Qualitative Evaluation of the Immediate Copy of the Rey–Osterrieth

Complex Figure: Comparison Between Vascular and Degenerative MCI

Patients
Emilia Salvadori1,*, Francesca Dieci2, Paolo Caffarra3, Leonardo Pantoni4
1
NEUROFARBA Department, Neuroscience Section, University of Florence, Florence, Italy
2
Outpatients Center for Cognitive Disorders and Dementia, AUSL of Parma, Parma, Italy
3
Unit of Neuroscience, Department of Medicine and Surgery, Parma, Italy
4
L. Sacco’ Department of Biomedical and Clinical Sciences, University of Milan, Milan, Italy
*Corresponding author at: NEUROFARBA Department, Neuroscience section, University of Florence, Largo Brambilla 3, 50134 Florence, Italy.
Tel.: +0039-055-7947995; fax: +0039-055-2758181.
E-mail address: emilia.salvadori@unifi.it (E. Salvadori).
Editorial Decision 16 January 2018; Accepted 22 January 2018

Abstract
Objective: The immediate copy of the Rey–Osterrieth Complex Figure (ROCF) is considered a visuo-spatial test. However, reproducing
this complex structure possibly involves also executive functions, such as planning and organizational strategies. In a previous study, we
found a high rate of impaired performances in this test in a sample of subcortical vascular mild cognitive impairment patients. Executive
functions contribution in the immediate copy of the ROCF can be assessed with the Boston Qualitative Scoring System (BQSS). We aimed
at examining whether BQSS executive scores of ROCF immediate copy: (1) differ between vascular (v-MCI) and degenerative MCI
(d-MCI) patients; (2) can at least partly explain the high rate of abnormal ROCF immediate copy performances in v-MCI patients.
Method: Thirty d-MCI patients (age 75.2 ± 4.4) and 27 v-MCI (age 73.2 ± 6.9) were enrolled. The performances of patients were scored
using the BQSS executive scores (Fragmentation, Planning, Organization, Perseveration) during the accomplishment of ROCF immediate copy.
Results: Comparing d-MCI and v-MCI performances, d-MCI patients scored worse on ROCF delayed recall (9.9 ± 4.7 vs. 13.4 ± 5.9,
p = .020) and MMSE (23.9 ± 2.6 vs. 27.8 ± 2.3, p = .001) while v-MCI patients had more frequently impaired performances in ROCF
immediate copy (40% vs. 81%, p = .001) and showed worse scores on Fragmentation (2.4 ± 0.9 vs. 1.8 ± 1.3, p = .035), Planning (2.4 ±
0.8 vs. 1.8 ± 1, p = .039), Organization (4.8 ± 1.3 vs. 3.6 ± 2.1, p = .017), and Perseveration (3.5 ± 0.8 vs. 2.9 ± 1.2, p = .048).
Conclusions: The performance of v-MCI patients in ROCF immediate copy seemed to be more affected by executive dysfunction than
the performance obtained by d-MCI. When analyzing ROCF performances, a qualitative approach allows to evaluate patients’ strategies dur-
ing the reproduction, and thus to discriminate between executive and visuo-constructional abilities.
Keywords: Executive functions; Perception/spatial processing; Mild cognitive impairment; Alzheimer’s disease; Cerebrovascular disease/accident and stroke

Introduction

The Rey–Osterrieth Complex Figure (ROCF) is a task widely used for the assessment of visuo-spatial abilities and visual
memory. The task, originally designed by Rey (1941) and later standardized by Osterrieth (1944), requires the subject to copy
a complex geometrical figure (immediate copy condition) and, after an interval that varies according to different administra-
tion procedures, to reproduce the figure from memory without forewarning (delayed recall condition).
The ROCF is composed of several units that can be perceptually divided into global or local elements (Fig. 1A). Global
elements are the large rectangle, the diagonals, the horizontal and vertical lines, and the large triangle to the right of the rect-
angle. Global elements are fundamental for the organization of the figure, and a successful reproduction of the ROCF requires
to draw first of all these elements, as they represent the structural framework of the other local elements. The local elements

© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: [email protected].
doi:10.1093/arclin/acy010

Downloaded from https://academic.oup.com/acn/advance-article-abstract/doi/10.1093/arclin/acy010/4840584

by Mount Royal College user
on 09 February 2018
 2 E. Salvadori et al. / Archives of Clinical Neuropsychology (2018); 1–10

Fig. 1. (A) Rey–Osterrieth Complex Figure: bold lines represent the global elements. Examples productions illustrating Fragmentation additional rules: (B)
hatching of the most part of the lines was considered as a graphic style, and it was not scored as fragmentation; (C) lack of conjunctions between different
lines of the same element (such as angles and diagonals) was not scored as fragmentation.

