Introduction
A key aspect of neuropsychological assessment is to quantify cognitive performances, and to compare these to normative data sets using adjustment for age, sex, education, and/or ethnicity (Lezak et al., Reference Lezak, Howieson, Bigler and Tranel2012). While different standardized scores are applied, all use the normal distribution as the reference, provided the data of the test at hand are normally distributed. Such standard scores either express the performance in number of standard deviations below or above the normative statistical mean (e.g. Z, T, or Wechsler Standard Scores) or reflect the area under the curve for a specific cut-off, indicating the number of people in the normative sample performing below an individual’s score (i.e., percentiles) (see Figure 1). This psychometric approach is based on instruments and tests developed in an Anglo-Saxon or European tradition (Casaletto & Heaton, Reference Casaletto and Heaton2017), as opposed to more qualitatively driven assessment procedures that have been (and still are) influential in the North-Eastern European context, such as the assessment procedures developed by Alexander Luria (see Mikadze et al., Reference Mikadze, Ardila and Akhutina2019, for an overview). Even though specific tests and instruments require cultural, linguistic, or country-specific adaptations, the quantitative method is widely applied among neuropsychologists across the globe (Casaletto & Heaton, Reference Casaletto and Heaton2017; Chan et al., Reference Chan, Shum and Cheung2003; Delgadillo et al., Reference Delgadillo, Usuga, Landa, Valdés, Lasprilla and Irani2022; Lucas, Reference Lucas, Laher and Cockcroft2013).
Figure 1.
Overview of the different labeling and classification systems used internationally in relation to the normal distribution (percentages under the curve, standard deviations from the mean, and percentiles) and derived standard scores (Z, T, Wechsler Age-Adjusted Scaled Scores and IQ/Index scores). The consensus labels for use in the Netherlands and Flanders are highlighted in gray (cf. Bouma et al., Reference Bouma, Mulder, Lindeboom and Schmand2012).
Figure 1 Long description
A normal distribution curve illustrating percentages under the curve, standard deviations from the mean, and percentiles. The curve is symmetrical and bell-shaped, with the mean at the center. The x-axis represents standard deviations ranging from -4 to +4, while the y-axis represents percentages under the curve. Key points include cumulative percentages and percentiles, Z scores, T scores, Wechsler Standard Scores, and IQ/Index scores. Different labeling and classification systems are shown, with consensus labels for the Netherlands and Flanders highlighted in gray. The curve segments are labeled with specific ranges and corresponding classifications from various studies.
While there is global agreement on this psychometric-quantitative approach to neuropsychological assessment, large differences exist in the verbal, qualitative labels used to describe individual test scores, as well as the criteria for the boundaries to do so. For the interpretation of neuropsychological tests, the boundaries used in the Wechsler intelligence tests – resulting in 7 categories – are often applied (Wechsler, Reference Wechsler2008), but narrower categories have been introduced as well, especially for describing performances in the lower ranges (e.g., Schoenberg & Rum, Reference Schoenberg and Rum2017; Schretlen et al., Reference Schretlen, Testa and Pearlson2010). Different verbal labels have been used to describe the individual categories, with the “average range” as benchmark (e.g., low average, below average) or by applying labels such as “borderline,” “extremely low,” or “exceptionally low” (see Figure 1 for an overview). Using different labeling systems is confusing not only for (clinical) neuropsychologists and other healthcare professionals but also for the patients themselves and their loved ones. To overcome this, the American Academy of Clinical Neuropsychology (AACN) recommended the use of consensus labels for describing test scores with a normal distribution: exceptionally low, below average, average, high average, above average, and exceptionally high (Guilmette et al., Reference Guilmette, Sweet, Hebben, Koltai, Mahone, Spiegler, Stucky and Westerveld2020).
These recommendations aim to promote greater clarity and consistency in the field of neuropsychology. However, they are based on a consensus statement from a North American organization. It is therefore relevant to discuss the possible adaptation of these labels for broader international use, that is, in other cultures and other languages. To our knowledge, there is no global standard on reporting the result of neuropsychological tests. The handbook by Lezak and colleagues (2012) has been (and continues to be) influential in an international perspective, but many local handbooks and guidelines exist that follow a country’s or culture’s own best practice, historical development, or consensus on report writing and the labeling of neuropsychological test results. In some countries, the AACN consensus labels can be easily applied. For instance, the South African Clinical Neuropsychological Association recommended the use of the English-language AACN consensus labels and boundaries (Maganlal et al., Reference Maganlal, Truter and Shuttleworth-Edwards2021). However, in countries where English is not the primary or major language, adaptation of these labels may require more elaboration with respect to the exact meaning of the (translated) label or their more subtle implications.
Toward a global consensus?
