Memory Dysfunction in Patients Diagnosed with Schizophrenia: A Critical Examination of the Literature

Based on the findings of experimental and idiographic research, it has been well documented that there is a memory deficiency with patients diagnosed with schizophrenia (Aleman, Hijman, Haan, & Kahn, 1999). This has been studied by examining neurological structures, recognition capability, mood states, sex and age differences. The investigation of drug treatment has also been tested with the anticipation of ameliorating the deficiency.

When schizophrenia was first studied by Kraeplin and Bleuler, noted as the pioneers of this field of research, memory impairment was not of any great concern (Aleman, Hijman, Haan, & Kahn, 1999). Recently, it has been greatly noted that there is a deficiency in memory when examining schizophrenic patients (Aleman, et. al., 1999). Research, through means of Meta analysis have yielded results that indicate that memory deficiency is enhanced when the schizophrenic patient is depressed. This was found in conjunction with evidence supportive of the notion that memory capacity is less than that of subjects lacking a psychotic diagnosis (Aleman, et. al., 1999). Aleman, Hijman, Haan, & Kahn (1999) investigated this through an extensive literature search. Past research supplied statistical information dealing with free recall, cued recall, and recognition. The results of 70 studies were compared and effect sizes were calculated. Although the studies were empirically based, the results yielded by Meta analysis tended to be more descriptive and supplied preliminary information for further investigation. Meta analysis allowed for scientific assumptions to be made. In this case, the acceptance of the theory that schizophrenics have a memory problem which may be reduced further under certain conditions.

Wexler, Stevens, Bowers, Sernyak, and Goldman-Rakic (1998) have researched schizophrenic memory under verbal and non-verbal conditions. The researchers gathered a sample population of 77 subjects. Thirty-eight of the participants had received a diagnosis of schizophrenia based on the criteria found in the DSM-III-R. The control group, consisting of the remaining 39 subjects, did not have a psychotic diagnosis and were considered representative of the general population. Verbal measures were gathered through the use of the Word List Recall Test and the Word Serial Recall Test. Non-Verbal scores were yielded from the Tone Delayed Discrimination Test and the Tone Serial Position Test. These tools were chosen based on their noted validity and reliability, supported by past research.

Prior to completing any of the tests for memory, the participants were screened for their capability to form perceptions and maintain attention. This was done to factor out the interplay of extraneous variables and strengthen construct validity. It was also noted that the participants did not differ in age, sex, education, parental education, or handedness (Wexler, Stevens, Bowers, Sernyak, & Goldman-Rakic, 1998, p. 1094). Although this was stated, the authors failed to clarify exactly what the age, sex, education, parental education, or handedness were, giving an unclear description. This hurt the external validity and reliability (generalizability) because it failed to state what type of patient among the schizophrenic population were examined. The study did note that subjects were considered either left or right handed according to their tendency to write with either hand five out of seven times during everyday tasks, however, failed to mention the participants dominant hand. Overall, it was not the premise of the study to focus on handedness, but it was a technical question, raised by the researcher, that went without explanation in this case.

The results of the study indicated that overall; schizophrenics perform poorly on memory tasks in general compared to control subjects. Significantly supported was the finding that schizophrenics performed worse on verbal tasks than non-verbal tasks. Although not mentioned in the text, it may be due to the fact that subjects were read a list of words in the verbal task. Non-Verbal stimuli came in the form of a tone. Perhaps the administrator during the verbal condition enhanced the results through his or her tone of voice, presence of authority (making the subject nervous), or presenting an expectancy of poor performance. This was a concept not noted in discussing the results. The discussion did report the validity of the measures taken and the variable accounted for during the trials. In general, the findings were achieved through professional and credible methodology.

Another study with a sample of 60 participants (30 schizophrenics and 30 normals) established the general deficiency of memory in schizophrenic compared to control subjects through the Revised Wechsler Memory Scale (Huron, Danion, Giacomoni, Grange, Robert, & Rizzo, 1995). In this case, the focus of the study was on recognition memory. Subjects were presented with 80 simple words written on cards. The words chosen were either of high or low frequency, as established by research conducted by Gardiner (as cited in Huron, Danion, Giacomoni, Grange, Robert, & Rizzo, 1995, p. 1739). After a fifteen-minute break from the learning task, participants were presented with a list of the 80 words. They were instructed to circle the words that they recognized from the card presentation. Besides mere recognition, the subjects were asked to differentiate words that had stimulated a memory response of any type (coded as "know") versus being just remembered as included in the learning task (coded as "remember"). Upon completion of the examination of the word list, participants were asked to describe the memory that was spurred by the word presentation. This yielded responses that were rated by the researcher as personal, interitem, bizarre, and miscellaneous.