can be further hierarchically divided into units made up of shapes or lines that form a coherent gestalt within the main figure
(e.g., the small rectangle with the inside cross, the small triangle above large rectangle, the circle with three dots) or simple
details made of single line segments.
In the immediate copy condition of the ROCF, the complexity of the figure requires an integrative cognitive ability, based
on the ability to gather an overall view and to organize the figure into a meaningful perceptual unit. Therefore, the reproduc-
tion of such a complex structure involves also cognitive processes such as planning and organizational strategies that are
related to executive functions (Elderkin-Thompson et al., 2004; Freeman et al., 2000; Shin, Park, Park, Seol, & Kwon, 2006).
Among many scoring systems developed for the ROCF, the most used is the traditional Osterrieth method, a quantitative
scoring system that provides a 36-point summary score based on the presence and accuracy of 18 units of the figure (Caffarra,
Vezzadini, Dieci, Zonato, & Venneri, 2002a; Osterrieth, 1944). The single-score method does not take into account the differ-
ent importance of the figure’s elements, both from a perceptual and structural point of view, nor the logical processes underly-
ing the order in which elements are reproduced. Therefore this method is not able to capture the strategies such as planning
and organizational approach to complete the figure. Qualitative scoring systems have been proposed for the evaluation of
executive strategies, and several studies have shown that these strategies can differentiate healthy controls from patients with
an executive impairment, in neurological, psychiatric and also pediatric populations (Elderkin-Thompson et al., 2004;
Eslinger & Grattan, 1990; Freeman et al., 2000; Scarpina, Ambiel, Albani, Pradotto, & Mauro, 2016; Shin et al., 2006;
Somerville, Tremont, & Stern, 2000; Stern et al., 1999). Other studies found that organizing the ROCF into a meaningful per-
ceptual unit during the immediate copy condition enhances its subsequent recall from memory (Savage et al., 1999, 2000;
Shorr, Delis, & Massman, 1992). As executive functions mediate the reorganization of randomized stimuli into a meaningful
cluster, they also reinforce the encoding and consolidation process, and thus the long-term memory retrieval.
Among the qualitative scoring systems developed for the ROCF, the Bennett-Levy (1984) and Bylsma, Bobholz,
Schretlen, and Correa (1995) methods are based on the Gestalt principles of perceptual organization, and reflect the order in
which elements are drawn and their fragmentation. Other simple scoring systems that are focused on the organizational quality
have been devised by Hamby, Wilkin, and Barry (1993) and Savage and colleagues (1999). Further methods have been spe-
cifically developed for children (e.g., Developmental Scoring System; Waber & Holmes, 1986) which evaluates organization
and production style, and the Organizational Strategy Score (Anderson, Anderson, & Garth, 2001) based on the sequence of
elements’ drawn and their relevance within the structure. Despite the availability of several approaches to qualitatively evalu-
ate the ROCF performance, these systems focus only on one or two qualitative features, such as organization or symmetry, or
are not appropriate for use with adults. At present, the most complete qualitative scoring system for the ROCF available for
adults is the Boston Qualitative Scoring System (BQSS) that, among a comprehensive set of qualitative ratings, provides five
scores (Planning, Fragmentation, Neatness, Perseveration, and Organization) specifically developed for the evaluation of the
executive functions (Stern et al., 1999).
Concerning the psychometric properties of the BQSS executive scores, previous studies showed a good to excellent inter-
rater reliability (Folbrecht, Charter, Walden, & Dobbs, 1999; Stern et al., 1999), and good discriminant validity in differentiating
healthy controls from patients with Parkinson’s disease, obsessive–compulsive disorder, attention deficit hyperactivity disorder,
schizophrenia, alcohol abuse, and traumatic brain injury (Cahn et al., 1996; Dawson & Grant, 2000; Eslinger & Grattan, 1990;
Freeman et al., 2000; Javorsky, Rosenbaum, & Stern,1999; Mahurin, Eckert, Velligan, Hazelton, & Miller, 1997; Scarpina et al., 2016;

Downloaded from https://academic.oup.com/acn/advance-article-abstract/doi/10.1093/arclin/acy010/4840584

by Mount Royal College user
on 09 February 2018
 E. Salvadori et al. / Archives of Clinical Neuropsychology (2018); 1–10 3

Schreiber, Javorsky, Robinson, & Stern, 1999; Stern et al., 1999). The study by Somerville and colleagues (2000) found that
BQSS executive scores significantly correlated with some traditional executive measures, were less correlated with discriminant
measures (verbal and visual memory retention), and significantly differentiated patients with varying degrees of executive dys-
function. This study provided preliminary support for the construct validity of the BQSS executive scores, and for the useful-
ness of the ROCF as a measure of executive functioning.
A qualitative analysis of patients’ performance in a complex neuropsychological task, such as the ROCF, could be of high rele-
vance for the identification of different patterns of cognitive deficits particularly in those patients that are in the early stages of the
disease, and could thus benefit from specific preventive and therapeutic approaches. Mild cognitive impairment (MCI) is an inter-
mediate state between normal cognitive status and dementia; and is thought to anticipate dementias of various etiologies (Gauthier
et al., 2006; Winblad et al., 2004). The term MCI has been initially conceived to refer to a preclinical state of Alzheimer’s disease
(AD) and, therefore, as a consequence of medial temporal system dysfunction, a memory impairment was considered the core
cognitive feature of degenerative MCI (d-MCI) (Albert et al., 2011; Petersen et al., 2014). More recently, the concept of MCI has
been expanded to include cognitive impairment in domains other than memory, and other clinical subtypes of MCI have been pro-
posed as prodromal forms of a variety of dementias. Preclinical stages of vascular dementia (VaD) are also recognized (Pantoni &
Gorelick, 2011), among which those consequent to cerebral small vessel disease (SVD), are the most common (Pantoni, Poggesi,
& Inzitari, 2009). In patients with SVD, ischemic lesions are particularly located in the subcortical areas, and cause the disruption
of corticostriatal loops subserving the frontal lobes functions (Cummings, 1993) and thus affecting executive and attentional pro-
cessing as observed in subcortical vascular MCI (Sachdev, Brodaty, & Looi, 1999).
The starting point of the present study comes from the results published in a paper on the development of a neuropsycholog-
ical battery for Vascular Mild Cognitive Impairment (v-MCI) (Salvadori, Poggesi, Pracucci, Inzitari, & Pantoni, 2015). In that
study, 201 patients with MCI and SVD were enrolled. As expected, the vast majority of patients had normal performances on
Mini Mental State Examination, underlining an overall mild degree of cognitive impairment. Distribution of cognitive perfor-
mances confirmed that attention-executive dysfunction was one of the prominent features of vascular cognitive impairment:
tests assessing speed of information processing as sustained and divided attention, resulted in an elevated percentage of abnor-
mal performance, while language and prose memory tests were mostly normal. Conversely, the immediate copy of the ROCF
resulted the most difficult test, with impaired performances in 65% of patients (Salvadori et al., 2015). The impairment in high
level visuo-constructional abilities could be in agreement with previous data about the heterogeneity of the neuropsychological
profile in vascular cognitive impairment. However, we hypothesized that a lack of strategic approach was another possible
explanation for the results emerged in this cohort of patients, consequent to the observed executive dysfunction.
In the present study, we aimed at examining whether the analysis of the qualitative features of the immediate copy of the
ROCF could help to confirm the following hypotheses: d-MCI patients should have a predominance of memory disorders,
thus resulting in a worse performance in the delayed recall of ROCF, whereas the immediate copy of the ROCF is expected
to be more impaired in v-MCI patients due to an executive dysfunction related to frontal lobe damage. Furthermore, BQSS
executive scores should be able to highlight the executive component of the immediate reproduction of the ROCF, and thus
differentiate degenerative and vascular MCI patients.

