Schizophrenia Research
Volume 100, Issue 1 , Pages 153-160, March 2008

Assessment of single and bound features in a working memory task in schizophrenia

  • David Luck

      Affiliations

    • INSERM U. 666, Physiopathologie clinique et expérimentale de la schizophrénie, Département de Psychiatrie, Hôpital Civil, 67091 Strasbourg Cedex, France
    • Corresponding Author InformationCorresponding author. Present address: Brain Imaging Group, Douglas Hospital Research Centre, McGill University, 6875 Boul. LaSalle, Verdun, Canada, H4H 1R3. Tel.: +1 514 761 6131x3389 (office); fax: +1 514 888 4064.
    • ,
  • Jack R. Foucher

      Affiliations

    • INSERM U. 666, Physiopathologie clinique et expérimentale de la schizophrénie, Département de Psychiatrie, Hôpital Civil, 67091 Strasbourg Cedex, France
    • ,
  • Isabelle Offerlin-Meyer

      Affiliations

    • INSERM U. 666, Physiopathologie clinique et expérimentale de la schizophrénie, Département de Psychiatrie, Hôpital Civil, 67091 Strasbourg Cedex, France
    • ,
  • Martin Lepage

      Affiliations

    • Brain Imaging Group, Douglas Hospital Research Centre, Verdun, Quebec, Canada
    • ,
  • Jean-Marie Danion

      Affiliations

    • INSERM U. 666, Physiopathologie clinique et expérimentale de la schizophrénie, Département de Psychiatrie, Hôpital Civil, 67091 Strasbourg Cedex, France

Received 27 August 2007; received in revised form 30 October 2007; accepted 5 November 2007. published online 19 December 2007.

Article Outline

Abstract 

If disturbance of binding in long term memory is well established in schizophrenia, data concerning working memory maintenance are less clear. Feature binding in working memory was investigated in 19 patients with schizophrenia and 19 healthy controls. Binding was assessed by comparing two conditions in which participants had to retain four letters and four spatial locations. These features were presented either bound or separate. Results showed that both groups had better performances for bound than separate features, despite the fact that patients performed significantly worse than controls. When maintenance for isolated features was assessed, patients were severely disturbed for spatial locations but not for letters. Such a result suggests that reduced working memory performance in patients with schizophrenia for bound features is probably a consequence of a spatial deficit rather than a specific deficit of the binding process. Thus, not all form of binding are disturbed in schizophrenia.

Keywords: Binding, Schizophrenia, Spatial, Verbal, Working memory

 

Back to Article Outline

1. Introduction 

Schizophrenia is associated with a wide range of cognitive dysfunctions. One of the most pronounced concerns episodic memory (Aleman et al., 1999, Heinrichs and Zakzanis, 1998, Pelletier et al., 2005, Saykin et al., 1991). The representation of episodes includes perceptive dimensions of physical objects, as well as the time and place they occurred. However, information are not retained separately in memory, but are stored as a coherent whole. This type of associative process is called binding. Several studies have described impairments of associative recognition in episodic memory in patients with schizophrenia (Danion et al., 1999, Lepage et al., 2006, Rizzo et al., 1996, Waters et al., 2004). Danion and colleagues suggested that these deficits may result from inefficient binding processes in working memory (Stone et al., 1998), which in turn disturb most integrated cognitive functions.

