Correlation between amplitude-frequency parameters of electroencephalogram and attention in children with schizophrenia.

Salimova Ksenia Ramizovna

Moscow State University of Psychology and Education, Sretenskaya street, 29, Moscow, RF

Abstract

Background German psychiatrist Eugen Bleuler was the first who coined a term “schizophrenia” in 1908. The disease may start at any age, including early childhood and typically involve impairment of thinking, memory, speech and perception. As potential clinical markers of the disease, among other syndromes it should also be noted impairment of filtration of information coming from sense-organs into the brain. These processes constitute different types of attention: involuntary and voluntary, sensory, emotional, intellectual and motor one. It has been shown that people with schizophrenia have severe attentional deficiency displaying decrease in the volume and selectivity of their attention as compared with healthy subjects. The voluntary attention is impaired from the early stages of the disease, while involuntary attention remains safe longer than voluntary one. 

Although there are several different hypothesis trying to explain pathogenesis of schizophrenia, recognition of biomarkers of this disease in children still remains an actual problem of modern biomedical research. Among most handy and non-invasive biomarkers schizophrenia are parameters of electroencephalogram (EEG) that are accepted by researchers as criteria for differential diagnosis. However, only few studies in children with schizophrenia analyze correlation of EEG variables with parameters of attention.

The aim of the present study was analysis of quantitative EEG (QEEG) data and attention parameters in children with schizophrenia before and after pharmacological treatment.

Methods The sample consisted of 26 children with early children schizophrenia who were treated with neuroleptics. The control group consisted of 26 healthy children. The correlations between QEEG absolute powers of delta (1.5-3.0 Hz), theta (3.0-7.5 Hz), alpha (7.5-12.5 Hz) and beta (12.5-20.0 Hz) frequency bands and attention parameters were investigated before and after neuroleptical therapy. To evaluate attention variables we used Burdon test, Stroop test and Schulte tables as well.

Results and conclusion

EEG in children with schizophrenia differed from control group in several amplitude-frequency characteristics. The patients displayed reduction of EEG amplitude in alpha-band, increase in the index of beta-1 and beta-2 activity and elevation of slow EEG activity. Beta-activity level correlates with the severity of disease and the impairment of the  executive control functions  and possibly reflects hyperexcitability of cortical networks in children with schizophrenia, which somewhat improves after antipsychotic treatment.

Background.

Our knowledge about children schizophrenia goes back to knowledge about schizophrenia in adult psychiatry. In the works of clinicians of 19-20 century, an extensive knowledge about children psychosis with hallucinations and abnormalities of behavior have been stored. In the 20th century, psychiatrists from Germany, Austria, Denmark, Switzerland and other European countries made a major contribution to studies of early children schizophrenia. They described different affective disorders and disorders of movement that were typical for this disease. However, all the history of children schizophrenia studies is related to discussion about existence of early children schizophrenia takes place. Most of psychiatrists consider children and adult schizophrenia as a common disease which can start at any age including the earliest one. In both cases, the main psychopathological symptoms, types development and outcomes are similar. The clinical pattern in children is also similar to clinical picture in adults.

Schizophrenic disorders are characterized by fundamental and typical dysfunctions of thinking and perception as well as inadequate or lowered affect. As a rule, consciousness and intellectual abilities remain safe though with the course of time there can  appear some cognitive dysfunctions. Schizophrenic disorders affect the main mental functions that provide a normal person with a sense of individuality. There can be auditory hallucinations in the patients. The perception is also disturbed: sounds and colors can seem too bright or important and different from usual way of seeing them, unimportant characteristics of things and situations can be felt more important than other things and situation in whole.

Thereby, thinking becomes fuzzy, interrupted and indistinct, and the speech can be incomprehensible. Interrupting or mixing of thoughts is also usual for such kind of patients and they have feeling of forced thoughts removal.

There are two different systems to diagnose schizophrenia. The first one is ICD-10 which is used in Russia, and the other is DSM- IV which is used in USA and Great Britain. Schizophrenia is not rare disease. According to foreign authors, in big cities the number of diseased people can be 8 or even 10 per 1000 habitants.

