Revisiting Rorschach test and Minnesota Multiphasic Personality Inventory-II patterns in Kraepelinian vs. DSM-wise Schizophrenia: How They Differ and What It Means for Diagnosis

Abstract

Objective

This retrospective chart review study aimed to investigate the differences in the Rorschach test and Minnesota Multiphasic Personality Inventory (MMPI)-II profiles among patients with Kraepelinian schizophrenia, those with DSM-wise schizophrenia, and controls. Kraepelinian schizophrenia is characterised by a chronic, deteriorative disease course and a predominance of negative symptoms.

Methods

Patients with Kraepelinian schizophrenia were selected based on medical record reviews. We then compared their Rorschach test and MMPI-II results with those of the DSM-wise schizophrenia group and the control group.

Results

The Rorschach test revealed a significant increase in DV2 score and a decrease in D score in patients with Kraepelinian schizophrenia compared to those with DSM-wise schizophrenia. In the MMPI-II profiles, patients with Kraepelinian schizophrenia exhibited an elevated L relative to those with DSM-wise schizophrenia.

Conclusion

Our results suggested the value of revisiting psychological tests in clinically delineated subgroups, such as Kraepelinian schizophrenia. Although patients fall under the same diagnostic category of schizophrenia, considering different phenotypes is important when interpreting psychological test outcomes. Additionally, our study indicated that both schizophrenia groups did not show as many abnormalities as expected compared to controls. This highlights the potential value of revisiting established profiles of certain psychological tests and calls for further research on other psychological tests.

Keywords

Rorschach test MMPI-II psychological assessment schizophrenia Kraepelinian

Introduction

Schizophrenia is a psychiatric disorder characterised by hallucinations, delusions, disorganised language and behaviour, blunted affect and negative symptoms. Since its initial conceptualisation by Emil Kraepelin, who pioneered psychiatric nosology, the concept of schizophrenia has undergone major and minor changes.¹ Despite the changes, schizophrenia remains a significant psychiatric disorder, accounting for a substantial worldwide disease burden.² Recent psychiatric research has employed several methodologies, including brain imaging,^3,4 electrophysiology⁵ and cutting-edge computational methods,⁶ for identifying the biological markers for schizophrenia. However, these studies have thus far yielded incomplete and inconsistent results.^7,8

Some researchers argue that the current diagnostic concept of schizophrenia itself has fundamental limitations, especially given the lack of consistency in research findings on the biological pathophysiology of schizophrenia.⁹ It has been suggested that the schizophrenia diagnosis, as outlined by the Diagnostic and Statistical Manual of Mental Disorders (DSM), encompasses several distinct conditions with differing disease pathologies.¹⁰ The debate over whether schizophrenia represents a single psychiatric disorder has been intertwined with the evolution of the diagnostic concept since its inception.^11,12 Kraepelin originally conceived schizophrenia to be a cognitive disorder that developed during adolescence and followed a chronic course. Hence, he coined the term ‘dementia praecox’ to differentiate schizophrenia from manic-depressive illness.¹³ Conversely, Eugen Bleuler did not view schizophrenia as a lifelong disorder. Therefore, he subsequently renamed it ‘schizophrenia’, emphasising the disrupted integration of the psyche from a cross-sectional view.¹⁴

The modern diagnostic criteria for schizophrenia, based on the DSM, seem to primarily draw from Bleuler's perspective. From DSM-III to DSM-5, the diagnostic criteria have shifted towards emphasising positive symptoms instead of a deteriorating disease course.^15,16 Although several issues are associated with the current diagnostic concept of schizophrenia, we will address them in another paper. Fortunately, the importance of cognitive function in schizophrenia has been increasingly recognised.^17–19 However, the evolving focus of schizophrenia diagnosis throughout history raises questions about the reliability of several pre-existing tools that have been standardised or validated decades ago. For example, the standardised tools for patients with schizophrenia emphasised positive symptoms in the past. Therefore, expecting these tools to be equally applicable to Kraepelinian schizophrenia, which places a greater emphasis on a deteriorating disease course, would be challenging.

These issues inevitably arise in the application of psychological tests, especially when the tools were developed long ago and have been in use for a while. It is well-known that the Minnesota Multiphasic Personality Inventory (MMPI)^20,21 and the Rorschach test, whose validity and utility for assessing psychosis have been supported by many studies, are utilised extensively in clinical settings.^22,23 Since the initial conceptualisation of schizophrenia, psychological tests have been invaluable for identifying cognitive distortions and formal thought disorders in patients, aiding clinicians in making diagnoses.^24,25 In South Korea, comprehensive psychological tests – including full battery assessments – routinely incorporate these methods for diagnosing psychiatric disorders, such as schizophrenia. However, as previously mentioned, these tools have a significant limitation in that they are mainly standardised for patients with schizophrenia diagnosed via the DSM. Most existing clinical studies using the Rorschach test or MMPI have targeted patients with schizophrenia diagnosed according to the DSM, or occasionally, patients with psychotic symptoms.

Given the recent emphasis on the importance of cognition in schizophrenia²⁶ and the identification of a cognitive endophenotype,²⁷ it is crucial to re-evaluate the results of the Rorschach test and MMPI in patients with schizophrenia. This is particularly important because these tests, developed and standardised in the past, continue to play a significant role in contemporary full battery assessments. Therefore, in this study, we examined the results of clinical psychological tests in a subgroup of patients with schizophrenia whose diagnosis can be considered Kraepelinian schizophrenia. We could thoroughly examine the differences in test results between DSM-based and Kraepelinian schizophrenia by investigating a relatively homogeneous patient group derived from a broad, heterogeneous diagnostic group. This might also assist us in reflecting on the current psychiatric nosology, particularly concerning the schizophrenia spectrum and other psychotic disorders.

