From Towards a Science of Consciousness 3         Section III: Neural Correlates        CogNet Proceedings

Neural Correlates of Hallucinogen-induced Altered States of Consciousness

F. X. Vollenweider, Alex Gamma, and M. F. I. Vollenweider-Scherpenhuyzen

The study of hallucinogens ("psychedelics") and related substances offers a promising avenue to investigate biological correlates of altered states of consciousness (ASC) (Vollenweider 1998a). In combination with functional brain imaging techniques and pharmacological methodologies, these compounds are remarkable molecular probes into the biochemistry and functional organization of the brain in nonordinary states. The study of hallucinogens in humans is important firstly because they profoundly affect a number of brain functions that characterize the human mind, including cognition, volition, emotion, ego and self-awareness, which cannot be reliably studied in behavioral animal models. Secondly, because they elicit a clinical syndrome resembling in several aspects the first manifestation of schizophrenic disorders (Gouzoulis-Mayfrank et al. 1998). The various forms of ego alterations are especially prominent features of psychedelic and naturally occurring psychoses. These alterations may range from a slight loosening of ego boundaries to a dissolving of ego into an ecstatic oneness with the cosmos. The dissolution of the self as a center of reference, however, can also evoke anxiety and feelings of fragmentation, confusion and disorganization resembling the core features of schizophrenic ego disorders. Hence, studies of the neuronal mechanisms of hallucinogen action should provide not only novel insights into the pathophysiology of psychiatric disorders and their treatment, but, in a broader sense, into the biology of consciousness as a whole, for example, into the biology of ego structuring processes.

In the present contribution, I wish to summarize some of our recent results and advances in hallucinogen research, which are the result of human studies conducted in our group. We have developed different strategies to explore the pharmacological effects of hallucinogens on brain functions. The basic approaches are to investigate hallucinogen-induced metabolic changes with fluorodeoxyglucose (FDG) and to characterize functional interactions of neurotransmitter systems by assessing hallucinogen-induced displacement of specific radiolabeled receptor ligands using positron emission tomography (PET) (Vollenweider 1998a). A human model of sensory gating deficits, the cortico-striato-thalamo-cortical (CSTC) loop model, will be introduced to provide a perspective how current knowledge about hallucinogen drug action could be visualized within a synthetic framework to explain its subjective effects in humans. Hypotheses derived from this model were tested in several FDG-PET studies, and correlational analyses between hallucinogen-induced brain activity changes and phenomenological dimensions of ASC were computed to elucidate the neuronal correlates of ASC.

Measurement of psychological dimensions of ASC


In the context of the present theme-relating psychological and biological effects of hallucinogens-the assessment and characterization of altered states of consciousness (ASC) is of fundamental importance. Among several rating scales, the APZ questionnaire has become the standard in Europe for measuring specific states of consciousness and has been used on a routine basis by our group. The APZ questionnaire was developed based on a large prospective study with 393 subjects tested with cannabinoids, dimethyltryptamine, psilocybin, mescaline, harmaline, nitrosoxide, hypnoses, autogenic training, and meditation techniques (Dittrich 1998). It measures three primary and one secondary etiology-independent dimensions of ASC. The first dimension designated "oceanic boundlessness" (OSE) measures derealization phenomena and ego-dissolution, which are associated with a pleasurable emotional state ranging from heightened mood to sublime happiness and exaltation. Ego-dissolution can start with a mere loosening of ego-boundaries but may end up in a feeling of merging with the cosmos, where the experience of time has changed or completely vanished. If fully developed, this state might be comparable to a mystical experience. The second dimension "Anxious ego-dissolution" (AIA) measures thought disorder, ego-disintegration, loss of autonomy and self-control variously associated with arousal, anxiety, and paranoid feelings. The third subscale "visionary restructuralization" (VUS) refers to auditory and visual illusions, hallucinations, synesthetic phenomena, as well as altered experience of meaning and ideas of reference. The intercultural consistency of the APZ dimensions OSE, AIA and VUS has been rigorously tested in a subsequent study, the International Study on Altered States of Consciousness (ISASC), and the dimensions have been shown to be altered consistently in a manner that is independent of the particular treatment, disorder, or condition that led to the ASC (Dittrich 1998). So far, the APZ questionnaire has been used by our group to characterize the psychological effects of hallucinogens (psilocybin) (Vollenweider et al. 1997c), dissociative anesthetics (ketamine) (Vollenweider et al. 1997a, Vollenweider et al. 1997b), stimulants (amphetamine) (Vollenweider et al. 1998b), and entactogens (MDMA) (Vollenweider et al. 1998c) (figure 9.1).