Working memory (WM) is defined as a limited capacity system devoted to temporary storage and on-line manipulation of information (Baddeley, 1986, Goldman-Rakic, 1996). Baddeley, 2000, Baddeley, 2003) had recently added a new component to this tripartite WM model. This new component, the episodic buffer, is proposed to be an integrating storage system for temporary representations being entered or retrieved from episodic memory. Episodes are achieved by integrating single items into chunks, and chunks into episodes (Rossi-Arnaud et al., 2005). Although deficits in WM are well documented in individuals with schizophrenia (see Keefe, 2000 for a review), little is known about the maintenance of bound items. Recent investigations of this issue have produced discrepant results. For instance, Gold et al. (2003) examined retention performance for coloured oriented bars. Participants were instructed to focus on either the colours or the orientations (single-feature conditions), or on both features (binding condition). Patients with schizophrenia showed no greater impairment for the binding condition compared to single-feature conditions, leading the authors to conclude that patients are able to bind features into integrated object representations. By contrast, Burglen et al. (2004) successively presented three familiar object drawings located in a 3×3 grid. In different blocks of trials, participants had to memorize either objects or locations, or their combination. They found that the rate of combination errors exceeded the rate of feature errors in patients, but was equal in controls. They thus argued that these results may highlight a specific binding deficit for WM storage or a non specific effect of memory load in patients with schizophrenia (Mitchell et al., 2000). In both of these studies, binding was assessed by comparing single-feature conditions to a binding condition. A potential limitation was that subjects had to memorize and evaluate twice as much information in the binding condition than is required for a single-feature condition. To address this issue, Prabhakaran et al. (2000) developed a procedure where binding is evaluated by comparing two conditions comprised of an equal number of features, but differing in the presence or absence of a link between them. Items were composed of letters and spatial locations, presented either bound or separate. In both conditions, probes consisted of a letter presented into a location. Subjects were instructed to respond affirmatively if both items were familiar, regardless of their initial combination in the bound condition. Thus, probe items are either congruent or incongruent with the target display. Prabhakaran et al. found that normal subjects showed better performance for bound than separate items and for congruent than incongruent positive probes. They concluded that this benefit results from the integration in WM of verbal and spatial information.

The aim of the present study was to investigate the WM storage of bound features in schizophrenia, using an advanced procedure from which straightforward conclusions could be drawn. To this end, we adapted the procedure of Prabhakaran et al. (2000) for use with our patient population. Given the Prabhakaran et al. findings, we expected healthy controls to better perform bound items relative to separate items. Consistent with the binding deficit hypothesis in schizophrenia, we also anticipated that patients would exhibit equal performance for bound and separate items. In other words, a significant group by binding conditions interaction should be observed, whereas an absence of this type of interaction would be interpreted as intact binding processes in patients with schizophrenia.

Back to Article Outline

2. Materials and methods 

2.1. Subjects 

Demographic and clinical data are summarized in Table 1. Nineteen outpatients (14 male and five female) participated in the study. All of them met DSM-IV criteria for schizophrenia (4th edition; DSM–IV; American Psychiatric Association, 1994), as determined by consensus of the current treating psychiatrist and senior psychiatrists belonging to the research team (disorganized, n=2; paranoid, n=16; residual, n=1). Symptom severity was determined using the Positive and Negative Syndrome Scale for Schizophrenia (PANSS). Patients were clinically stabilized for at least one month at the time of testing. All patients, except two, were treated with antipsychotic drugs. Ten patients were receiving typical antipsychotics (flupenthixol, haloperidol, levomepromazine, zuclopenthixol) and seven atypical antipsychotics (amisulpride, carpipramine, olanzapine, risperdone).

Table 1. Groups' demographic and clinical data
Patients (n=19)Controls (n=19)
Age (years)35.4734.89
(2.40)(6.38)
Education level (years)11.3711.37
(0.50)(0.17)

PANSS
Positive20.47
(1.80)
Negative24.58
(1.41)
General49.16
(3.35)
Duration of illness (years)11.50
(2.02)
Duration of treatment (years)5.75
(1.24)

CPZ/eq (mg/day)
Typicals142.87
(89.62)
Atypicals92.87
(41.78)

Nineteen healthy subjects were matched to patients for age, education duration and gender. They were recruited through local advertisements. The groups did not differ in age and years of education (t36<1 in both case). They differed in IQ, as measured using a short form of the Wechsler Adult Intelligence Scale—Revised (Crawford et al., 1992) (patients: mean=83.00, S.D.=3.11; controls: mean=94.05, S.D.=2.23) (t36=2.81; p<0.001).

No participants had a history of traumatic brain injury, epilepsy, alcoholism or substance abuse, other diagnosable neurological conditions or organic mental disorder, nor were being treated with antidepressants, benzodiazepines or lithium. Additionally, controls were excluded if they reported past or present psychiatric disorder, as well as if a first-degree relative had sought psychiatric diagnosis or treatment.

The Strasbourg Consultative Committee for the Protection of Human Subjects in Biomedical Research approved the study. Each participant signed an informed consent form prior to the experiment and received financial compensation for its participation.