Recently, a working group of the Norwegian Neuropsychological Society (NNS) has formulated guidelines that are based on the Guilmette et al. (Reference Guilmette, Sweet, Hebben, Koltai, Mahone, Spiegler, Stucky and Westerveld2020) recommendations. Their process analysis emphasized the need for a broad representation of professionals from different regions and ethnicities (i.e., the Sami indigenous people in Norway), and that members of the local neuropsychological societies should be involved. From a linguistic perspective, their findings also illustrated that literal translations of the English labels into other languages may not always result in easy-to-interpret descriptions. For instance, the Norwegian word for “average” uses the same word stem as the word for the statistical mean as a single point on a curve rather than as an area in the normal distribution. Similarly, the Norwegian word for “exceptional” is associated with an outstanding performance, also making the direct translation of this label confusing for poor performances. The eventual guidelines made available by the NNS thus carefully took these linguistic challenges into account, as well as comments and recommendations provided by their members regarding phrasing and wording (Raudeberg et al., Reference Raudeberg, Stangeland Maroni, Løvstad, Holth Skogan, Hansen, Olsen and Guilmette2025).
In the Netherlands, using the AACN labels in the Dutch language was already proposed shortly after publication of the Guilmette et al. (Reference Guilmette, Sweet, Hebben, Koltai, Mahone, Spiegler, Stucky and Westerveld2020) paper. In a position paper by Hendriks et al. (Reference Hendriks, Mol and Kessels2020), translated labels were proposed for describing normally distributed test outcomes in the Dutch language: uitzonderlijk laag, benedengemiddeld, laaggemiddeld, gemiddeld, hooggemiddeld, bovengemiddeld, and uitzonderlijk hoog. This proposal initiated a debate regarding the optimal terminology within the Dutch linguistic context. Concerns were raised that introducing these new labels might compete with the existing labels published in the widely used Dutch-language handbook on neuropsychological assessment (Bouma et al., Reference Bouma, Mulder, Lindeboom and Schmand2012), potentially causing confusion. Furthermore, feedback indicated that the Hendriks et al. (Reference Hendriks, Mol and Kessels2020) labels might be less distinct than the Bouma et al. (Reference Bouma, Mulder, Lindeboom and Schmand2012) labels. Specifically, the use of an adverb like “exceptionally” was considered to reflect the American culture, whereas in Dutch, the equivalent of “very” (zeer) would be considered more appropriate (Schmand & Bouma, Reference Schmand and Bouma2020). Additionally, the label “low average” (laaggemiddeld) could be interpreted as implying a worse performance than “below average” (benedengemiddeld) in Dutch (Van Zandvoort, Reference Van Zandvoort2020). These discussions highlighted regional and cross-cultural nuances that had not yet been fully considered, stressing the need for a more systematic approach to achieve consensus on reporting neuropsychological test outcomes in the Dutch language.
Coming to a consensus for the Netherlands and Flanders (Belgium)
Dutch (Nederlands) is a Germanic language that is an official language in six countries: the Netherlands and Belgium in Europe, Curaçao, Aruba, and Sint Maarten in the Caribbean, and Suriname in South America. Dutch is also an official language on the Dutch Caribbean islands Bonaire, Saba, and Sint Eustatius (which are not independent countries but “special municipalities” of the Netherlands). Most speakers of Dutch live in the Netherlands (about 17 million) and the Flanders region of Belgium (6.5 million), with a total estimate of about 24 million people speaking Dutch worldwide (Dutch Online Academy, 2025).
In the Netherlands, the Dutch Association of Psychologists (Nederlands Instituut van Psychologen; NIP) is the professional association for psychologists that is concerned with guidelines and recommendations for professional clinical practice, with its Section Neuropsychology concerned with neuropsychological clinical practice. In Flanders (Belgium), the Flemish Society for Clinical Psychology (Vlaamse Vereniging voor Klinisch Psychologen; VVKP), Division Neuropsychology is its equivalent. In Flanders, the Flemish Forum for Diagnostics (Vlaams Forum voor Diagnostiek; VFD) is an interdisciplinary working group that is concerned with psychodiagnostics, in which the VVKP takes part.