Materials and methods

Study participants

Study participants who underwent clinical psychological tests at Asan Medical Center between 1 January 2013 and 30 June 2021 were assessed retrospectively. Clinical psychological tests included the Korean Wechsler Adult Intelligence Scale (K-WAIS-IV), Rorschach test and the Korean version of the MMPI-II. The participants in the patient group were selected by carefully reviewing their medical records. Among the patients diagnosed with schizophrenia according to the DSM diagnostic criteria, those who met the criteria for Kraepelinian schizophrenia were selected. These criteria included: 1) no remission of psychotic symptoms and 2) patients following a deteriorative course over time. On the other hand, if the patients did not meet the criteria for Kraepelinian schizophrenia, they were classified as DSM-wise schizophrenia.

On the contrary, the control group consisted of individuals who underwent military fitness tests. Among them, participants with psychotic symptoms that were reported at least once in their entire medical history, or for whom psychosis was clinically suspected, or those findings suggesting formal thought disorder or impairment of reality testing were confirmed in projective psychological tests were all excluded. To establish a control-to-case ratio of approximately 2:1, control participants were selected from those who underwent clinical psychological tests between 1 January 2017 and 30 June 2021. To enhance comparability between groups, we excluded patients with comorbid conditions that could be diagnosed concurrently with schizophrenia under the DSM and significantly affect the results of the Rorschach test and MMPI-II. Consequently, individuals with intellectual disabilities, developmental issues, or autism spectrum disorder were excluded from both the patient and control groups. This study was approved by the Institutional Review Board of Asan Medical Center (IRB number: 2021-1193). The requirement for obtaining informed consent was waived by the Institutional Review Board owing to the retrospective nature of the study. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.

Psychological tests

Rorschach test

The Rorschach test was developed by Hermann Rorschach, a student of Eugen Bleuler who coined the term schizophrenia. It is a projective test that evaluates the process of perception and association through the participant's response to ambiguous inkblots. This test has been expanded to include the perception of the association process and personality dynamics, encompassing emotions and interpersonal relationships. Historically, the test has been used to identify psychological processes that are related to thought and perceptual disturbances. It also helps in the differential diagnosis of schizophrenia,²⁸ as it can detect formal thought disorder by scoring peculiarities in the content and form of the test participant's spoken language. Hermann Rorschach claimed that the test could differentiate patients with psychosis from normal controls.²⁹ Subsequent studies have indicated that the Rorschach test effectively distinguishes psychosis with fair reliability, validity, and diagnostic efficiency.^30,31 Additionally, a meta-analysis has previously demonstrated the Rorschach test's utility in evaluating psychosis or formal thought disorder.³²

For study participants, the Rorschach test was performed according to the Rorschach Comprehensive System^22,33 by experienced clinical psychologists or clinical psychology trainees under the guidance of clinical psychologists. The test results were quantified into various outcome measures, which were further classified into several subgroups based on their characteristics. These subgroups included the special score section reflecting unusual or disorganised language or thought; the core section, which was first examined for a structured test interpretation; the affect section reflecting the role of emotions in the psychological organisation and functioning; the interpersonal section, reflecting how an individual views others or behaviour patterns; the ideation section reflecting general thinking patterns; the mediation section, reflecting the extent to which an individual perceives external reality when making decisions as opposed to being influenced by their mind; the processing section reflecting the type and efficiency of processing related to problem-solving; and the self-perception section reflecting how one thinks about their experiences. In addition, there are few indexes, such as the schizophrenia index (SCZI), depression index (DEPI) or coping deficit index (CDI), which are calculated from multiple variables. Table 1 presents the list of all Rorschach variables investigated.

Table 1.

List of the Rorschach test variables investigated.

Score	Description	Score	Description
Special score
DV1	Deviant verbalisation, level 1	DV2	Deviant verbalisation, level 2
INC1	Incongruous combinations, level 1	INC2	Incongruous combinations, level 2
DR1	Deviant response, level 1	DR2	Deviant response, level 2
FAB1	Fabulised combinations, level 1	FAB2	Fabulised combinations, level 2
ALOG	Inappropriate logic	CONTAM	Contamination
Core
R	Number of response	L	Lambda
EB(M)	Human movement response	EB (sumC)	Sum weighted colour
eb (FM + m)	Animal + Inanimate movements	eb (sumC’)	Achromatic colour response
Texture	Texture response
Affect
FC	Form-colour response	CF + C	Restrained forms of affective discharge
PureC	Pure colour response	Afr	Affective ratio
Space	White space response
Interpersonal
Food	Food response	Isol	Isolation index
Ideation
INTELL	Intellectualization index	M-	M with negative form quality
WSum6	Weighted sum of 6 special scores
Mediation
Popular	Popular response	S-%	Space use
X+%	Conventional form use	X-%	Distorted form use
Xu%	Unusual form use
Processing
Zf	Z frequency	Zd	Processing efficiency
W	Whole response	D	Common detail response
Dd	Unusual detail response	DQv	Synthesised response
Self-perception
3r + (2)/R	Egocentricity index	Fr + rf	Reflection response
FD	Form dimension response	An + Xy	Anatomy + X-ray contents
MOR	Morbid content
Index
SCZI	Schizophrenia index	DEPI	Depression index
CDI	Coping deficit	S-CON	Suicide constellation

Minnesota multiphasic personality inventory-ii

The MMPI was developed by Hathaway and McKinley in the 1940s. It is one of the significant personality tests widely used in mental health settings.³⁴ The original MMPI had been refined over the years.³⁵ Furthermore, much effort was dedicated to the development of items that were printed on the cards. The MMPI was initially developed by collecting questions that patients with various mental disorders responded to differently from the general population: the scales were constructed by contrasting the responses of the criterion group with a ‘normal’ group, and the construction of a scale was attempted using items with different proportions of endorsement among the groups.³⁶ Furthermore, the MMPI was revised to maintain continuity with the existing research base.²¹ Based on a larger normative sample, the MMPI-II provides 12 supplementary scales along with 15 additional content scales.³⁶ It also includes some revised MMPI contents. Currently, the MMPI-II, consisting of 567 yes-or-no questions, is in use.