Methods

Participants

The study was based on the collaboration of two Italian centers: Florence (NEUROFARBA Department, Neuroscience
Section, University of Florence) and Parma (Center for Cognitive Disorders and Dementia), and each center enrolled an inde-
pendent sample of MCI patients.
Vascular MCI (v-MCI) patients were a subsample of the VMCI-Tuscany study cohort enrolled in Florence. The VMCI-
Tuscany study is a multicenter, prospective, observational study aimed at evaluating the effect of a large set of clinical, neuro-
imaging, and biological markers of SVD in predicting the transition from MCI to dementia (Poggesi et al., 2012). To be
included, patients had to be diagnosed as affected by MCI with SVD according to the following criteria: (1) MCI defined ac-
cording to Winblad and colleagues (2004) criteria and operationalized according to Salvadori and colleagues (2016); (2) evi-
dence on MRI of moderate to severe age-related white matter hyperintensities (WMH) on T2 weighted fluid attenuated
inversion recovery (FLAIR) images according to a modified version of the Fazekas scale (Pantoni et al., 2005). The diagnosis
of v-MCI required at least one score borderline (an adjusted score between the outer and inner 95% confidence limits for the
fifth percentile of the normal population according to published normative data) among the neuropsychological tests included
in the VMCI-Tuscany neuropsychological battery (Salvadori et al., 2015), and preservation of independence in functional

Downloaded from https://academic.oup.com/acn/advance-article-abstract/doi/10.1093/arclin/acy010/4840584

by Mount Royal College user
on 09 February 2018
 4 E. Salvadori et al. / Archives of Clinical Neuropsychology (2018); 1–10

abilities. Exclusion criteria for the VMCI-Tuscany study were the inability or refusal to undergo cerebral MRI, and inability
to give an informed consent. Patients with degenerative MCI (d-MCI) due to Alzheimer Disease (prodromal AD), according
to the National Institute on Aging and the Alzheimer’s Association (NIAA) clinical criteria (Albert et al., 2011) were enrolled
in the Parma center. Core criteria for the diagnosis of d-MCI included a prominent impairment in episodic memory (an age
and education adjusted score 1 SD below the mean of the normal population according to published normative data), and
preservation of independence in functional abilities. Exclusion criteria for d-MCI patients were the occurrence of other medi-
cal, traumatic or brain diseases that could account for the decline in cognition, with particular attention to parkinsonism, mul-
tiple vascular risk factors or the presence of extensive cerebrovascular disease on MRI, prominent behavioral or language
disorders early in the course of disease, or very rapid cognitive decline that occurred over weeks or months.

Cognitive and functional evaluation

Due to some difference in the composition of the neuropsychological battery between the two labs, cognitive data derived
from the shared tests which included:

• Immediate copy and delayed recall of the ROCF (Caffarra et al., 2002a). To perform the immediate copy of the
ROCF, the figure was placed in front of the subject, who was requested to copy the figure as accurately as possible
without any time limit. When copying the figure was completed, the stimulus was removed from sight. After a 10-
min delay, subjects were asked to reproduce the figure from memory without forewarning. Score range 0–36: higher
scores represent better performance.
• Mini Mental State Examination (Measso et al., 1993) for global cognitive functioning. Score range 0–30: higher
scores represent better performance.
• Color Word Stroop Test (Caffarra, Vezzadini, Dieci, Zonato, & Venneri, 2002b) for selective attention and executive
functions. The interference effect was evaluated based on execution time (in seconds) and number of errors: higher
scores represent worse performance.
• Phonemic (P-F-L) and semantic (animals–fruits–cars) verbal fluency tests (Novelli, Papagno, Capitani, Laiacona, &
Vallar, 1986) for both language and executive functions. For both tests, the final score was the total number of words
produced for the three initials or categories, respectively: higher scores represent better performance.

For all neuropsychological tests, scoring methods, age and education adjustments of raw scores, and evaluations of the perfor-
mance was based on the above mentioned validation and normative studies for the Italian population.
Functional status was measured by means of:

• Activities of Daily Living scale (Katz, Ford, Moskowitz, Jackson, & Jaffe, 1963): the total score was the number of
preserved items (score range 0–6; higher scores represent less disability).
• Instrumental Activities of Daily Living scale (Lawton & Brody, 1969): the total score was the number of impaired
items (score range 0–8; higher scores represent more disability).

Qualitative evaluation of the ROCF

To highlight the executive strategies used by patients during the reproduction of the immediate copy of the ROCF, we applied
a qualitative evaluation based on the BQSS executive scores (Stern et al., 1999):

• Fragmentation. A measure of integration of information that evaluates if individual elements are drawn as whole
units. Score range: 0 (extreme fragmentation) to 4 (no fragmentation).
• Planning. A measure of overall planning ability based on the order in which elements are drawn, their placement on the
page and within the figure, and the overall integrity of the production. Score range: 0 (poor planning) to 4 (good planning).
• Organization. The arithmetic sum of the Fragmentation and Planning scores, providing an overall measure of organi-
zational skills. Score range: 0 (poor organization) to 8 (good organization).
• Perseveration. A measure of the extent of recognizably inappropriate repetition that may take one of two forms: rep-
etition of components within a cluster (within-cluster) or replication of any element of the figure (element repetition).
Self-corrected or changed lines were not considered perseverative. Score range: 0 (extreme perseveration) to 4 (no
perseveration).