To come to a recommendation, the NIP Section Neuropsychology set up a task force “Uniformity in the quantitative descriptions of scores on performance tasks” in 2023. This task force set out a survey among the members of the NIP Section Neuropsychology, with the question to indicate the preference for either the Bouma et al. (Reference Bouma, Mulder, Lindeboom and Schmand2012) classification or the Hendriks et al. (Reference Hendriks, Mol and Kessels2020). Additionally, the option was given to indicate “no preference” or propose another classification system, and responders were asked to motivate their preference. 253 members responded, who had a diverse background in (clinical) neuropsychology, most of whom are working with (older) adults and a minority working in the field of child and adolescence neuropsychology. The majority of the responders indicated a preference for Bouma et al. (Reference Bouma, Mulder, Lindeboom and Schmand2012) (58.1%), 23.7% preferred Hendriks et al. (Reference Hendriks, Mol and Kessels2020), 14.2% had no preference, and 4% suggested an alternative. A large majority of the respondents stressed the importance of coming to a uniform labeling for the interpretation of test scores, and many also highlighted the importance that the labels should be easy to use not only for (clinical) neuropsychologists but also for other medical specialists and patients themselves. The motivations in favor of the labels by Bouma et al. (Reference Bouma, Mulder, Lindeboom and Schmand2012) stressed that the difference between the Dutch translations of the labels “above average” (bovengemiddeld) and “high average” (hooggemiddeld) and between “below average” (benedengemiddeld) and “low average” (laaggemiddeld) proposed by Hendriks et al. (Reference Hendriks, Mol and Kessels2020) was unclear. Some responders also deemed the label “exceptionally low” (extreem laag) too confronting for patients (see NIP, 2025).
In Flanders, Belgium, the VFD also set up a task force “Descriptive labels” in 2023. Members came from different disciplines, including (neuro)psychologists, physical therapists, and logopedics, and from different settings (both researchers, academic teachers, and clinicians). The aim of the task force was to come to a classification system that was both statistically and linguistically sound and feasible in clinical practice. They performed a literature review on how to describe normally distributed test scores, both nationally and internationally. The task force subsequently reviewed all labels from a Flemish perspective, identifying labels that may possibly result in misunderstanding. They thus adopted a different approach than the Netherlands task force. This VFD task force came to the conclusion that the Bouma et al. (Reference Bouma, Mulder, Lindeboom and Schmand2012) descriptive labels and boundaries would also work best in the Flemish context (VFD, 2024).
Next, the recommendations by the NIP and VFD task forces were published as a monodisciplinary (consensus-based) guideline in the Netherlands (NIP, 2025). For the Flanders region, the VFD only advised to use the new consensus labels, which is less mandatory than a guideline from a legal perspective, due to the multidisciplinary nature of the VFD (see, e.g., Hurwitz, Reference Hurwitz1995). Moreover, both task forces recommended including the labels and boundaries used for interpreting the test outcomes in the supplement or appendix of a patient’s (neuro)psychological report as a table or figure.
Discussion
With the publication of this report, implementation in the clinical practice of neuropsychologists is the next step in harmonizing reporting on neuropsychological test outcomes. Here, it should be stressed that neuropsychological assessment and diagnostics include more than merely describing test outcomes using uniform qualitative labels. Interpretation of the full assessment should consider the patient’s medical and personal history, the referral question (either from another healthcare professional or from the patients themselves), and the context or setting (e.g., medicolegal vs. healthcare focused). Furthermore, using uniform labels in reports from neuropsychologists does not automatically improve their understanding by either patients and their loved ones or other healthcare professionals.
In a recent study performed in a single center in the U.S., Smetana et al. (Reference Smetana, Picard, Parker, Bajo and Broshek2025) investigated the understanding of the AACN labels in a group of patients (n = 62), family members (n = 31), and other healthcare professionals (n = 34, mostly neurologists, psychiatrists, and internal medicine specialists). All participants were asked to rank the AACN labels as well as a patient-centered set of labels developed by the researchers from weakest to strongest. Interesting differences were found in the interpretations of these labels across these groups. For instance, most patients preferred the label “very high” to “exceptionally high,” whereas family members preferred “exceptionally high” to “very high.” Most healthcare professionals also preferred “exceptionally high” to either “very high” or “very superior.” Healthcare professionals also preferred “exceptionally low” to “very low” or “impaired.” Both patients and family members preferred “very low” for poor performances to “exceptionally low.” Based on these findings, Smetana et al. (Reference Smetana, Picard, Parker, Bajo and Broshek2025) suggested adopting a patient-centered approach in the communication with patients and their significant others to avoid misunderstanding, suggesting the following patient-centered labels: very high, well above average, high average, average, low average, well below average, and very low.
