https://link.springer.com/article/10.1007/s12207-025-09538-7 Skip to main content Advertisement Advertisement Springer Nature Link Log in Menu Find a journal Publish with us Track your research Search Cart 1. Home 2. Psychological Injury and Law 3. Article ChatGPT Helps Students Feign ADHD: An Analogue Study on AI-Assisted Coaching * Original Research * Open access * Published: 24 March 2025 * Volume 18, pages 97-107, (2025) * Cite this article Download PDF You have full access to this open access article [12207] Psychological Injury and Law Aims and scope Submit manuscript ChatGPT Helps Students Feign ADHD: An Analogue Study on AI-Assisted Coaching Download PDF * Anselm B. M. Fuermaier ORCID: orcid.org/0000-0002-2331-0840^1 & * Isabella J. M. Niesten^2 * 1091 Accesses * 18 Altmetric * 1 Mention * Explore all metrics Abstract This preregistered study aimed to assess whether AI-generated coaching helps students to successfully feign attention-deficit/ hyperactivity disorder (ADHD) in adulthood. First, based on questions generated by 22 students, we conducted an extensive ChatGPT query to develop a concise AI-generated information sheet designed to coach students in feigning ADHD during a clinical assessment. Second, we evaluated the effect of this coaching in an experimental analogue study in which 110 university students were randomly assigned to one of three groups: (1) a control group (n = 42), (2) an ADHD symptom-coached simulation group (n = 35), and (3) an AI-coached simulation group (n = 33). All participants underwent a clinical neuropsychological assessment that included measures of ADHD symptoms, functional impairments, selective attention, and working memory. Our preregistered data analysis revealed that the AI-coached simulation group consistently moderated their symptom overreporting and cognitive underperformance compared to the symptom-coached group in small to medium size, resulting in lower detection sensitivity. We conclude that publicly accessible AI tools, such as current versions of chatbots, can provide clear and effective strategies for feigning ADHD during clinical neuropsychological assessments, posing a significant threat to the validity assessments. We recommend that researchers and clinicians exercise caution when sharing assessment materials, example items, and scoring methodologies. Similar content being viewed by others [placeholde] GOLIAH (Gaming Open Library for Intervention in Autism at Home): a 6-month single blind matched controlled exploratory study Article Open access 22 March 2017 [44202_2024] Group cognitive behavioural therapy (CBT) for university students with attention-deficit/hyperactivity disorder (ADHD): a feasibility study Article Open access 04 March 2024 [978-3-319-] Using Serious Games to (Re)Train Cognition in Adolescents Chapter (c) 2017 Explore related subjects Discover the latest articles and news from researchers in related subjects, suggested using machine learning. * ADHD * Coaching * Cyberpsychology * Intelligence Augmentation * Psychological Testing * Artificial Intelligence Use our pre-submission checklist Avoid common mistakes on your manuscript. Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental disorder, which is characterized by developmentally inappropriate levels of inattentiveness, hyperactivity, and impulsivity. It is estimated that about half of the children with ADHD continue to meet diagnostic criteria for ADHD as adults, resulting in a global prevalence rate of persistent ADHD in adulthood of 2.58% (Song et al., 2021). The diagnostic assessment of ADHD in adults is primarily based on a comprehensive diagnostic interview aligned with major diagnostic classification systems such as the DSM-5 and ICD-11 (e.g., American Psychiatric Association, 2013 ; World Health Organization, 2019). Many clinicians rely heavily on self-reports of symptoms and perceived impairments when making the diagnosis (Fuermaier et al., 2024a, b; Nelson et al., 2014; Weis et al., 2019;). However, diagnostic guidelines recommend supplementing the interview with various tools, including self- and informant-reported questionnaires for symptoms and impairments, as well as neuropsychological performance tests, if deemed helpful (see Fuermaier et al., 2024a, b), to objectively assess cognitive functions (Sibley, 2021). A thorough, multimethod assessment is essential given the susceptibility of ADHD to distorted symptom presentation and symptom fabrication (see Dandachi-FitzGerald et al., 2024) for an explanatory framework of poor symptom validity). In this respect, symptom validity tests (SVTs) refer to the accuracy of symptomatic complaints, and performance validity tests (PVTs) refer to the validity of actual ability task performance (Larrabee, 2012). Performance validity is assessed either by stand-alone tests (developed for the sole purpose to assess validity) or embedded measures (derived from standard measures of cognition). A large body of evidence from extensive research over the past two decades on the clinical evaluation of adult ADHD highlights significant base rates (ranging from 9 to up to 27%) of symptom overreporting and cognitive underperformance as indicated by SVTs and PVTs (e.g., see Dong et al., 2023; Fuermaier et al., 2024a, b; Hirsch et al., 2022; Mascarenhas et al., 2023; Ovsiew et