Downloaded from https://academic.oup.com/acn/advance-article-abstract/doi/10.1093/arclin/acy010/4840584

by Mount Royal College user
on 09 February 2018
 E. Salvadori et al. / Archives of Clinical Neuropsychology (2018); 1–10 5

Despite the Boston manual offers a set of well-defined scoring criteria (Stern et al., 1999), the complexity of the qualitative
analysis entails a consensus between two independent raters. To standardize the evaluation procedure, v-MCI patients, after
consent, were filmed during the immediate copy task, and the raters (two expert neuropsychologists) revised and evaluated
the videos by consensus meeting. Some specific aspects of the evaluation of the Fragmentation and Planning scores required
the shared definition of additional rules detailed below.

Fragmentation additional rules

• Hatching of the most part of the lines was considered as a graphic style, and it was not scored as fragmentation
(Fig. 1B), unless the patient came back to trace again over the line after starting another element.
• Lack of conjunctions between different lines of the same element (such as angles and diagonals) was not scored as
fragmentation (Fig. 1C).
• Considering that the Fragmentation raw score is based on the total number of fragmentations (from 0 to 9) of seven
elements, we decided to take into account also if any of the involved elements was omitted. The Fragmentation raw
score was then adjusted for the number of omissions according to the procedure shown in Table 1.

Planning additional rules

The evaluation of a moderately (score = 2) or significantly (score = 1) poor planning is based on the placement of the figure
on the page, the recognizability of some elements, and the disorganized sequence of the drawing elements. This last criterion
is further described as the lack of a “logical and systematic order”. In a moderately poor planning the order has to be “not
haphazard”, while in a significantly poor planning the order has to be “not completely haphazard”, and the distinction of those
two conditions was somewhat difficult and liable to different interpretations. According to our operationalization of this crite-
rion, the order was considered “not haphazard” if:

• The order of drawing gave priority to the completion of structural elements (configural or clusters).

• The elements were drawn according to a spatial proximity criterion.

More specifically, each step of the drawing has been evaluated separately and considered “logical” if a structural element (config-
ural or cluster) was completed before another was begun or, if it was not completed, the patient began another element that was
spatially close or inside the previous one. Each step that violates this criterion was considered “illogical” and was counted.
The definitions of “not haphazard” or “not completely haphazard” order have been then operationalized as follows:

– In case of 1 illogical step the order was considered “not haphazard”.

– In case of 2 illogical steps the order was considered “not haphazard” if Rectangle A and Triangle F were completed,
otherwise the order was considered “not completely haphazard”.
– In case of 3 illogical steps the order was considered “not completely haphazard” and a significantly poor planning
(score=1) was assigned if Rectangle A was recognizable, otherwise an extremely poor planning (score = 0) was rated.

Statistical analysis

Univariate statistical analyses (independent sample t tests and Pearson’s chi square tests) have been used to compare the v-
MCI and d-MCI groups in respect of demographic characteristics (age, years of education, and sex), cognitive functioning
(immediate and delayed reproduction of the ROCF, MMSE, Color Word Stroop Test, and phonemic and semantic fluency),
Table 1. Fragmentation score correction grid for the number of omissions
Number of omissions Correction factor Fragmentation adjusted score
0–1 0 Fragmentation raw score + Correction factor
2 −1
3 −2
4–5 −3
6–7 −4

Downloaded from https://academic.oup.com/acn/advance-article-abstract/doi/10.1093/arclin/acy010/4840584

by Mount Royal College user
on 09 February 2018
 6 E. Salvadori et al. / Archives of Clinical Neuropsychology (2018); 1–10

functional status, and Boston executive scores (Fragmentation, Planning, Organization, and Perseveration). For the latter
analyses, the effect sizes were estimated by means of the Cohen’s d (Cohen, 1988) and the unbiased Hedges’ g (a variation of
Cohen’s d that corrects for biases due to small sample sizes) (Hedges & Olkin, 1985), and the 95% confidence intervals
around the effect size estimates were also computed (Cummings, 2012).
To investigate the convergent validity of the BQSS scores, the association between the latter and the ROCF immediate
copy original score, and the Color Word Stroop Test score, was evaluated by means of non-parametric correlation analyses
(Spearman’s Rho coefficient) in the total sample.
All statistical analyses were performed using SPSS 20.

Results

Twenty-seven v-MCI patients (mean age 73.2 ± 6.9, mean years of education 9.2 ± 3.9, males 67%) and 30 d-MCI pa-
tients (mean age 75.2 ± 4.4, mean years of education 9 ± 4, males 37%) were enrolled. The groups were not significantly dif-
ferent in age, education, and functional status, but there was a significant prevalence of males in v-MCI patients (Table 2).
Concerning vascular risk factors distributions in the 27 v-MCI patients: 23 (85%) had hypertension, 21 (78%) hypercholester-
olemia, 14 (52%) reported smoking habits, 12 (44%) had history of stroke, and 14 (52%) alcohol consumption.
As shown in Table 2, d-MCI patients were more impaired in terms of general cognitive status as measured with Mini
Mental State Examination compared to v-MCI (mean adjusted scores 23.9 ± 2.6 vs. 27.8 ± 2.3, respectively) and percentage
of patients with an impaired performance (41% vs. 4%, respectively).
Considering the ROCF total score according to the original Osterrieth quantitative evaluation, compared to d-MCI, v-MCI
patients showed a significant worse performance at the immediate copy of the ROCF, both in terms of mean adjusted scores
(30.6 ± 4.2 vs. 24.9 ± 7.9, respectively) and percentage of patients with an impaired performance (40% vs. 81%, respectively)
(Table 2). Conversely, in the delayed recall the mean of the adjusted scores resulted significantly lower in d-MCI patients
compared to v-MCI ones (9.9 ± 4.7 vs. 13.4 ± 5.9, respectively) (Table 2).
For the remaining cognitive tests, d-MCI sample had a higher percentage of impaired performance at the semantic fluency
task compared to v-MCI (30% vs. 8%, respectively) (Table 2).
Table 3 illustrates the comparisons of the Boston executive scores between v-MCI and d-MCI groups. All the scores taken
into account were significantly lower in v-MCI patients compared to d-MCI ones: Fragmentation (2.4 ± 0.9 vs. 1.8 ± 1.3),
Planning (2.4 ± 0.8 vs. 1.8 ± 1), Organization (4.8 ± 1.3 vs. 3.6 ± 2.1), and Perseveration (3.5 ± 0.8 vs. 2.9 ± 1.2). With

