Effect of organizational strategy on visual ... - Wiley Online Library

Psychiatry and Clinical Neurosciences 2008; 62: 427–434

doi:10.1111/j.1440-1819.2008.01821.x

Regular Article

Effect of organizational strategy on visual memory in patients with schizophrenia Myung-Sun Kim,

PhD,1*

Yoon Namgoong,

MA1

and Tak Youn,

MD, PhD2

1 Department of Psychology, Sungshin Women’s University, Seoul and 2Department of Psychiatry, Chonnam National University Medical School, Kwangju, Korea

Aims: The aim of the present study was to examine how copy organization mediated immediate recall among patients with schizophrenia using the Rey– Osterrieth Complex Figure Test (ROCF). Methods: The Boston Qualitative Scoring System (BQSS) was applied for qualitative and quantitative analyses of ROCF performances. Subjects included 20 patients with schizophrenia and 20 age- and gendermatched healthy controls. Results: During the copy condition, the schizophrenia group and the control group differed in fragmentation; during the immediate recall condition, the two groups differed in configural presence and planning; and during the delayed recall condition, they differed in several qualitative measurements, including configural presence, cluster presence/placement, detail presence/placement, fragmentation, planning, and neatness. The two groups also differed in several

TUDIES HAVE SHOWN that deficits in a broad range of cognitive functions such as attention, executive function, learning, and memory are core features of schizophrenia, but a consensus has not been reached about whether patients with schizophrenia have generalized cognitive deficits1 or selective deficits in specific cognitive functions including memory, learning, and executive functions.2 Researchers have intensively studied verbal memory in schizophrenia patients, and have

S

*Correspondence: Myung-Sun Kim, PhD, Department of Psychology, Sungshin Women’s University, 3 Dongsun, Sungbuk, Seoul 136-742, Korea, Email: [email protected] Received 11 September 2007; revised 28 December 2007; accepted 25 April 2008.

quantitative measurements, including immediate presence and accuracy, immediate retention, delayed retention, and organization. Although organizational strategies used during the copy condition mediated the difference between the two groups during the immediate recall condition, group also had a significant direct effect on immediate recall.

Conclusion: Schizophrenia patients are deficient in visual memory, and a piecemeal approach to the figure and organizational deficit seem to be related to the visual memory deficit. But schizophrenia patients also appeared to have some memory problems, including retention and/or retrieval deficits. Key words: Boston Qualitative Scoring system, organizational strategy, Rey–Osterrieth Complex Figure, schizophrenia, visual memory.

produced substantial evidence that these patients have a verbal memory deficit.2,3 In contrast, few studies have examined visual memory in schizophrenia patients.4 These studies have consistently reported that patients with schizophrenia are deficient in visual memory,5,6 but the mechanisms underlying this deficit are poorly understood.4,5 Some researchers have reported that non-verbal memory deficit is associated with problems related to consolidation,7 and others have suggested that visual memory impairment is associated with a deficit related to organization and encoding,6 or with a combined deficit in organization and retention.4 Glahn et al. recently examined non-verbal memory impairment in patients with schizophrenia and found that they performed worse than patients with bipolar

© 2008 The Authors Journal compilation © 2008 Japanese Society of Psychiatry and Neurology

427

428

M-S. Kim et al.

disorder or normal controls, but observed differential impairment only when organizational demands were significant.5 These results indicate that non-verbal memory impairment in schizophrenia patients is related to organizational and encoding deficits, necessitating a direct examination of the organizational processes used in visual memory among these patients. The Rey–Osterrieth Complex Figure Test (ROCF) is widely used to evaluate visual memory and visuospatial construction abilities.8 The ROCF task involves copying a geometric figure and then reproducing it from memory, either immediately or after a delay. Subjects can reproduce the figure in many ways, and the ROCF involves several visual components that can be perceived as large-scale organizational features (gestalts) or small details. Because of these complexities of the figure, ROCF testing can yield important, qualitative information about the subject’s strategies and organizational approach.9 Many studies have reported that the ROCF is useful to assess executive functions, including organization and planning; for example, it has been used to evaluate organizational strategy in patients with obsessive–compulsive disorder.10,11 Few studies have applied the ROCF to investigate how organizational strategy affects recall in schizophrenia patients. Seidman et al. used it to compare visual memory and organizational function among patients with chronic schizophrenia and chronic bipolar disorder, and healthy controls,4 and found that patients with schizophrenia used a more detailoriented style during copy and recall and had significantly worse visual memory than patients with bipolar disorder or healthy controls. These results indicate that patients with schizophrenia have a visual memory disorder characterized by both organizational processing deficits and retention difficulties. Kalinowski et al. also administered the ROCF to patients with schizophrenia and healthy controls and found that patients with schizophrenia were worse at organization and recall, which could result from their more detail-oriented style.12 Previous studies of how organizational strategy affects recall in patients with schizophrenia using the ROCF have generally applied quantitative methods. For example, Seidman et al. used a scoring system developed by Osterrieth to measure organization; it assigns a score from 1 to 6 depending on the subject’s strategy (a low score indicates a gestalt approach and a high score indicates a disorganized approach).4


