** AhISee **   PSYCHOLOGY > Crossed Lines  |<|Up|?|Menu|Search/Map|Translate|>| 

Crossed Lines: What can atypical anti-psychotics tell us about schizophrenia?

This essay enquires into the contribution the new atypical anti-psychotics have made to our current understanding of the neuropharmacological aspects of schizophrenia - if they have. The atypical anti-psychotics have received much publicity for their reduced side effects compared to typical anti-psychotics. But is it true? And do they actually work? The essay concludes 'probably not'

This essay has been reformatted for on-line presentation.


Essay Outline



Introduction: what is the issue?

Schizophrenia is a disorder affecting (by some estimates) more than 1.0% of the American population, with over 300,000 acute episodes reported each year in the United States alone (Stahl 1997). Rates in Europe and Asia are calculated using different criteria than the USA, and perhaps for this reason are lower, being of the order of 0.2 to 1.0% lifetime prevalence (Feldman, Meyer & Quenzer 1997); this still represents a very large number of people. Schizophrenia is intensely distressing for sufferers, families, and carers alike, and is usually a chronic disorder (Kolb & Whishaw 1996). For all these reasons, ongoing efforts are being made to find effective treatments in the 'magic bullet' tradition (Feldman, Meyer & Quenzer (1997)), which would alleviate the symptoms of schizophrenia without causing unpleasant side effects.

Any such treatment would have benefits for all those afflicted and affected by the disorder, and also of course for the shareholders of any company fortunate enough to develop a 'cure'. Unfortunately, '...the biological basis of schizophrenia remains unknown...' (Stahl 1997); it is hard to develop a treatment for a disease which has unknown causes. Moreover it is a diagnosis covering a very heterogeneous set of symptoms; the simplified model put forward by Crow (1980)) divides these into two categories, positive symptoms(those 'added on' to 'normal' behaviour), and negative symptoms (as if 'subtracted' from 'normal' behaviour). What contribution can neuropharmacology, particularly by way of study of the newer atypical anti-psychotics, make to our current understanding of the neurological basis of schizophrenia?

The aetiology of schizophrenia is very problematic; despite around a hundred years of experimentation and theorising since Bleuler's (1908) original introduction of the term, there is no generally-agreed account for the origins of schizophrenia (Davison & Neale 1998). Proposals include epidemiological factors, genetic factors, developmental factors, biopsychosocial factors, vulnerability-stress factors, structural abnormalities and functional abnormalities, with various conflicting patterns of evidence supporting and contradicting each of these proposals (Feldman, Meyer & Quenzer (1997); Sarason & Sarason (1996); Davison & Neale (1998)).

Neuropharmacology concerns itself with '...drug-induced modification of nervous system function...' (Feldman, Meyer & Quenzer (1997)). Where schizophrenia and other psychoses are concerned, the neuropharmacological aim is to re-establish 'normal' function by changing brain chemistry by administration of drugs, specifically the group of drugs known as neuroleptics. The neuroleptics are drugs of dramatic effect:

...antipsychotic drugs have been termed 'neuroleptics' in that these drugs' actions imitate a neurological disease... Textbook of Psychiatry: American Psychiatric Press (1988)

They can very significantly change the behaviour of those patients to whom they are administered. Notably, the traditional or 'typical' anti-psychotic drugs have specific (and undesirable) side effects (Davison & Neale 1998); more recently, 'atypical' anti-psychotics have been described as having lesser or different, less troublesome, side effects (Stahl 1997). Is it possible, then, to infer something about the changes that occur from 'normal' functioning in the 'schizophrenic' brain by examining the effects of neuroleptic drugs on brain function? Atypical anti-psychotics are defined in terms of their differences from typicals, so we will first examine the effects on brain function of the typical anti-psychotics.


Typical and atypical anti-psychotics

The original and archetypal typical anti-psychotic is chlorpromazine, which is one of the phenothiazines, a group of chemicals synthesised as part of the search for new organic dyes in the late 1800's (Feldman, Meyer & Quenzer (1997)). Many phenothiazines are antihistamines; chlorpromazine was found by chance to have anti-psychotic effects when its antihistaminic properties were being studied by way of administration to a ward of schizophrenics (Stahl 1997). This group of organic chemicals has a characteristic nucleus composed of two benzene rings joined with nitrogen and sulphur atoms, making up a three-ring structure; various other compounds attach to this, giving rise to differing activities and pharmacological potencies (Feldman, Meyer & Quenzer (1997)).