Table 2. Comparisons of demographic characteristics and cognitive tests scores between v-MCI and d-MCI groups

Score range v-MCI d-MCI p

N = 27 N = 30
Age, years (Mean ± SD) 73.2 ± 6.9 75.2 ± 4.4 .859a
Years of education (Mean ± SD) 9.2 ± 3.9 9±4 .202a
Sex, males (%) 18 (67%) 11 (37%) .024b
Activities of Daily Living (preserved items) 0–6 (Mean ± SD) 5.9 ± 0.2 5.9 ± 0.2 .624a
Instrumental Activities of Daily Living (impaired items) 0–8 (Mean ± SD) 0.7 ± 1.3 0.5 ± 1 .589a
Mini Mental State Examination 0–30 Adjusted score 27.8 ± 2.3 23.9 ± 2.6 .001a
(% Impaired performance) 1 (4%) 12 (41%) .002b
Rey–Osterrieth Complex Figure Immediate copy 0–36 Adjusted score 24.9 ± 7.9 30.6 ± 4.2 .002a
(% Impaired performance) 22 (81%) 12 (40%) .001b
Rey–Osterrieth Complex Figure Delayed recall 0–36 Adjusted score 13.4 ± 5.9 9.9 ± 4.7 .020a
(% Impaired performance) 12 (48%) 14 (47%) .921b
Color Word Stroop Test (time) Adjusted score 33.9 ± 22.6 24.3 ± 13.6 .054a
(% Impaired performance) 13 (48%) 9 (30%) .160b
Phonemic fluency Adjusted score 31.5 ± 8.6 31.1 ± 7.3 .833a
(% Impaired performance) 4 (16%) 3 (11%) .570b
Semantic fluency Adjusted score 37.3 ± 6.5 33.5 ± 8.5 .080a
(% Impaired performance) 2 (8%) 9 (30%) .042b
v-MCI = vascular mild cognitive Impairment; d-MCI = degenerative mild cognitive Impairment
a
Independent sample t tests.
χ tests.
b 2

Bold values are statistically significant p values.

Italic values are statistical symbols and parameters.

Downloaded from https://academic.oup.com/acn/advance-article-abstract/doi/10.1093/arclin/acy010/4840584

by Mount Royal College user
on 09 February 2018
 E. Salvadori et al. / Archives of Clinical Neuropsychology (2018); 1–10 7

Table 3. Comparisons of Boston Qualitative Scoring System executive scores between v-MCI and d-MCI groups
Score range v-MCI d-MCI pa db db
N = 27 N = 30 (effect size) (95% CI)
Fragmentation 0–4 1.78 ± 1.34 2.43 ± 0.86 .035 0.58 0.05–1.11
Planning 0–4 1.85 ± 1.03 2.37 ± 0.81 .039 0.56 0.03–1.10
Organization 0–8 3.63 ± 2.08 4.80 ± 1.35 .017 0.67 0.14–1.21
Perseveration 0–4 2.96 ± 1.25 3.53 ± 0.78 .048 0.55 0.02–1.08
v-MCI = vascular mild cognitive impairment; d-MCI = degenerative mild cognitive impairment.
a
Independent sample t tests.
b
Cohen’s d (value and 95% confidence intervals), equal to unbiased Hedge’s g.
Bold values are statistically significant p values.
Italic values are statistical symbols and parameters.

Table 4. Associations between the Boston Qualitative Scoring System executive scores and the Rey–Osterrieth Complex Figure immediate copy original
score, and the Color Word Stroop Test score in the total sample (N = 57)

Fragmentation Planning Organization Perseveration

Rey–Osterrieth Complex Figure Immediate copy .326 .536 .503 .402
p = .013 p = .001 p = .001 p = .002
Color Word Stroop Test (time) −.142 −.285 −.239 .082
p = .291 p = .032 p = .074 p = .542
Non-parametric correlations, Spearman’s Rho.
Bold values are statistically significant p values.
Italic values are statistical symbols and parameters.

respect to effect size measures, Hedges’ g indexes were always equal to Cohen’s d indexes and resulted moderate in sizes
with significant 95% confidence intervals (Table 3).
As shown in Table 4, all the Boston executive scores were significantly associated with the ROCF immediate copy original
score, while only the Planning score correlated with the Color Word Stroop Test score in the total sample (N = 57).

