were removed as well. The contaminating effect of using psychiatric patients in THC research is discussed and the importance of using carefully screened normal subjects in studies of neurophysiological abuse drug effects is stressed.

Habituation and motion sickness.

The vestibular, cerebellar, and reticular systems are central in importance, in motion sickness and habituation, to the effects of motion. Nuclear medicine single photon emission computed tomography (SPECT) studies of cerebral blood flow and power spectral electroencephalographic recordings during motion sickness were used to determine alterations in the central nervous system. The rotating chair with and without visual stimulation was used to study the rate of habituation and the effect of antimotion sickness medications on this rate. An increase of theta waves over the frontal cortex indicated a decreased activation of the higher centers during motion sickness. Motion sickness also produces an increase of blood flow in the central cerebellum that has connections to the reticular system. This increase in cerebellar activity is relayed to the reticular system whereby neural recruitment builds up to trigger the vomiting center, producing motion sickness. Habituation may be a conditioned compensatory activation of the reticular neurons that prevents this disruption of normal activation. The rate of habituation when motion sickness was prevented by scopolamine was slowed, indicating that, if the central nervous system is not challenged by disruption of normal activation, it does not produce the compensatory reactions that result in habituation.

Persistent topographic quantitative EEG sequelae of chronic marihuana use: a replication study and initial discriminant function analysis.

In a previous pilot study using psychiatric patients we reported that daily marihuana users had significant elevations of (1) Absolute Alpha Power, (2) Relative Alpha Power, and (3) Interhemispheric Alpha Coherence over both frontal and frontal-central areas when contrasted with subjects who did not use marihuana. We referred to this phenomenon as Hyperfrontality of Alpha. The study presented here is a successful replication of our previous findings using new samples of subjects and identical methods. Post hoc analyses based on the combined sample from both studies suggest that variables of psychiatric diagnoses and medication did not bias our results. In addition, a discriminant function analysis using quantitative EEG variables as candidate predictors generated a 95% correct THC user versus nonuser classification accuracy which received a successful jackknife replication.

Topographic mapping of quantitative EEG variables in chronic heavy marihuana users: empirical findings with psychiatric patients.

EEG studies of marihuana use dating back to 1945 were reviewed. The earlier studies depended upon visual analysis of the tracing, and while some minor frequency and amplitude variations occurred in some subjects, there was no consistent THC induced change noticeable across subjects. Quantitative EEG studies of acute exposure to THC came later and produced reliable findings of a placebo controlled dose dependent THC induced increase in relative power of alpha, combined with decreased alpha frequency and a reduction of beta activity. These findings were reported for data collected from central-occipital derivations only. In our present investigation, we report that chronic heavy THC users have EEGs characterized by (1) increased absolute power of all frequencies over all cortical areas (2) hyperfrontality of elevated relative and absolute power and coherence values of alpha activity, and (3) a decrease in relative power of all non-alpha frequencies. Methodological issues were discussed and some suggestions were made for continuing research in this area.

Combinations of evoked potential amplitude measurements in relation to psychiatric diagnosis.

The purposes of this investigation were (1) to determine the extent to which single nonredundant measures, derived from optimal combinations of evoked potential (EP) amplitude measurements, could differentiate between various groups of psychiatric patients and between patients and nonpatients, and (2) to assess the replicability of such discriminations. Somatosensory, visual, and auditory EPs were recorded from 15 locations in 253 unmedicated patients and 99 nonpatients. Multivariate statistical methods were used to reduce the amplitude measurements to sets of factor scores that met specified criteria for entry into discriminant analyses between pairs of the following groups: nonpatients, neuroses, personality disorders, schizophrenias, schizotypal/borderlines, major depressives, and manics. Sensitivity, specificity, and predictive value were assessed for discriminant scores. The discriminant scores yielded many differences between groups, most of which were replicable in split-half analyses. Among these were differences between all six patient groups and nonpatients, and between schizophrenics and nonpsychotics. Split-half discriminant analyses were also performed with 12 measures that were obtained by taking the means of factor scores grouped by sensory modality, time, and spatial location; these variables, used without preselection, provided several replicable diagnostic discriminations.

Distinctive somatosensory evoked potential features in obsessive-compulsive disorder.

The purpose was to determine whether obsessive-compulsive disorder (OC) is associated with a distinctive pattern of somatosensory evoked potential (SEP) measurements, based on amplitudes of the N60 (negativity 60 msec poststimulus), P90, and N130 peaks. Previously, we found higher N60 amplitudes in 14 OC patients than in controls or other neurotics. Here, SEPs of the same 14 OCs were compared with those of 28 other neurotics, 99 nonpatients, 49 chronic schizophrenics, 27 "other" schizophrenics, 20 DSM-II latent schizophrenics, and 42 major depressives. Utilizing all available subjects, each peak amplitude was factor analyzed across 14 leads; the resulting factor scores were compared between OCs and every other group. The factor scores of OCs differed from those of all other groups, and patterns of factor score differences were similar in all comparisons. Most discriminations were replicable in split-half analyses. OC appears to be associated with a relatively distinctive pattern of middle latency SEP peak amplitude measurements.

Psychiatric diagnostic discriminations with combinations of quantitative EEG variables.

The possible psychiatric diagnostic utility of certain quantitative EEG measures was evaluated by further analysis of previously reported data from 242 unmedicated patients and 94 non-patients. Time series of amplitude, frequency and wave symmetry measures for 12-lead EEGs (eyes closed and open) were factor analyzed across leads. Factor scores meeting specified criteria in multivariate analyses were entered into discriminant analyses comparing pairs of the following groups: non-patients, neurotics, personality disorders, overt schizophrenics, latent schizophrenics, major depressives and manics. The following discriminations were obtained with at least 50 per cent sensitivity, and diagnostic confidence rates from 69 to 92 per cent: (a) non-psychotic patients (neuroses, personality disorders) from overt schizophrenics, latent schizophrenics or manics; (b) major depressives from latent schizophrenics or manics; (c) non-patients from schizophrenics (overt and latent), depressives or manics. Most discriminations were replicable in split-half analyses. Possible utility of EEG measures in differential diagnosis is supported.