Although it was initially assumed that each clinical scale could selectively discriminate a specific diagnosis, this initial development purpose was not adequately met. This is particularly true for schizophrenia, as some researchers have suggested that MMPI variables themselves lack clinical validity for diagnosing schizophrenia.³⁷ Nevertheless, the sixth and eighth clinical scales of the MMPI are still considered useful for detecting altered perception or disordered thought in patients with psychosis, although the Rorschach test is regarded as having better validity.³⁸ In this study, participants received the MMPI-II questionnaire before the formal psychological test, and all the participants responded with sufficient time. The MMPI-II results consist of nine validity scales and 10 clinical scales. In this study, the scores on seven out of nine validity scales as well as all 10 clinical scales were obtained, compared, and analysed.

Statistical analysis

First, demographic variables, including sex, age, and other factors, were compared between the Kraepelinian schizophrenia, DSM-wise schizophrenia, and control groups. Analysis of variance, Kruskal–Wallis H-test,³⁹ and Chi-squared test were applied depending on the type and normality of each variable. Then, the scores from the Rorschach test, as described above, were compared between these three groups. We adjusted for sex and age differences between groups. Since sex and age showed no multi-collinearity (variance inflation factor < 5), both variables were included as covariates. Additionally, the MMPI test profiles were also compared between these groups. Since these were T-scores, no additional adjustment was needed. The Bonferroni correction was used for addressing the issue of multiple comparisons. Because we aimed to focus on the difference between Kraepelinian and DSM-wise schizophrenia groups, we performed pairwise comparison. We first compared DSM-wise schizophrenia patients and controls, if there was statistically significant difference found in three groups comparisons and then also compared two schizophrenia groups. For pairwise comparisons, linear regression was applied where covariate adjustment was required; otherwise, Tukey's HSD test⁴⁰ or Wilcoxon rank sum test⁴¹ was applied. All statistical analyses were performed with Python Version 3.9.6 and SciPy 1.11.1.⁴²

Results

Study participants

A total of 38 patients with Kraepelinian schizophrenia, 50 patients with DSM-wise schizophrenia, and 87 controls without any finding suggestive of psychosis were included in this study. The average age of the Kraepelinian schizophrenia group was 39.1 [10.6] years, while that of the DSM-wise schizophrenia group was 32.9 [12.2] years and 21.2 [2.69] in control group (H = 81.5, p < 0.001). In terms of sex, 44.7% and 54.0% of study participants were female in the Kraepelinian schizophrenia and DSM-wise schizophrenia groups, respectively, while the control group consisted solely of male patients (chi = 59.1, p < 0.001). These demographic differences in age and sex were apparent due to the nature of the military fitness test. Additionally, the general intelligence measured by the K-WAIS-IV was 91.2 [14.3] in the Kraepelinian schizophrenia group, 90.2 [13.4] in the DSM-wise schizophrenia group, which was lower than 95.5 [13.2] in the control group (F = 2.89, p < 0.001). The demographic details are presented in Table 2.

Table 2.

Demographics and clinical characteristics in patients and controls.

	K-SPR(N = 38)	DSM-SPR(N = 50)	Control(N = 87)	Statistics	p-value
Age	39.1 [10.6]	32.9 [12.2]	21.2 [2.69]	81.5^a	< 0.001^d
Sex (Female)	17 (44.7)	27 (54.0)	0 (0)	59.1^c	< 0.001^d
FSIQ	91.2 [14.3]	90.2 [13.4]	95.5 [13.2]	2.89^b	0.058
VCI	99.5 [14.1]	100.2 [12.4]	100.7 [12.3]	0.13^b	0.879
PRI	95.9 [13.4]	92.7 [14.4]	98.8 [14.5]	2.97^b	0.054
WMI	96.6 [14.6]	95.6 [12.0]	98.7 [15.5]	0.82^b	0.440
PSI	81.8 [14.4]	81.4 [16.6]	89.3 [14.8]	5.64^b	0.004^d

Mean [Standard deviation]; Number (Ratio); K-SPR: Kraepelinian schizophrenia; DSM-SPR: DSM-wise schizophrenia; FSIQ: full-scale intelligence quotient; VCI: verbal comprehension index; PRI: perceptual reasoning index; WMI: working memory index; PSI: processing speed index; ^aKruskal–Wallis H-test; ^banalysis of covariance; ^cChi-squared test; ^dstatistically significant p < 0.05.

Rorschach test

Only the DV2 and D scores showed significant differences between the three groups after conservatively correcting for multiple comparisons using the Bonferroni correction (F = 12.38, p < 0.001; F = 27.98, p < 0.001). No other variables or indices showed a significant difference between Kraepelinian schizophrenia patients, DSM-wise schizophrenia patients, and controls. In the pairwise comparison between DSM-wise schizophrenia patients and controls, the D score in the processing section was elevated for DSM-wise schizophrenia, compared to controls (7.567 [4.133] vs. 6.628 [4.358], t = 5.886, p < 0.001). However, the Kraepelinian schizophrenia patients showed a decreased score of 6.400 [4.461], which was rather similar to that of the controls. In the pairwise comparison between Kraepelinian and DSM-wise schizophrenia groups, DV2 in the core section was only elevated in Kraepelinian schizophrenia compared to DSM-wise schizophrenia (0.133 [0.346] vs. 0 [0], t = -2.553, p = 0.013). Table 3 presents comparisons of the Rorschach variables between the three groups in detail.

Table 3.

Comparisons of the Rorschach test variables between patients and controls.