Chlorpromazine was the first drug found to have some definite clear effect on reducing the symptoms of schizophrenia, particularly those known (in Crow's (1980) system) as positive symptoms. As a result of its introduction, there was a revolution in the treatment of schizophrenia, with many traditional hospitals being closed and psychiatric patients being returned (for better or worse) to 'care in the community' whilst being maintained on a dose of chlorpromazine. These results led to a major research drive to find out why drugs of this class caused their effects. Chlorpromazine was demonstrated by X-ray crystallography to be morphologically similar to dopamine, a principal inhibitory neurotransmitter, and thus it was proposed that it preferentially occupied dopamine receptor sites (see Figure 1   Open in popup window).

This finding was part of the evidence that led to the proposal of the dopamine hypothesis, a claim that the symptoms of schizophrenia were caused by (in the original formulation) an excess of dopamine in the brain. However, the absence of high levels of dopamine metabolites (particularly homovanillic acid, HVA) in schizophrenic patients led to a modification of the theory to suggest that the symptoms were associated with excessive numbers of - or sensitivity in - dopamine receptors in the schizophrenic brain (Davison & Neale 1998).


Chlorpromazine effects

Unfortunately chlorpromazine and related agents are not unimodal in their effects; as well as having the desired result of calming schizophrenic patients and diminishing the expression of their positive symptoms, with chronic application they also have side effects. The principal effects caused are:

Taken together with the desired effects of the phenothiazines, this pattern of responses maps more-or-less neatly onto three principal dopamine neuronal pathways in the brain. These are, respectively:

There is a fourth dopamine pathway, the mesocortical dopamine pathway, from the midbrain tegmentum to the limbic cortex. There is however no consensus as to whether the effects of phenothiazines on this pathway are beneficial; it is regarded as a cognitive pathway, and thus reducing its activity may not be of advantage, particularly in negative symptom patients (Feldman, Meyer & Quenzer (1997)). The general run of the various dopamine pathways is shown in schematic form in Figure 2   Open in popup window, below:


Dopamine and the phenothiazines

Overall there is a strong argument for the involvement of dopamine pathways in the action of phenothiazines, which can be expressed as follows:

  1. The phenothiazines are known to act at dopamine receptor sites

  2. The administration of phenothiazines causes a pattern of (eventual) changes in expression of positive schizophrenic symptoms, as well as a pattern of side effects (as described above)

  3. The dopamine pathways, where dopamine receptors are especially prevalent and so where we can presume the phenothiazines are particularly active, relate to the pattern of therapeutic and side effects seen in chronic administration of the drugs.

Thus, a picture has been constructed of the logic behind the dopamine hypothesis and the presumed action of the typical anti-psychotics. It must be emphasised that there are other classic anti-psychotic drugs apart from the phenothiazines, but pressure of space means that this group has been selected here as embodying typical properties; all the classical anti-psychotics tend to block dopaminergic synapses (Stahl 1997).


Dopamine receptor sites

Dopaminergic systems present contradictory aspects; dopamine constitutes up to 80% of the total brain catecholamine content, yet dopaminergic cells are rather rare, there being no more than about a million in the whole human brain (Feldman, Meyer & Quenzer (1997)). Moreover, although the dopamine hypothesis preceded any detailed understanding of the dopamine (DA) receptor, work commencing in the 1970's (Kebabian & Calne 1979) showed that there was more than one type of DA receptor; at least six types are now identified, though these are still categorised as generally similar to the first two recognised, that is D1-like and D2-like receptors.

The D1-like include D1 and D5 receptors, the D2-like include D2S and D2L, D3, and D4. These can be broadly characterised in that activated D1-like receptors stimulate adenylate cyclase enzymatic response; activated D2-like receptors either inhibit it or do not affect it (Kalat 1996). The obvious question arises, if the phenothiazines affect dopaminergic synapses by blocking them, then which DA receptor is involved? The balance of evidence is that it is the D2-type receptors that are involved in the anti-psychotic effects of the typical drugs (Feldman, Meyer & Quenzer (1997), Davison & Neale (1998), Sarason & Sarason (1996), Stahl (1997)). An obvious corollary is that the therapeutic effect may be connected with inhibition of the second messenger cAMP system (Vallar & Meldolesi 1989).