EEG dysfunctions in schizophrenia.

EEG is one of the functional and noninvasive methods of diagnostic. As early as in 1938 year, Lemer described EEG of patients with schizophrenia and noticed the decrease in alpha-band power.  Later, Itil and colleagues also described increase in high frequency activity and reduction of alpha-activity of EEG. Numerous studies attempted to correlate positive and negative symptoms of schizophrenia with amplitude-frequency characteristics of EEG. For example, Jons compared 28 patients with schizophrenia with normal subjects and found that positive symptoms correlated with increase in power of low alpha-band activity (8,5-10 Hz) while negative symptoms were connected with the increase in power of delta-band frequency diapason and decrease in power of high-frequency alpha-band activity (10,5-12 Hz) [John, 2009]. For positive symptoms, he noted that reduction of theta-band activity which along with alpha-band power reduction can serve as marker of accentuation of cognitive dysfunctions in schizophrenia.

Attention and its dysfunctions in schizophrenia.

Selection of information in the brain that is necessary for specific behavioral programs control is called attention [Luria, 2004]. Attention characterizes the dynamic of any psychic process, it guarantees the selectivity of simple and complicated mental activity.

It is known that there are two levels of attention: unconditioned and conditioned one. The first one is present from birth while the second one is formed during the development of other psychic functions and depends on social environment [Vigotsky, 1960]. Conditioned type of attention is tightly connected with speech system. Conditional type of attention is more vulnerable in schizophrenia unconditioned one. Even if it seems that a patient doesn’t pay any attention on environment and doesn’t answer any questions, it is capable of noticing the smallest details of scenery. Active attention is usually weak because of lack of interest to any activity [Cononova, 1963]. Many authors note that people with schizophrenia are unable to inhibit reactions to insignificant stimulus durin performance of task for separation of significant stimulus from insignificant one [Minas, Park, 2007]. It was also shown that patients with schizophrenia can perform such attention tests like Conners test and Stroop test worse than healthy subjects [Frangou, 2010]. In these tests there was no difference between children suffering from schizophrenia and adult patients.

Materials and methods.

There were studied 26 children at the age from 8 to 14 years suffering from schizophrenia. All of them were diagnosed by psychiatrists from Scientific Centre of Mental Health. According the ICD-10 they had diagnosis of children type of schizophrenia. All patients received drug treatment (haloperidol, triphtasin, aminasin, cyclodol, depakin). Attention and EEG- measures were evaluated before the beginning of therapy and 2 month after the therapy. The EEG data were compared with the corresponding data by normal subjects from database of Scientific Centre of Mental Health. Thirty seven healthy children at the age from 8 to 14 years served as control during measurements of attention . The table 1 shows the experimental design.

Methods.

All EEG studies were made with 16-chanel electroencephalograph, electrodes were located according to the international system 10:20, as referent electrode was located on ear. There were 16 active electrodes: F3, F4, F7, F8, C3, Cz, C4, P3, Pz P4, O1, O2, T3, T4, T5, T6. During EEG registration, children were sitting in blackout room. The frequency of digitization was 128 Hz. Bandwidth was from 0,5 to 30 Hz.  EEG recording was monopolar. Subjects were awake and quiet with eyes opened or closed. The computer processing of data was made with the help of Fourier analysis performed by brain mapping software «Brainsys». All artifacts were removed before the analysis. Statistic analysis was also made with the help of «Brainsys». Between-group comparison was made with help of built-in Z-statistics. Wilcoxon test was used for comparison of EEG from patients before and after treatment. The results are given as maps of p-significance (significance level – p<0.05). Same subjects we were evaluated in the following tests for attention:

1. Burdon test. Burdon test was suggested in 1895 by B. Burdon. It is used to diagnose the stability, to reveal degree sustained attention, its efficiency and stability. The test employs special form with numbers and figures given in rows in random order. There were figures composed of three, four and five dots and the subject had recognized and mark figures composed of four dots. Time of test and number of mistakes was analyzed.