	K-SPR(N = 38)	DSM-SPR(N = 50)	Control(N = 87)	Statistics^a	adjustedp-value
Special score (Thought Disturbance)⁵
DV1	0.200 [0.484]	0.189 [0.462]	0.035 [0.240]	0.354	1.000
DV2	0.133 [0.346]	0 [0]	0 [0]	12.38	< 0.001^b
INC1	0.100 [0.305]	0.405 [0.686]	0.395 [0.740]	1.460	1.000
INC2	0.233 [0.679]	0.108 [0.393]	0.151 [0.448]	1.662	1.000
DR1	0.400 [0.855]	0.297 [0.777]	0.058 [0.281]	0.227	1.000
DR2	0.833 [2.437]	0.108 [0.393]	0 [0]	3.310	1.000
FAB1	0.067 [0.254]	0.216 [0.479]	0.326 [0.641]	2.502	1.000
FAB2	0.100 [0.403]	0.216 [0.630]	0.081 [0.382]	2.199	1.000
ALOG	0.767 [1.406]	1.162 [1.756]	0.151 [0.604]	5.513	0.231
CONTAM	0.200 [0.664]	0.162 [0.834]	0.023 [0.152]	1.105	1.000
Core
R⁵	14.63 [5.262]	18.16 [7.455]	18.05 [6.031]	1.878	1.000
L⁶	1.648 [2.034]	0.928 [1.053]	1.527 [2.740]	1.063	1.000
EB (M)⁴	2.500 [2.813]	3.444 [2.942]	2.860 [2.291]	2.055	1.000
EB (sumC)⁴	1.667 [1.621]	3.111 [3.169]	2.983 [2.454]	2.764	1.000
eb (FM + m)⁵	2.700 [2.215]	4.472 [2.635]	3.770 [2.896]	3.453	1.000
eb (sumC’)⁵	1.833 [1.599]	2.333 [2.426]	3.057 [3.013]	2.902	1.000
Texture⁵	0.167 [0.461]	0.139 [0.351]	0.402 [0.769]	1.328	1.000
Affect⁵
FC	0.600 [0.724]	1.361 [1.355]	1.460 [1.301]	4.154	0.826
CF + C	1.167 [1.392]	2.111 [2.605]	1.943 [1.961]	1.647	1.000
PureC	0.400 [0.675]	0.639 [1.496]	0.517 [0.951]	0.399	1.000
Afr	0.379 [0.185]	0.432 [0.183]	0.800 [3.820]	0.110	1.000
Space	1.467 [1.332]	1.556 [2.144]	1.966 [1.839]	0.204	1.000
Interpersonal
Food⁴	0.467 [1.042]	0.972 [1.665]	0.407 [0.726]	4.580	0.553
Isol³	0.087 [0.118]	0.156 [0.149]	0.145 [0.128]	3.345	1.000
Ideation
INTELL²	2.300 [3.771]	1.886 [3.027]	1.941 [2.112]	0.547	1.000
M-³	0.933 [1.596]	1.028 [1.404]	0.835 [1.404]	0.108	1.000
WSum6⁴	14.40 [23.75]	12.56 [20.38]	4.221 [5.914]	2.621	1.000
Mediation
Popular²	3.333 [1.988]	3.750 [2.511]	3.250 [1.488]	3.092	1.000
X+%³	0.458 [0.186]	0.438 [0.181]	0.427 [0.139]	0.687	1.000
X-%³	0.314 [0.158]	0.296 [0.171]	0.289 [0.153]	0.775	1.000
S-%³	0.219 [0.273]	0.108 [0.175]	0.163 [0.185]	2.058	1.000
Xu%³	0.178 [0.112]	0.239 [0.135]	0.266 [0.115]	1.201	1.000
Processing
Zf¹	7.833 [3.733]	10.57 [3.720]	10.72 [4.129]	4.611	0.540
Zd³	−1.350 [3.896]	−1.556 [5.629]	−1.071 [4.071]	0.157	1.000
W¹	6.367 [3.672]	8.833 [5.106]	8.860 [4.257]	2.715	1.000
D⁷	6.400 [4.461]	7.567 [4.133]	6.628 [4.358]	27.98	< 0.001^b
Dd⁰	2.100 [1.807]	2.533 [2.763]	2.694 [2.619]	0.382	1.000
DQv¹	1.433 [1.524]	1.433 [2.176]	1.221 [1.537]	0.054	1.000
Self-Perception
3r + (2)/R⁴	0.263 [0.205]	0.301 [0.185]	0.290 [0.161]	1.862	1.000
Fr + rf⁴	0.167 [0.592]	0.083 [0.368]	0.174 [0.654]	0.339	1.000
FD⁴	0.133 [0.346]	0.389 [0.549]	0.593 [0.886]	2.064	1.000
An + Xy⁴	1.133 [1.833]	0.833 [1.207]	1.523 [1.794]	1.199	1.000
MOR³	0.633 [1.671]	0.667 [1.095]	1.471 [2.130]	1.124	1.000
INDEX
SCZI⁴	3.000 [1.486]	2.750 [1.500]	2.233 [1.369]	4.133	1.000
DEPI⁴	3.500 [0.974]	3.583 [0.967]	3.884 [1.296]	0.883	0.843
CDI³	3.433 [1.431]	3.306 [0.951]	3.400 [1.177]	0.665	1.000
S-CON⁴	4.800 [1.627]	4.944 [1.433]	5.000 [1.564]	1.006	1.000

Mean [Standard deviation]; K-SPR: Kraepelinian schizophrenia; DSM-SPR: DSM-wise schizophrenia; ⁰30, ¹29, ²25, ³24, ⁴23, ⁵22, ⁶21, ⁷1 participants’ data are missing; ^aanalysis of covariance; ^bstatistically significant p < 0.05.

MMPI-II

No elevated (T > 65) scale was found in patients with either Kraepelinian schizophrenia or DSM-wise schizophrenia, while three clinical scales, D, Pt and Si, were elevated in controls. Figure 1 shows the MMPI profiles of study participants in summary. When comparing the three groups, five validity scales and seven clinical scales showed significant differences after applying the Bonferroni correction. Generally, the control group showed more significantly elevated validity and clinical scales than the two schizophrenia groups. Except for only one scale, all other scales showed a significant difference between the schizophrenia groups and controls, but not between two types of schizophrenia. In the pairwise comparison between Kraepelinian and DSM-wise schizophrenia groups, only the L scale showed a significant difference, with an elevated L in Kraepelinian schizophrenia (52.1 [10.8] vs. 46.9 [9.30], W = 2.067, p = 0.039). Controls had a mean L score of 44.7 [8.96], which was similar to that of DSM-wise schizophrenia patients. Table 4 provides a detailed comparison of MMPI variables between the three groups.