Researchers ran separate ANOVA tests for remember words and know words with the number of correct responses (recognition). The results suggest that low frequency words and high frequency words have the same effect on schizophrenic memory. In both conditions, the schizophrenics reported less words having a remembering effect than the control subjects. When considering low frequency words and the know response, both the schizophrenic group and the control group scored approximately the same. High frequency scores for the know response yielded similar number of correct responses, but schizophrenics performed poorer than the comparison group.

When analyzing the subjective response of personal, interitem, bizarre, and miscellaneous memories, schizophrenics had less associations in all categories except for the area of bizarre, in which case they far exceeded the control subjects. Although it is generally accepted among individuals within and out of the psychological realm that schizophrenics experience more bizarre thoughts than the general population, it may be this assumption that caused the researcher to judge the responses given as bizarre. Otherwise, the associations concerning interitem relationships of words, and personal, emotionally charged words were supportive of the findings dealing with basic recognition and preliminary data.

Efforts to further study recognition and recall were undertaken with a group of 40 individuals diagnosed with schizophrenia (Brebion, Smith, Amador, Malaspina, & Gorman, 1997). In this study, subjects were given tasks that dealt with free recall, implicit recall, recognition and short-term memory, dependent on depression scores. Free recall scores were collected from the subjects through a presentation of 16 words they were told to memorize. After presented with all the words, the subjects were given an empty sheet of paper and asked to recall the words the best to their ability. There were no control groups for comparison used in this study. Subjects were asked to read a paragraph from a book for one minute. The recall task was then repeated for the 16 words. The nature of the book read by subjects was not disclosed in the literature.

Implicit recall was tested immediately following the free recall task. During this part of the study, participants were given a list of prefixes that could be formed into legitimate words by adding words learned in the free recall trial. The numbers of words formed with words from the free recall list were recorded. The study failed to mention whether or not prefixes were used more than once in the word completion exercise. This information could imply the difficulty of forming 16 distinct words or 16 words that share the same prefix.

Having participants choose words presented in the free recall activity tested recognition. An index of accuracy was formed through the use of the Two-High Threshold Theory for both delayed and immediate responses. This theory was not mentioned in the introduction section of the research, therefore, couldn't be explained in this critique.

Short-term memory was analyzed through the forward and reverse presentation of numbers of the WAIS. The exercise involved the repetition of the presented series of digits. Once all the trials were completed for the different memory measures, SPSS for windows processed the scores. The author stated this procedure to be accepted and decided not to further explain its use in this study. No comment can be made at this time since this has been the preferred method for statistical analysis at my last two academic institutions.

Prior to the administration of the manipulated measures, the Hamilton Depression Rating Scale assessed subjects for depression. The hypothesis was grounded on the premise that degree of depression would yield poor memory scores. Although depression's influence on memory was discussed in the introduction of the study, there was no mention of the Hamilton Depression Scale. This does not allow the reader to judge the scale's validity or reliability; no support was given for its inclusion.

Brebion, Smith, Amador, Malaspina and Gorman (1997) found that depression has a negative influence on memory. This effect was noted as valid for subjects who displayed positive or negative symptoms. This allowed for support of the assumption that anti-depressants could raise schizophrenic's living standard. It fails to answer the question of, so what? Although this study has been written with an extensive literature review and somewhat credible method, it does not really add new knowledge for the trade. It has been documented that depression decreases functioning on many levels, that is the nature of the disorder (see American Psychiatric Association, 1994). It only makes sense that this would decrease performance with schizophrenic patients.