2.2. Apparatus and stimuli 

White coloured stimuli were displayed on a black background on a 17” LCD screen of a PC running an E-Prime software (v. 1.1, Psychology Software Tools Inc.) program. The target display was composed of four letters and/ or four spatial locations according to the experimental condition. Letters were consonants (“W” excluded) and were shown in upper-case (36; Times New Roman). Spatial locations were enclosed by parentheses (36; Times New Roman) and randomly presented among 12 different locations displayed on a virtual centred ellipse. The single probe consisted of a single lower-case letter and/or a single location enclosed by braces. Letters and locations were shown in a different case and shape in order to control for a possible priming effect.

2.3. Design 

A graphical representation of the procedure is presented in Fig. 1, Fig. 2. Each trial started with the presentation of a white central cross (1000 ms), followed by a target display of items (4000 ms). In each binding condition, the target display consisted of four letters and four spatial locations. In the bound condition (B+), letters were included within locations, whereas in the separate condition (B−) letters were centrally and separated from locations. After display offset, participants had to maintain items over a retention interval (5000 ms) during which only the central cross was presented. The central cross became green 500 ms before the probe presentation. In both conditions, the probe consisted of a single letter included in a location. Targets were previously seen items. Thus, in B+, probes could be congruent with the display if the same letter was presented in the same location, or incongruent if a letter was presented in a location occupied by another letter. In both cases, participants were required to answer affirmatively. Lures were equally balanced between letters, locations, or both and none of these belonged to the display. Each condition was comprised of an equal number of targets and lures in a pseudo-random order. Responses were collected by pressing on pre-assigned buttons, according to their laterality. Right-handed subjects responded affirmatively by pressing the right button with their right index and negatively with their left index on the left button. For left-handed subjects, buttons were inversed. Participants were instructed to respond as quickly as possible without sacrificing accuracy. Participants had 3000 ms to answer. After an unfilled delay of 2000 ms, a new trial began.

Two baseline single-feature conditions were added in order to assess capacities in processing only verbal or spatial items. In the Letters (L) condition, subjects had to decide if the single probe letter belonged to four centrally presented letters in the target display or not. The spatial locations (S) condition was identical to L, except that only locations were shown, instead of letters. Both conditions were comprised of an equal number of targets and lures in a pseudo-random order.

2.4. Procedure 

The WM task was divided into two sessions, separated by a minimum of one day and a maximum of one week. Each session started with a detailed description of the task and the instructions associated with each condition. This was followed by an interactive demonstration using three trials per condition, in order to check participant’s understanding of the whole procedure. Then, a short training session was given, in order to familiarize participants with the experimental task. The four conditions were pseudo-randomized, each training block starting with one of the two single-feature conditions.

Each WM session consisted of 120 trials arranged in four blocked conditions of 30 trials. Before each block, instructions were presented. The order of conditions was pseudo-randomized according to the same rules as the training session. Once completed, participants were given the opportunity to comment and ask questions about the task, and then were debriefed.

2.5. Data analysis 

Test data were analyzed using Statistica 6.0 (Statsoft). Accuracy scores and response times were used as dependent measures. Three omnibus analyses of variance (ANOVA) were conducted, with group (controls vs. patients) entered as a between-group factor. Binding conditions (B− vs. B+), congruence (B+ congruent vs. incongruent positive probes), and single-feature conditions (L vs. S) were used as within-group factors. Analyses of covariance with IQ as a covariate were also conducted. Because the two series of analyses yielded the same results, only the results of ANOVAs were presented.

When needed, Honestly Significantly Different Tukey test (HSD) comparisons were performed for post hoc analyses. In all analyses, the alpha level was set at 0.05.

Back to Article Outline

3. Results 

The mean proportions and response times of correct responses for binding conditions, congruence and single-feature conditions are summarized in Table 2, Table 3, Table 4 respectively.