Figure 1.

MMPI-II profiles of patients and controls. K-SPR: Kraepelinian schizophrenia; DSM-SPR: DSM-wise schizophrenia; mean value of each scale is presented on the graphs; Kruskal–Wallis H-test or analysis of covariance; statistically significant *p < 0.05; **p < 0.01; ***p < 0.001. MMPI: Minnesota Multiphasic Personality Inventory.

Table 4.

Comparisons of the MMPI-II scales between patients and controls.

	K-SPR(N = 38)	DSM-SPR(N = 49)⁰	Control(N = 87)	Statistics	adjustedp-value
F	52.9 [13.5]	54.4 [11.1]	59.8 [13.1]	13.69^a	0.018^c
F(B)	51.7 [13.4]	50.7 [10.3]	62.6 [12.1]	19.99^b	< 0.001^c
F(P)	52.5 [11.9]	48.5 [8.51]	50.4 [9.76]	2.539^a	1.000
L	52.1 [10.8]	46.9 [9.30]	44.7 [8.96]	12.34^a	0.036^c
K	53.4 [12.7]	49.1 [10.8]	42.6 [8.63]	26.09^a	< 0.001^c
S	53.1 [13.5]	50.0 [11.1]	41.6 [9.50]	28.89^a	< 0.001^c
FBS¹	49.2 [10.0]	53.4 [11.3]	62.5 [10.4]	24.37^b	< 0.001^c
Hs	48.3 [10.0]	49.6 [9.52]	59.9 [11.3]	23.13^b	< 0.001^c
D	50.7 [10.9]	54.2 [11.6]	69.8 [12.5]	45.93^b	< 0.001^c
Hy	48.8 [9.92]	51.7 [10.7]	59.0 [9.42]	17.28^b	< 0.001^c
Pd	52.7 [9.65]	55.9 [12.8]	58.5 [12.3]	3.224^b	0.718
Mf	49.9 [10.5]	50.7 [9.33]	56.6 [9.32]	9.295^b	0.003^c
Pa	56.7 [13.8]	60.1 [15.3]	62.0 [13.8]	4.203^a	1.000
Pt	53.4 [11.2]	56.5 [12.2]	69.2 [13.7]	27.02^b	< 0.001^c
Sc	55.0 [12.2]	56.5 [11.7]	64.8 [12.5]	11.91^b	< 0.001^c
Ma	49.1 [11.9]	51.3 [12.4]	48.5 [9.61]	1.751^a	1.000
Si	53.0 [10.9]	56.1 [12.3]	69.9 [12.8]	33.99^b	< 0.001^c

All scores are T-scores; mean [standard deviation]; K-SPR: Kraepelinian schizophrenia; DSM-SPR: DSM-wise schizophrenia; ⁰1 participant is missing MMPI data; ¹2 participants’ data are missing; ^aKruskal–Wallis H-test; ^banalysis of covariance; ^cstatistically significant p < 0.05. MMPI: Minnesota Multiphasic Personality Inventory.

Discussion

In this study, we compared the results of the Rorschach test and MMPI-II in patients with Kraepelinian and DSM-wise schizophrenias, as well as controls. These tests have been widely used for detecting cognitive distortions and thought disorders in patients with schizophrenia. Particularly, in patients with schizophrenia, it has been demonstrated that the Rorschach is a valid and reliable instrument for thought disorder.^23,43 Many previous studies have reported the clinical usefulness of the Rorschach test in clinical settings for patients with psychosis.^28,31,44 A key distinction of this study from previous research is that we defined patients with Kraepelinian schizophrenia, which differs from the standard DSM diagnostic criteria for schizophrenia, by a careful and comprehensive review of their medical records. Studies focusing on specific phenotypes of schizophrenia are recommended because the diagnosis of schizophrenia is known to be heterogeneous. Furthermore, a taxonomic review of psychiatric diagnoses is required.^9,45

First, in the Rorschach test, patients with schizophrenia have long been known to show an increase in special scores, such as DV1 or DV2, which reflect disorganised language or thought. Notably, level 2 special scores, reflecting severely disorganised or derailed language, poor association processes or illogical reasoning, show a marked increase. Additionally, scores known as thought disorder indicators, such as INTELL, M- or WSum6, also rise in patients with schizophrenia. WSum6, calculated as the weighted sum of special scores, is especially supported as useful and valid in distinguishing patients with schizophrenia. High scores of WSum6 are associated with poor concentration, confused thoughts and unreasonable reasoning.⁴⁶ The X-% represents the proportion of responses with severe morphological distortions and can also be considered to suggest psychosis. For example, in a study designed to assess cognitive functioning using the Rorschach test in a clinical sample of adolescents, the psychotic group, including patients with schizophrenia, had higher X-%, M- and WSum6 than the two groups – one with patients at clinically high risk for psychosis and the other with non-psychotic patients.²⁸ Among few indices, the SCZI is known as an index to evaluate thought disorder, which is valid in determining the presence of psychosis.^31,47 Furthermore, its validity has been supported by recent studies.^31,48,49 In a recent study, patients with schizophrenia showed higher D, DQ and FQ-, while having lower DQ+, reflecting their distorted perception and poor recognition.⁵⁰ Another study showed that patients with schizophrenia showed significant differences in several different variables, including W, S, Dd, DQ, FQ, H, Fd and also special scores.⁵¹

The results of this study, however, varied greatly from those mentioned in previous studies. First, there was no significant difference in most Rorschach scores, including many special scores and ideation section scores, such as INTELL or X-%, between the two schizophrenia groups and controls. The DSM-wise schizophrenia group only showed a higher D than the control group. This discrepancy might be attributed to the fact that not only patients with Kraepelinian schizophrenia, who are generally considered chronic, but also those with DSM-wise schizophrenia, tended to produce responses of poor quality and quantity that were difficult to evaluate and score properly. Another possibility is that our sample size was insufficient to detect the differences, particularly given the wide range – indicated by a large standard deviation – of scores among participants. As a result, some Rorschach test variables, which had been previously considered as diagnostic for schizophrenia, did not significantly deviate from controls. Nevertheless, it is still noteworthy that DV2, which indicates a failure to select appropriate words, demonstrated a significant difference between the Kraepelinian and DSM-wise schizophrenia groups. This distinction suggests that linguistic anomalies might be important pathophysiological traits in some phenotypes of schizophrenia.⁵² A previous study also revealed that unusual verbalisation was a fairly reliable measure in patients with schizophrenia.⁵³ Interestingly, Kraepelinian schizophrenia patients exhibited decreased D scores compared to DSM-wise schizophrenia, with scores resembling those of controls. This might be because this chronic subtype of schizophrenia patients generally produces a fewer number of responses.