Researchers have developed several other quantitative methods to score the ROCF.13,14 Although many of these methods have stressed the importance of executive functioning on ROCF performance, most have not provided a comprehensive system for evaluating executive functions that affect ROCF performance.8 Only one commercially available qualitative scoring system has been developed to assess executive functioning in adults: the Boston Qualitative Scoring System (BQSS) developed by Stern et al.15 The BQSS provides comprehensive sets of qualitative ratings in addition to quantitative summary scores. The BQSS for the ROCF also includes executive function variables that are composed of five scores: planning, fragmentation, neatness, perseverance, and organization. BQSS executive function variables are significantly associated with traditional executive measures, including perseverative errors on the Wisconsin CardSorting Test and the total number of responses on the Controlled Oral Word Association Test.8 We investigated the characteristics of visual memory deficit in patients with schizophrenia. Specifically, we used the BQSS for the ROCF to determine whether visual memory deficits were mediated by poor organizational strategies used by these patients while copying the figure, with the goal of clarifying the mechanisms that underlie visual memory deficit in these patients.

METHODS Participants Subjects included 20 patients with schizophrenia living at the Institute for Social Return and 20 age- and gender-matched healthy controls. All patients satisfied the DSM-IV16 criteria for schizophrenia based on the Structured Clinical Interview for DSM-IV (SCIDIV).17 The severity of schizophrenia symptoms was evaluated using the Positive and Negative Syndrome Scale (PANSS).18 Of the 20 patients, eight had paranoid schizophrenia, 10 had undifferentiated schizophrenia, and two had residual schizophrenia. All patients were being treated with neuroleptic medication at the time of testing; 12 were taking typical antipsychotics (chlorpromazine equivalent dose: 529.16 ⫾ 423.17 mg)19 and eight were taking atypical antipsychotics (risperidone, n = 4; olanzapine, n = 1; clozapine, n = 3). Among 20 patients, eight were taking anticholinergic medication (benztropine, n = 3, mean dose, 1.44 ⫾ 1.01 mg/day; procyclidine,



Visual memory deficit in schizophrenia

429

Table 1. Subject characteristics Schizophrenia (n = 20) Variable Gender Male Female Age (years) Education (years) IQ BDI score Age of onset (years) Duration of illness (years) PANSS score Positive symptom Negative symptom General pathology

Mean

Control (n = 20)

SD

Mean

12 8

SD

t (38)

7.69 2.23 8.55 5.84

-1.11 4.34*** 6.54*** -2.44*

12 8

41.9 11.95 99.55 15.05 26.94 14.44

6.84 2.70 9.69 13.86 7.31 7.39

15.83 16.50 30.17

2.23 2.59 5.04

39.35 15.35 118.45 6.85

*P < 0.05, ***P < 0.001. BDI, Beck Depression Inventory; PANSS, Positive and Negative syndrome Scale.

n = 5, mean dose, 7.00 ⫾ 2.74 mg/day). Exclusion criteria for patients included any history of head injury, neurological disorders, substance abuse or any other psychiatric disorders. The healthy controls were recruited through the Internet. We used the Structured Clinical Interview for DSM-IV, Non-Patient (SCID-NP)17 to ensure that none of the 20 controls had any history of psychiatric, medical, or neurological disorders, or drug or alcohol abuse. All participants were right-handed. All participants provided written informed consent after receiving a complete description of the study, and this study was approved by the Sungshin Women’s University Institutional Bioethics Review Board. Table 1 summarizes the demographic characteristics of the schizophrenia group and the control group. The groups differed significantly in the level of education (t = 4.34, d.f. = 38, P < 0.001), IQ (t = 6.54, d.f. = 38, P < 0.001), and depression (t = -2.44, d.f. = 38, P < 0.05).