Table 2. Performance (proportions and response times) as a function of binding conditions (separate, B−; bound, B+) in patients with schizophrenia and healthy controls
B–B+
MeanSEMRangeMeanSEMRange
Mean proportions of correct responses
Patients with schizophrenia0.750.030.56–0.970.790.020.59–0.97
Healthy controls0.870.020.75–0.970.890.010.76–1.00

Mean response times (ms)
Patients with schizophrenia1891.8596.221000.57–2905.761838.2796.231082.72–2851.07
Healthy controls1534.1869.65998.24–2069.601500.8557.641007.52–1858.94
Table 3. Performance (proportions and response times) as a function of B+ positive probes (congruent and incongruent) in patients with schizophrenia and healthy controls
B+ congruentB+ incongruent
MeanSEMRangeMeanSEMRange
Mean proportion of correct responses
Patients with schizophrenia0.810.040.53–1.000.690.040.40–1.00
Healthy controls0.920.020.73–1.000.840.020.71–1.00

Mean response times (ms)
Patients with schizophrenia1869.0687.691158.27–2601.252099.46145.441047.07–3474.67
Healthy controls1472.6163.51837.36–1934.581717.8576.281032.42–2260.33
Table 4. Performance (proportions and response times) as a function of single-feature conditions (letters, L; spatial locations, S) in patients with schizophrenia and healthy controls
LS
MeanSEMRangeMeanSEMRange
Mean proportion of correct responses
Patients with schizophrenia0.940.010.85–1.000.720.030.52–0.97
Healthy controls0.980.0040.95–1.000.880.020.70–1.00

Mean response times (ms)
Patients with schizophrenia1442.3571.16706.77–2108.631542.6171.16809.50–2269.50
Healthy controls1046.2552.06692.60–1519.361296.0375.74836.05–2098.86

3.1. Binding conditions 

Analyses of rates showed a main effect of group (F(1,36)=16.97; p<0.001), with patients’ scores being lower than those of controls. There was a significant effect of binding conditions (F(1,36)=4.33; p=0.05), with overall performance being higher in B+ than in B−. But there was no significant group×binding conditions interaction (F(1,36)=0.76; p=0.39). With respect to responses times, patients globally took more time to respond than controls (F(1,36)=9.43; p=0.004). Both groups tended to be faster for bound than for separate items (F(1,36)=3.54; p=0.07). Again, the interaction was not significant (F(1,36)=0.19; p=0.66).

3.2. Congruence 

The analysis of proportion of hits in B+ indicated that patients were globally less accurate than controls (F(1.36)=9.90; p=0.003). Both groups were more accurate for the congruent than for the incongruent positive probes (F(1,36)=28.96; p<0.001). The congruence X group interaction failed to reach significance (F(1.36)=1.90; p=0.18). Similarly, the analysis of response time showed that patients took significantly more time to respond than controls (F(1.36)=8.71; p=0.006). Both groups responded faster for the congruent than for the incongruent positive probes (F(1.36)=28.66; p<0.001), with no significant group X congruence interaction (F(1,36)=0.03; p=0.87).

3.3. Single-feature conditions 

The ANOVA based on rates showed significantly lower rates for patients than for controls (F(1,36)=23.42; p<0.001). Both groups showed better performance in L than in S (F(1,36)=74.48; p<0.001). The group X single-feature conditions interaction was significant (F(1,36)=11.34; p=0.002). Subsequent HSD tests indicated that both groups performed better for letters than locations (p<0.002 in both cases). The two groups did not differ in L (p=0.55), but patients showed lower rates than controls in S (p<0.001). Analyses of response times indicated that patients took significantly more time to respond than controls (F(1,36)=12,28; p=0.001). Response times were shorter in L than in S in both groups (F(1,36)=14.76, p<0.001). The group and the single-feature conditions factors did not interact significantly (F(1,36)=2.69; p=0.11).

3.4. Complementary analyses 

In order to establish if the low performance obtained for binding conditions reflected spatial deficits, we then used a median split on the overall spatial performance in the condition S. We divided the group of patients into two subgroups: a high-performance group (n=9), and a low-performance group (n=10). Neither subgroup differed in sociodemographic, in clinical data, or in IQ (t<1 in all cases). We focused our analyses on correct response proportions. The group (controls, high-performance patients, low-performance patients) X binding conditions ANOVA indicated a significant effect of group (F(2,35)=23.89; p<0.001), with low-performance patients being lower than both controls and high-performance patients (p<0.001 in both cases). Controls and high-performance patients did not significantly differ from each other (p=0.36). The effect of conditions reached significance (F(1,35)=4.85; p=0.03), but both factors did not interact (F(2,35)=0.44; p=0.65). The group×congruence analysis showed a significant effect of group (F(2,35)=10.38; p<0.001). Subsequent HSD tests revealed that controls and high-performance patients did not differ from each other (p=0.57), but were in turn higher than low-performance patients (p<0.02 in both cases). The significant effect of the congruence ((F(1,35)=30.30; p<0.001), as well as the absence of a group X congruence interaction (F(2,35)=1.07; p=0.35) indicated that all groups performed better for congruent than incongruent positive probes.