For the MMPI, the results were more unexpected. Based on previous studies, patients with schizophrenia were anticipated to have higher F, Pt and Sc scales than controls.^54,55 Specifically, F scores, known to be influenced by symptoms of psychosis, are highly associated with schizophrenia.⁵⁶ In addition to the Sc, the Pa scale is useful for differentiating patients with schizophrenia from non-clinical samples.⁵⁷ However, in this study, most validity and clinical scales were significantly increased in the control group, but not in the two schizophrenia groups. All scales were within normal ranges for these groups, suggesting that patients with schizophrenia might not experience substantial psychological distress. It has been previously proposed that patients with schizophrenia might display surprisingly normal profiles, differing from expectations.⁵⁸ Given these generally normal scales, it could be interpreted that patients with schizophrenia might be more familiar with their current psychiatric symptoms and not experience subjective psychological pain, especially during remission periods. Moreover, in the comparison between two schizophrenia groups, only the L scale showed a significant difference. Although the L scale was elevated in Kraepelinian schizophrenia patients compared to DSM-wise schizophrenia patients, it remained below 60. Thus, there might not be clinically significant meaning in these MMPI profile differences. It's possible that the MMPI profile isn’t valuable in distinguishing two different schizophrenia groups.

However, several limitations should be considered when interpreting our study results. First, standardised diagnostic criteria for Kraepelinian schizophrenia are not present. Hence, in this study, patients were selected according to an arbitrary criterion of not having symptom remission and a deteriorative disease course. Additionally, due to the nature of a retrospective study, these criteria were applied through medical record review rather than face-to-face interviews. However, from a historical and customary point of view, the criteria used in this study are sufficiently consistent with Kraepelin's original definition.^59–61 Furthermore, despite the retrospective nature of this study, the medical records of the entire treatment period were thoroughly reviewed to ensure the diagnostic accuracy of each patient with schizophrenia.

Second, the control group did not comprise truly healthy controls. It was challenging to recruit healthy participants as the study participants were retrospectively recruited from those who underwent clinical psychological tests at the university hospital. Therefore, we set the control group as participants who received military fitness tests. These participants could complain of some psychological distress and even experience depression or anxiety. However, participants with a medical history of psychotic symptoms, suspicious findings of thought disorder and reality testing impairment were excluded from clinical interview records or psychological tests. Therefore, such a control group was plausible considering the study's purpose to understand the results of clinical psychological tests in patients with schizophrenia. Nevertheless, the results of this study still must be carefully interpreted.

Third, due to the characteristics of the control group, there were marked differences in age and sex between the patient and control groups. As the control group included participants who had been evaluated for enlistment in the military services, all of them were young males. However, there were still significant differences in major profiles of the Rorschach test between the two groups after adjusting demographic variables. Furthermore, the MMPI profiles were T-scores and could therefore be adjusted initially.

Fourth, psychological test variables are not typically interpreted in isolation. This is especially true for Rorschach variables, which should be interpreted in conjunction with other related variables. Consequently, merely comparing a single variable to that of controls might not accurately reflect the genuine psychological patterns in schizophrenia patients. Nevertheless, certain variables possess inherent diagnostic value. We believe that direct head-to-head comparisons of these variables can still provide meaningful insights.

Lastly, we did not perform a power analysis to justify the sample size for this research. We recognise that a scientifically calculated number of participants is necessary to draw statistically significant and confirmatory findings. This is particularly important given the numerous variables under exploration in our study, including most of the Rorschach and MMPI variables. However, due to limitations in participant recruitment, we inevitably skipped the sample size justification. Nevertheless, we applied multiple comparisons corrections, which allowed us to draw conservative, preliminary findings, even without a pre-determined sample size.

In this study, both the Rorschach test and MMPI-II results revealed differences compared to previously known profiles found in patients with schizophrenia. Patients with either Kraepelinian or DSM-wise schizophrenia generally exhibited minimal differences in profiles from both psychological tests compared to controls. Specifically, for the MMPI, neither schizophrenia group consistently showed significant increases across clinical scales. An overall subdued response on the Rorschach test and a general flatness on most MMPI scales collectively suggest that patients with schizophrenia have a devastated and impoverished psyche. From a psychopathological perspective, instead of manifesting positive symptoms, these patients tend to show more negative symptoms.

These findings underscore the need for clinicians to exercise caution when interpreting psychological tests for patients with schizophrenia. The most recognised test findings might represent symptoms observed during the acute phase of the disease rather than the disease's innate characteristics. Furthermore, since DSM-wise schizophrenia has been focused on positive symptoms, traditional psychological test results may be skewed toward detecting these symptoms. Additionally, given the differences between Kraepelinian and DSM-wise schizophrenia, although minimal, clinical psychological test results may vary between patients in the later stages or with a deteriorating course of the disease compared to those in the acute phase. Lastly, although further research is recommended, the notable difference in the DV2 scale also suggested significant diagnostic implications regarding language-related impairments, particularly in the Kraepelinian schizophrenia group.