[place figure 9.1 about here; figures not yet available]

Another psychometric scale that has proved useful for assessing specific dimensions of ego disorders is the "Ego Pathology Inventory" (EPI) developed by Scharfetter (Scharfetter 1981). The EPI has been divided empirically into five dimensions describing ego pathology and related behavior: ego identity, ego demarcation, ego consistency, ego activity and ego vitality. The "ego identity" scale includes items of doubts, changes or loss of one's identity in respect to "gestalt," physiognomy, gender, genealogical origin and biography. The "ego demarcation" scale refers to one's uncertainty or lack of differentiation between ego and non-ego spheres concerning thought process, affective state and body experience. The "ego consistency" scale comprises dissolution, splitting and destruction in experiencing a coherent self, body, thought process, chain of feelings and a structured external world. The "ego activity" scale refers to the deficit in one's ability, potency or power for self-determined action, thinking, feeling and perceiving. The "ego vitality" scale includes the experience or fear of one's death, of the fading away of liveliness, of the demise of mankind or the universe. As seen in figure 9.2, acute first-break schizophrenics differ from psilocybin and ketamine subjects most markedly in the extent of their impairment of ego activity and ego vitality. Notwithstanding, the similar range of values for the ego identity, ego demarcation and ego consisteny score indicates the similarity between hallucinogen-induced and endogenous psychotic states.

[place figure 9.2 here; figures not yet available]

The CSTC model of sensory information processing and ASC Based on the available neuroanatomical evidence and pharmacological findings of psychedelic drug action, we propose a cortico-subcortical model of psychosensory information processing that can be used as a starting point to analyze and integrate the effects of different chemical types of hallucinogens at a system level. The model advances that psychedelic states can be conceptualized as complex disturbances arising from more elementary deficits of sensory information processing in cortico-striato-thalamo-cortical (CSTC) feedback loops. The model is not entirely new, it incorporates the idea that psychotic symptoms might relate to a dopaminergic and/or dopaminergic-glutamatergic neurotransmitter dysbalance in mesolimbic and/or mesolimbic-corticostriatal pathways (Carlsson and Carlsson 1990), but it extends this hypothesis, insofar that the serotonergic and GABAergic neurotransmission are also brought into the scheme (Vollenweider 1992, Vollenweider 1994). In short, five CSTC loops have been identified and each loop, functioning in parallel, is thought to mediate a different set of functions; the motor, the oculomotor, the prefrontal, the association and the limbic loop. The limbic loop is involved in memory, learning and self-nonself discrimination by linking of cortical categorized exteroceptive perception and internal stimuli of the value system. The limbic loop originates in the medial and lateral temporal lobe and hippocampal formation and projects to the ventral striatum. Projections from this region then converge on the ventral pallidum and feed back via the thalamus to the anterior cingulate and the orbitofrontal cortex (figure 9.3).

[Place figure 9.3 here; figures not yet available]

The CSTC model posits that the thalamus acts as a filter or gating mechanism for the extero- and interoceptive information flow to the cerebral cortex and that deficits in thalamic gating may lead to a sensory overload of the cortex, which in turn may cause the sensory flooding, cognitive fragmentation and ego-dissolution seen in drug-induced ASC and endogenous psychotic states. The filter capability of the thalamus is thought to be under the control of cortico-striato-thalamic (CST) feedback loops. Specifically, it is hypothesized that the striatum and pallidum exert an inhibitory influence on the thalamus. Inhibition of the thalamus should result in a decrease of sensory input to the cortex and in a reduction of arousal, protecting the cerebral cortex from sensory overload and breakdown of its integrative capacity. The striatal activity is modulated by a number of subsidiary circuits and neurotransmitter systems, respectively. The mesostriatal and mesolimbic projections provide an inhibitory dopaminergic input to the striatum. Under physiological conditions, the inhibitory influence of the dopaminergic systems on the striatum is, however, thought to be counterbalanced by the glutamatergic excitatory input from cortico-striatal pathways. This assumption implies that an increase in dopaminergic tone (e.g., by amphetamine) as well as a decrease in glutamatergic neurotransmission (e.g., by ketamine) should lead to a reduction of the inhibitory influence of the striatum on the thalamus and result in a opening of the thalamic "filter" and, subsequently, in a sensory overload of the cerebral cortex and psychotic symptom formation. Finally, the reticular formation, which is activated by input from all sensory modalities, gives rise to serotonergic projections to the components of the CST loops. Excessive activation of the postsynaptic elements of the serotonergic projection sites (e.g., by psilocybin) should also result in a reduction of thalamic gating and, consequently, in a sensory overload of frontal cortex and psychosis.