What the study by Smetana et al. (Reference Smetana, Picard, Parker, Bajo and Broshek2025) showed is that overall, the adverb “very” was preferred to “exceptionally” in English for weak performances, with a similar finding in the Dutch language (cf. Schmand & Bouma, Reference Schmand and Bouma2020; NIP, 2025). Still, in the Dutch-language context, no formal research has been done on the interpretation of the consensus (or any) labels for describing neuropsychological test results by patients, their significant others, or other healthcare professionals. Furthermore, it has recently been argued that the new AACN consensus labels have been constructed with a bias toward a forensic or medicolegal perspective (Loring et al., Reference Loring, Hermann, Meador, Lah, Goldstein and Bilder2025), as inconsistencies across labels used may affect the objectivity of the assessment (Guilmette et al., Reference Guilmette, Hagan and Giuliano2008). In clinical practice and in individuals with brain diseases specifically, there is often a need to classify test performances as “impaired” or “abnormal” compared to an individual’s prior (estimated) baseline. From this perspective, it could be argued that fine-grained labels such as “below average” or “very/exceptionally low” would, consequently, be less informative than, for instance, the label “impaired” in many settings (Loring et al., Reference Loring, Hermann, Meador, Lah, Goldstein and Bilder2025).
When it comes to implementing uniform classification in clinical practice, it is also crucial to include professionals from other disciplines with whom neuropsychologists collaborate and who refer patients to our services, such as neurologists or psychiatrists. Grand-round papers or handbook chapters expressing the field’s “expert opinion” on neuropsychological assessment may be helpful here, combined with clinical lessons for these disciplines (see, e.g., Cullum et al., Reference Cullum, Schaffert and Denney2024; Kessels & Hendriks, Reference Kessels, Hendriks, Friedman and Markey2023). Informing other disciplines may also overcome the problem that diagnostic criteria and consensus guidelines for specific diseases or disorders apply different boundaries and labels for the interpretation of neuropsychological test outcomes. For instance, vascular cognitive impairment (VasCog-2-WSO Criteria Consortium, 2025), MCI due to Parkinsons’s disease (Geurtsen et al., Reference Geurtsen, Hoogland, Goldman, Schmand, Tröster, Burn and Litvan2014), or HIV-associated neurocognitive disorder (Antinori et al., Reference Antinori, Arendt, Becker, Brew, Byrd, Cherner, Clifford, Cinque, Epstein, Goodkin, Gisslen, Grant, Heaton, Joseph, Marder, Marra, McArthur, Nunn, Price and Wojna2007) all apply their own cut-off scores and labels. In the field of mild cognitive impairment (MCI), multiple approaches are even being used, both for research and in clinical practice (see, e.g., Fountain-Zaragoza et al., Reference Fountain-Zaragoza, Braun, Horner and Benitez2021; Polcher et al., Reference Polcher, Wolfsgruber, Peters, Frölich, Wiltfang, Kornhuber, Hüll, Rüther, Lewczuk, Maier, Jessen and Wagner2022). The here-proposed uniform consensus labels using 7 categories could also be used to harmonize cognitive labels and classifications, adopting a transdiagnostic perspective.
Conclusion
Although the AACN international consensus labels provide a valuable foundation for describing normally distributed neuropsychological test results, direct translation into other languages and cultures often proves inadequate. Experiences from Norway, the Netherlands, and Flanders-Belgium illustrate that linguistically and culturally sensitive adaptations are essential for clinical clarity and patient-centered communication. These regions therefore developed their own terminologies, grounded in the AACN classification but tailored to local linguistic nuances, clinical practice, and communication with patients and other healthcare professionals.
These regional initiatives demonstrate that achieving global consensus on terminology remains challenging due to cultural, linguistic, and professional differences, even though uniformity within language areas (such as between the Netherlands and Flanders) appears feasible. While the AACN labels provide a strong conceptual foundation, they are not universally applicable without contextualization; simple translation does not suffice. Moreover, labels must be understandable and acceptable to both patients and healthcare providers, as certain terms (such as “exceptionally low”) may be perceived as confronting or clinically misleading. Overall, neuropsychology across the globe must strive for terminology that is unambiguous, psychometrically sound, and culturally appropriate, which can be achieved using regional task forces as illustrated in the current paper. The development of these adapted labels represents an important milestone, but their effective implementation in clinical practice, communication, and disorder-specific guidelines is yet to be realized. Ultimately, standardized labeling is not a goal in itself, but a fundamental prerequisite for more reliable, transdiagnostic classifications of cognitive impairment in both clinical practice and research.
You have
Access
Statement of Research Significance
Research Question(s) or Topic(s): Consensus criteria for the uniform labeling of normally distributed test scores have been proposed by the American Academy of Clinical Neuropsychology. However, adapting these labels for use in an international context faces both linguistic and cultural challenges. Here we present the process of reaching consensus on uniform labels in the Dutch language for use in the Netherlands and Belgium. Main Findings: The direct translations of the English-language labels into Dutch have been criticized both in the Netherlands and in the Dutch-speaking part of Belgium (Flanders), making adaptations necessary. Consensus labels in Dutch have been proposed by task forces in each country, which were subsequently formally adopted by the Netherlands Institute of Psychologists and the Flemish Society of Clinical Psychologists. Study Contributions: The process outlined in this commentary may help to come to a global standard of labeling neuropsychological test scores.