Gold, Carpenter, Randolph, Goldberg, & Weinberger (1997) conducted an in depth study that investigated auditory working memory along with performance. The research, in this case, was founded on a plethora of neuro-psychiatric tests juxtaposition verbal and word association tasks. The major tests discussed were the letter number span (LN) and the Wisconsin Card Sorting Test (WCST). When participants worked with the LN, they sorted a series of numbers and letters, randomly presented, into an order of smallest to largest. Under the assumption that dysfunction in the frontal lobe of the schizophrenic brain causes a deficit in task performance, the use of the WCST was instituted. In order for subjects to successfully complete the WCST, they must draw from past response to continue to the next correct response. Of course, it is not guaranteed that the response will be correct, but the recall is necessary for each succeeding question. Although the explanation of its substructures and complexities has been simplified for this critique, schizophrenic memories were analyzed mainly through the use of the WCST.

The investigators reported that the basic findings of this study were supportive of the hypothesis that schizophrenics have dysfunctional auditory working memories. Structurally, it was found that poor performance on the WCST co-varied with auditory working memory deficits. Overall, providing implications that support the findings of so far mentioned research.

In general, the study presented was complicated and involved many levels of testing. The method seemed credibly supported by past research noted in the introduction. The results explained the complexities of the study and funneled the information into several simple statements that found memory in schizophrenia to be deficient.

Vinogradov, Willis-Shore, Poole, Marten, Ober, and Shenaut (1997) attempted to study source monitoring, which is involved in memory, through the use of the WCST. The Wisconsin Card Sorting Test was used to examine the complex thought activity processed by the frontal lobe of the brain. In this research, much like the previous noted study, tasks that required higher functioning and mental processing were considered representative of the part of the brain that is responsible for the activity. These assumptions were tested through neuropsychological test batteries, not with standardized medical equipment. By comparing the results of subjects diagnosed with schizophrenia to those of assumed normal participants, investigators intended to make judgement on memory deficits.

As an aside, prior to continuing the examination of this particular study, I feel it is necessary to raise an issue that seems to be present in the studies that use neurological test batteries. It seems that the tests are not really testing the construct stated. Although, the exercises all require responses dependent on memory, it seems that performance is really being studies by these tests. If this were the case, this would be a major validity issue. As stated with the previous study that dealt with the WCST, there may be test anxiety, expectations by the subjects to do poorly, or other extraneous variables related to performance. By reading the articles, it is obvious that care was taken by the researchers to have solid cause for the measures taken, however, no matter how professionally executed, interactions of unwanted variables are possible.

These issues are evident in the research currently being examined. Researchers had subjects study sentences and instructed memorization. Sentences presented alternated complete sentences with ones that the subject was to complete. Afterwards, the subjects were tested on the words that were presented in one sentence paired with the ones fabricated in the completion tasks by the subjects. Understandably, this is a measure of memory; however, it is also a measure of performance. It is very difficult to differentiate which construct is really being analyzed.

To gather information related to neurology, subjects were given the neurological battery of tests. This included the Neurological Signs Inventory that, as with the WSCT, measure frontal lobe dysfunction. The Brief Psychiatric Rating Scale and the Shipley Institute of Living Scale, an IQ test, were also administered. In this case, it was found that there is some degree of memory dysfunction in schizophrenia. An interesting trend that was noted occurred during the testing trial for the word completion task. Apparently, participants with schizophrenia that recognized the words they had supplied in the sentence completion task held the experimenter responsible for the original presentation of the word. They were dissociated with the word they had interjected originally. The nature of this finding was not discussed in much further depth.

Typically, upon the commencement of a study, subjects are asked basic questions of age and sex. To put this information to use, the question of sex differences among schizophrenics has been addressed in reference to memory impairment (Goldstein, Seidman, Goodman, Koren, Lee, Weintraub, & Tsuang, 1998). In this case the participants consisted of males and females that were either schizophrenic or considered by the researchers, as normal (not diagnosed with schizophrenia). The subjects in the experimental and control groups were matched for age, sex, social economic status, and handedness. A neuropsychological battery was used to assess the degree of many factors of functioning, including memory.

Although the results failed to make significant implications, when analyzing the data descriptively, it was noted that males who were diagnosed schizophrenic performed more poorly than females under the same diagnosis. Overall, patients diagnosed with schizophrenia displayed memory deficiencies independent of sex. The author regards memory deficits regarding schizophrenic patients as being an accepted assumption.