Back to Article Outline

4. Discussion 

The study yielded three main results. First, performance was lower for separate than for bound items, and participants better matched congruent than incongruent positive probes. Second, patients with schizophrenia exhibited overall lower performance compared to controls. Third, patients showed a selective deficit for isolated spatial cues, but not for isolated verbal cues.

Participants were more accurate and tended to be faster when verbal and spatial information were bound than separate. Such a benefit was observed in others studies using a similar procedure (Campo et al., 2005, Prabhakaran et al., 2000, Wu et al., 2007). Wu et al. (2007) suggested that this benefit results from a lower task difficulty for bound items. Thus, the binding processing leads to a decreased number of objects in the bound condition (four objects in the bound condition while eight objects in the separate condition), despite an equal amount of information in both the bound and the separate conditions (four letters and four locations). On the other hand, Campo et al. (2005) proposed that the benefit for bound items results in less interfering between the processing of verbal and spatial information, which in turn facilitates the processing of both information. For bound items, participants compared probe information better and faster when probes were congruent with those previously displayed. They were however less accurate and slower when probe information was incongruent with the target display. This last result suggests that participants had greater difficulty recognizing features when their combination changed. The features were automatically bound and maintained, even if participants were instructed to monitor only features. As a whole, and consistent with Prabhakaran et al. data, our results confirmed the hypothesis that verbal and spatial information are held in the bound condition in an integrated fashion in WM.

All these analyses revealed lower performance in patients with schizophrenia. Moreover, the group factor never interacted with within-group factors, such as binding conditions and congruence. This result suggests that patients with schizophrenia have preserved binding capacities. Indeed, Carr et al. (1998) reported intact capacities for feature binding in perception and these findings were extended to visual short term memory by Gold et al. (2003). Our experiment both confirmed this result by showing a non specific deficit for bound information and extended it to other kinds of information, with longer retention delays.

Our results demonstrated intact verbal performance, but impaired spatial processing in patients with schizophrenia. Such a deficit has been widely described in the literature (Fleming et al., 1997, Park and Holzman, 1992, Park and Holzman, 1993, Park et al., 1999, Spindler et al., 1997, Tek et al., 2002), even if little is known about the underlying mechanism of this impairment. One explanation relies on a deficit occurring during perceptual processing. Some authors considered that perceptual/ encoding but not necessarily retention impairments are the feature core of deficit in WM (Hartman et al., 2002, Javitt et al., 2007, Tek et al., 2002; see Lee and Park, 2005 for a meta-analysis). For instance, Hartman et al. (2002) showed that patients need five more time to encode information in WM to reach similar performance than controls. When the encoding time is adapted, patients did not exhibit a deficit during delay condition. On the other hand, Tek et al. (2002) showed that a spatial WM impairment was observed even when between-group differences in perceptual processing were controlled, suggesting a deficit of the retention of spatial information. According to them, spatial WM impairment in schizophrenia is multifactorial. Here, this selective impairment of spatial processing could explain patients’ overall lower performance in the binding conditions. In order to verify this hypothesis, the group of patients was divided into a median-split, according to their capacities of processing isolated spatial cues. If high-performance patients did not differ from controls, low-performance patients were in turn less accurate. In that way, performance decrements for patients in the binding conditions were probably due to a difficulty in processing spatial information, rather than a deficit in processing multiple feature information.

A few considerations should be taken into account when interpreting our results. First, it remains possible that our procedure was not sensitive enough to highlight subtle alterations, such as binding deficits, in patients with schizophrenia. Here, participants had to evaluate the items regardless their initial combination in the bound condition. It remains possible that a task in which participants have to explicitly consider the link between items would be more constraining for patients, and thus more able to reveal a deficit of binding. Second, the difference observed with Burglen et al. (2004) may also be attributed to the type of stimuli being presented (letters vs. objects drawings). Processing objects drawings needs additional processes such as object recognition and semantic associations, all of which may contribute to an encoding processing “overload” when combined with spatial processing in the combination task. In the current task, participants simply need to encode letters/locations. As a safe conclusion, it can be stated that feature binding for letters/location is not impaired.