Introduction
A key aspect of neuropsychological assessment is to quantify cognitive performances, and to compare these to normative data sets using adjustment for age, sex, education, and/or ethnicity (Lezak et al., Reference Lezak, Howieson, Bigler and Tranel2012). While different standardized scores are applied, all use the normal distribution as the reference, provided the data of the test at hand are normally distributed. Such standard scores either express the performance in number of standard deviations below or above the normative statistical mean (e.g. Z, T, or Wechsler Standard Scores) or reflect the area under the curve for a specific cut-off, indicating the number of people in the normative sample performing below an individual’s score (i.e., percentiles) (see Figure 1). This psychometric approach is based on instruments and tests developed in an Anglo-Saxon or European tradition (Casaletto & Heaton, Reference Casaletto and Heaton2017), as opposed to more qualitatively driven assessment procedures that have been (and still are) influential in the North-Eastern European context, such as the assessment procedures developed by Alexander Luria (see Mikadze et al., Reference Mikadze, Ardila and Akhutina2019, for an overview). Even though specific tests and instruments require cultural, linguistic, or country-specific adaptations, the quantitative method is widely applied among neuropsychologists across the globe (Casaletto & Heaton, Reference Casaletto and Heaton2017; Chan et al., Reference Chan, Shum and Cheung2003; Delgadillo et al., Reference Delgadillo, Usuga, Landa, Valdés, Lasprilla and Irani2022; Lucas, Reference Lucas, Laher and Cockcroft2013).
Overview of the different labeling and classification systems used internationally in relation to the normal distribution (percentages under the curve, standard deviations from the mean, and percentiles) and derived standard scores (Z, T, Wechsler Age-Adjusted Scaled Scores and IQ/Index scores). The consensus labels for use in the Netherlands and Flanders are highlighted in gray (cf. Bouma et al., Reference Bouma, Mulder, Lindeboom and Schmand2012).
Figure 1 Long description
A normal distribution curve illustrating percentages under the curve, standard deviations from the mean, and percentiles. The curve is symmetrical and bell-shaped, with the mean at the center. The x-axis represents standard deviations ranging from -4 to +4, while the y-axis represents percentages under the curve. Key points include cumulative percentages and percentiles, Z scores, T scores, Wechsler Standard Scores, and IQ/Index scores. Different labeling and classification systems are shown, with consensus labels for the Netherlands and Flanders highlighted in gray. The curve segments are labeled with specific ranges and corresponding classifications from various studies.
While there is global agreement on this psychometric-quantitative approach to neuropsychological assessment, large differences exist in the verbal, qualitative labels used to describe individual test scores, as well as the criteria for the boundaries to do so. For the interpretation of neuropsychological tests, the boundaries used in the Wechsler intelligence tests – resulting in 7 categories – are often applied (Wechsler, Reference Wechsler2008), but narrower categories have been introduced as well, especially for describing performances in the lower ranges (e.g., Schoenberg & Rum, Reference Schoenberg and Rum2017; Schretlen et al., Reference Schretlen, Testa and Pearlson2010). Different verbal labels have been used to describe the individual categories, with the “average range” as benchmark (e.g., low average, below average) or by applying labels such as “borderline,” “extremely low,” or “exceptionally low” (see Figure 1 for an overview). Using different labeling systems is confusing not only for (clinical) neuropsychologists and other healthcare professionals but also for the patients themselves and their loved ones. To overcome this, the American Academy of Clinical Neuropsychology (AACN) recommended the use of consensus labels for describing test scores with a normal distribution: exceptionally low, below average, average, high average, above average, and exceptionally high (Guilmette et al., Reference Guilmette, Sweet, Hebben, Koltai, Mahone, Spiegler, Stucky and Westerveld2020).
These recommendations aim to promote greater clarity and consistency in the field of neuropsychology. However, they are based on a consensus statement from a North American organization. It is therefore relevant to discuss the possible adaptation of these labels for broader international use, that is, in other cultures and other languages. To our knowledge, there is no global standard on reporting the result of neuropsychological tests. The handbook by Lezak and colleagues (2012) has been (and continues to be) influential in an international perspective, but many local handbooks and guidelines exist that follow a country’s or culture’s own best practice, historical development, or consensus on report writing and the labeling of neuropsychological test results. In some countries, the AACN consensus labels can be easily applied. For instance, the South African Clinical Neuropsychological Association recommended the use of the English-language AACN consensus labels and boundaries (Maganlal et al., Reference Maganlal, Truter and Shuttleworth-Edwards2021). However, in countries where English is not the primary or major language, adaptation of these labels may require more elaboration with respect to the exact meaning of the (translated) label or their more subtle implications.
Toward a global consensus?