Conclusion

In this study, we observed that patients with schizophrenia, whether diagnosed by the classic Kraepelinian criteria or the modern DSM criteria, unexpectedly exhibited minimal deviations in the Rorschach test and MMPI-II results when compared to controls. This finding contrasts with previous studies. Although there were some limitations in the study design, our results suggest that it would also be beneficial to examine and revisit other classic psychological tests. Moreover, small disparities were noted in psychological test variables between the two distinct schizophrenia groups, with a significant difference in DV2 score from the Rorschach test. Given the recent perspective that posits cognitive dysfunction as a core pathophysiological feature of schizophrenia, there might be value in revisiting psychological tests in clinically delineated subgroups focusing on cognition, such as Kraepelinian schizophrenia. Therefore, further research on other well-known psychological tests in patients with Kraepelinian schizophrenia is highly recommended.

Footnotes

Acknowledgements

None.

Author contributions

Conceptualisation: Young Tak Jo and Ji Soo Lee; methodology: Young Tak Jo and Ji Soo Lee; formal analysis: Young Tak Jo; investigation: Young Tak Jo and Ji Soo Lee; resources: Young Tak Jo and Ji Soo Lee; writing—original draft: Young Tak Jo, Ji Soo Lee and Myung Joo Lee; writing—review & editing: Myung Joo Lee and Yeon Ho Joo.

Consent

Given that this study was a retrospective chart review, the risk to participants was minimal. Moreover, obtaining consent from all participants was not feasible due to its retrospective nature. Therefore, the Institutional Review Board at Asan Medical Center waived the requirement for written informed consent.

Data availability

Data sharing does not apply to this article due to local restrictions.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical considerations

This study was approved by the Institutional Review Board of Asan Medical Center (IRB number: 2021-1193) on 13 August 2021.

Funding

This study was supported by the National Research Foundation of Korea (NRF-2022R1F1A1067605).

ORCID iD

Young Tak Jo

References

Jablensky

. The diagnostic concept of schizophrenia: its history, evolution, and future prospects. Dialogues Clin Neurosci 2010; 12: 271–287.

Charlson

Ferrari

Santomauro

, et al. Global epidemiology and burden of schizophrenia: findings from the global burden of disease study 2016. Schizophr Bull 2018; 44: 1195–1203.

Joo

Yoon

, et al. Aberrant executive control and auditory networks in recent-onset schizophrenia. Neuropsychiatr Dis Treat 2020; 16: 1561–1570.

Joo

Kim

, et al. White matter impairments in patients with schizophrenia: a multisite diffusion MRI study. Progr Neuro-Psychopharmacology Biol Psychiat 2021; 111: 110381.

Perrottelli

Giordano

Brando

, et al. EEG-based measures in at-risk mental state and early stages of schizophrenia: a systematic review. Front Psychiatry 2021; 12: 653642.

de Filippis

Carbone

Gaetano

, et al. Machine learning techniques in a structural and functional MRI diagnostic approach in schizophrenia: a systematic review. Neuropsychiatr Dis Treat 2019; 15: 1605–1627.

Tandon

Keshavan

Nasrallah

. Schizophrenia, “just the facts”: what we know in 2008: part 1: overview. Schizophr Res 2008; 100: 4–19.

Tandon

Keshavan

Nasrallah

. Schizophrenia, “just the facts” what we know in 2008. 2. Epidemiology and etiology. Schizophr Res 2008; 102: 1–18.

Carpenter

. Deconstructing schizophrenia. Neuropsychobiology 2018; 75: 126–128.

10.

Joo

. Changes in the DSM diagnostic concept of schizophrenia: from 1980 to 2013. J Korean Neuropsych Assoc 2021; 60: 241–252.

11.

Insel

. Rethinking schizophrenia. Nature 2010; 468: 187–193.

12.

Jablensky

. The diagnostic concept of schizophrenia: its history, evolution, and future prospects. Dialogues Clin Neurosci 2010; 12: 271–287.

13.

Angst

. Historical aspects of the dichotomy between manic–depressive disorders and schizophrenia. Schizophr Res 2002; 57: 5–13.

14.

Berrios

. Eugen Bleuler’s place in the history of psychiatry. Schizophr Bull 2011; 37: 1095–1098.

15.

Moscarelli

. A major flaw in the diagnosis of schizophrenia: what happened to the Schneider's first rank symptoms. Psychol Med 2020; 50: 1409–1417.

16.

Kendler

. An historical framework for psychiatric nosology. Psychol Med 2009; 39: 1935–1941.

17.

McCutcheon

Keefe

RSE

McGuire

. Cognitive impairment in schizophrenia: aetiology, pathophysiology, and treatment. Mol Psychiatry 2023; 28: 1902–1918.

18.

Guo

Ragland

Carter

. Memory and cognition in schizophrenia. Mol Psychiatry 2019; 24: 633–642.

19.

Keefe

RSE

Fenton

. How should DSM-V criteria for schizophrenia include cognitive impairment? Schizophr Bull 2007; 33: 912–920.

20.

Hathaway

McKinley

. The Minnesota multiphasic personality inventory. Rev. ed., 2nd printing ed. University of Minnesota Press, 1943.

21.

Hathaway

McKinley

Butcher

. MMPI-2: Minnesota multiphasic personality inventory-2. University of Minnesota Press, 1989.

22.

Exner

Jr . The Rorschach: A Comprehensive System: Basic Foundations, Vol. 1. John Wiley & Sons, 1993.

23.

Kleiger

. Disordered thinking and the Rorschach: Theory, research, and differential diagnosis. Routledge, 2013.

24.

Kelley

Klopfer

. Application of the Rorschach method to research in schizophrenia. Rorschach Res Exc 1939; 3: 55–66.

25.

Rubin

. The Minnesota multiphasic personality inventory as a diagnostic aid in a veterans hospital. J Consult Psychol 1948; 12: 251–254.

26.

Hallmayer

Kalaydjieva

Badcock

, et al. Genetic evidence for a distinct subtype of schizophrenia characterized by pervasive cognitive deficit. Am J Hum Genet 2005; 77: 468–476.

27.

