Pii: s0306-9877(02)00406-

Medical Hypotheses (2003) 60(3), 382–386 ª 2002 Elsevier Science Ltd. All rights reserved.
The neurochemical hypothesis of‘theory of mind’ This paper aims to explore the neurochemical basis of the ability to represent one’s own or other’s mental states such as intentions, beliefs, wants and knowledge, an ability often referred to as ‘theory of mind’.
Based on neurochemical and psychopharmacological investigations in autism and schizophrenia, pathologies inwhich this ability is impaired, it is hypothesized that ‘theory of mind’ abilities are contingent on the integrity of theserotonergic and dopaminergic system. This hypothesis is discussed in light of the system’s neurochemicalproperties and role in cognition. It is suggested that specific abnormalities to this system can account for differencesin the profile of ‘theory of mind’ impairments that may exist among patients belonging to different pathologies.
ª 2002 Elsevier Science Ltd. All rights reserved.
our ability to mentalize. In providing evidence for thishypothesis, the nature of the DS system will be reviewed Theory of mind (ToM), sometimes used interchangeably in patients with autism and schizophrenia. This will be with other terms such as mentalizing capacity, is the followed by a review of the role of the DS system in ability to represent one’s own or another’s mental states cognition and a discussion of its relevance to mentalizing such as intentions, beliefs, wants and knowledge. This abilities. The implications of this hypothesis to explain- cognitive ability has a central role in human social ing differences in ToM impairments across disorders will interaction and is believed to underlie our ability to then be presented. Finally, the paper will conclude with a understand and predict other people’s behavior. Im- summary and some predictions based on the hypothe- pairment to this ability has long been recognized in in- sized role of the DS system in mentalizing.
dividuals with autism (1), and more recently in patientswith schizophrenia (2). Theory of mind deficits have alsobeen observed in other disorders such Williams syn- drome and Prader-Willi syndrome (3).
Neurochemical and psychopharmacological investi- gations reveal that patients with these disorders, partic- Theory of mind impairment is most widely investigated ularly autism and schizophrenia, have prominent deficits in autism and schizophrenia. While both autism (4,5) in the dopaminergic–serotonergic (DS). In addition, the and schizophrenia (6,7) have dysregulated DS systems, DS system has been shown to play a substantial role in schizophrenia is largely linked to dysfunctions in the cognitive functioning. It is hypothesized that the DS dopaminergic system (8,9), and autism is largely linked system is likely to be the neurochemical basis subserving to a dysfunction in the serotonergic system (10,11). Withrespect to schizophrenia, the most widely accepted hy-pothesis is that schizophrenic patients suffer from two different disturbances in the dopaminergic transmission system. The first is hyperactivity in the mesolimbicsubcortical component of the dopamine system, and the Correspondence to: Ahmad Abu-Akel, P.O. Box 1217, Arara 30026, Israel.
second is hypoactivity of the mesocortical component of Phone: +972-4-6352187; Fax: 972-4-6355632; the system (8). The hyperactivity of dopamine in the The neurochemical hypothesis of ‘theory of mind’ mesolimbic component may be due to a dysfunction in rewarding events’ (p. 1593) (21). It is therefore likely that the feedback inhibitory loop exerted by cortical over the dopamine system is involved in the mechanism With respect to autism, neurochemical investigations With respect to the serotonin, Buhot (22) concludes report higher mean levels of 5-HT (5-hydroxytrypta- that serotonin receptors (perhaps all) affect various mine; serotonin) in the whole-blood of children with cognitive functions such as memory and executive autism compared to non-autistic children (10,11). Pa- functioning. This conclusion finds support in studies tients whose whole-blood 5-HT is greater than 270 showing that serotonin has a role in the treatment and/ ng/ml are considered hyperserotonemic (12). Elevated or pathogenesis of cognitive disorders such as schizo- serotonin levels are found in between 30% and 50% of phrenia and autism, and the projection of this system to children and adolescents with autism, and it has been regions subserving these cognitive functions such as the suggested to be the result of a defect in 5-HT biochem- prefrontal cortex. Moreover, serotonin abnormalities istry of the dentatothalamocortical pathway (13). Using have also been shown to affect language abilities (23).
