Schizophrenia and Working Memory
This paper will cover some of the neurological characteristics of schizophrenia and the ways that working memory is disordered within individuals who are afflicted with the disease. It will explore the new evidence cognitive neuroscience has provided on the topic and possible treatments that could come from further research.
First, I want to provide an overview of the classification of symptoms of schizophrenia. The fourth version of the Diagnostic and Statistical Manual of Mental Disorders diagnoses and classifies two separate types of schizophrenia, positive (type I) and negative (type II). Positive-symptom schizophrenics experience delusions, hallucinations, and disorganized speech, while negative-symptom schizophrenics experience significant lack of emotion, speech, or motivation; both types are classified by social and occupational dysfunction and decline, as well as continuous signs of the disorder for at least six months prior to diagnosis.(Hansell & Damour, p. 460) Often, patients exhibit both positive and negative symptoms. Within positive and negative schizophrenia, the other subtypes of the disorder include paranoid (prominent delusions or auditory hallucinations), disorganized (prominent disorganized speech, behavior, affect), catatonic (prominent psychomotor symptoms, muteness, echolalia, echopraxia), undifferentiated (active schizophrenia unlike the other subtypes), and residual (after schizophrenic episode, some remaining negative symptoms and mild positive symptoms). (Hansell & Damour, p. 467)
There are several abnormalities in the dopamine (DA), glutamate, serotonin, and GABA neurotransmission systems that have been observed in the neurology of schizophrenic patients that may help explain the positive and negative symptoms of the disorder. (Dean, 2000) The most prominent neurological explanation for symptoms of schizophrenia until about ten years ago was known as the dopamine hypothesis. (van Rossum, 1966) The association of reduced positive symptoms with patients who used D2 DA receptor antagonists, known as neuroleptic drugs, led researchers to conclude that it was an excess of dopamine that caused the primary problems in schizophrenia. Higher baseline densities of pre-synaptic dopamine accumulation in the striatum of the brain observed in many schizophrenic patients as compared to healthy subjects are further evidence of dopamine’s significant role in this illness. (Lyon, Abi-Dargham, Moore, Lieberman, Javitch, & Sulzer, 2009) However, dopamine is now seen as only one part of the explanation of schizophrenia symptoms because some individuals with schizophrenia do not show changes in frequency or intensity of psychotic episodes with neuroleptic drugs.(Hansell & Damour, p. 475) In addition, antipsychotic medicines that focus on other neurotransmitters such as serotonin show effective results that prove more than just dopamine neurotransmission is disordered in schizophrenia. (Lewis, et al., 1999) Research using positron emission tomography and single positron emission computerized tomography suggests that the coexistence of positive and negative symptoms of schizophrenia may be a result of both sub-cortical excesses and cortical deficits of dopamine; the under-regulation of sub-cortical D2 receptors is correlated with the positive psychotic symptoms of schizophrenia, while the over-regulation of cortical D1 receptors in the dorsolateral prefrontal cortex is correlated with deficits in working memory. (Abi-Dargham, 2004) Studies have also shown an over-regulation of GABA transmission in the schizophrenic dorsolateral prefrontal cortex may also be related to working memory impairments.(Lewis, Volk, & Hashimoto, 2004) These findings suggest that there may be chemical imbalances, consisting of both excesses and deficits, of neurotransmitters on cortical and sub-cortical levels that would explain the mix of positive and negative symptoms of schizophrenia.