al., 2023; Phillips et al., 2023 ). Various factors can contribute to these forms of distorted symptom presentations and test performance in ADHD assessments (Dandachi-FitzGerald et al., 2024), with secondary gain motivations for exaggerating or feigning symptoms being especially significant among young adults, particularly university students (Booksh et al., 2010; Harrison & Edwards, 2010). These benefits can include extended time on exams or assignments, special accommodations or bursaries, access to stimulant medications, excuses for academic underperformance or unreliable behavior in social settings, and increased attention from peers (Dandachi-Fitzgerald et al., 2020; Fuermaier et al., 2021; Harrison, 2017; Rabiner, 2013). An important issue to consider is that individuals attempting to feign cognitive impairment and/or mental disorders, at least in many cases, prepare themselves in order to avoid being caught by validity measures (for an overview, see Garcia-Willingham et al., 2018). In most cases, this includes the preparation with information about the detection procedures and/or typical symptoms of the condition. In forensic contexts, research found that attorneys (or law students) felt ethically obligated to coach their clients about validity indices in psychological assessments (Wetter & Corrigan, 1995; Youngjohn, 1995). Similarly, in the context of adult ADHD assessments, research has found that examinees seek out online information to help them produce test results that mimic those of individuals genuinely diagnosed with ADHD while evading detection (Harrison, 2015). To capture the effects of coaching (also phrased as self-preparation, see Rogers, 2018a), research in analogue studies differentiates between symptom-coaching (disorder-based; providing information about the disorder) and test-coaching (strategy-based; providing insights about assessment tools and strategies to avoid detection, see Rogers, 2018a). Analogue (simulation) studies are experimental studies in which a clinical sample of individuals with genuine pathology is compared to one or more nonclinical samples of individuals who are randomly assigned to either a control group (instructed to show normal behavior) or to one of several experimental groups (instructed to feign, see Rogers et al., 2018a). In earlier analogue research, strategy-based coaching to avoid detection of feigning was given more relevance and was shown to have superiority in a study of Rogers et al. (1993) on feigned schizophrenic disorders. However, more recent findings indicate that neither disorder-based coaching nor detection-based coaching significantly improves success in feigning (Boskovic et al., 2022; Dunn et al., 2003; Fuermaier et al., 2017a, b). The scientific examination of coaching's impact on feigning ADHD is essential, as mental health information, including ADHD-related content, is nowadays readily accessible online (Zhao et al., 2022). Recent research using an analogue design has shown that individuals can easily find key information about ADHD and assessment tools through online searches, including both written and video-based content, which could be a serious threat to test security (Winter & Braw, 2024 ). These findings are concerning, as they highlight the potential risk of undermining the validity of an ADHD assessment when such online information is accessed and utilized effectively. Information accessible on the web serves as the foundation for content generated by artificial intelligence (AI) chatbots. Recent technological advances have made AI more functional, efficient, and accessible to a broader population, with tools such as ChatGPT (OpenAI, 2023) becoming especially popular among students for academic assistance. As AI chatbots evolve rapidly, it is essential for research to monitor its potential role in coaching individuals who may attempt to feign ADHD. Such coaching can influence feigning strategies (Crisan et al., 2023) and is crucial for consideration in both research and clinical practice. AI-generated coaching material has the potential to combine symptom-coaching with test-coaching, by providing the examinee concise and efficient information about the test materials being used, and give concrete behavioral advice on how to evade detection (see "Materials" section for a detailed presentation of AI-generated coaching material). A recent study of Lavigne et al. (2024) underlines this concern by showing that AI chatbots presented responses about how to feign on validity measures that were classified, at least for a proportion of the responses, as a serious threat to their test security. In this study, we conducted an extensive ChatGPT (version 4) query using a comprehensive set of questions generated by a sample of university students. Students were instructed to formulate questions for ChatGPT to explore its potential in coaching individuals on feigning ADHD during a brief clinical neuropsychological assessment. The AI-generated responses can be interpreted as a combination of advanced symptom- and test-coaching, as they provided information not only on ADHD and diagnostic instruments but also on how these instruments assess response validity and strategies to evade detection of feigning. The assessment in question was representative of typical ADHD diagnostic