Recently, a working group of the Norwegian Neuropsychological Society (NNS) has formulated guidelines that are based on the Guilmette et al. (Reference Guilmette, Sweet, Hebben, Koltai, Mahone, Spiegler, Stucky and Westerveld2020) recommendations. Their process analysis emphasized the need for a broad representation of professionals from different regions and ethnicities (i.e., the Sami indigenous people in Norway), and that members of the local neuropsychological societies should be involved. From a linguistic perspective, their findings also illustrated that literal translations of the English labels into other languages may not always result in easy-to-interpret descriptions. For instance, the Norwegian word for “average” uses the same word stem as the word for the statistical mean as a single point on a curve rather than as an area in the normal distribution. Similarly, the Norwegian word for “exceptional” is associated with an outstanding performance, also making the direct translation of this label confusing for poor performances. The eventual guidelines made available by the NNS thus carefully took these linguistic challenges into account, as well as comments and recommendations provided by their members regarding phrasing and wording (Raudeberg et al., Reference Raudeberg, Stangeland Maroni, Løvstad, Holth Skogan, Hansen, Olsen and Guilmette2025).
In the Netherlands, using the AACN labels in the Dutch language was already proposed shortly after publication of the Guilmette et al. (Reference Guilmette, Sweet, Hebben, Koltai, Mahone, Spiegler, Stucky and Westerveld2020) paper. In a position paper by Hendriks et al. (Reference Hendriks, Mol and Kessels2020), translated labels were proposed for describing normally distributed test outcomes in the Dutch language: uitzonderlijk laag, benedengemiddeld, laaggemiddeld, gemiddeld, hooggemiddeld, bovengemiddeld, and uitzonderlijk hoog. This proposal initiated a debate regarding the optimal terminology within the Dutch linguistic context. Concerns were raised that introducing these new labels might compete with the existing labels published in the widely used Dutch-language handbook on neuropsychological assessment (Bouma et al., Reference Bouma, Mulder, Lindeboom and Schmand2012), potentially causing confusion. Furthermore, feedback indicated that the Hendriks et al. (Reference Hendriks, Mol and Kessels2020) labels might be less distinct than the Bouma et al. (Reference Bouma, Mulder, Lindeboom and Schmand2012) labels. Specifically, the use of an adverb like “exceptionally” was considered to reflect the American culture, whereas in Dutch, the equivalent of “very” (zeer) would be considered more appropriate (Schmand & Bouma, Reference Schmand and Bouma2020). Additionally, the label “low average” (laaggemiddeld) could be interpreted as implying a worse performance than “below average” (benedengemiddeld) in Dutch (Van Zandvoort, Reference Van Zandvoort2020). These discussions highlighted regional and cross-cultural nuances that had not yet been fully considered, stressing the need for a more systematic approach to achieve consensus on reporting neuropsychological test outcomes in the Dutch language.
Coming to a consensus for the Netherlands and Flanders (Belgium)
Dutch (Nederlands) is a Germanic language that is an official language in six countries: the Netherlands and Belgium in Europe, Curaçao, Aruba, and Sint Maarten in the Caribbean, and Suriname in South America. Dutch is also an official language on the Dutch Caribbean islands Bonaire, Saba, and Sint Eustatius (which are not independent countries but “special municipalities” of the Netherlands). Most speakers of Dutch live in the Netherlands (about 17 million) and the Flanders region of Belgium (6.5 million), with a total estimate of about 24 million people speaking Dutch worldwide (Dutch Online Academy, 2025).
In the Netherlands, the Dutch Association of Psychologists (Nederlands Instituut van Psychologen; NIP) is the professional association for psychologists that is concerned with guidelines and recommendations for professional clinical practice, with its Section Neuropsychology concerned with neuropsychological clinical practice. In Flanders (Belgium), the Flemish Society for Clinical Psychology (Vlaamse Vereniging voor Klinisch Psychologen; VVKP), Division Neuropsychology is its equivalent. In Flanders, the Flemish Forum for Diagnostics (Vlaams Forum voor Diagnostiek; VFD) is an interdisciplinary working group that is concerned with psychodiagnostics, in which the VVKP takes part.