I want to provide an overview of working memory before further exploring the working memory of schizophrenics. Working memory is synonymous with short-term memory; it is the process by which we maintain pieces of information for relatively short periods of time, and quickly remove the information from memory unless rehearsed or refreshed. (Purves, et al., pp. 405-6) The two primary models of working memory are the Baddeley and the Cowan Models. Alan Baddeley’s model (1974) proposes that a single, multi-component model of working memory is comprised of three components: a verbal storage system called a phonological loop, a visual storage system called a visuospatial sketchpad, and a central executive that allocates attentional resources. (Baddeley, 2003) The phonological loop has a store that can hold memory traces for a few seconds before fading and has a rehearsal process similar to sub-vocal speech; Baddeley proposed its primary function was to facilitate the acquisition of language.(Baddeley, 2003) The visuospatial sketchpad retains up to about three or four objects in working memory at a time, providing humans with a measure of non-verbal intelligence and the capacity to hold and manipulate visuospatial representations. (Baddeley, 2003) Robert Logie proposed to modify this particular system theory after studying visuospatial neglect in patients with right hemisphere damage by distinguishing ‘fractionation’ between the visual storage component, the visual cache, and “the inner scribe”, a dynamic retrieval and rehearsal process. (Logie & Salway, 1995; Baddeley, 2003) Finally, the central executive, “the most important but least understood component of working memory”, was modified to adopt the Norman and Shallice model of attentional control (1986) by dividing control between behavior by habit patterns or schemas implicitly guided by environmental cues and the “supervisory activating system” that takes over when simple routine behavioral execution is inappropriate.(Baddeley, 2003) A fourth component, known as the episodic buffer, was proposed as a system that temporarily stores a “multimodal code” that combines information from the other systems of working memory and from long-term memory into a “unitary episodic representation,” with conscious awareness assumed to be the principal retrieval mode from the episodic buffer.(Baddeley, 2000)
Unlike the Baddeley model, the model proposed by Nelson Cowan is based on the possibility that working memory and long-term memory depend on the same memory representations. There are two levels of working memory according to this model. The first level is made up of an unlimited number of long-term memory representations that are in an “activated state” that declines rapidly without rehearsal. The second level consists of activated representations that receive attention; similar to Baddeley’s model, the Cowan model suggests that a central executive pays attention to specific representations. (Purves, et al., pp. 407-8) It is suggested by both models that the dorsolateral prefrontal cortex controls working memory processes in posterior brain regions. (Purves, et al., p. 410; Baddeley, 2003) Therefore, in order to understand the negative symptoms associated with disordered working memory, cognitive neuroscientists have focused on the dorsolateral prefrontal cortex in schizophrenia.
Cognitive Neuroscience has contributed greatly to our understanding of how schizophrenia is detected, studied, and treated. Conversely, the study of schizophrenia has helped cognitive neuroscientists learn more about working memory. As I began to discuss earlier, GABA neurotransmission in the dorsolateral prefrontal cortex is overregulated in some patients with schizophrenia. This has led some to entertain the possibility that a treatment for dysfunctional working memory in schizophrenics who cannot use neuroleptic drugs might be effective if it targets the inhibitory neurotransmissions at chandelier neuron axon terminals that cause over-regulation of the axon initial segment of pyramidal neurons within the dorsolateral prefrontal cortex. (Lewis, Volk, & Hashimoto, 2004) According to a 2001 study, a calcium binding protein, parvalbumin, exists within these particular chandelier neurons, but chandelier cells of patients with schizophrenia were found to have significantly decreased levels of parvalbumin mRNA, as well as having lowered mRNA levels for GABA membrane transporters responsible for the reuptake of released GABA into the presynaptic axon terminals. (Volk, Austin, Pierri, Sampson, & Lewis, 2001) The implication is an inhibition of GABA neurotransmission in this cortical area, and therefore a potential explanation for functional impairment of working memory. The chandelier neuronal firing patterns facilitate the rhythmic discharge of pyramidal cells, and a single cell can synchronize 250-300 pyramidal cells to fire together; as Lewis, Volk, and Hashimoto concluded, disturbances in these particular chandelier cells “might be expected to produce a reduction in measures of synchronized cortical activity of the type required for working memory” and that “agents with selective agonist activity at [these particular] GABA receptors… may enhance the synchronization of pyramidal neuron activity and thereby improve dorsolateral prefrontal cortex functional output in schizophrenia.” (Lewis, Volk, & Hashimoto, 2004, p. 148) These findings support the aspects of the Baddeley and Cowan models that suggest the dorsolateral prefrontal cortex plays a central role in working memory. The findings also encourage research into GABA-transmission selective agonist treatments for certain subgroups of schizophrenic patients who suffer from a lack of parvalbumin and GABA membrane transporter mRNA.