To conclude, previously mentioned associative impairments in episodic memory were not manifested in this binding WM task. We also found evidence for preserved performance when patients with schizophrenia maintained bound information over a short time, suggesting a dissociation between integrative processes. Episodes are supposed to be achieved by integrating items into chunks, and chunks into episodes (Rossi-Arnaud et al., 2005). It may also be suggested that the first step of integration into chunks is preserved in schizophrenia and that deficits occur at the next step, when chunks are integrated into episodes. Future research should thus dissociate binding processes specific to each form of memory.

Back to Article Outline

Role of the funding source 

Funding for this study was provided by Ministère de la Recherche et des Nouvelles Technologies (MNRT) grant; David Luck was supported by a salary award from the Fondation pour la Recherche (FRM); the FRM and MNRT had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.

Back to Article Outline

Contributors 

Authors Luck, Foucher, Offerlin-Meyer, Lepage and Danion all contributed to drafting the manuscript. Author Danion was both principal investigator on the grants and project director of the study. Author Luck designed the experimental protocol. Authors Luck and Foucher were involved in experiment programming and undertook statistical analyses. Authors Luck and Offerlin-Meyer were involved in data collection and data management. Author Offerlin-Meyer performed IQ assessments. All authors have contributed to and have approved the final manuscript.

Back to Article Outline

Conflict of interest 

All authors declare that they have no conflicts of interest.

Back to Article Outline

Acknowledgements 

This study was supported by operating grants from the Ministère de la Recherche et des Nouvelles Technologies. David Luck was supported by a salary award from the Fondation pour la Recherche Médicale. We thank the CAT Route nouvelle staff and the Pr. Pierre Vidailhet for their help with recruitment and clinical assessments.