As a criticism, the study had a small sample size (n = 68). Perhaps the findings would have been significant with a larger sample. In this case, the researchers took special measures in supporting their instruments by noting success in past research and considering interactions of extraneous variables. Overall, the researchers considered many of the factors that go into constructing an empirically based study. They mentioned reliability and consistencies as well as discussed the strengths and weaknesses of the methodology after the fact of testing. Of course, the results further support the notion that schizophrenic memory seems impaired. Sex differences, however, seems not to be an issue.

The issue of age is important in the study of schizophrenia. Schizophrenics who suffer negative symptoms (Type II) are typically diagnosed at an earlier age than those who experience positive symptoms (Type I) (see, American Psychiatric Association, 1994). The assumption that age may be a factor with the deficits experienced by schizophrenics give cause for the investigation of memory in patients that were diagnosed early in life (Merete, & Rund, 1999). Through the use of measures previously mentioned in studies already discussed, such as the Brief Psychiatric Rating Scale, IQ tests, and the WCST, investigators examined the memories of schizophrenics with adolescent onset. As with the findings with the general schizophrenic population, it was found that those suffering from schizophrenia during their late adolescent years have dysfunctional memories. This presented evidence that age was not a factor when discussing schizophrenic memory. It was noted, however, that the findings of memory deficit were similar to a vast body of research that examined the same construct through medical equipment, specifically, the MRI.

In order to examine the relationship of memory and diagnosis of schizophrenia, research must investigate the biological structures that process memory. By measuring the physiology of a process, researchers are able to rule out expectations and extraneous variables more than in the use of observational studies or idiographic techniques. Unfortunately, as with all research, there are still problems present. Biological measurements are collected through the use of medical equipment, such as the EEG, MRI, and PET. Although it is rare, the instruments may be dysfunctional and supply incorrect data, which may not be observed by the researcher.

Stevens, Goldman-Rakic, Gore, Fulbright, and Wexler (1998) tested cortical dysfunction in schizophrenic patients that completed memory tasks through the MRI (magnetic resonance imaging system). This instrument scans the cortex of the brain (location of higher functioning) and projects its activity in the form of an image onto a computer monitor. This type of examination validates tests such as the WCST. If the imagery of brain activity correlates to the results suggested by the WCST, the findings are less probable the cause of an extraneous variable. The results of this study pinpoint the area of the brain that is associated with memory deficit. The results of the MRI imply that impairment of verbal working memory (the construct being tested in this particular study) may be related to the cortical hemodynamic response in the inferior frontal gyrus (Stevens, Goldman-Rakic, Gore, Fulbright, and Wexler, 1998).

Brain function in relation has been studied through the use of a positron emission tomography scan (PET) (Carter, Perlstein, Ganguli, Brar, Mintun, & Cohen, 1998). This test measures the blood flow in the brain, therefor, displaying activity. In conjunction with the N-back task, noted for stimulating the prefrontal cortex as a result of working memory in the general population, the PET measured the brain activity of schizophrenics. The results showed that schizophrenic brain activity when the N-back task was performed were slower than that of control subjects. Both groups had difficulty with high load memory tasks, as measured by blood flow. It was found that dorsolateral prefrontal cortex activation was less for schizophrenia patients, especially as approaching increased memory load. Retarded functioning of the posterior parietal cortex was also noted in high memory load tasks. Critically speaking not much could be discussed on the findings of the PET scan. It is an instrument that measures physiological function and is hard to disprove. The findings of positron emission tomography scan and the magnetic resonance imagery may be necessary for the validation of psychoneurological tests such as the Wisconsin Card Sorting Test and the N-back task. In this study, it was found through the PET scan that the prefrontal cortex was stimulated, an assumption on which the N-back task is based.

Overall, the studies thus far have shown there is a deficit in schizophrenia. Upon accepting that theory, it could be assumed that the psychological community would want to ameliorate the impairment. Since the studies are in agreement and have replicated the deficiency under numerous conditions, it would be logical that treatment could be achieved through some means. Based on the physiology of the impairment, neuroleptics have been suggested to be the answer.