a-[11C]methyl-L-tryptophan as a tracer for serotonin To put it together, these studies reveal three impor- synthesis with positron emission tomography, Chugani tant aspects that are highly relevant for the hypothe- and co-workers report unilateral alterations of serotonin sized role of the DS system in mentalizing. First, both the synthesis in the dentatothalamocortical pathway in pa- dopaminergic and the serotonergic systems innervate tients with autism. Asymmetries of serotonin synthesis regions that are critical for mentalizing. These brain re- were found in the frontal cortex, thalamus, and dentate gions which include the prefrontal cortex, the temporo- nucleus of the cerebellum. The authors suggest that parietal junction and the anterior cingulate cortex have these serotonergic abnormalities in the dentatothal- been independently shown in several imaging and lesion amocortical pathway could be one mechanism under- studies to be involved during tasks requiring mentaliz- lying the pathophysiology of autism.
ing abilities (24). Second, abnormalities in either thedopaminergic and the serotonergic systems lead to thedisruption of cognitive abilities such as language use Role of the dopaminergic–serotonergic system in that are dependent on ToM abilities (25), or cognitive abilities that influence ToM abilities such as executive On the surface, these data provide at best circumstantial functioning (26). Third and perhaps more importantly, is evidence for the possible role of the DS system in men- the putative property of the dopaminergic system in talizing abilities. However, there is independent evi- signaling predictions about consequences of future dence for the role of dopamine and serotonin in events. The dopamine system appears a natural mecha- cognitive functioning from which the hypothesized role nism upon which ToM abilities have evolved.
of the DS system in mentalizing can be inferred. With It should be emphasized, however, that while both respect to dopamine, studies reveal that manipulation of the serotonergic and the dopaminergic systems can in- the dopaminergic system affects performance on cogni- dependently have adverse effects on ToM abilities, the tive tasks that are dependent on the prefrontal cortex coupling of both systems is more adequate in account- (14,15), and the anterior cingulate cortex (16). Equally, ing for our ability to mentalize for at least two reasons.
deficits in working memory (17) and executive func- First, serotonin has a modulatory effect on dopaminergic tioning (18) have been linked to dopamine abnormalities transmission (27–29). For example, blockade of 5-HT2 in the prefrontal cortex of patients with schizophrenia.
receptors results in an increase of extracellular dopamine Moreover, language deficits in patients with schizo- levels (28). Second, psychopharmacological studies re- phrenia have been linked to dopamine abnormalities in port better cognitive outcomes for patients with schizo- mesocortical areas such as the temporo-parietal region phrenia receiving medications that bind to both (19) and the prefrontal cortex (20).
dopamine and serotonin receptors (i.e., atypical neuro- Moreover, as mentioned earlier, the ability to men- leptics), as opposed to patients receiving medications talize enables us to predict other people’s behavior. One whose predominant mode of action is the blockade of application of this cognitive ability is that, during inter- dopamine receptors (i.e., typical neuroleptics) (6,30).
action, humans make guesses or predictions about theintentions, dispositions and knowledge of each other.
An interaction is said to be successful, or rewarding, when erroneous predictions are minimal. This ability topredict is most likely to depend on the dopamine system Current ToM paradigms examining mentalizing abilities which is known to be involved in learning and signaling are not designed to discern differences in ToM impair- ‘changes or errors in the predictions of future salient and ments that could exist among patients with different ª 2002 Elsevier Science Ltd. All rights reserved.
Medical Hypotheses (2003) 60(3), 382–386 disorders. For example, patients with schizophrenia Abnormalities in the DS system may also become irre- perform similarly to patients with autism. However, coverable in patients with schizophrenia receiving strong evidence suggests that there are substantial dif- medications at very high doses and/or for a long period ferences in the nature of ToM impairment between these of time (43). This condition is also present in negative two populations (31). For the purposes of this paper, the symptom schizophrenia, perhaps due to neurodevelop- most important difference pertains to the severity of the mental abnormalities. In fact, like in autism, this sub- impairment in that while ToM impairment is a trait in group of schizophrenia exhibits severe ToM deficits autism (32), it is a state in schizophrenia (2,33). That is, remitted schizophrenic patients perform as well as Alternatively, it is suggested that a possible pharma- normals on ToM tasks (34). It is claimed that the neu- cotherapy outcome in schizophrenia is the restoration of rochemical hypothesis proposed herein provides a ‘lost’ ToM abilities. It is important to point out that ref- framework where differences in the nature of ToM im- erence to pharamcotherapeutic treatment be confined to pairment can be accounted for. It should be noted, atypical neuroleptics such as risperidone and olanzapine however, that in reference to the transient nature of ToM which are designed to correct abnormalities, primarily, impairments in schizophrenia, patients with chronic in the DS system (6). Unlike typical neuroleptics such as negative feature schizophrenia are excluded. There is haloperidol, atypical neuroleptics have been shown to converging evidence strongly suggesting that, similar to improve cognitive functioning in patients with schizo- autism, ToM impairment in this subgroup of schizo- phrenia such as attention, memory and executive func- phrenia is nontransient. First, patients belonging to this tioning (6,30) and to reduce language deficits (45). These group reportedly have social and cognitive development cognitive and language abilities have been shown in abnormalities in early childhood (35) during which pe- many studies to be linked to ToM abilities (46).