First results testing the CSTC model


Although the CSTC model is an oversimplification, it provides a set of testable hypotheses. According to the CSTC model we have hypothesized that both the reduction of glutamatergic transmission by the NMDA antagonist ketamine and stimulation of the serotonergic system by the mixed 5-HT2/1 agonist psilocybin should lead to a sensory overload and metabolic activation of the frontal cortex (hyperfrontality). This hypothesis has been tested in healthy volunteers using positron emission tomography (PET) and the radiolig and [18F] fluorodeoxyglucose (FDG). In fact, it was possible to confirm the central hypothesis of a frontocortical activation in psychedelic states. Both ketamine and psilocybin led to a marked metabolic activation of the frontal cortex, including the anterior cingulate, and a number of overlapping metabolic changes in other brain regions (Vollenweider et al. 1997b, Vollenweider et al. 1997c). The observed hyperfrontality is interesting in several ways. First, the marked stimulation of the frontal cortex, the anterior cingulate, the temporomedial cortex and the thalamus seen in psilocybin and ketamine subjects accords with the thalamic filter theory suggesting that a disruption of the cortico-striato-thalamic (CST-) loop should lead to a sensory overload of the frontal cortex and its limbic relay stations. Second, hallucinogen-induced hyperfrontality is of particular interest because it appears to parallel similar findings in acutely ill schizophrenic and nonschizophrenic psychotic patients (Ebmeier et al. 1995, Sabri et al. 1997). Third, the hyperfrontality after ketamine and psilocybin also supports the idea that the psychedelics used in these studies may mediate their effects through a common final pathway or neurotransmitter system, downstream to their primary locus of action (see Vollenweider 1998a for a review).

Patterns of cortical activity in ASC


A multivariate analysis of metabolic and psychological data and a relative large sample size (e.g., 50-100 subjects) are imperative if the common neuroanatomical substrates of ASC are to be identified accurately. Therefore, a number of additional placebo-controlled FDG-PET experiments with S-ketamine, R-ketamine and amphetamine were performed in normal subjects (Vollenweider et al. 1997a, Vollenweider et al. 1998b). To identify the interactive organization of the brain in resting state and ASC, normalized metabolic PET data from placebo and corresponding drug conditions were subjected to a factor analysis, and factor scores for each individual subjects was computed (n = 106). Surprisingly, this computation revealed that the cortical-subcortical organization (based on a five-factor solution) during ASC was very similar to that seen under placebo, indicating that the functional integrity of interrelated brain regions (factors), which might be interpreted as functional "units" or "modules," is not disrupted in ASC (see figure 9.4). According to their content, the factors were labeled "fronto-parietal cortex," "temporal cortex," "occipital cortex," "striatum" (which included the caudate nucleus and putamen) and "thalamus." Subsequent comparison of the factor scores of drug and placebo condition revealed that subjects had significantly higher scores on the "frontal-parietal" and "striatal" network and lower scores on the "occipital cortex" during hallucinatory states than in resting states. This finding indicates that the neuronal activity within these modules and the more global relationship between these modules differs markedly between ASC and normal waking state.

[place figure 9.4 here; figures not yet available]

Multiple regression analysis of psychological scores (APZ scores) and factor scores (normalized metabolic activity) revealed, firstly, that the dimension OSE (oceanic boundlessness) relates to changes in metabolic activity in the frontal-parietal cortex, occipital cortex and striatum. Secondly, VUS (visionary restructuralization including hallucinatory phenomena) is associated with activity changes in the same network as the OSE dimension, but additionally relates to temporal activity. Thirdly, AIA (anxious ego-dissolution) is primarily associated with metabolic changes in the thalamus (table 9.1). The observed association between AIA and increased relative metabolic activity in the thalamus is underscored by the finding of a positive correlation between ego-identity impairment and the thalamic factor.