2. Schulte tables are used for evaluation of stability and volume of attention as well as the working rate. Three different tables included numbers from 1 to 25 located in random order. The subject had to find all numbers in increasing order. The time of test was analyzed.

3. Stroop Test. The test is used to diagnose the cognitive style and to evaluate distribution of attention. Two different stimulus maps were used. The first one contained squares of different colors and the subject had to name all the color row by row. In the second table, the words were written with different colors and the subject had to name all the colors but not to read the words. Time of the test the number of mistakes were analyzed.

Between group comparison was made by the use of Student’s T-test.

Results

Attention tests

The result by the patients significantly differed from normal subjects. Figure 1 and 2 represent test time and number of mistakes in the attention tests before and after treatment, respectively. Both parameters are lowered in patients as compared with control as before as after treatment.

In figure 3, time and number of mistakes by patients before and after treatment are shown. No significant difference in the attention measure was found.

EEG studies of patients with schizophrenia.

Figure 4 show logarithm power EEG-spectrum in schizophrenic patients and control subjects before treatment. The power of delta-band is increased significantly in all points of recording, but most increase was seen in central and frontal areas. There was no significant decrease in alpha- and betha-1 activity.

Figure 5 represents power of EEG-spectrum in schizophrenic patients after treatment compared with control subjects. The power of low-band is increased significantly in all points of recording. There is also decrease in alpha-band power, primarily, in occipital and left temporal areas as well as decrease in betha-1 activity in the same areas.

Figure 6 presents maps of p-scores of EEG in schizophrenic patients before and after treatment. One can see that slow-band activity is increased in frontal and left temporal areas. There is also significant increase in theta-activity (5-7 Hz) in central, frontal and left temporal disposal. At the same time, the presence of alpha-activity (8-9 Hz) is decreased in temporal area increased in central, frontal and parietal areas. Betha-1 and betha-2 activities are decreased in all areas.

Figure 7 shows correlation of Stroop test time (first stimulus map) with the spectrum of logarithm EEG-power in group of patients before therapy. There is positive correlation between slow alpha-band power (8-9 Hz) in central, left frontal and temporal areas. Positive correlation with level of betha-1 activity can be seen in left parietal, temporal and frontal areas. Positive correlation with the level of betha-2 activity can be seen in right frontal and central areas.    

Figure 8 displays correlation of Stroop test time (first stimulus map) with EEG-power spectrum in group of patients after therapy. One can see positive correlation between test time and theta- and delta-band power. There is also positive correlation with level of alpha activity in occipital area and betha-1 activity also in occipital area. Negative correlation with betha-2 band power can be seen in left frontal area and the positive one is presented in occipital and temporal areas.

Discussion

The results revealed abnormalities of attention measures and aberrations of EEG in children with early schizophrenia as compared with healthy subjects. Before treatment the patients make more mistakes in all attention tests and spend more time performing test than healthy children. The worst scores were seen in  patients in the first stimulus table in Stroop test. It was more difficult for the subjects to name the color of the words then the color of squares. Stroop found that associative links between verbal stimulus and the reading reaction are more steady than between color stimulus and reaction of naming colors [Stroop, 1935]. Difficulties in the first stimulus table can show impossibility to form simple associations, that can indicate cognitive disorganization, brokenness of thoughts and their disorderliness. Data about violation of active attention in patients with schizophrenia, their inability to do tasks without outside stimulation, and weakness of attention concentration [Cononova, 1963] suggest that they have difficulties with the second stimulus table because they can’t differentiate more important for healthy people stimulus (the semantic meaning of the word) from less important (the color of the word), that can be confirmed with data from other studies [Minas, Park, 2007].