riod ToM abilities develop (36). Second, early neurode- In sum, it can be hypothesized that the trait-state velopmental abnormalities have been linked to patients distinction between schizophrenia and autism with re- with this type of schizophrenia (37,38). Third, it is spect to ToM impairments can be attributed to specific widely reported that patients with chronic schizophrenia disturbances in the DS system. Recoverability of men- are refractory to medication, and hence cognitive abili- talizing abilities appears to be contingent on targeting ties including ToM impairments would likely be irre- both the dopaminergic and serotonergic systems – medications that primarily target the dopamine system One can assume that the trait-state nature of ToM (47,48), or the serotonin system (41) fail in producing impairment in autism and schizophrenia is due to the significant cognitive improvements. The duration of the specific abnormalities in the neurochemistry of the DS disruption to the DS system (or duration of illness) is system of each disorder. Psychopharmacological trials in another factor that can predict recoverability.
patients with autism and schizophrenia provide clues infavor of this assumption. Unlike those in schizophrenia, psychopharmacological interventions including thoseusing atypical neuroleptics (39,40) were unsuccessful in Neurochemical dysfunctions in pathologies of belief such improving core problems of social relatedness and as schizophrenia and autism provide an insight to the communication/language deficits for which ToM abili- neurochemical basis of ToM. It is proposed that a dis- ties is a prerequisite. Such trials were also unsuccessful ruption to the dopamine system itself or to neurochem- in improving symptoms or the course of autism in the ical processes that effect/modulate its functioning such majority of patients, expect for the induction of minimal as the serotonin system, could lead to the generation of palliative effects of specific behaviors (41).
erroneous predictions about the content of the mind of But what could retard the recovery of the DS system others, as is the case for patients with schizophrenia, or in autism? One possibility is that recoverability of the DS the inability to generate any predictions whatsoever as is system is related to etiological differences. Patients with the case in patients with autism. A corollary hypothesis is schizophrenia contrast with patients with autism in that that the severity of ToM impairment depends on the they initially develop ToM normally without any delay extent to which the DS system is recoverable. Given the (2,34), i.e., these patients do not have impairment to current state of advancement in psychopharmacology, it ToM until the onset of their illness. The onset of appears that early disruptions to the DS render restora- schizophrenia is typically between 20 and 35 years of tion of ToM abilities more difficult. Alternatively, dura- age (42), whereas autism appears, by definition, before 3 tion of illness could equally retard recovery. Where years of age. This could be taken to mean that the recoverable, atypical neuroleptics can be seen as an agent protracted neurochemical disruption to the DS system in that has a specific and unique therapeutic effect on the autism has made it refractory to medication treatment.
Medical Hypotheses (2003) 60(3), 382–386 ª 2002 Elsevier Science Ltd. All rights reserved.
The neurochemical hypothesis of ‘theory of mind’ This hypothesis is believed to add an important di- memory and motor performance in monkeys. J Neurosci mension to existing ToM neurobiological models (49,50). These models, which map this ability at the 15. Muller U., von Cramon D. Y., Pollman S. D1-versus D2-receptor modulation of visuospatial working memory gross neuroanatomical level, cannot account for differ- in humans. J Neurosci 1998; 18: 2720–2728.
ences in the nature of ToM impairment across disorders.