Instruments We administered the ROCF to evaluate visual memory using three test conditions: a copy condition, an immediate recall condition (3 min after the copy condition), and a delayed recall condition (30 min after the copy condition).20

We used the BQSS15 to qualitatively analyze the ROCF data; this method scores ROCF data by dividing the figure into three sets of elements that are hierarchically arranged in terms of structural importance: configural elements, clusters, and details. These elements are then scored according to presence, accuracy, placement, and fragmentation. Configural elements are evaluated first, followed by clusters, and finally by details. All elements are initially scored for presence. If an element is present, then judgments are made about the element’s accuracy, placement, and fragmentation, depending on which element of the figure is being evaluated. In addition to 17 specific qualitative scores for each condition, the method yields six summary scale scores: copy presence and accuracy, immediate presence and accuracy, delayed presence and accuracy, immediate retention, delayed retention, and organization. Each summary score is derived from combinations of qualitative scale scores and provides a more quantitative evaluation of each participant’s overall ROCF performance. Interrater agreement on the performance of five randomly selected participants was relatively high; the kappa value of interrater agreement was 0.86 (P < 0.001). We administered the Beck Depression Inventory (BDI)21 and the Korean version of the Wechsler Adult Intelligence Scale (K-WAIS)22 to evaluate the severity of depression and IQ, respectively.


430

M-S. Kim et al.


In the delayed recall condition, significant differences appeared between the schizophrenia group and the healthy control group for several qualitative variables: configural presence (F1,35 = 23.04, P < 0.001), cluster presence (F1,35 = 7.68, P < 0.01), cluster placement (F1,35 = 8.16, P < 0.01), detail presence (F1,35 = 6.76, P < 0.05), detail placement (F1,35 = 8.96, P < 0.01), fragmentation (F1,35 = 4.06, P < 0.05), planning (F1,35 = 17.16, P < 0.001), and neatness (F1,35 = 4.50, P < 0.05). Patients with schizophrenia scored lower than healthy controls on all these qualitative scales, indicating impaired ability to recall the figure’s overall appearance and details.

Statistical analysis We conducted a multivariate analysis of variance to investigate group differences in ROCF performances using covariates of IQ, level of education, and BDI score, and we used hierarchical multiple regression to examine the mediating effect of organizational strategy on visual memory as measured by the ROCF.

RESULTS ROCF qualitative scales Table 2 lists the means and standard deviations of 17 qualitative scores on ROCF copy, immediate recall, and delayed recall conditions for each group. In the copy condition, the schizophrenia group and the control group differed only in fragmentation (F1,35 = 6.29, P < 0.05); scores were lower among patients with schizophrenia. In the immediate recall condition, the schizophrenia group and the control group differed in scores for configural presence (F1,35 = 5.95, P < 0.05), and planning (F1,35 = 3.59, P < 0.05); both scores were lower among patients with schizophrenia.

ROCF quantitative scales Table 3 lists the means and standard deviations of quantitative ROCF summary scores between the schizophrenia group and the control group. Patients with schizophrenia scored lower than healthy controls on immediate presence and accuracy (F1,35 = 4.70, P < 0.05), immediate retention (F1,35 = 8.10, P < 0.05), delayed retention (F1,35 = 4.19, P < 0.05), and organization (F1,35 = 5.84, P < 0.05), suggesting

Table 2. Qualitative score in ROCF copy, immediate recall, and delayed recall conditions Copy condition Schizophrenia (n = 20)

Immediate recall condition

Control (n = 20)

Schizophrenia (n = 20)

Control (n = 20)

Delayed recall condition Schizophrenia (n = 20)

Control (n = 20)

Variable

Mean

SD

Mean

SD

Mean

SD

Mean

SD

Mean

SD

Mean

SD

Configural presence Configural accuracy Cluster presence Cluster accuracy Cluster placement Detail presence Detail placement Fragmentation Planning Neatness Vertical expansion Horizontal expansion Reduction Rotation Perseveration Confabulation