Maes

. Major neurocognitive psychosis: a novel schizophrenia endophenotype class that is based on machine learning and resembles Kraepelin’s and Bleuler’s conceptions. Acta Neuropsychiatr 2023; 35: 123–137.

28.

Ilonen

Heinimaa

Korkeila

, et al. Differentiating adolescents at clinical high risk for psychosis from psychotic and non-psychotic patients with the Rorschach. Psychiatry Res 2010; 179: 151–156.

29.

Teles

. Hermann Rorschach: from klecksography to psychiatry. Dement Neuropsychol 2020; 14: 80–82.

30.

Hilsenroth

Fowler

Padawer

. The Rorschach schizophrenia Index (SCZI): an examination of reliability, validity, and diagnostic efficiency. J Pers Assess 1998; 70: 514–534.

31.

Jorgensen

Andersen

Dam

. The diagnostic efficiency of the Rorschach depression Index and the schizophrenia Index: a review. Assessment 2000; 7: 259–280.

32.

Mihura

Meyer

Dumitrascu

, et al. The validity of individual Rorschach variables: systematic reviews and meta-analyses of the comprehensive system. Psychol Bull 2013; 139: 548–605.

33.

Exner

Jr . The Rorschach: A comprehensive system: Interpretation, Vol. 2. John Wiley & Sons, 1991.

34.

Piotrowski

. Assessment practices in the era of managed care: current status and future directions. J Clin Psychol 1999; 55: 787–796.

35.

Colligan

. History and development of the MMPI. Psychiatr Ann 1985; 15: 524–533.

36.

Helmes

Reddon

. A perspective on developments in assessing psychopathology: a critical review of the MMPI and MMPI-2. Psychol Bull 1993; 113: 453–471.

37.

Ritsher

. Association of Rorschach and MMPI psychosis indicators and schizophrenia spectrum diagnoses in a Russian clinical sample. J Pers Assess 2004; 83: 46–63.

38.

Dao

Prevatt

Horne

. Differentiating psychotic patients from nonpsychotic patients with the MMPI–2 and Rorschach. J Pers Assess 2008; 90: 93–101.

39.

Kruskal

Wallis

. Use of ranks in one-criterion variance analysis. J Am Stat Assoc 1952; 47: 583–621.

40.

Tukey

. Comparing individual means in the analysis of variance. Biometrics 1949; 5: 99–114.

41.

Wilcoxon

. Individual comparisons by ranking methods. Biometrics Bulletin 1945; 1: 80–83.

42.

Virtanen

Gommers

Oliphant

, et al. Scipy 1.0: fundamental algorithms for scientific computing in python. Nat Methods 2020; 17: 261–272.

43.

Viglione

Meyer

. An overview of Rorschach psychometrics for forensic practice. In: Gacono

Evans

(eds) Handbook forensic Rorschach assess. Routledge/Taylor & Francis Group, 2008, pp.21–53.

44.

Perry

Minassian

Cadenhead

, et al. The use of the Ego Impairment Index across the schizophrenia spectrum. J Pers Assess 2003; 80: 50–57.

45.

Ahmed

Strauss

Buchanan

, et al. Schizophrenia heterogeneity revisited: clinical, cognitive, and psychosocial correlates of statistically-derived negative symptoms subgroups. J Psychiatr Res 2018; 97: 8–15.

46.

Groth-Marnat

Wright

. Handbook of Psychological Assessment. John Wiley & Sons, 2016.

47.

Lindgren

. A comparison of the MMPI-2 and Rorschach inkblot technique in assessing schizophrenia. University of Maryland, 1995.

48.

Smith

Baity

Knowles

, et al. Assessment of disordered thinking in children and adolescents: the Rorschach perceptual-thinking Index. J Pers Assess 2001; 77: 447–463.

49.

Stokes

Pogge

Grosso

, et al. The relationship of the Rorschach schizophrenia Index to psychotic features in a child psychiatric sample. J Pers Assess 2001; 76: 209–228.

50.

Kishimoto

Yamamuro

Iida

, et al. Distinctive Rorschach profiles of young adults with schizophrenia and autism spectrum disorder. Neuropsychiatr Dis Treat 2016; 12: 2403–2410.

51.

Mondal

Kumar

. Rorschach inkblot test and psychopathology among patients suffering from schizophrenia: a correlational study. Indian J Psychiatry 2021; 30: 74–83.

52.

Park

, et al. Linguistic anomalies in the language of patients with schizophrenia. Schizophrenia Res: Cognition 2023; 31: 100273.

53.

Adair

Wagner

. Stability of unusual verbalizations on the Rorschach for outpatients with schizophrenia. J Clin Psychol 1992; 48: 250–256.

54.

Graham

. MMPI-2: Assessing Personality and Psychopathology. Oxford University Press, 2011.

55.

Hsu

Dai

, et al. Understanding MMPI-2 response structure between schizophrenia and healthy individuals. Front Psychiatry 2022; 13: 918999.

56.

Carter

Parnas

Cannon

, et al. MMPI variables predictive of schizophrenia in the Copenhagen high-risk project: a 25-year follow-up. Acta Psychiatr Scand 1999; 99: 432–440.

57.

Greene

. The MMPI-2: An interpretive manual. Allyn & Bacon, 2000.

58.

Bosch

Van Luijtelaar

Van Den Noort

, et al. The MMPI-2 in chronic psychiatric illness. Scand J Psychol 2014; 55: 513–519.

59.

Molina

Hernández

Sanz

, et al. Subcortical and cortical gray matter differences between Kraepelinian and non-Kraepelinian schizophrenia patients identified using voxel-based morphometry. Psychiat Res: Neuroimag 2010; 184: 16–22.

60.

Decker

. How Kraepelinian was Kraepelin? How Kraepelinian are the neo-Kraepelinians? — from Emil Kraepelin to DSM-III. Hist Psychiatry 2007; 18: 337–360.

61.

Kendler

. Kraepelin and the differential diagnosis of dementia praecox and manic-depressive insanity. Compr Psychiatry 1986; 27: 549–558.