The atypical anti-psychotics

Stahl (1997) makes clear the point that the blocking of two of the four dopamine pathways in the brain (the nigrostriatal and tuberinfundibular pathways) is definitely not beneficial, whilst blocking the third (the mesocortical pathway) is of doubtful utility. The problem with the typical anti-psychotics is that they affect them all. Given that there are various different types of DA receptors, the possibility arises that they are unevenly distributed in the brain and - if that should be the case - it might be possible to selectively block them. In this way, it could be possible to alter the therapeutically relevant pathways without causing undesirable side effects - just the claim that is put forward for the atypical anti-psychotics. As with the typicals, here the pressure of space means that we will concentrate on one archetypal atypical anti-psychotic, clozapine.



What is atypical about an atypical anti-psychotic? Kinon & Lieberman (1996) in a comprehensive review, point out that the term 'atypical' is problematic. Some of the criteria used to ascribe atypicality include:

Criteria for atypicality in anti-psychotics
1 Decrease or absence of acute EPSE and TD
2 Increase of therapeutic efficacy toward positive, negative or cognitive symptoms
3 Decrease or absence of capacity to elevate prolactin


Table 1: atypical criterion table, taken from Kinon & Leiberman (1996)

Clozapine was first recognised as atypical for its high therapeutic efficiency without concomitant extra-pyramidal side-effects, arising from its low propensity to block D2 receptors (Kerwin 1994). Kerwin also points out that the simplest definition of 'atypical' is '...an effective neuroleptic that does not cause catalepsy in rats...'. But how does clozapine achieve its effects? The available evidence gives a contradictory picture. In part, clozapine has low occupancy of the nigrostriatal sites, in the range 30-60% (Farde et al 1992), thus accounting for its low EPE. However, Pilowsky et al (1992) showed that here was no direct connection between D2 occupancy and therapeutic effect, thus introducing other possible sites of action for clozapine. The effects of clozapine are amongst the most complex known (Stahl 1997). As well as DA receptors, clozapine also affects the serotinergic 5HT2, muscarinic acetylcholine, adrenergic, NMDA glutamate, GABA, and cholecystokin systems.

Notably, while affecting only some 20-60% of D2 receptors it also affects some 85-90% of 5HT2 receptors (Stahl 1997), and this finding has led to proposals that this balance of blockade of different systems (or perhaps some other pairing out of the nine currently-known affected NT receptors) is responsible for its enhanced therapeutic profile (Feldman, Meyer & Quenzer (1997)). Thus, serotonin-dopamine antagonists (SDA) have been developed to mimic this aspect of clozapine's function without some of its side-effects.


Clozapine side effects

Clozapine also has atypical side effects (Feldman, Meyer & Quenzer (1997)). These include those resulting from its broad effects in adrenergic and cholinergic systems, such as orthostatic hypotension, tachycardia, hypersalivation, weight gain, and most seriously aganulocytosis, the destruction of bone marrow (the aetiology of this being unknown). This means that clozapine is often only administered as a drug of last resort, and efforts are being made to develop drugs with its advantages but without the dangerous aspects.


Other possibilities

Neurochemical hypotheses of the aetiology of schizophrenia now involve the dopamine hypothesis, the neurodevelopmental model, a noradrenalin (norepinephrine) model, a serotonin hypothesis, a glutamate hypothesis and more but the evidence available does not presently enable clear acceptance or rejection of any of them (Feldman, Meyer & Quenzer (1997), Kerwin 1994). It is difficult to make psychological sense out of the maze of potential interactions; the argumentation (and potential supporting evidence) here is complex, because NT systems in their inhibitory, excitatory, and facilitatory interactions are knitted into a complex skein. The dynamic, interactive, and self-regulating feedback networks inherent to the CNS further complicate what is already difficult to understand.