3.95 2.15 4.00 2.45 3.10 3.40 3.90 2.95* 2.20 2.65 3.80 3.65 3.85 3.65 3.75 3.95

0.22 0.81 0.00 0.89 0.72 0.68 0.31 0.76 0.95 0.75 0.70 0.75 0.67 0.49 0.44 0.22

4.00 2.15 3.95 3.25 3.45 3.70 3.70 3.85 3.20 3.05 3.80 3.60 4.00 3.70 3.95 4.00

0.00 0.67 0.22 0.91 0.51 0.57 0.92 0.37 0.83 0.39 0.52 0.75 0.00 0.47 0.22 0.00

2.65* 1.70 1.65 1.30 2.25 1.10 2.70 3.30 1.95* 2.55 3.80 3.55 3.80 3.80 3.40 3.30

0.88 1.30 0.59 1.30 1.40 0.30 1.75 0.92 1.10 0.69 0.89 1.05 0.52 0.41 0.88 1.13

3.75 2.45 2.50 2.20 2.85 1.45 3.10 3.80 3.10 2.95 3.85 3.25 3.95 3.85 3.00 3.90

0.44 0.76 0.69 0.89 0.75 0.60 1.21 0.41 0.64 0.51 0.37 1.20 0.22 0.37 0.92 0.31

2.75*** 2.00 1.90** 1.35 1.90** 0.95* 2.05** 3.35* 2.05*** 2.55* 3.80 3.30 3.80 3.95 3.00 3.40

0.85 0.86 0.55 1.04 1.20 0.51 1.70 0.93 1.05 0.83 0.89 1.34 0.52 0.22 1.03 0.88

3.85 2.30 2.50 1.65 2.75 1.45 3.35 3.75 3.25 2.90 3.75 2.95 4.00 3.75 3.00 3.80

0.37 0.73 0.67 0.88 0.79 0.60 1.04 0.44 0.72 0.31 0.44 1.39 0.00 0.44 0.92 0.41

*P < 0.05, **P < 0.01, ***P < 0.001. ROCF, Rey–Osterrieth Complex Figure Test.



Visual memory deficit in schizophrenia

431

Table 3. ROCF summary score Schizophrenia (n = 20)

Control (n = 20)

Variable

Mean

SD

Mean

SD

F

Copy presence and accuracy Immediate presence and accuracy Delayed presence and accuracy Immediate retention Delayed retention Organization

15.90 8.40 8.90 47.38 11.80 5.15

1.92 2.58 2.15 14.96 33.76 1.57

17.05 12.35 11.75 27.47 8.85 7.05

1.76 2.16 2.15 10.35 13.62 0.94

1.52 4.70* 0.37 8.10* 4.19* 5.84*

*P < 0.05. ROCF, Rey–Osterrieth Complex Figure Test.

impaired ability in the immediate and delayed recall of a complex figure and poor organizational strategies.

Mediating effect of organizational strategy on visual memory We tested direct and indirect models of the group effects to investigate whether the visual memory deficit among patients with schizophrenia was mediated by poor organizational strategies during copying. These models have been described previously by Savage et al.23 First, we used a composite score of immediate presence and accuracy and immediate retention of summary scores as the dependent variable and the copy organization score as the mediating variable. Baron and Kenny proposed that for all linkages to be supported in a mediated model, significant associations must appear between the independent variable (group), the dependent variable (composite score), and the mediator (copy organization score).24 We found significant correlations between group and organization scores (r = -0.60, P < 0.001), between group and composite scores (r = -0.63, P < 0.001), and between composite score and organization score (r = 0.58, P < 0.001). Figure 1 presents a path model representing direct and indirect effects based on standardized coefficients (b) and significance levels. In the direct model, group differences in ROCF immediate recall are expressed directly as a significant regression coefficient between group and immediate recall (b = -0.63, P < 0.001). In the mediated model, the group effect on immediate recall is expressed indirectly based on the influence of copy organizational

strategy. Organizational strategy appeared to mediate the difference between the two groups under the ROCF immediate recall condition (b = 0.31, P < 0.05), but group also had a significant direct effect on ROCF immediate recall (b = -0.44, P < 0.01).

Relationship between ROCF quantitative scales and the duration of illness We found no significant associations between quantitative BQSS scores and the duration of illness in the schizophrenia group. We also found no significant

Direct model Group

β= -0.63 P