settings and included self-reports of symptoms and impairments, neuropsychological tests of attention and working memory, and validity measures in the form of SVTs and PVTs. The objectives of this study were twofold: (1) to assess the quality of information provided by ChatGPT for feigning ADHD, evaluated through a discussion of AI-generated content and its potential implications, and (2) to examine the practical utility of this information in an analogue study, where students were instructed to feign ADHD, analyzed through empirical data. In this analogue study, we introduced two experimental simulation conditions, in which individuals are coached either with DSM-5 diagnostic criteria for ADHD (symptom-coaching), or with an AI-generated coaching sheet (created by ChatGPT). We expected that both simulation groups would exhibit higher symptom reports and lower test performance compared to the control group. Further, we hypothesized that ChatGPT would effectively coach individuals on how to feign ADHD. This effectiveness would be evidenced if the AI-coached simulation group demonstrated more subtle symptom reporting and less excessively poor test performance than the symptom-coached group. Implications for test security, communication about psychological instruments, and the design of ADHD assessments will be discussed. Methods Participants One-hundred and twenty-five first-year psychology students of the University of Groningen were recruited via the first-year participant pool SONA, and were randomly allocated to one of three conditions: a control condition, a symptom-coached condition, and an AI-coached condition. Fifteen participants had to be excluded from data analysis because of incomplete data in large parts of the assessment due to administrative errors (n = 1), failed pre-experimental check (n = 1), failed post-experimental check (n = 12), or being nonrepresentative in age (31 years, n = 1). We did not define the exclusion of age at preregistration because we did not expect this high age in a first-year student sample. The remaining sample of 110 participants (90 females, 20 males) had a mean age of 19.6 years (SD = 1.3). None of the participants reported a formal diagnosis of ADHD in the present or past. However, 9 participants reported other psychiatric diagnoses, including dyslexia, mood disorders, anxiety disorders, autism, obsessive-compulsive disorder, personality disorder, PTSD, and eating disorder. We decided at preregistration not to exclude these conditions as they are presentative for a student population. The characteristics of the remaining sample of 110 individuals are presented in Table 1, separated by the control group (CG, n = 42), symptom-coached simulation group (S-SG; n = 35), and AI-coached simulation group (AI-SG; n = 33). We exceeded our minimally desired sample size of 30 per group, but did not reach our aimed group size of 40 because of strict exclusion criteria for the simulation groups. Table 1 Characteristics of participants per group Full size table Materials We developed a concise and representative set of instruments commonly used in adult ADHD assessment settings. This battery includes self-report scales for ADHD symptoms and functional impairment, as well as two neuropsychological tests for evaluating working memory and selective attention. In addition to standard measures of symptoms, functional impairment, and cognitive functions, this battery allows for the calculation of three embedded indicators of symptom validity (SVTs) and two embedded indicators of performance validity (PVTs). Conners' Adult ADHD Rating Scale (CAARS) The self-report-long form of the CAARS (CAARS-S:L; Conners et al., 1999) was applied to assess the severity of current adult ADHD symptoms. The original scale includes 66 items, each rated on a 4-point Likert scale ranging from 0 (not at all/never) to 3 (very much/very frequently). Sum scores of respective items are computed to obtain subscale scores. For the purpose of the present study, three primary scales for inattentive symptoms, hyperactive symptoms, and impulsive symptoms were analyzed. Furthermore, an Inconsistency Index is computed which is described to indicate careless or random responding. In addition to its traditional scales, two non-overlapping and independently developed symptom validity indicators are computed. The Infrequency Index (CII; Suhr et al., 2011) is calculated by summing up a number of item scores which were shown to be infrequently endorsed by healthy individuals and adults with credible ADHD in previous research. Previous studies found that sensitivity estimates of the CII ranged from 17 to 69% (Becke et al., 2021; Cook et al., 2016, 2018; Fuermaier et al., 2016; Robinson & Rogers, 2018; Walls et al., 2017). Specificity estimates of the CII ranged from 65 and 69% (Becke et al., 2021; Fuermaier et al., 2016) to 86% (Robinson & Rogers, 2018) and 95% (Walls et al., 2017). Second, the ADHD Credibility Index (ACI; Becke et al., 2021, 2022) was computed from additional items that are embedded and dispersed in the original 66 items of the CAARS. All items are based on previously described detection strategies for feigned symptom presentations (Rogers, 2018b), that is, selective symptoms, supposed symptoms, exaggerated symptoms, and symptom combinations. In an analogue