Discussion

Patients with vascular MCI had a worse performance in the immediate copy of the ROCF compared to individuals with
degenerative MCI, despite their significant impairment in terms of general cognitive status and visual memory. When the per-
formance in the immediate copy of the ROCF was evaluated by means of a qualitative scoring method, namely the BQSS
executive indexes, vascular MCI patients showed more planning and organizational deficits than d-MCI ones.
In line with the hypothesis that the immediate copy of the ROCF could elicit strategic cognitive abilities related to the
executive functions, our results may indicate the presence of an executive impairment that is expected in MCI patients with a
subcortical cerebrovascular disease due to dysfunction of subcortical-frontal connections (O’Brien et al., 2003; Sachdev et al.,
1999). Several previous studies have examined the contribution of executive functions on the reproduction of the ROCF, as
well as the ability of the qualitative scoring methods to discriminate between patients groups and healthy controls, or among
different patients populations (Elderkin-Thompson et al., 2004; Eslinger & Grattan, 1990; Freeman et al., 2000; Scarpina
et al., 2016; Shin et al., 2006; Somerville et al., 2000; Stern et al., 1999). Available evidences show that patients with disor-
ders that possibly involve attention and executive functions (e.g., Parkinson’s disease, traumatic brain injury, obsessive-
compulsive disorder, attention deficit hyperactivity disorder) are characterized by a more disorganized approach when copying
the ROCF compared to controls (Eslinger & Grattan, 1990; Scarpina et al., 2016; Shin et al., 2006; Stern et al., 1999). In line
with the studies by Scarpina and colleagues (2016) and Elderkin-Thompson and colleagues (2004), our results seemed to con-
firm that the planning score is a robust index of executive functioning, and could represent a measure of prefrontal systems
dysfunctions. One preliminary study has found a higher rate of fragmentations, perseverations, and omissions in the drawings
in vascular patients compared with AD (Freeman et al., 2000), and our results further confirmed that vascular MCI patient re-
productions were more fragmented and contained more perseverations than those of degenerative MCI patients.
Concerning the validity of the BQSS scores as measures of executive functions, we found a small but significant correla-
tion between planning and time to complete the Stroop test. Our results are in line with preliminary evidences of a modest but
significant convergent validity between the BQSS executive scores and other widely known executive tasks in a sample of

Downloaded from https://academic.oup.com/acn/advance-article-abstract/doi/10.1093/arclin/acy010/4840584

by Mount Royal College user
on 09 February 2018
 8 E. Salvadori et al. / Archives of Clinical Neuropsychology (2018); 1–10

adult patients (Somerville et al., 2000). However, further confirmations of the clinically usefulness of the BQSS scores as
measures of executive functions are certainly needed. Furthermore, in the study by Somerville and colleagues (2000) BQSS
executive scores were unrelated to measures of basic attentional abilities. However, psychomotor retardation and slowing of
information processing could be part of the cognitive features of MCI patients with subcortical cerebrovascular disease
(O’Brien et al., 2003), and their contribution in the performance on the ROCF need to be further examined and compared
with the role of executive functions.
Limitations of our study need to be considered. First of all, the limited sample size reduced the statistical power, and thus
our results need to be taken cautiously, and further explored in larger population. Nevertheless, our preliminary findings are
to some extent supported by moderate effect sizes corrected for small samples biases.
A second limitation refers to the partially different composition of the neuropsychological batteries applied in the enrolling
centers, and thus the limited number of cognitive tests shared by the two protocols. The lack of a shared test of verbal mem-
ory reduced our possibility to compare and accurately characterize the cognitive profile of the two MCI groups. The availabil-
ity of only few shared tests of executive functions, predominantly measuring of inhibition and verbal retrieval processes,
further limited both the comparisons between the groups, and the possibility to extensively evaluate the convergent validity of
the BQSS executive scores.
A third limitation was the statistically significant disproportion on sex distributions between the two MCI groups. This
result was somewhat expected due to the different prevalence of cerebrovascular disease between men and women. Despite
the most part of the normative studies have found no significant sex differences in the ROCF performance, the disproportion
on sex distribution might have influenced the analyses on cognitive performances, and our results need to be confirmed in
sex-balanced populations.
Lastly, we do not have data on inter-rater agreement because we decided to apply a consensus evaluation, and to maximize
the standardization procedure of the qualitative scoring system. Moreover, the consensus meeting pointed out some potentially
misleading issues on the operationalization of the Fragmentation and Planning scoring guidelines. We then defined some addi-
tional rules whose implementation within the BQSS scoring criteria could be further examined and applied in future studies
on this qualitative scoring method.
In conclusion, our study demonstrated that when analyzing ROCF performance, it is important to discriminate between
executive and visuo-constructional abilities. The implementation of the BQSS for the evaluation of the ROCF into clinical
practice would enable to weigh up both these skills during the reproduction of this task. At present, no normative data are
available for the BQSS and, despite its good discriminant validity between healthy controls and dysexecutive patients, we are
not able to evaluate patients’ performance in the BQSS executive scores as normal or impaired according to a validated psy-
chometric cut-off. Furthermore, the availability of normative data for all the comprehensive set of scores of the BQSS, that
included among others also accuracy, neatness, placement, and retention scores, could additionally highlight the relative
weight of executive and visuo-constructional abilities in different patients populations. BQSS normative data could further
test the hypothesis of an executive impairment in subcortical vascular MCI patients, as it could be expected that many of these
patients would perform below the fifth centile of the normal population in the executive scores, but not in other accuracy and
retention scores. On the other hand, an opposite distribution of performances according to normative data would be expected
in degenerative MCI patients.
In a wider perspective, detailed, and possibly qualitative, analyses of patients’ performance in complex and multidimen-
sional neuropsychological tasks, such as ROCF, are able to provide useful cognitive information both for clinicians and re-
searchers. A better characterization of the cognitive profile of each patient could lead the clinician toward a diagnostic
hypothesis in the early phases of patient evaluation and, from a research point of view, could play an important role for the
identification of patterns of cognitive deficits of patients populations with different etiologies.

Funding

The VMCI-Tuscany study was supported by Tuscany region. E.S. has been supported by the RehAtt study funds. The
RehAtt study was funded by Tuscany region and Italian Ministry of Health under Grant Aimed Research Call 2010 (Bando
Ricerca Finalizzata 2010). The development of the neuropsychological battery of the Center for Cognitive Disorders and
Dementia in Parma was supported by funds (ex FIL 60%) to P.C. Local Funds for Research, University of Parma.