Numerous cognitive neuroscience studies have pointed to the irregularities of the dorsolateral prefrontal cortex in schizophrenia correlating with working memory impairment. (Abi-Dargham, 2004) Functional magnetic resonance imagery (fMRI) during subjective-rating working memory tasks (in this case, an N-back test) has shown abnormally reduced grey matter volume in the dorsolateral prefrontal cortices of schizophrenic patients, which in turn correlates with the poor performance on these working memory tasks.(Garlinghouse, Roth, Isquith, Flashman, & Saykin, 2010) However, the examinations used to make the diagnosis of the disease and its effects on working memory, as well as some qualities of fMRI studies with respect to the sample size, have come into question. According to one recent article, all fMRI studies of schizophrenia can fall prey to the potential overgeneralizations of having too small a pool of available test subjects, a problem that could be overcome by more scientific collaboration and combination of data sets. (Demirci & Calhoun, 2009)
Not all fMRI studies have fallen prey to this sampling problem. In response to findings suggesting that fMRI can predict whether schizophrenics with catechol-O-methyltransferase genotype val108/158met polymorphism who have taken eight-weeks of olanzapine, a neuroleptic drug, will have working memory deficits(Bertolino, Caforio, Blasi, De Candia, Latorre, & Petruzzella, 2004), a study found that patients with the homozygous, low-activity version of this specific genotype polymorphism have enhanced cognitive performance with neuroleptic drugs, while patients with high-activity COMT val108/158met homozygous genotype show significantly worse performance when treated with neuroleptic drugs.(Weickert, et al., 2004) This discounted the previous findings that made it appear as if fMRI alone could indicate cases of working-memory dysfunction in schizophrenia. More importantly, the study showed that there are schizophrenic patients whose working memory is actually impaired by the drugs they are told will help them. The response study claims to overcome the small subject pool problem by combining data with the Bertolino 2004 study and showing effective replication of the independent findings of an interaction between the genotype polymorphism and antipsychotic treatment found in the study to which it responded. (Weickert, et al., 2004, p. 681) This shows promise that the criticisms brought up by Demirci and Calhoun with regards to fMRI studies can be dealt with if scientists make sure that studies can be replicated and that they combine data sets if the subject pool is small.
My personal criticism of the studies I have discussed so far is that they are often not specific about the DSM-IV diagnosis of each patient as necessary. For example, in the Garlinghouse, et. al. 2010 article, the researchers simply state “all patients met the DSM-IV criteria for schizophrenia according to the Structured Clinical Interview DSM-IV and chart reviews” without specifying whether patients were type I or type II schizophrenics, or whether the patients experienced a combination of symptoms of both. (Garlinghouse, Roth, Isquith, Flashman, & Saykin, 2010) In the Weickert et al. study, the patients were described only as “schizophrenic or schizoaffective,” with even less specificity as to what subtype, or even how the diagnosis itself was made. (Weickert, et al., 2004) In fact, none of the studies I have mentioned thus far have explored the subtypes of schizophrenia that their subjects exhibit. As Demirci and Calhoun say in their article, scientists need to work together to combine studies and data.(Demirci & Calhoun, 2009) This kind of information could be valuable for further research in cognitive neuroscience, abnormal psychology, and psychopharmacology, as well as for understanding the subjects that are being analyzed.
In conclusion, I believe that cognitive neuroscience has contributed greatly to the study of schizophrenia and potential treatments for cognitive impairments to working memory. The complexities of neurotransmission in the dorsolateral prefrontal cortex, as well as the genotypes that specify different reactions to neuroleptic medicines, are now better understood, and give great hope that if scientists can work together by combining research, they will make great strides in treating the positive and negative symptoms of schizophrenia.
Works Cited
* Abi-Dargham, A. (2004). Do we still believe in the dopamine hypothesis? New data bring new evidence. International Journal of Neuropsychopharmacology (7), S1-S5.