It is also known that unconditioned attention is closely related with verbal system. It was shown presence of deterioration of energetic metabolism in Wernicke’s area in patients with schizophrenia [Мartin-de-Souza et al, 2009]. Moreover some authors suppose that the dysfunction of NMDA-receptors in schizophrenia leads to disorder between parallel and non-parallel speech processes in brain [Strelnikov, 2010]. In the second stimulus table (Stroop 2) we used as a key stimulus the stimulus connected with speech. This fact can also explain worse results in this test in patients with schizophrenia. The results from comparison of the first and the second stimulus table in patients before and after therapy also confirm this suggestion showing that patients with schizophrenia perform the Stroop test with the second stimulus table worse than with the first stimulus table.

The results from comparison of time from Schulte tables test show clear decrease in stability and volume of attention in patients with schizophrenia. It also corresponds the results suggesting deterioration of conditioned attention in that type of patients. [Cononova, 1963].

Interestingly, there was neither increase in time of all attention tests nor decrease in the number of mistakes due to treatment neuroleptics. It seems that the treatment does not effect negative symptoms of schizophrenia such as an attention deficit. [Ushakov, 1973].

The descending reticular system, that connects frontal parts of brain with upper brain stem, plays an important role in control of selective attention.  Impairment of this system can be seen as alteration in amplitude-frequency characteristics of EEG: decrease in alpha-band power and presence of beta-activity [Itil, 1977]. Comparison of EEG from the patients before treatment with healthy subjects shows the increase in delta-band power in all brain areas and decrease in power and index of alpha-activity. Thalamus is a key structure for interaction between cortical and subcortical structures. In schizophrenic patients compared with healthy subjects thalamus has smaller size [Pakkenberg, 1990] striatic dophamine influence on thalamus through striatopallidothalamic system is stronger. [Kuperberg, Heckers, 2000].  The impairment of thalamo-cortical connections which takes place in that type of patients [Ghosh et al, 1990] can lead to reduction or alteration in time and frequency parameters of EEG alpha-band. Weakening of inhibitory influence of frontal cortex on subcortical structures can account for presence of slow EEG-activity, and results in attention and memory deviations [Kuperberg, Heckers, 2000]. The increasing of beta-2 activity can be caused by artifacts.

The decrease in beta-1 and beta-2 EEG-activity in patients after therapy can be caused by decrease in cortex excitation that clinically can be seen as decrease in psychotic symptoms in the patients. This fact is in accord with effect of neuroleptics on total brain activity [Marder, Van Putten, 1995] and behavioral changes. Besides, there was an increase in slow-band EEG-activity in frontal and temporal zones that can be related to lower activity of subcortical structures, disinhibition of nonspecific synchronizing thalamus system and, therefore, further imbalance in cortical-subcortical relationships.

We could also notice after the  therapy an increase in index of slow EEG alpha-band power (8-9 Hz) and high-frequency alpha-band power (10-11 Hz) that can be a result of improvement of functioning of frontal brain of the patients.  

Correlation between time of Stroop test (second stimulus map) and amplitude-frequency characteristics of EEG can also reflect relative ineffectiveness of neuroleptic therapy in regard to attention measures. Before treatment, there was positive correlation between the time variable and the betha-band EEG-activity, however, after treatment there was correlation with low activity level. Decline in level of conditioned attention revealed before treatment could be due to of elevated excitability of frontal brain and can explain presence of these correlations before therapy [Kuperberg, Heckers, 2000].

Conclusions

1. Schizophrenic children make significantly more mistakes in Stroop test 1 and spend much time in Schulte table test as that healthy subjects of the same age.
2. Neuroleptic therapy produced no improvement of attention parameters.
3. Neuroleptic therapy produced increase in low-band and alpha-band EEG-activity and decrease in betha-band EEG-activity.
4. There was significant positive correlation between the time Stroop test 1 performance and index of beta- EEG-activity in non-treated schizophrenic patients. After neuroleptic therapy there was positive correlation between the time in this test and index of low-band EEG-activity.