Conflict of interest

None declared.

Downloaded from https://academic.oup.com/acn/advance-article-abstract/doi/10.1093/arclin/acy010/4840584

by Mount Royal College user
on 09 February 2018
 E. Salvadori et al. / Archives of Clinical Neuropsychology (2018); 1–10 9

Conflict of Interest

All authors reports no disclosures.

References

Albert, M. S., DeKosky, S. T., Dickson, D., Dubois, B., Feldman, H. H., Fox, N. C., et al. (2011). The diagnosis of mild cognitive impairment due to
Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for
Alzheimer’s disease. Alzheimer’s & Dementia, 7, 270–279. doi:10.1016/j.jalz.2011.03.008.
Anderson, P., Anderson, V., & Garth, J. (2001). Assessment and development of organizational ability: The Rey Complex Figure Organizational Strategy
Score (RCF-OSS). The Clinical Neuropsychologist, 15, 81–94.
Bennett-Levy, J. (1984). Determinants of performance on the Rey-Osterrieth complex figure test: An analysis and a new technique for single case assessment.
British Journal of Clinical Psychology, 23, 109–119.
Bylsma, F. W., Bobholz, J. H., Schretlen, D., & Correa, D. D. (1995). A brief, reliable approach to coding how subjects copy the Rey-Osterrieth Complex
Figure. Journal of the International Neuropsychological Society, 1, 125.
Caffarra, P., Vezzadini, G., Dieci, F., Zonato, F., & Venneri, A. (2002a). Rey-Osterrieth complex figure: Normative values in an Italian population sample.
Neurological Sciences, 22, 443–447. doi:10.1007/s100720200003.
Caffarra, P., Vezzadini, G., Dieci, F., Zonato, F., & Venneri, A. (2002b). Una versione abbreviata del test di Stroop. Dati normativi nella popolazione italiana.
Nuova Rivista di Neurologia, 12, 111–115.
Cahn, D. A., Marcotte, A. C., Stern, R. A., Arruda, J. A., Akshoomoff, N. A., & Leshko, I. C. (1996). The Boston qualitative scoring system for the Rey-
Osterrieth Complex Figure: A study of children with attention deficit hyperactivity disorder. The Clinical Neuropsychologist, 10, 397–406.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates.
Cumming, G. (2012). Understanding the new statistics: Effect sizes, confidence intervals, and meta-analysis. New York: Routledge.
Cummings, J. L. (1993). Frontal-subcortical circuits and human behavior. Archives of neurology, 50, 873–880.
Dawson, L. K., & Grant, I. (2000). Alcoholics’ initial organizational and problem-solving skills predict learning and memory performance on the Rey-
Osterrieth Complex Figure. Journal of the International Neuropsychological Society, 6, 12–19.
Elderkin-Thompson, V., Kumar, A., Mintz, J., Boone, K., Bahng, E., & Lavretsky, H. (2004). Executive dysfunction and visuospatial ability among depressed
elders in a community setting. Archives of Clinical Neuropsychology, 19, 597–611. doi:10.1016/j.acn.2003.08.009.
Eslinger, P. J., & Grattan, L. M. (1990). Influence of organizational strategy on neuropsychological performance in frontal lobe patients. Journal of Clinical
and Experimental Neuropsychology, 12, 54.
Folbrecht, J. R., Charter, R. A., Walden, D. K., & Dobbs, S. M. (1999). Psychometric properties of the Boston Qualitative Scoring System for the Rey-
Osterrieth Complex Figure. Clinical Neuropsychology, 13, 442–449.
Freeman, R. Q., Giovannetti, T., Lamar, M., Cloud, B. S., Stern, R. A., Kaplan, E., et al. (2000). Visuoconstructional problems in dementia: Contribution of
executive systems functions. Neuropsychology, 14, 415–426.
Gauthier, S., Reisberg, B., Zaudig, M., Petersen, R. C., Ritchie, K., Broich, K., et al. (2006). Mild cognitive impairment. Lancet, 367, 1262–1270.
Hamby, S. L., Wilkin, J. W., & Barry, N. S. (1993). Organizational quality on the Rey–Osterrieth and Taylor Complex Figure Tests: A new scoring system.
Psychological Assessment, 5, 27–33.
Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta-analysis. San Diego, CA: Academic Press.
Javorsky, D. J., Rosenbaum, J., & Stern, R. A. (1999). Utility of the Boston Qualitative Scoring System (BQSS) for the Rey-Osterrieth Complex Figure in
the evaluation of traumatic brain injury. Archives of Clinical Neuropsychology, 14, 789–790.
Katz, S., Ford, A. B., Moskowitz, R. W., Jackson, B. A., & Jaffe, M. W. (1963). Studies of illness in the aged. The index of ADL: A standardized measure of
biological and psychosocial function. JAMA, 185, 914–919.
Lawton, M. P., & Brody, E. M. (1969). Assessment of older people: Self-maintaining and instrumental activities of daily living. The Gerontologist, 9, 179–186.
Mahurin, R., Eckert, S., Velligan, D., Hazelton, B., & Miller, A. L. (1997). Rey figure as a predictor of everyday functional ability in schizophrenia. Journal
of Neuropsychiatry and Clinical Neurosciences, 9, 36–37.
Measso, G., Cavarzeran, F., Zappalà, C., Lebowitz, B. D., Crook, T. H., & Pirozzolo, F. J. (1993). The Mini-mental State Examination. Normative study of
an Italian random sample. Developmental Neuropsychology, 9, 77–85.
Novelli, G., Papagno, C., Capitani, E., Laiacona, M., & Vallar, G. (1986). Tre test clinici di ricerca e produzione lessicale: taratura su soggetti normali.
Archivio di Psicologia, Neurologia e Psichiatria, 47, 477–506.
O’Brien, J. T., Erkinjuntti, T., Reisberg, B., Roman, G., Sawada, T., Pantoni, L., et al. (2003). Vascular cognitive impairment. Lancet Neurology, 2, 89–98.
Osterrieth, P. A. (1944). Le test de copie d’une figure complexe. Archives de psychologie, 30, 206–356.
Pantoni, L., Basile, A. M., Pracucci, G., Asplund, K., Bogousslavsky, J., Chabriat, H., et al. (2005). Impact of age-related cerebral white matter changes on
the transition to disability – the LADIS study: Rationale, design and methodology. Neuroepidemiology, 24, 51–62.
Pantoni, L., & Gorelick, P. (2011). Advances in vascular cognitive impairment 2010. Stroke, 42, 291–293.
Pantoni, L., Poggesi, A., & Inzitari, D. (2009). Cognitive decline and dementia related to cerebrovascular diseases: Some evidence and concepts.
Cerebrovascular Diseases, 27, 191–196.
Petersen, R. C., Caracciolo, B., Brayne, C., Gauthier, S., Jelic, V., & Fratiglioni, L. (2014). Mild cognitive impairment: A concept in evolution. Journal of
Internal Medicine, 275, 214–228.
Poggesi, A., Salvadori, E., Pantoni, L., Pracucci, G., Cesari, F., Chiti, A., et al. (2012). Risk and determinants of dementia in patients with mild cognitive
impairment and brain subcortical vascular changes: A study of clinical, neuroimaging, and biological markers – The VMCI-Tuscany Study: Rationale,
design, and methodology. International Journal of Alzheimer’s Disease. doi:10.1155/2012/608013.
Rey, A. (1941). L’examen psychologique dans les cas d’encephalopathie traumatique. Archives de Psychologie, 28, 286–340.
Sachdev, P. S., Brodaty, H., & Looi, J. C. (1999). Vascular dementia: diagnosis, management and possible prevention. Medical Journal of Australia, 170, 81–85.