To come to a recommendation, the NIP Section Neuropsychology set up a task force “Uniformity in the quantitative descriptions of scores on performance tasks” in 2023. This task force set out a survey among the members of the NIP Section Neuropsychology, with the question to indicate the preference for either the Bouma et al. (Reference Bouma, Mulder, Lindeboom and Schmand2012) classification or the Hendriks et al. (Reference Hendriks, Mol and Kessels2020). Additionally, the option was given to indicate “no preference” or propose another classification system, and responders were asked to motivate their preference. 253 members responded, who had a diverse background in (clinical) neuropsychology, most of whom are working with (older) adults and a minority working in the field of child and adolescence neuropsychology. The majority of the responders indicated a preference for Bouma et al. (Reference Bouma, Mulder, Lindeboom and Schmand2012) (58.1%), 23.7% preferred Hendriks et al. (Reference Hendriks, Mol and Kessels2020), 14.2% had no preference, and 4% suggested an alternative. A large majority of the respondents stressed the importance of coming to a uniform labeling for the interpretation of test scores, and many also highlighted the importance that the labels should be easy to use not only for (clinical) neuropsychologists but also for other medical specialists and patients themselves. The motivations in favor of the labels by Bouma et al. (Reference Bouma, Mulder, Lindeboom and Schmand2012) stressed that the difference between the Dutch translations of the labels “above average” (bovengemiddeld) and “high average” (hooggemiddeld) and between “below average” (benedengemiddeld) and “low average” (laaggemiddeld) proposed by Hendriks et al. (Reference Hendriks, Mol and Kessels2020) was unclear. Some responders also deemed the label “exceptionally low” (extreem laag) too confronting for patients (see NIP, 2025).
In Flanders, Belgium, the VFD also set up a task force “Descriptive labels” in 2023. Members came from different disciplines, including (neuro)psychologists, physical therapists, and logopedics, and from different settings (both researchers, academic teachers, and clinicians). The aim of the task force was to come to a classification system that was both statistically and linguistically sound and feasible in clinical practice. They performed a literature review on how to describe normally distributed test scores, both nationally and internationally. The task force subsequently reviewed all labels from a Flemish perspective, identifying labels that may possibly result in misunderstanding. They thus adopted a different approach than the Netherlands task force. This VFD task force came to the conclusion that the Bouma et al. (Reference Bouma, Mulder, Lindeboom and Schmand2012) descriptive labels and boundaries would also work best in the Flemish context (VFD, 2024).
Next, the recommendations by the NIP and VFD task forces were published as a monodisciplinary (consensus-based) guideline in the Netherlands (NIP, 2025). For the Flanders region, the VFD only advised to use the new consensus labels, which is less mandatory than a guideline from a legal perspective, due to the multidisciplinary nature of the VFD (see, e.g., Hurwitz, Reference Hurwitz1995). Moreover, both task forces recommended including the labels and boundaries used for interpreting the test outcomes in the supplement or appendix of a patient’s (neuro)psychological report as a table or figure.
Discussion
With the publication of this report, implementation in the clinical practice of neuropsychologists is the next step in harmonizing reporting on neuropsychological test outcomes. Here, it should be stressed that neuropsychological assessment and diagnostics include more than merely describing test outcomes using uniform qualitative labels. Interpretation of the full assessment should consider the patient’s medical and personal history, the referral question (either from another healthcare professional or from the patients themselves), and the context or setting (e.g., medicolegal vs. healthcare focused). Furthermore, using uniform labels in reports from neuropsychologists does not automatically improve their understanding by either patients and their loved ones or other healthcare professionals.
In a recent study performed in a single center in the U.S., Smetana et al. (Reference Smetana, Picard, Parker, Bajo and Broshek2025) investigated the understanding of the AACN labels in a group of patients (n = 62), family members (n = 31), and other healthcare professionals (n = 34, mostly neurologists, psychiatrists, and internal medicine specialists). All participants were asked to rank the AACN labels as well as a patient-centered set of labels developed by the researchers from weakest to strongest. Interesting differences were found in the interpretations of these labels across these groups. For instance, most patients preferred the label “very high” to “exceptionally high,” whereas family members preferred “exceptionally high” to “very high.” Most healthcare professionals also preferred “exceptionally high” to either “very high” or “very superior.” Healthcare professionals also preferred “exceptionally low” to “very low” or “impaired.” Both patients and family members preferred “very low” for poor performances to “exceptionally low.” Based on these findings, Smetana et al. (Reference Smetana, Picard, Parker, Bajo and Broshek2025) suggested adopting a patient-centered approach in the communication with patients and their significant others to avoid misunderstanding, suggesting the following patient-centered labels: very high, well above average, high average, average, low average, well below average, and very low.