study, the ACI distinguished feigned from genuine symptom report with a sensitivity of 44% and a specificity of 98% (Becke et al., 2021). Weiss Functional Impairment Rating Scale (WFIRS) The Weiss Functional Impairment Rating Scale (WFIRS) is a self-report measure for impairments that commonly occur in patients with ADHD and that are assumed to be of diagnostic value (CADDRA, 2017; Fuermaier et al., 2024a). The WFIRS comprises 70 items that are divided into seven domains: Family (8 items), Work (11 items; not considered in this student sample), School (11 items), Life Skills (12 items), Self-concept (5 items), Social (9 items), and Risk (14 items). Each item is scored on a four-point Likert scale ranging from 0 to 3 (0 = never, not at all; 1 = sometimes, somewhat; 2 = often, much; 3 = very often, very much). An additional answering option is given with Not Applicable. A scale score per domain is calculated by summing up the responses to all items per domain, and dividing this sum by the number of endorsed items. The WFIRS was reported to have high internal consistency with Cronbach's alpha > 0.8 for each domain. Perceptual and Attention Functions-Selective Attention (WAFS) The Perceptual and Attention Functions-Selective Attention (WAFS; Schuhfried, 2013; Sturm, 2006) was administered as a measure of selective attention and to derive an embedded validity indicator. In this test, a total of 144 geometric stimuli (triangle, circle, and square) that may get darker or lighter or stay the same were presented. Participants are asked to react to 30 target stimuli (i.e., a circle becomes darker, a circle becomes lighter, a square becomes darker, and a square becomes lighter) as quickly as possible and ignoring distracting stimuli. In the present study, recorded outcome measures included reaction time (RT) in milliseconds, dispersion of reaction time (SDRT), number of omissions, and number of commissions. The internal consistency (Cronbach's a) of the main variables was reported to be 0.95. Subsequent research on the WAFS showed the omission and commission errors of the WAFS to be promising candidates as embedded validity indicators (Becke et al., 2023; Dong et al., 2023). Reliable Digit Span (RDS) The Reliable Digit Span (RDS, Greiffenstein et al., 1994) was derived from the Digit Span (DS) subtest of the Wechsler Adult Intelligence Scale-Third edition (Wechsler, 2008) and is computed by summing the longest forward string forward and the longest string backward without error on both trials. The RDS is an established embedded PVT with high specificity and moderate sensitivity. We followed the recommendations of Bing-Canar et al. (2022) suggesting a cut score indicating likely noncredible performance in the assessment of adult ADHD. Coaching Information Symptom-coaching Participants in the symptom-coaching condition were provided with an information sheet on ADHD diagnostic criteria as outlined in the DSM-5. We prepared a DSM-based information sheet as it can be found by a quick online search (can be accessed from the authors by reasonable request). AI-coaching An AI-generated information sheet of four pages was created using ChatGPT (free-to-use version 4, October 2024). To develop this, we prepared a set of questions that one might pose to ChatGPT when seeking advice on how to feign ADHD in a clinical neuropsychological assessment. To ensure a representative collection of questions, we asked 22 university students to list all the questions they would use to gather relevant information for feigning ADHD. This initial set was supplemented by additional questions from our research group, resulting in a total of 87 questions. Using this input, ChatGPT generated a 190-page document containing AI-generated responses. To create a more concise and practical information sheet, we asked ChatGPT to summarize the content into a maximum of 2000 words. We refined our request by specifying the diagnostic instruments used in our study. This context was provided because these instruments are widely utilized, readily accessible through online searches, and had already been referenced in ChatGPT's comprehensive document. Additionally, we prompted ChatGPT to include information on (1) the primary goal of the diagnostic assessment, (2) the specific diagnostic instruments used, (3) how to align responses and behavior with those of individuals genuinely diagnosed with ADHD, (4) how the instruments contribute to assessing response validity, and (5) strategies to avoid detection of feigning. Notably, we explicitly refrained from correcting or adding any information to the content generated by ChatGPT, as our goal was to obtain a purely AI-generated coaching material. The content of the AI-generated coaching sheet is discussed in the following sections and is available upon reasonable request from the corresponding author. For test security reasons, AI-generated coaching information is not enclosed in this article, but can be accessed from the authors by reasonable request. Pre- and Post-experimental Check Participants in the simulation condition completed a pre-experimental check to ensure that all participants appropriately understood the instructions to feign ADHD. Participants had to give brief verbal summary of the instructions and were prompted to read the instructions