Back to Article Outline

References 

  1. Aleman A, Hijman R, de Haan EH, Kahn RS. Memory impairment in schizophrenia: a meta-analysis. Am. J. Psychiatry. 1999;156:1358–1366
  2. American Psychiatric Association . In: Diagnostic and Statistical Manual of Mental Disorders, DSM-IV 4. 1994;p. 273–290
  3. Baddeley AD. Working Memory. Oxford: Oxford University Press; 1986;
  4. Baddeley AD. The episodic buffer: a new component of working memory?. Trends Cogn. Sci. 2000;4:417–423
  5. Baddeley AD. Working memory: looking back and looking forward.. Nat. Rev. Neurosci. 2003;4:829–839
  6. Burglen F, Marczewski P, Mitchell KJ, Van Der Linden M, Johnson MK, Danion J-M, et al. Impaired performance in a working memory binding task in patients with schizophrenia. Psychiatry Res. 2004;125:247–255
  7. Campo P, Maestu F, Ortiz T, Capilla A, Santiuste M, Fernandez A, et al. Time modulated prefrontal and parietal activity during the maintenance of integrated information as revealed by magnetoencephalography. Cereb. Cortex. 2005;15(2):123–130
  8. Carr VJ, Dewis SA, Lewin TJ. Illusory conjunctions and perceptual grouping in a visual search task in schizophrenia. Psychiatry Res. 1998;80:69–81
  9. Crawford JR, Allan KM, Jack AM. Short form of the UK WAIS-R: regression equation and their predictive validity in a general population sample. Br. J. Clin. Psychol. 1992;31:191–202
  10. Danion J-M, Rizzo L, Bruant A. Functional mechanisms underlying impaired recognition memory and conscious awareness in patients with schizophrenia. Arch. Gen. Psychiatry. 1999;56:639–644
  11. Gold JM, Wilk CM, McMahon RP, Buchanan RW, Luck SJ. Working Memory for Visual Features and Conjunctions in Schizophrenia. J. Abnorm. Psychology. 2003;112:61–71
  12. Goldman-Rakic PS. Regional and cellular fractionation of working memory. Proc. Natl. Acad. Sci. U. S. A. 1996;93:13473–13480
  13. Hartman M, Steketee MC, Silva S, Lanning K, McCann H. Working memory and schizophrenia: evidence for slowed encoding. Schizophr. Res. 2002;59:99–113
  14. Fleming K, Goldberg TE, Binks S, Randolph C, Gold JM, Weinberger DR. Visuospatial working memory in patients with schizophrenia. Biol. Psychiatry. 1997;41:43–49
  15. Heinrichs RW, Zakzanis KK. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology. 1998;12:426–445
  16. Javitt DC, Rabinowicz E, Silipo G, Dias EC. Encoding vs. retention: differential effects of cue manipulation on working memory performance in schizophrenia. Schizophr. Res. 2007;91(1-3):159–168
  17. Keefe RS. Working memory dysfunction and its relevance to schizophrenia. In:  Sharma T,  Harvey PD editor. Cognition in schizophrenia: Impairments, importance, and treatment strategies. Oxford, England: Oxford University Press; 2000;p. 16–50
  18. Lee J, Park S. Working Memory Impairments in Schizophrenia: a meta-analysis. J. Abnorm. Psychology. 2005;114(4):599–611
  19. Lepage M, Montoya A, Pelletier M, Achim AM, Menear M, Lal S. Associative memory encoding and recognition in schizophrenia: an event-related fMRI study. Biol. Psychiatry. 2006;1215–1223
  20. Mitchell KJ, Johnson MK, Raye CL, Mather M, D’Esposito M. Aging and reflective processes of working memory: binding and test load deficits. Psychol. Aging. 2000;15:527–541
  21. Park S, Holzman PS. Schizophrenics show spatial working memory deficits. Arch. Gen. Psychiatry. 1992;49:975–982
  22. Park S, Holzman PS. Association of working memory deficit and eye tracking dysfunction in schizophrenia. Schizophr. Res. 1993;11:55–61
  23. Park S, Puschel J, Sauter BH, Rentsch M, Hell D. Spatial working memory deficits and clinical symptoms in schizophrenia: a 4-month follow-up study. Biol. Psychiatry. 1999;46:392–400
  24. Pelletier M, Achim AM, Montoya A, Lal S, Lepage M. Cognitive and clinical moderators of recognition memory in schizophrenia: a meta-analysis. Schizophr. Res. 2005;74:233–252
  25. Prabhakaran V, Narayanan K, Zhao Z, Gabrieli JDE. Integration of diverse information in working memory within the frontal lobe. Nat. Neurosci. 2000;3:85–90
  26. Rizzo L, Danion J-M, Van der Linden M, Grange D. Patients with schizophrenia remember that an event has occurred, but not when. Br. J. Psychiatry. 1996;168:427–431
  27. Rossi-Arnaud C, Pieroni L, Baddeley A. Symmetry and binding in visuo-spatial working memory. Neuroscience. 2005;139:393–400
  28. Saykin AJ, Gur RC, Gur RE, Mozley PD, Mozley LH, Resnik SM, et al. Neuropsychological function in schizophrenia: selective impairment in memory and learning. Arch. Gen. Psychiatry. 1991;48:618–624
  29. Spindler KA, Sullivan EV, Menon V, Lim KO, Pfefferbaum A. Deficits in multiple systems of working memory in schizophrenia. Schizophr. Res. 1997;27:1–10
  30. Stone M, Gabrieli JDE, Stebbins GT, Sullivan EV. Working and strategic memory deficits in schizophrenia.. Neuropsychology. 1998;12:278–288
  31. Tek C, Gold J, Blaxton T, Wilk C, McMahon RP, Buchanan RW. Visual perceptual and working memory impairments in schizophrenia. Arch. Gen. Psychiatry. 2002;59:146–153
  32. Waters FA, Maybery MT, Badcock JC, Michie PT. Context memory and binding in schizophrenia. Schizophr. Res. 2004;68:119–125
  33. Wu X, Chen X, Li Z, Han S, Zhang D. Binding of verbal and spatial information in human working memory involves large-scale neural synchronization at theta frequency. Neuroimage. 2007;35(4):1654–1662

PII: S0920-9964(07)00517-8

doi:10.1016/j.schres.2007.11.004

Schizophrenia Research
Volume 100, Issue 1 , Pages 153-160, March 2008

Article Tools

* Image Usage & Resolution