16. Dolan R. J., Fletcher P., Frith C. D. et al. Dopaminergic This is particularly the case for autism and schizophrenia modulation of impaired cognitive activation in the – neuroimaging studies reveal that mentalizing deficits anterior cingulate cortex in schizophrenia. Nature 1995; in these two disorders originate from abnormalities in 17. Goldman-Rakic P. S. Prefrontal cortical dysfunction in common brain regions which include the amygdala, the schizophrenia: the relevance of working memory. In: B. J.
prefrontal cortex, and the temporo-parietal junctions Carroll, J. E. Barrett (eds). Psychopathology and the Brain.
(51). Psychopharmacological investigations and neuro- New York: Raven Press, 1991: 1–23.
chemical profiling in schizophrenia are profitable ave- 18. Okubo Y., Suhara T., Suzuki K. et al. Decreased prefrontal nues that could help understand the neurochemistry of dopamine D1 receptors in schizophrenia revealed by PET.
Nature 1997; 385: 634–636.
19. Goldberg T. E., Weinberger D. R. Thought disorder in schizophrenia: a reappraisal of older formulations and anoverview of some recent studies. Cogn Neuropsychiatry 20. Cohen J. D., Servan-Schreiber D. Context, cortex, and 1. Baron-Cohen S., Leslie A. M., Frith U. Does the autistic child dopamine: a connectionist approach to behavior and have a ‘theory of mind’? Cognition 1985; 21: 37–46.
biology in schizophrenia. Psychol Rev 1992; 2. Corcoran R., Mercer G., Frith C. Schizophrenia, symptomatology and social inference: investigating the 21. Schultz W., Dayan P., Montague P. R. A neural substrate of theory of mind in people with schizophrenia. Schizophr Res prediction and reward. Science 1997; 275: 1593–1599.
22. Buhot M. C. Serotonin receptors in cognitive behaviors.
3. Sullivan K., Tager-Flusberg H. Second-order belief Curr Opin Neurobiol 1997; 7: 243–254.
attribution in Williams syndrome: intact or impaired? Am J uller R.-A., Chugani D. C., Behen M. E. et al. Impairment of dentatothalamocortical pathway in autistic men: 4. Herault J., Martineau J., Perrot-Beaugerie A. et al.
language activation data from positron emission Investigation of whole blood and urine monoamines in tomography. Neurosci Lett 1998; 245: 1–4.
autism. Eur Child Adolesc Psychiatry 1993; 2: 211–220.
24. Adolphs R. The neurobiology of social cognition. Curr Opin 5. Croonenberghs J., Delmeire L., Verkerk R. et al. Peripheral markers of serotonergic and noradrenergic function in post- 25. Siegal M., Peterson C. C. Children’s theory of mind and the pubertal, Caucasian males with autistic disorder.
conversational territory of cognitive development. In: C.
Neuropsychopharmacology 2000; 22: 275–283.
Lewis, P. Mitchell (eds). Children’s Early Understanding of 6. Meltzer H. Y., McGurk S. R. The effects of clozapine, Mind: Origins and Development. Hillsdale: Erlbaum, 1994: risperidone, and olanzapine on cognitive function in schizophrenia. Schizophr Bull 1999; 25: 233–255.
26. Russell J., Mauthner N., Sharpe S., Tidswell T. The ‘windows 7. Keefe R. S. E., Harvey P. D. Understanding Schizophrenia.
task’ as a measure of strategic deception in preschoolers and autistic subjects. Br J Dev Psychol 1991; 9: 331–349.
8. Weinberger D. R. Implications of normal brain development 27. McCormick D. A. Neurotransmitter actions in the for the pathogenesis of schizophrenia. Arch Gen Psychiatry thalamus and the cerebral cortex and their role in neuromodulation of thalamocortical activity. Prog 9. Lipska B. K., Weinberger D. R. Cortical regulation of the mesolimbic dopamine system: implications for 28. Devaud L. L., Hollingsworth E. B., Cooper B. R. Alterations in schizophrenia. In: P. W. Kalivas, C. D. Barnes (eds). Limbic extracellular and tissue levels of biogenic amines in rat brain Motor Circuits and Neuropsychiatry. London: CRC Press, induced by the serotonin (2) receptor antagonist, ritanserin.
10. Cook E. H. Autism: review of neurochemical investigation.
29. Milan M. J. Improving the treatment of schizophrenia: focus on serotonin (5-HT)-sub(1A) receptors. J Pharmacol Exp Ther 11. Cook E. H., Leventhal B. L. The serotonin system in autism.
Curr Opin Pediatr 1996; 8: 348–354.