What the study by Smetana et al. (Reference Smetana, Picard, Parker, Bajo and Broshek2025) showed is that overall, the adverb “very” was preferred to “exceptionally” in English for weak performances, with a similar finding in the Dutch language (cf. Schmand & Bouma, Reference Schmand and Bouma2020; NIP, 2025). Still, in the Dutch-language context, no formal research has been done on the interpretation of the consensus (or any) labels for describing neuropsychological test results by patients, their significant others, or other healthcare professionals. Furthermore, it has recently been argued that the new AACN consensus labels have been constructed with a bias toward a forensic or medicolegal perspective (Loring et al., Reference Loring, Hermann, Meador, Lah, Goldstein and Bilder2025), as inconsistencies across labels used may affect the objectivity of the assessment (Guilmette et al., Reference Guilmette, Hagan and Giuliano2008). In clinical practice and in individuals with brain diseases specifically, there is often a need to classify test performances as “impaired” or “abnormal” compared to an individual’s prior (estimated) baseline. From this perspective, it could be argued that fine-grained labels such as “below average” or “very/exceptionally low” would, consequently, be less informative than, for instance, the label “impaired” in many settings (Loring et al., Reference Loring, Hermann, Meador, Lah, Goldstein and Bilder2025).
When it comes to implementing uniform classification in clinical practice, it is also crucial to include professionals from other disciplines with whom neuropsychologists collaborate and who refer patients to our services, such as neurologists or psychiatrists. Grand-round papers or handbook chapters expressing the field’s “expert opinion” on neuropsychological assessment may be helpful here, combined with clinical lessons for these disciplines (see, e.g., Cullum et al., Reference Cullum, Schaffert and Denney2024; Kessels & Hendriks, Reference Kessels, Hendriks, Friedman and Markey2023). Informing other disciplines may also overcome the problem that diagnostic criteria and consensus guidelines for specific diseases or disorders apply different boundaries and labels for the interpretation of neuropsychological test outcomes. For instance, vascular cognitive impairment (VasCog-2-WSO Criteria Consortium, 2025), MCI due to Parkinsons’s disease (Geurtsen et al., Reference Geurtsen, Hoogland, Goldman, Schmand, Tröster, Burn and Litvan2014), or HIV-associated neurocognitive disorder (Antinori et al., Reference Antinori, Arendt, Becker, Brew, Byrd, Cherner, Clifford, Cinque, Epstein, Goodkin, Gisslen, Grant, Heaton, Joseph, Marder, Marra, McArthur, Nunn, Price and Wojna2007) all apply their own cut-off scores and labels. In the field of mild cognitive impairment (MCI), multiple approaches are even being used, both for research and in clinical practice (see, e.g., Fountain-Zaragoza et al., Reference Fountain-Zaragoza, Braun, Horner and Benitez2021; Polcher et al., Reference Polcher, Wolfsgruber, Peters, Frölich, Wiltfang, Kornhuber, Hüll, Rüther, Lewczuk, Maier, Jessen and Wagner2022). The here-proposed uniform consensus labels using 7 categories could also be used to harmonize cognitive labels and classifications, adopting a transdiagnostic perspective.
Conclusion
Although the AACN international consensus labels provide a valuable foundation for describing normally distributed neuropsychological test results, direct translation into other languages and cultures often proves inadequate. Experiences from Norway, the Netherlands, and Flanders-Belgium illustrate that linguistically and culturally sensitive adaptations are essential for clinical clarity and patient-centered communication. These regions therefore developed their own terminologies, grounded in the AACN classification but tailored to local linguistic nuances, clinical practice, and communication with patients and other healthcare professionals.
These regional initiatives demonstrate that achieving global consensus on terminology remains challenging due to cultural, linguistic, and professional differences, even though uniformity within language areas (such as between the Netherlands and Flanders) appears feasible. While the AACN labels provide a strong conceptual foundation, they are not universally applicable without contextualization; simple translation does not suffice. Moreover, labels must be understandable and acceptable to both patients and healthcare providers, as certain terms (such as “exceptionally low”) may be perceived as confronting or clinically misleading. Overall, neuropsychology across the globe must strive for terminology that is unambiguous, psychometrically sound, and culturally appropriate, which can be achieved using regional task forces as illustrated in the current paper. The development of these adapted labels represents an important milestone, but their effective implementation in clinical practice, communication, and disorder-specific guidelines is yet to be realized. Ultimately, standardized labeling is not a goal in itself, but a fundamental prerequisite for more reliable, transdiagnostic classifications of cognitive impairment in both clinical practice and research.
Acknowledegments
The authors would like to thank all involved. The NIP task force consisted of Rients Huitema, Petra Hurks, Rudolf Ponds, Martine Wilbers, and Marie-Noëlle Witjes-Ané. The Guidelines Committee of the NIP Division Neuropsychology consisted of Anne Buunk, Rhody Haalboom, Adinda Janse, and Vera van Stek-Smits. The VFD task force consisted of Veerle Briers, Eva Dierckx, Noortje Hermans, Sarah Huysegems, Steven Joris, Mark Schittekatte, Marlies Tierens, Tine Van Damme, Christel Van Vreckem, and Annelies Wagemans.
Funding Statement
Open access funding provided by Radboud University Nijmegen.
Competing interests
No authors have any conflicts of interest to declare.