Table 9.1


The relationship between the APZ scores OSE, VUS and AIA, and the EPI score ego-identity impairment and normalized metabolic activity in the five functional brain modules (factors) identified by factor analysis. F1 (fronto-parietal factor), F2 (occipital factor), F3 (temporal factor), F4 (striatal factor), F5 (thalamic factor). Factors which significantly contribute to the computation are indicated by asterisks *p<0.05.

OSE

= 0.32 F1*

- 0.20 F2*

+ 0.11 F3

+ 0.20 F4*

+ 0.05 F5

VUS

 <

= 0.20 F1*

- 0.27 F2*

+ 0.17 F3*

+ 0.32 F4*

+ 0.10 F5

AIA

= 0.00 F1

+ 0.09 F2

+ 0.01 F3

+ 0.17 F4

+ 0.28F5*

Identity

= 0.04 F1

- 0.04 F2

+ 0.05 F3

+ 0.10 F4

+ 0.20 F5*

The present data show that the positively experienced form of ego-dissolution, OSE, can be clearly differentiated in terms of neurometabolic activity from the more fragmented and anxious ego-dissolution AIA. The OSE dimension, which relates to the pleasurable experience of dissolution of ego-boundaries, possibly culminating in transcendental or "mystical" states, and the alterated perception of time and space, substantially loads on the fronto-parietal factor. Indeed, according to current views, the frontal cortex in conjunction with parietal and limbic areas, is critical for the construction and maintenance of a coherent self. In its executive faculty, the frontal cortex, including the anterior cingulate, has an active role in structuring time, directing attention to relevant extero- or interoceptive stimuli and initiating and expressing appropriate behaviors (Milner et al. 1985, Fuster 1989, Posner and Petersen 1990). The parietal cortex is important for determining the relationship of the self to extrapersonal space, based on visuospatial input from the dorsal stream (Pribram 1991). It is noteworthy that the fronto-parietal factor also includes somatosensory and motor cortical areas, which contribute essential information to the formation of body image and physical representation of the self. As an interrelated network, the areas of the fronto-parietal factor are sometimes called "Central Neural Authority" (Hernegger 1995) to express the idea that they constitute a functional system crucially involved in ego-structuring processes and the formation of a coherent self defined in time and space. Based on these theoretical concepts, it appears well plausible that overstimulation of the Central Neural Authority may lead to profound alterations of self-experience and space/time perception, as reflected by the increased OSE scores in hallucinogen-induced ASC.

Anxious ego-dissolution (AIA) and ego-identity impairment appear to depend mainly on thalamic activity. This finding is in line with the view that dysfunction of the thalamic filter could lead to sensory overload, cognitive fragmentation and psychosis, as it is postulated by the CSTC model. Interestingly, increased thalamic activity with newly exacerbation of psychotic symptoms was also observed in neuroleptic-stabilized patients after ketamine administration.

Outlook


Hallucinogen research offers the exciting possibility to explore how our experience of self and ego relates to the complex interplay of neural networks in the brain, and, more generally, to narrow the gap between the mental and the physical. Due to the similarity of hallucinogen-induced states and endogenous psychoses, it can also enhance our understanding of the pathophysiology of neuropsychiatric disorders. The CSTC model presented here provides a useful starting point from which to approach the functional organization of the brain in drug-induced or naturally occurring ASC. It should be noted, however, that the present correlations, which are based on aggregated observations over time (APZ ratings, metabolism) and space (brain regions), though probably correct in the order of magnitude, might be inadequate at a finer level of resolution. New methodologies with a high time resolution such as 3D-electromagnetic tomography (Pasqual-Marqui et al. 1994) will allow us to also capture the temporal microdynamics of brain processes in ASC.

Acknowledgments


This study was supported in part by the Swiss National Science Foundation (Grants: 32-28746, 32-32418, and 32-04090).

References


Carlsson, M., and A. Carlsson. 1990. Schizophrenia: A subcortical neurotransmitter imbalance syndrome? In Schizophrenia Bull 16:425-432.

Dittrich, A. 1998. The standardized psychometric assessment of altered states of consciousness (ASCs) in humans. in Pharmacopsychiat 31:80-84.