* Baddeley, A. (2000). The episodic buffer: a new component of working memory? Trends in Cognitive Sciences , 4 (11), 417-23.
* Baddeley, A. (2003). Working memory: looking back and looking forward. Nature Review: Neuroscience , 4, 829-839.
* Ban, T. (2007). Fifty years chlorpromazine: A Historical Perspective. Neuropsychiatric Disease and Treatment , III (4), 495-500.
* Bertolino, A., Caforio, G., Blasi, G., De Candia, M., Latorre, V., & Petruzzella, V. (2004). Interaction of COMT val^108/158met genotype and olanzapine treatment on prefrontal cortical function in patients with schizophrenia. American Journal of Psychiatry (161), 1798-1805.
* Dean, B. (2000). Signal transmission, rather than reception, is the underlying neurochemical abnormality in schizophrenia. The Australian and New Zealand Journal of Psychology , 34 (4), 560-9.
* Demirci, O., & Calhoun, V. (2009). Functional Magnetic Resonance Imaging - Implications for Detection of Schizophrenia. European Neurological Review , 4 (2), 103-7.
* Garlinghouse, M., Roth, R., Isquith, P., Flashman, L., & Saykin, A. (2010, February). Subjective rating of working memory is associated with frontal lobe volume in schizophrenia. Schizophrenia Research , 1-5.
* Hansell, J., & Damour, L. (2007). Abnormal Psychology: Second Edition. Jefferson City: John Wiley & Sons, Inc.
* Lewis, D., Volk, D., & Hashimoto, T. (2004). Selective alterations in prefrontal cortical GABA neurotransmission in schizophrenia: a novel target for the treatment of working memory dysfunction. Psychopharmacology (174), 143-150.
* Lewis, R., Kapur, S., Jones, C., DaSilva, J., Brown, G., Wilson, A., et al. (1999). Serotonin 5-HT2 receptors in schizophrenia: A PET study using setoperone in neuroleptic-naive patients and normal subjects. American Journal of Psychiatry , 156, 72-8.
* Logie, R., & Salway, A. F. (1995). Visuo-spatial Working Memory. British Journal of Psychology (86), 253-69.
* Lyon, G., Abi-Dargham, A., Moore, H., Lieberman, J., Javitch, J., & Sulzer, D. (2009, June 12). Presynaptic Regulation of Dopamine Transmission in Schizophrenia. Retrieved April 29, 2010 from Schizophrenia Bulletin: http://www3.interscience.wiley.com/cgi-bin/lnx?FROM=REF&TYPE=JCIT&TTL=Schizophr.+Bull.&VID=&IID=&ISSN=&PPF=&SNM=Lyon&ATL=Presynaptic+regulation+of+dopamine+transmission+in+schizophrnenia&PUB=&PYR=2009&PPL=&ADOI=10.1111%2Fj.1471-4159.2010.06604.x&BIBID=142_139&EVIEW=Y&ENABLEISI=N&ENABLECAS=Y&ENABLEPM=Y
* Purves, D., Brannon, E., Cabeza, R., Huettel, S., LaBar, K., Platt, M., et al. (2008). Principles of Cognitive Neuroscience. Sunderland, MA: Sinauer Associates, Inc.
* van Rossum, J. (1966). The significance of dopamine-receptor blockade for the mechanism of action of neuroleptic drugs. Archives Internationales de Pharmacodynamie en et Therapie , 160 (2), 492-4.
* Volk, D., Austin, M. C., Pierri, J. N., Sampson, A. R., & Lewis, D. (2001). GABA transporter-1 mRNA in the prefrontal cortex in schizophrenia: decreased expression in a subset of neurons. American Journal of Psychiatry (57), 237-45.
* Weickert, T., Goldberg, T., Mishara, A., Apud, J., Kolachana, B., Egan, M., et al. (2004). Catechol-o-methyltransferase val^108/158met Genotype Predicts Working Memory Response to Antipsychotic Medications. Society of Biological Psychology (56), 677-682.