A cortical dopamine theory

Recently evidence has been accumulating that perhaps points to an important involvement of the cortex in the neuropathology of schizophrenia. Van Tol et al (1991) and Seeman (1992) described the D4 receptor (which is common in the cortex) as a potential site of action of clozapine. However, Roth, Tandra, Burgess & Sibley (1994) found that there was no way reliably to distinguish between typical and atypical anti-psychotics based on their affinities for the D4 receptor. Lidow, Williams and Golman-Rakic (1998) suggest an alternate mechanism involving upregulation of cortical D2 receptors (which are highly occupied by clozapine in the cortex; Pilowsky et al 1997) without downregulating D1 receptors at the same time. They argue that the clozapine-D4 hypothesis is not supported on evidence including the lack of psychosis in individuals genetically lacking D4 receptors; clozapine unlike chlorpromazine occupies cortical sites but not striatal sites. The psychological implication here is straightforward. Schizophrenia is a thought disorder (Feldman, Meyer & Quenzer (1997)), the cortex is generally regarded as implementing higher-order cognitive and social functions (Kolb & Whishaw 1997), and thus there is intuitive appeal in a theory that links a thought disorder with molecular biological dysfunction in the cortex.


Does clozapine actually work in clinical use?

There is not universal agreement that the claims made for clozapine can be substantiated; various practitioners have found little or no beneficial effect. Baldachino & McKnight (1994), for example, report on 14 patients who fitted the profile of not responding well to typical anti-psychotics and were prescribed clozapine over a three-year period. They point out that '... the treatment outcome reported in scientific studies would have a significant impact on the number of long-term rehabilitation beds required if it were reflected in a clinical setting...'. However, out of the 14, only three remained on the drug at the end of the period; three had moved to living in the community, of whom two had already stopped taking clozapine. One patient did move into the community as a result of clozapine administration. Their conclusion was that the clinical outcome in practice does not reflect the controlled study outcome, and that the perceived benefit in various patients was '...due to a decrease in their arousal...' (Baldachino & McKnight 1994); in other words, they troubled staff less. Kerwin (1994) pointed out that '...the acute drowsiness noted for the drug was considered as an advantage in acute cases...'.


Critical discussion

What sorts of things can psychopharmacology tell us? Given a particular disorder of the brain, psychoneuropharmacology can look at a disordered brain, examine its workings at a molecular level, and if irregularities of function can be identified, suggest potential agents to modify or rectify those irregularities. Clearly, the whole process depends on accurate clinical diagnosis, description and discrimination - but this is problematic. For a start, the schizophrenic brain is inside the patient's head, and only available post-mortem, so most work is done with rats. There are other problems.


What is schizophrenia?

A diagnosis of schizophrenia cannot stand alone; the 'disease' does not exist outside the individual showing symptoms of schizophrenia, unlike a pathogen. Abstracting a disease - 'schizophrenia' - from a set of symptoms is reification, the conversion of an idea to a thing. But is schizophrenia one disease? There are at least two distinct patterns, positive and negative symptoms, which in any modern analysis would hardly be clustered together as one disease. We might remember the criticisms of Freud's theories, that they cannot be disproven or falsified (in the Popperian mode) because they make causal claims that embody both positive and negative outcomes of both positive and negative causes - rather like schizophrenia.


Problems of diagnosis

If a claim is to be made that (say) D2 receptors are associated with positive symptoms of schizophrenia, then final testing of this needs not a set of animals but access to suitable schizophrenic patients, diagnosed as schizophrenic by DSM IV or ICD 10 criteria. But these criteria embed certain assumptions about what constitutes 'schizophrenia', which in part include responsiveness to drugs that are D2 antagonists. This embeds a clear circularity and a potential self-fulfilling prophecy.


Problems of cause and effect

Examining the neuropathology of the 'schizophrenic brain' in the hope of developing a 'cure' is only useful if we assume a particular casual chain - that disordered behaviour is the result of disordered brains, Figure 3   Open in popup window.

But of course this is not the only feasible causal chain. As long as we admit the possibility that mental states (whatever they are) affect the physiology of the brain (something well established in psychoneuroimmunology, for example, and see Feldman, Meyer & Quenzer (1997)) -- then this other chain, from disordered mentality to disrupted biology is possible: Figure 4   Open in popup window. Or indeed, why should there not be a feedback mechanism between biology, brain, behaviour, mental acts and attitudes, thus: Figure 5   Open in popup window

Should this last suggestion have any truth in it, it's conceivable that anti-psychotics may sometimes perpetuate any disorder, rather than necessarily helping to reorder the brain. In other cases they may break the chain, allowing the system to reorder itself. Of course, all these causal chains are much easier to speculate on than prove, or disprove.