Ebmeier, K. P., S. M. Lawrie,D. H. Blackwood, E.C. Johnstone, and G. M. Goodwin. (1995). Hypofrontality revisited: a high resolution single photon emission computed tomography study in schizophrenia. In J Neurol Neurosurg Psychiat 58:452-456.

Fuster, J. M. 1989. The prefrontal cortex. New York: Raven Press.

Gouzoulis-Mayfrank, E., L. Hermle, B. Thelen, and H. Sass. 1998. History, rationale and potential of human experimental hallucinogen drug research in psychiatry. In Pharmacopsychiat 31:63-68.

Hernegger R. 1995. Wahrnehmung und Bewusstsein. Ein Diskussionsbeitrag zur Neuropsychologie. Berlin: Spectrum Akademischer Verlag.

Milner, B., M. Pertrides, and M. L. Smith. 1985. Frontal lobes and the temporal organisation of memory. in Hum Neurobiol 4:137-142.

Pasqual-Marqui, R. D., C. M. Michel, and D. Lehmann. 1994. Low resolution electromagnetic tomography: a new method for localizing electrical activity in the brain. In Int J Psychophysiol 18:49-65.

Posner, M. I., and S. E. Petersen. 1990. The attention system of the human brain. In Ann Rev Neurosci 13:25-42.

Pribram, K. H. 1991. Brain and Perception. Hillsdale, N.J.: Lawrence Erlbaum Associates.

Sabri, O., R. Erkwoh, M. Schreckenberger, A. Owega, H. Sass, and U. Buell. 1997. Correlation of positive symptoms exclusively to hyperperfusion or hypoperfusion of cerebral cortex in never-treated schizophrenics. In Lancet 349:1735-1739.

Scharfetter, C. 1981. Ego-pychopathology: the concept and its empirical evaluation. In Psychol Med 11:273-280.

Vollenweider, F. X. 1992. Die Anwendung von Psychotomimetika in der Schizophrenieforschung unter besonderer Berücksichtigung der Ketamin/PCP-Modell-Psychose [The use of psychotomimetics in schizophrenia research with special amphasis on the PCP/ketamine model psychosis]. In SUCHT 38:389-409.

Vollenweider, F. X. 1994. Evidence for a cortical-subcortical dysbalance of sensory information processing during altered states of consciousness using PET and FDG. In Pletscher A, Ladewig D (eds.), 50 Years of LSD: State of the Art and Perspectives of Hallucinogens. London, Parthenon Publishing, pp. 67-86.

Vollenweider, F. X. 1998a. Advances and pathophysiological models of hallucinogen drug actions in humans: a preamble to schizophrenia research. In Pharmacopsychiat 31:92-103.

Vollenweider, F. X., A. Antonini,K.L. Leenders, and K. Mathys. 1998b. Effects of high amphetamine doses on mood and cerebral glucose metabolism in normals using positron emission tomography (PET). [In Press] In Psychiatry Research: Neuroimaging.

Vollenweider, F. X., A. Antonini, K.L. Leenders, I. Oye, D. Hell, and J. Angst. 1997a. Differential Psychopathology and patterns of cerebral glucose utilisation produced by (S)- and (R)-ketamine in healthy volunteers measured by FDG-PET. In Eur Neuropsychopharmacol 7:25-38.

Vollenweider, F. X., K. L. Leenders, C. Scharfetter, A. Antonini, P. Maguire, J. Missimer, and J. Angst. 1997b. Metabolic hyperfrontality and psychopathology in the ketamine model of psychosis using positron emission tomography (PET) and [F-18]-fluorodeoxyglocose (FDG). In Eur Neuropsychopharmacol 7:9-24.

Vollenweider, F. X., K. L. Leenders, C. Scharfetter, P. Maguire, O. Stadelmann, and J Angst. 1997c. Positron emission tomography and fluorodeoxyglucose studies of metabolic hyperfrontality and psychopathology in the psilocybin model of psychosis. IN Neuropsychopharmacology 16:357-372.

Vollenweider, F. X., M. Liechti, A.Gamma, and T. Huber, T. 1998c. Psychological and cardiovascular effects and short-term sequelae of MDMA ("Ecstasy") on MDMA-naive healthy volunteers. [In Press] In Neuropsychopharmacology.