again up until they showed sufficient understanding of the given instructions (and were excluded otherwise). In a post-experimental check, it was ensured that participants did, indeed, act according to the experimental instructions. The post-experimental check consisted of four questions asking participants whether they were compliant with instructions, with how much effort (ranging from 1, hardly, to 5, very much) they rated their attempt to feign ADHD, their perceived success (ranging from 1, hardly, to 5, very much), and an open question on their applied strategies to feign ADHD. Participants were excluded based on the post-experimental check when they indicated not having been compliant with instructions and/or rated their effort to feign ADHD with <= 3. Procedure This study followed the ethical principles outlined in the Declaration of Helsinki and received approval from the Ethics Board of the Department of Psychology, University of Groningen, the Netherlands (file number: PSY-2324-S-0407; approval date: July 1, 2024). The design and analysis plan was preregistered at AsPredicted (#204901; https://aspredicted.org/fdjm-tsb8.pdf). Participants were informed about the study in advance, were provided written informed consent, and were debriefed upon completion. The study was advertised within the first-year participant pool (SONA) of the Department of Psychology as research on ADHD assessment, without mentioning feigning to prevent prior preparation. Participation was voluntary, could be withdrawn at any stage during the assessment, and was compensated with study credits. Assessments took place in November and December 2024. All participants were assessed by one of three trained psychology Master students of the University of Groningen under the supervision of their two thesis supervisors and authors of this report. Participants were randomly allocated to either the control condition, symptom-coached, or AI-coached simulation condition. Participants in the control condition were instructed to perform the tests (WAFS, RDS) to the best of their abilities, and complete the questionnaires (CAARS, WFIRS) to the best of their knowledge. Instruments were given in the fixed order WAFS, RDS, CAARS, and WFIRS. Cognitive tests were always administered first to avoid potential negative effects of cognitive exhaustion if they were placed last. Participants in the simulation condition received the instruction to complete the assessment protocol as if they pretended to suffer from ADHD. To get acquainted with this instruction, participants in the simulation conditions received a one-page vignette that emphasized the potential benefits of feigned ADHD and instructions for taking on this role realistically. After this vignette, participants in the simulation conditions were given their respective coaching information, i.e., DSM diagnostic criteria or AI-generated coaching. Participants were instructed to read the instructions carefully (with the vast majority of participants returning from the instructions within 10 min). Afterwards, they completed the tests and questionnaires. A pre- and post-experimental check assessed whether experimental instructions were understood sufficiently, and whether participants adhered to the instruction of their experimental condition. Statistical Analysis We preregistered this study (https://aspredicted.org/fdjm-tsb8.pdf; # 204901) and reported explicitly if we deviated from the preregistered analysis plan. Questionnaire scores and test performance were presented in descriptive statistics per group. Because of nontrivial violations of assumptions of normality and sphericity, groups were compared with nonparametric statistics instead of with ANOVAs that rely on parametric assumptions. The Kruskal-Wallis test was deemed an adequate nonparametric alternative to an ANOVA, and was computed to assess differences between the three groups. Bonferroni-corrected pairwise Dunn's tests follow significant omnibus effects. In addition to the existence of group effects, we calculated effect sizes for the magnitude of findings. In this field of research, effect size Cohen's d is a common metric to indicate the magnitude of effects in pairwise comparisons. We calculated Cohen's d and interpreted the values based on Cohen's standards as negligible (d < 0.20), small (0.20 <= d < 0.50), medium (0.50 <= d < 0.80), and large (d >= 0.80; Cohen, 1988). Detection rates of feigned ADHD (sensitivity) were calculated for all SVTs and PVTs separately for the symptom-coached and AI-coached simulation groups, and the results were compared using Chi-squared tests. Other classification statistics, such specificity, positive predictive value, or negative predictive value, cannot be derived because of a lack of a clinical group with confirmed credible data. Results AI-Generated Coaching Information ChatGPT generated an information sheet of approximately 1400 words in less than four pages. The document was structured as requested and was easy for laypeople to follow. It contained accurate information about the purpose of the assessment instruments, including example items and scoring methods. The AI-generated text also provided specific details, such as how individuals with ADHD typically respond to these items, behave during neuropsychological tests, and the domains