In order to test the theory, Risperidone, noted to be the least volatile, yet effective neuroleptic available through prescription, was studied for its effects on working memory (Green, Marshall, Wirshing, Ames, Marder, McGurk, Kern, and Mintz, 1997). Specifically, Risperidone was being tested to replace the current use of Haloperidol. In this case, many factors went into the controls of the study. Patients were analyzed using psychoneurological tests and were monitored under extreme conditions that limited the interaction of other psychoactive drugs. Careful measures were taken to perform the double blind experiment. Ultimately, it was found that Risperidone yielded results that were considered more beneficial than Haloperidol. This remained constant across both positive and negative symptoms. Further discussions concerning the careful measures taken in the study were noted as being accurate and allowing for the exclusion of variables that would invalidate the study. Overall, the study seemed credible.

In general the problem of memory impairment in schizophrenia has been noted as a condition that exists in addition to the other deficiencies that the diagnosed experiences. Unfortunately, the only solution found at this point is chemical treatment. As with most drug interventions, the effects are temporary and symptoms return when discontinued. A problem with schizophrenics and other psychopathologies is that patients tend to discontinue medication upon their taking effect. They feel better and ignore the prescription.

Recognizing the problem is beneficial for clinical intervention and for better understanding of the client. Memory dysfunction, along with the other ailments of the diagnosis should be further studied for treatment. It seems that there are plenty of similar findings and methods of testing for the dysfunction. Less popular, as far as finding literature related, is in the treatment of the impairment. This is where the research should be heading. After reading the literature available, this is the obvious next step.

Aleman, A., Hijman, R., Haan, E. F., & Kahn, R. S. (1999). Memory impairment in schizophrenia: A meta-analysis. American Journal of Psychiatry, 156, 1358-1366.

American Psychiatric Association (1994). Diagnostic and Statistical Manual-IV. Washington, D.C.:A.P.A.

Brebion, G., Smith, M.J., Amador, X., Malaspina, D., & Gorman, J.M. (1997). Clinical correlates of memory in schizophrenia: Differential links between depression, positive and negative symptoms, and two types of memory impairment. American Journal of Psychiatry, 154 (11), 1538-1543.

Carter, C.S., Perlstein, W., Ganguli, R., Brar, J., Mintun, M., & Cohen, J.D. (1998). Functional hypofrontality and working memory dysfunction in schizophrenia. American Journal of Psychiatry, 155 (9), 1285-1287.

Gold, J.M., Carpenter, C., Randolph, C., Goldberg, T.E., & Weinberger, D.R. (1997). Auditory working memory and Wisconsin Card Sorting Test performance in schizophrenia. Archives of General Psychiatry, 54 (2), 159-165.

Goldstein, J. M., Seidman, L. J., Goodman, J. M., Koren, D., Lee, H., Weintraub, S., & Tsuang, M. T. (1998). Are there sex differences in neuropsychological functions among patients with schizophrenia? American Journal of Psychiatry, 155, 1358-1364.

Green, M.F., Marshall, B.D., Wirshing-Williams, C., & Ames, D. (1997). Does risperidone improve verbal working memory in treatment-resistant schizophrenia? American Journal of Psychiatry, 154 (6), 799-804.

Huron, C., Danion, J. M., Giacomoni, F., Grange, D., Robert, P., & Rizzo, L. (1995). Impairment of recognition memory with, but not without, conscious recollection in schizophrenia. American Journal of Psychiatry, 152, 1737-1742.

Merete, O., & Rund, B. R. (1999). Neuropsychological deficits in adolescent-onset schizophrenia compared with attention deficit hyperactivity disorder. American Journal of Psychiatry, 156, 1216-1222.

Stevens, A.A., Goldman-Rakic, P.S., Gore, J.C., Fulbright, R.K., & Wexler, B.E. (1998). Cortical dysfunction in schizophrenia during auditory word and tone working memory demonstrated by functional magnetic resonance imaging. Archives of General Psychiatry, 55 (12), 1097-1103.

Vinogradov, S., Willis-Shore, J., Poole, J. H., Marten, E., Ober, B. A., & Shenaut, G. K. (1997). Clinical and neurocognitive aspects of source monitoring errors in schizophrenia. American Journal of Psychiatry, 154, 1530-1537.

Wexler, B.E., Stevens, A.A., Bowers, A.A., & Sernyak, M.J. (1998). Word and tone working memory deficits in schizophrenia. Archives of General Psychiatry, 55 (12), 1093-1096.