Problems of inference from animals to humans

Schizophrenia, as the author has emphasised repeatedly, is a thought disorder (Feldman, Meyer & Quenzer (1997)). Yet the basic screening method for detecting potential anti-psychotic drugs uses rats (Kerwin 1994). Can we reasonably expect rats to have thought disorders? The answer, obviously, is 'no'. Various stratagems are adopted; schizophrenia-analogues are induced in animals by amphetamine or PCP psychosis (Feldman, Meyer & Quenzer (1997)). However, amphetamine psychosis in humans is distinguishable from schizophrenia by an absence of low Responsive Search Score in a visual tracking task; plainly, the two are not identical. PCP psychosis is not principally dopaminergic in its effects (Stahl 1997). Thus, a major problem of the ecological validity of extrapolating from animal studies is introduced.


Recent work on the cerebellum

We still know very little about how the brain works; only a few years ago, large parts of the brain now known for specific functions were labelled as 'association areas'. Recent study has been addressing the cerebellum, which used to be thought to be involved only in fine motor functions. Schmahmann (1998), in a review of the implication of the cerebellum in cognitive disorders, points to the possibility that cerebellar disorders can lead to disorder of thought - the cerebellar cognitive affective syndrome. Middleton & Strick (1998) in another review concur that this has implications for schizophrenia. If this should prove to be substantiated, then yet another set of neuropathological considerations enter.


Cognitive aspects of thought disorders

For this author, one of the critical elements illustrating the inadequacy of current hypothesising on the aetiology of schizophrenia is the delay that occurs between the administration of an anti-psychotic drug (which changes NT levels in minutes) and the development of changes in extrinsic behaviour (which takes months). The author knows of no satisfactory proposal to explain this.

The cognitive revolution in psychology was based on an information-processing paradigm based on analogy with computer systems. The brain is not a computer system, but nonetheless psychopharmacology seems to ignore the potential distinction between two levels of analysis - hardware and software. A thought disorder (as in schizophrenia) can be analogised as a software-level error. Giving an anti-psychotic drug, however, is the equivalent of rearranging the wiring. This would not achieve much in a typical computer system, when faced with a program error. The whole validity of the information-processing approach is based on the recognition of the programmability and flexibility of the neural substrate in (at least) higher-order organisms - the capacity to learn and modify behaviour, to abstract from the intrinsic stimuli. Increasingly it is recognised that those systems once thought to be autonomous are, in fact, in a complex feedback/feedforward relationship with levels of cognition that were once seen as quite abstracted and detached from the pure biological level - an example would be the interactions between appraisal, attitude and mood with the activity and effectivity of the immune system (Feldman, Meyer & Quenzer (1997)). Thus, the attempt to permanently change a thought disorder by messing with the wiring may be inherently doomed to failure.


Summary and conclusion

This of necessity has been a very selective, simplified and compressed review of but a few of the issues surrounding psychopharmacology and schizophrenia. Let us return to the opening statement: what is 'the contribution of atypical anti-psychotics to our current understanding of the neuropharmacological basis of schizophrenia'? The truthful answer is 'not a lot' - though this is not due to any fault in psychopharmacology, but because of the tremendous complexities inherent in the issues.

What we do know is that the cluster of symptoms labelled as 'positive schizophrenia' is (perhaps) associated with a set of changes in the mesolimbic dopaminergic pathway - or perhaps the mesocortical, or perhaps it is the cortex or cerebellum that is central. It may be that the dopaminergic system is critical, or the 5HT2 system, or glutamate, or something else. It is definite that something changes somewhere in the brain, though we do not know if that change is causal or a response. We also know that neuroleptics in general affect the brain, and that atypical anti-psychotics reportedly sometimes in some cases work in patients that have not responded to typical anti-psychotics, and perhaps (sometimes) help sufferers from negative systems. The available treatments are not however any sort of a cure, merely a palliative for some of the worst symptoms and an aid to hospital staff in patient management, in that giving neuroleptic drugs to 'schizophrenics' quietens them down, shuts them up, and (sometimes) helps them to feel better.