Psychiatric drugs and the somatosensory P400 wave.

The effects of antipsychotic and antidepressant medications on the somatosensory evoked potential (SEP) late wave (P400) were investigated in a diagnostically heterogeneous group of 51 psychiatric patients. SEPs were recorded from five scalp locations while subjects counted electrical stimuli to one of four randomly stimulated fingers. In seven subjects, tested both on and off medication, no significant drug effects were observed. P400 amplitudes of 14 patients being treated with antipsychotic medication were greater than those of unmedicated patients matched for age, sex, and diagnosis. P400 amplitudes did not differ between eight patients being treated with antidepressants and eight unmedicated matched patients. Accuracy in stimulus counting (index of task performance) was significantly correlated with P400 amplitude; more accurate counting was associated with higher P400 amplitude. When 14 medicated and 14 unmedicated subjects, matched for task performance, age, sex, and approximate diagnoses, were compared, P400 amplitude did not differ. These findings suggest that higher P400 amplitudes associated with medication probably reflected improved task performance with medication, rather than a direct action of the drugs on P400. The results suggest the conclusion that lower P400 amplitude in psychosis is not secondary to psychiatric drugs.

Failure to replicate evoked potential observations suggesting corpus callosum dysfunction in schizophrenia.

Somatosensory potentials (SEPs) evoked by vibrotactile finger stimulation have been reported to be the same in both hemispheres in schizophrenics, whereas they are asymmetrical in normals, with the contralateral hemisphere leading the ipsilateral (Jones and Miller, 1981). These findings were taken to indicate that the corpus callosum is nonfunctional in schizophrenics. To attempt replication of these results, vibrotactile SEPs of 6 schizophrenics and 6 normal controls were recorded with both bipolar and monopolar derivations. Asymmetrical bipolar SEPs were obtained in both schizophrenics and controls; previous observations of schizophrenic-control differences were not replicated. Acceptable evidence of ipsilateral early SEPs was not obtained; the test procedure seems inappropriate for measuring callosal conduction time.

Rheoencephalographic concomitants of cognitive engagement in nonpatients and schizophrenics. An exploratory study.

Noninvasive cranial impedance plethysmographic, or rheoencephalography (REG), characteristics are reported to be well correlated with cerebral blood flow. This paper reports an exploratory study, in which we recorded REGs from four locations over each hemisphere in 16 nonpatients and 16 medicated schizophrenics under four conditions: eyes closed, eyes open, and following performance of the Seashore Rhythms and Wisconsin Sorting Tests. REG (Jacquy F index) values were greater anteriorly than posteriorly under all conditions. F index differed regionally between card sorting and eyes open conditions. With card sorting, F values of nonpatients increased more in the left than right hemisphere; schizophrenics showed the reverse hemispheric effect. Card sorting performance scores were positively correlated with regional F index changes in nonpatients, but not in schizophrenics. The results suggest that the REG may reflect regional cerebral effects associated with cognition and psychopathology.

Relationships between psychiatric diagnosis and some quantitative EEG variables.

Psychiatric diagnostic correlates of EEG measures were investigated in 242 unmedicated patients and 94 nonpatients. Computer analysis of eyes-open and eyes-closed EEGs (12 leads) provided time series of amplitude, frequency, and wave-symmetry measures; mean levels and temporal variability indexes were also factor analyzed across leads. Comparisons between various age- and sex-matched groups disclosed numerous differences. Compared with those of nonpatients, schizophrenics' EEGs showed lower amplitude variability, greater frequency variability, greater wave symmetry, and less reactivity to eye opening. The EEGs of schizophrenic subgroups, including latent schizophrenics, were similar and also like those of manics. The EEGs of manics and major depressives differed markedly. The EEGs of subjects with personality disorders and schizophrenics differed from nonpatients in opposite directions. Groups could be classed according to the level of EEG activation: high included manics and schizophrenics; normal, nonpatients and neurotics; and low, major depressives and those with personality disorders.

Influence of antipsychotic and antidepressant drugs on evoked potential correlates of psychosis.

This study aimed to (i) determine the changes in evoked potentials (EPs) associated with hospital treatment involving either antipsychotic (AP) drugs, tricyclic (TTR) drugs, or no medication (NM); (ii) determine which EP changes may be direct drug effects; (iii) evaluate effects of testing patients receiving AP and TR drugs on psychiatric correlates of EPs. Somatosensory (SEP) to left and right median nerve stimuli, visual (VEP), and auditory (AEP) EPs were recorded from one eye and 14 scalp leads. Patients were tested before and during AP (N = 28), TR (N = 10), or NM (N = 15) treatments; results of 35 age- and sex-matched nonpatients were compared with predrug and drug test results of 35 patients. Main findings included (i) with AP and TR drugs, SEP peak P30 amplitude tended to decrease and SEP N130 to increase; (ii) with AP drugs and NM, VEP amplitudes decreased; (iii) AEP amplitudes increased with AP drugs; (iv) changes in amplitude measures did not differ significantly between AP, TR, and NM treatments; (v) marked later SEP amplitude differences between patients and nonpatients were about the same for pretreatment and drug tests; (vi) patient-control differences were greater with drugs for VEPs and less with drugs for AEPs. It was concluded that later SEP amplitude correlates of psychopathology are little influenced by AP and TR drugs, but that VEP and AEP results are modified.