Institute of Behavioural Sciences Semmelweis University
Nagyvárad tér 4., H-1089 Budapest Hungary
E-mail address: bodrob@net.sote.hu (Róbert Bódizs)

Beside the well-known positive symptoms patients suffering from schizophrenia are characterized by cognitive impairments and neuropsychological deficits as well as various forms of sleep disturbances. In volume 38 Robert Göder and colleagues (2004, pp. 591–99) reported a correlation between visuospatial memory impairments and the reduction in the amount of slow- wave sleep (SWS) and in sleep efficiency. Visuospatial memory performance and the amount of SWS did not correlate in healthy controls. This is the first study investigating the interrelationship between specific measures of night sleep and performance in different aspects of memory in individuals with schizophrenia and in healthy matched controls. Authors conclude that there is a functional interrelationship between the regulation of SWS and performance in visuospatial memory in schizophrenia. Without questioning this general conclusion, we disagree with the discussion part of the Göder et al. (2004) paper which, according to our view, is biased in favour of the sleep-related memory consolidation theories.

Correlations cannot delineate causal relationships and without manipulating sleep variables there is no scientific proof for the effects of sleep on visuospatial memory performance assessed by the Rey-Osterrieth Complex Figure Test. As there is no data on evening recall performance in this study, it is not clear whether memory consolidation or encoding was decreased in schizophrenic patients in comparison to controls. Moreover, if time spent in SWS would be a measure of sleep-related memory consolidation, then a lack of a correlation between SWS time and the Rey- Osterrieth Complex Figure Test performance in non- schizophrenic persons is peculiar. The presumption that a critical basic amount of SWS (available in non-schizophrenic persons) is needed for consolidation of visuospatial memory (p. 597) is very poorly founded. Göder et al. (2004) report a considerable overlap between SWS time of schizophrenic patients and normal controls (0–83 and 21–42 min, respectively). What could be the critical basic amount of SWS for visuospatial memory consolidation in what age groups and for what performance level? The pre- sumption of the critical basic amount of SWS is further complicated by the overlap in the morning visuospatial recall performance in the two groups: 6–29 and 13–34 points for schizophrenic and normal control subjects, respectively. Authors mention the correlations between gray matter decrease and neuropsychological dysfunc- tions in the introduction (Salgado-Pineda et al., 2003). The association between third ventricle enlarge- ment and neuropsychological deficit in schizophrenic patients is a similar observation (Bornstein et al., 1992). Taken into account that ventricle-to-brain ratio and SWS time are inversely related in schizophrenic patients (Benson et al., 1996), the correlation between the reduction in SWS and impairment in visuospatial memory could be a trait-like relationship, based on structural brain features, independent of the memory consolidating effects of night-time sleep. The finding that SWS time and ventricular brain ratio was unre- lated in drug-naive schizophrenic patients (Lauer et al., 1997), do not justify the complete neglect of this line of reasoning, because patients in the Göder et al. (2004) study were on stable antipsychotic medication.

We have already reported a positive correlation of the parahippocampal and temporolateral sleep slow oscillation with the Rey-Osterrieth Complex Figure Test performance in epileptic patients. Recall was tested at 5 and 30 min after initial copy of the figure without intervening sleep. Moreover, sleep recordings were done days or weeks after memory testing (Bódizs et al., 2002). This result could suggest a general, trait-like correlation between visuospatial memory performance and the sleep slow oscillation or SWS, at least in patient populations affected by certain neuropsychiatric disorders. Schizophrenia is characterized by decreased amplitude of the so-called slow-wave activity (0.5–4 Hz) during non-REM sleep (Hoffman et al., 2000). Decreased amplitude of the slow- wave activity during non-REMsleep is an EEGfeature of reduced SWS time. Moreover, this slow-wave activity covers the slow oscillation (<1 Hz), which was shown to correlate with Rey-Osterrieth Complex Figure Test per- formance in epileptic patients. The slow oscillation was shown to be associatedwith other cognitive performances in healthy older human subjects (Anderson and Horne, 2003) and with synaptic transmission efficacy in animal models (Amzica and Steriade, 1995). We think there is sufficient evidence to suggest a trait-like (morphologically based) correlation between slow-wave sleep and visuospatial memory performance in schizophrenic patients. In spite of these relevant scientific evidences Göder et al. (2004) base their interpretation exclusively on papers and theories of sleep-related memory consolidation, ignoring all other information. Of course there is no direct evidence for reduced brain volumes or sleep slow oscillation measures in their data, but there is also no proof for the effect of sleep on memory consolidation by using the correlational approach. We do not question the possible relevance of sleep-relatedmemory consolidation deficits in schizophrenia, but without testing daytime memory performance there is no proof for such a theory. Indeed these results fit very well with the stable interrelationship between neuropsychological performance, SWS and their common determinant factor, brain morphology. Of course replacing the interpretation of the correlational data of Göder et al. (2004) with our proposed correlational interpretation alternative does not necessarily increase the power of the findings. However, we think that the deliberate interpretation of a correlational data- set is based on a review of all possible relationships between the variables. The popularity of sleep-related memory consolidation theories may lead researchers to force their papers in the Procust bed of the according views, neglecting some sufficiently supported and maybe more evident interpretations.

References

Amzica F, Steriade M. Disconnection of intracortical synaptic linkages disrupts synchronization of a slo
w oscillation. Journal of Neuroscience 1995;15:4658–77.

Anderson C, Horne JA. Prefrontal cortex: links between low frequency delta EEG in sleep and neuropsychological performance in healthy, older people. Psychophysiology 2003;40: 349–57.

Benson KL, Sullivan EV, Lim KO, Lauriello J, Zarcone Jr VP, Pfefferbaum A. Slow wave sleep and computed tomographic measures of brain morphology in schizophrenia. Psychiatry Research 1996;60:125–34.

Bódizs R, Békésy M, Szűcs A, Barsi P, Halász P. Sleep-dependent hippocampal slow activity correlates with waking memory performance in humans. Neurobiology of Learning and Memory 2002;78:441–57.

Bornstein RA, Schwarzkopf SB, Olson SC, Nasrallah HA. Third- ventricle enlargement and neuropsychological deficit in schizophrenia. Biological Psychiatry 1992;31:954–61.

Go ¨der R, BoigsM, Braun S, Friege L, Fritzer G, Aldenhoff JB, Hinze- Selch D. Impairment of visuospatial memory is associated with decreased slow wave sleep in schizophrenia. Journal of Psychiatric Research 2004;38:591–9.

Hoffman R, Hendrickse W, Rush AJ, Armitage R. Slow-wave activity during non-REM sleep in men with schizophrenia and major depressive disorders. Psychiatry Research 2000;95: 215–25.

Lauer CJ, Schreiber W, Pollmacher T, Holsboer F, Krieg JC. Sleep in schizophrenia: a polysomnographic study on drug-naive patients. Neuropsychopharmacology 1997;16: 51–60.

Salgado-Pineda P, Baeza I, Perez-Gomez M, Vendrell P, Junque C, Bargallo N, Bernardo M. Sustained attention impairment correlates to gray matter decrease in first episode neuroleptic-naive schizophrenic patients. Neuroimage 2003;19:365–75.