What we do not know - and psychopharmacology alone is unlikely to tell us - is

Overall, it may be apparent that the author is deeply unconvinced by claims that there is any one simple cause-effect relation in schizophrenia. There is substantial evidence that this may not be one 'disease', but a clinically-useful but aetiologically-nonsensical clustering together of similar symptoms with dissimilar causes. Until these are clearly discriminated there seems little realistic prospect of progress beyond the palliative level -- though this is not to discount the tremendous value of any palliative option that is available for this distressing disorder. It is of course easy to be cleverly critical in an essay such as this, at a safe remove from any ward or lab, but we need to actually understand why schizophenia happens in order to help victims and sufferers, even if this means a Kuhnian-style paradigm-shift with its associated career costs and fall-out.



Baldachino, A., & McKnight, R. A review of the use of clozapine in clinical practice. The Pharmaceutical Journal (1994), Vol. 252, pp.
Carpenter, R (1996) Neurophysiology 3rd Edition; London: Arnold
Crow, T. (1980) Molecular pathology of schizophrenia: more than one disease process? British Medical Journal, 280, 66-88
Davison, G., & Neale, J. (1998). Abnormal Psychology. Wiley: New York.
Feldman, R., Meyer, J. & Quenzer, L. (1997)) Principles of Neuropsychopharmacology. New York: Sinauer.
Gardos, G., Casey, D., Cole, J., Perenyi, A., Kocsis, E., Arato, M., Samson, J. & Conley, C. (1994) Ten-year outcome of tardive dyskinesia. American Journal of Psychiatry, 151, 836-841.
Gazzaniga, M. (1995) Consciousness and the cerebral hemispheres IN M.Gazzaniga, (Ed.), The Cognitive Neurosciences MIT: Cambridge, Massachusetts.
Grace, A. Phasic versus tonic dopamine release and the modulation of dopamine system responsivity: a hypothesis for the aetiology of schizophrenia. Neuroscience (1991) 41: 1-24.
Kalat, J (1995) Biological Psychology 5th Edition CA: Brooks/Cole
Kebabian, J., & Calne, D. (1979) Multiple receptors for dopamine. Nature, 2177, 93-96.
Kerwin, R. The new atypical antipsychotics. British Journal of Psychiatry (1994), 164, 141-148.
Kinon, B. & Lieberman, J. Mechanisms of action of typical antipsychotic drugs: a critical analysis. Psychopharmacology (1996) 124: 2-34.
Kolb, B & Whishaw, I. (1996) Fundamentals of Human Neuropsychology, 4th ed. Freeman: New York
Lidow, M., Williams, G., and Goldman-Rakic, P. The Cerebral Cortex: a case for a common site of action of antipsychotics. Trends in Pharmaceutical Science (1998), vol 19, pp
Middleton, F. & Strick, L. Cerebellar output: motor and cognitive channels. Trends in Cognitive Science (1998) Vol 2, No. 9, 348-354
Reber, A. (1995) The Penguin Dictionary of Psychology. Penguin: London.
Roth, B., Tandra, S., Burgess, L., Sibley, D. & Meltzer, H. D
4 dopamine receptor binding affinity does not distinguish between typical and atypical antipsychotic drugs. Psychopharmacology (1995) 120: 365-368.
Schmahmann, J. Dysmetria of thought: clinical consequences of cerebellar dysfunction on cognition and effect. Trends in Cognitive Sciences. Vol2, No. 9: 362-371
Stahl, S. (1997). Essential Psychopharmacology: neuroscientific basis and practical applications. Cambridge: CUP.
Vallar, L., & Mendolesi, J. (1989) Mechanisms of transduction at the dopamine D2 receptor. Trends in Pharmacological Science, 10, 74-77.

|<|Up|?|^|Menu|Search/Map|Translate|> GOTO 


ahisee?? AhISee!

Valid HTML 4.01!

Entire Site Copyright ę 2001 J. Roberts and AhISee.com - All Rights Reserved.
Problems, questions or comments about the content? Go to the contacts page
Problems, questions or comments about this web site? Send mail to our webmaster