References

1. Cononova М. P: Handbook on phsycological research of  deseased school-going aged children. М. Gov. publishers of medical literature; 1963.
2. Frangou S: Cognitive function in early onset schizophrenia: a selective review. Frontiers in Human Neuroscience 2010, 3:1-4.
3. Ghosh A., Antonini A., McConnel S.K., Shatz C.J: Requirements for subplate neurons in the formation of thalamocortical connections. Nature 1990, 347:179-181.
4. Itil T.M: Qualitative and quantitative eeg findings in schizophrenia. Schizophrenia bulletin1977, 3:61-79.
5. John J. P: Fronto-temporal dysfunction in schizophrenia: A selective review. Indian J. Psychiatry 2009, 51:180-190.
6. Kuperberg G.,  Heckers S: Schizophrenia and cognitive function. Current opinion in Neurobiology 2000, 10:205-210.
7. Luria A.R: Psychological lectures. Spb.: Peter; 2004.
8. Marder S.R., Van Putten T: Antipsychotic medications. The American Psychiatric Press Textbook of Psychopharmacology, in press.
9. Martins-de-Souza D., Gattaz W.F., Schmitt A., Novello J.C., Turck C.W., Dias-Neto E: Proteome analysis of schizophrenia patients Wernicke’s area reveals an energy metabolism dysregulation. BMC Psychiatry 2009, 30:9-17.
10. Minas R.K., Park S: Attentional window in schizophrenia and schizotypal personality:Insight from negative priming studies. Appl Prev Psychol 2007, 12: 140-148.
11. Pakkenberg B: Pronounced reduction of total neuron number in mediodorsal thalamic nucleus and nucleus accumbens in schizophrenics. Arch Gen Psychiatry 1990, 47:1023-1028.
12. Strelnikov K: Schizophrenia and language – Shall we look for a deficit of deviance detection? Psychiatry Res 2010. [Epub ahead of print].
13. Stroop R.J: Studies of interference in serial verbal reaction.  Journal of Experimental Psychology 1935, 18:643-662.
14. Ushakov G.K: Children psychiatry. М.:Medicine; 1973
15. Vigotski L.S: The development of high psychic functions. М.: Pedagogica; 1960.

Legends to figures

Figure 1. Mean time (the upper panel) and number of mistakes (the lower panel) in the attention tests performed before treatment in schizophrenic patients in comparison with normal subjects. All the differences are statistically significant, p<0,05.

Figure 2. Mean time (the upper panel) and number of mistakes (the lower panel) in the attention tests performed after treatment in schizophrenic patients in comparison with normal subjects. All the differences are statistically significant, p<0,05.

Figure 3. Mean time (the upper panel) and number of mistakes (the lower panel) in the attention tests performed in schizophrenic before and after treatment. No significant differences were found.

Figure 4. Logarithm of power of EEG-spectrum in schizophrenic patients before treatment compared with control subjects (Z-statistics). Here and further: p<0,05; color means EEG-frequency (dark blue – delta, blue – theta, orange – alpha, red – betha-1, pink-betha-2); abscissa represents frequency in Hz, ordinate standard deviation.

Figure 5. Logarithm of power of EEG-spectrum in schizophrenic patients before treatment compared with control subjects (Z-statistics).

Рiс. 6. P-scores (0,5 to 30 Hz) comparing the group after treatment with the group before treatment. EEG data were compared with Wilcoxon criteria. Color codes value of р-score.

Figure. 7. Correlation between logarithm EEG-power autospectrum  and time of Stroop test 1 in group of patients before treatment. Here and below: color means frequency diapasons (dark blue – delta, blue – thetha, orange – alpha, red – betha-1 and pink betha-2); Abscissa represents frequencies (Hz), ordinate – correlation coefficient (p<0,05).

Figure 8. Correlation of logarithm of EEG-power autospectrum  and time of Stroop test 1 in group patients after treatment.

Figures

Figure 1

.

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Tables

Table 1. The experimental design.

Group	Age	Methods employed, in brackets – number of subjects
Patients before treatment	8-14	EEG (43), attention tests  (43)
Patients after treatment	8-14	EEG  (26), attention tests  (25)
Normal subjects	8-14	Attention tests  (37)