Downloaded from https://academic.oup.com/acn/advance-article-abstract/doi/10.1093/arclin/acy010/4840584

by Mount Royal College user
on 09 February 2018
 10 E. Salvadori et al. / Archives of Clinical Neuropsychology (2018); 1–10

Salvadori, E., Poggesi, A., Pracucci, G., Inzitari, D., & Pantoni, L. (2015). Development and psychometric properties of a neuropsychological battery for mild
cognitive impairment with small vessel disease: The VMCI-Tuscany Study. Journal of Alzheimer’s Disease, 43, 1313–1323. doi:10.3233/JAD-141449.
Salvadori, E., Poggesi, A., Valenti, R., Pracucci, G., Pescini, F., Pasi, M., et al. (2016). Operationalizing MCI criteria in small vessel disease: The VMCI-
Tuscany Study. Alzheimer’s & Dementia, 12, 407–418. doi:10.1016/j.jalz.2015.02.010.
Savage, C. R., Baer, L., Keuthen, N. J., Brown, H. D., Rauch, S. L., & Jenike, M. A. (1999). Organizational strategies mediate nonverbal memory impairment
in obsessive-compulsive disorder. Biological Psychiatry, 45, 905–916.
Savage, C. R., Deckersbach, T., Wilhelm, S., Rauch, S. L., Baer, L., Reid, T., et al. (2000). Strategic processing and episodic memory impairment in obses-
sive compulsive disorder. Neuropsychology, 14, 141–151.
Scarpina, F., Ambiel, E., Albani, G., Pradotto, L. G., & Mauro, A. (2016). Utility of Boston Qualitative Scoring System for Rey-Osterrieth Complex Figure:
Evidence from a Parkinson’s Diseases sample. Neurological Sciences, 37, 1603–1611. doi:10.1007/s10072-016-2631-9.
Schreiber, H. E., Javorsky, D. J., Robinson, J., & Stern, R. A. (1999). Rey-Osterrieth Complex Figure performance in adults with attention deficit hyperactiv-
ity disorder: A validation study of the Boston Qualitative Scoring System. The Clinical Neuropsychologist, 13, 509–520.
Shin, M. S., Park, S. Y., Park, S. R., Seol, S. H., & Kwon, J. S. (2006). Clinical and empirical applications of the Rey-Osterrieth Complex Figure Test.
Nature Protocols, 1, 892–899. doi:10.1038/nprot.2006.115.
Shorr, J. S., Delis, D. C., & Massman, P. J. (1992). Memory for the Rey-Osterrieth Figure: Perceptual clustering, encoding, and storage. Neuropsychology, 6, 43–50.
Somerville, J., Tremont, G., & Stern, R. A. (2000). The Boston Qualitative Scoring System as a measure of executive functioning in Rey-Osterrieth Complex
Figure performance. Journal of Clinical and Experimental Neuropsychology, 22, 613–621. doi:10.1076/1380-3395(200010)22:5;1-9;FT613.
Stern, R. A., Javorsky, D. J., Singer, E. A., Singer Harris, N. G., Somerville, J. A., Duke, L. M., et al. (1999). The Boston Qualitative Scoring System for the
Rey-Osterrieth Complex Figure. Odessa, FL: Psychological Assessment Resources.
Waber, D. P., & Holmes, J. M. (1986). Assessing children’s memory productions of the Rey–Osterrieth Complex Figure. Journal of Clinical and
Experimental Neuropsychology, 8, 563–580.
Winblad, B., Palmer, K., Kivipelto, M., Jelic, V., Fratiglioni, L., Wahlund, L. O., et al. (2004). Mild cognitive impairment – beyond controversies, towards a
consensus: Report of the International Working Group on Mild Cognitive Impairment. Journal of Internal Medicine, 256, 240–246. doi:10.1111/j.1365-
2796.2004.01380.x.

Downloaded from https://academic.oup.com/acn/advance-article-abstract/doi/10.1093/arclin/acy010/4840584

by Mount Royal College user
on 09 February 2018