30. Berman I., Klegon D., Fiedosewicz H., Chang H. The effect 12. Cook E. H., Leventhal B. L., Heller W. et al. Autistic children of novel antipsychotics on cognitive function. Psychiatry and their first-degree relatives: relationships between serotonin and norepinephrine levels and intelligence.
31. Abu-Akel A., Bailey A. L. The possibility of different forms of J Neuropsychiatry Clin Neurosci 1990; 2: 268–274.
theory of mind impairment in psychiatric and 13. Chugani D. C., Muzik O. et al. Altered serotonin synthesis in developmental disorders. Psychol Med 2000; 30: 735–738.
the dentatothalamocortical pathway in autistic boys. Ann 32. Mitchell P. Introduction to Theory of Mind: Children, Autism 14. Arnsten A. F., Cai J. X., Steere J. C., Goldman-Rakic P. S.
33. Sarfati Y., Hardy-Bayle M.-C. How do people with Dopamine D2 receptor mechanisms contribute to schizophrenia explain behavior of others? A study of theory age-related cognitive decline: the effects of quinpirole on of mind and its relationship to thought and speech ª 2002 Elsevier Science Ltd. All rights reserved.
Medical Hypotheses (2003) 60(3), 382–386 disorganization in schizophrenia. Psychol Med 1999; 29: 43. Miller R. The time course of neuroleptic therapy for psychosis: role of learning processes and implications for concepts of 34. Pickup G. J., Frith C. D. Theory of mind impairments in psychotic illness. Psychopharmacology 1987; schizophrenia: symptomatology, severity and specificity.
44. Bowler D. M. ‘Theory of mind’ in Asperger’s syndrome.
35. Forster A., Lewis S., Owen M., Murray R. Premorbid J Child Psychol Psychiatry 1992; 33: 877–893.
adjustment and personality psychosis. Br J Psychiatry 1991; 45. Salome F., Boyer P., Fayol M. The effects of psychoactive drugs and neuroleptics on language in normal subjects and 36. Leslie A. Pretense and representation: the origins of ‘Theory schizophrenic patients: a review. Eur Psychiatry 2000; 15: of Mind’. Psychol Rev 1987; 94: 412–426.
37. Murray R. M., O’Callaghan E., Castle D. J., 46. Baron-Cohen S., Tager-Flusberg H., Cohen D. J.
Lewis S. W. A neurodevelopmental approach to the Understanding Other Minds: Perspectives from Developmental classification of schizophrenia. Schizophr Bull 1992; Cognitive Neuroscience, 2nd edn. Oxford: Oxford University, 38. Lafargue T., Brasic J. Neurodevelopmental hypothesis of 47. Spohn H. E., Strauss M. E. The relation of neuroleptic and schizophrenia: a central sensory disturbance. Med anticholinergic medication to cognitive functions in 39. Fisman S., Steele M. Use of risperidone inpervasive developmental disorders: a case series. J Child Adolesc 48. Mortimer A. M. Cognitive function in schizophrenia – do neurolpetics make a difference? Pharmacol Biochem Behav 40. McDougle C. J., Holmes J. P., Carlson D. C., Pelton G. H., Cohen D. J., Price L. H. A double-blind, placebo-controlled 49. Baron-Cohen S., Ring H. A model of the mindreading: study of risperidone in adults with autistic disorder and neuropsychological and neurobiological perspectives.
other pervasive developmental disorders. Arch Gen In: P. Mitchell, C. Lewis (eds). Origins of Understanding of Mind. New Jersey: Erlbaum, 1994: 183–207.
41. Hunsinger D. M., Nguyen T., Zebraski S. E., Raffa R. B. Is 50. Frith U., Frith C. The biological basis of social interaction.
there a basis for novel pharmacotherapy of autism? Life Sci Curr Dir Psychol Sci 2001; 10: 151–155.
51. Abu-Akel A. (2001) A new neurobiological mapping of 42. McKenna P. J. Schizophrenia and Related Syndromes. New ‘theory of mind’. Unpublished Manuscript. PLD Research York: Oxford University Press, 1994.
Medical Hypotheses (2003) 60(3), 382–386 ª 2002 Elsevier Science Ltd. All rights reserved.

Source: http://cogpsy.skku.ac.kr/cwb-data/data/article/neurochemical_hypothesis2003.pdf