in which they often exhibit difficulties. Notably, the text did not include cutoffs indicating clinical relevance or suspicious responses, except for the number of ADHD symptoms required for diagnosis under the DSM-5 criteria. Furthermore, it offered insights into how the instruments detect feigning, such as by analyzing response consistency, extreme questionnaire scores, and unusually poor test performance. Aligned with this information, the text explicitly and repeatedly cautioned against exaggerating symptoms or performing excessively poorly on tests. Instead, it emphasized the importance of consistency and subtlety in presenting alleged problems to appear more credible and convincing. Effects of AI-Generated Coaching in the Analogue Study Table 2 provides descriptive statistics for all instruments, separately for the control group, the symptom-coached simulation group (S-SG), and the AI-coached simulation group (AI-SG). Significant group differences were identified for all but one measure (CAARS Inconsistency). As expected from an experimental analogue design and in line with preregistration predictions, individuals in both simulation groups altered their typical behavior, demonstrating significantly higher scores in ADHD-related symptoms and impairments and lower performance scores on neuropsychological tests (with the exception of self-concept on the WFIRS, as shown in Table 3). Table 2 Symptom scores and test performance per group Full size table Table 3 Post hoc pairwise comparisons (Dunn-Bonferroni) with effect sizes (Cohen's d) Full size table When comparing the two simulation groups, the AI-SG exhibited more nuanced symptom and impairment reports, as well as less excessively poor test performance compared to the S-SG, with the exception of reaction times (RT) on the WAFS, where the AI-SG demonstrated the longest reaction times. However, these differences did not reach significance under Bonferroni-corrected thresholds. Effect sizes for pairwise comparisons between the AI-SG and S-SG ranged from negligible to large, with most falling in the small to medium range. Notable effects in the medium to large range were observed for CAARS-Hyperactivity, RTSD, and commissions on the WAFS (favoring the AI-SG), as well as RT on the WAFS (favoring the S-SG). The lack of significance for these small to borderline large effects could be attributed to insufficient statistical power. A post hoc power analysis for Bonferroni-corrected pairwise comparisons revealed adequate power, based on our sample size, only for large effects (85% for d = 0.8), while power was insufficient for medium or smaller effects (44% for d = 0.5). Sensitivity rates of SVTs and PVTs are presented in Table 4. The detection rates for SVTs (overreporting) and PVTs (underperformance) were consistently higher for the S-SG than for the AI-SG, which indicates more excessive symptom report and overly poor test performance of the S-SG compared to the AI-SG. However, of note, the differences turned significant only for the commissions of the WAFS. Table 4 Detection rates (sensitivity) of symptom and performance validity measures Full size table Discussion Information on ADHD and validity assessments is readily accessible online and, when used effectively, may compromise the integrity of validity assessments in clinical and forensic settings (Winter & Braw, 2024; Zhao et al., 2022). Recent advancements in AI technology exacerbate this concern, as AI tools are rapidly evolving, widely accessible, and easy to use, even without prior expertise (Lavigne et al., 2024). The findings of the current study support this growing concern. We conducted an extensive ChatGPT query to develop a concise AI-generated information sheet designed to coach students in feigning ADHD during a clinical assessment, and evaluated the effect of this coaching in an experimental analogue study compared to traditional symptom-coaching. AI-Generated Coaching Material We demonstrated that ChatGPT provides useful information for students attempting to feign ADHD in a clinical neuropsychological assessment. The AI-generated script presented information in a concise and accessible format and explicitly highlighted the importance of consistency and avoiding excessive exaggeration to enhance the plausibility of faking attempts. Additionally, the current version of ChatGPT provides information on commonly used instruments, including example items and their scoring methods. However, we noted that the AI-generated script did not include cut-off scores for SVTs and PVTs. This highlights the critical importance of exercising caution when providing detailed descriptions of SVTs and PVTs in scientific articles to protect test security. Moreover, it is important to acknowledge that we specified the diagnostic instruments used in our study within our prompt to ChatGPT to generate a summary of the coaching material, along with additional instructions on how to structure the summary (see Methods). We, therefore, conclude that generating relevant information from ChatGPT is relatively easy only when the user possesses some prior knowledge about the topic or the assessment process. Without such background knowledge, the vast amount of information produced by ChatGPT can be overwhelming, making it challenging to identify and extract relevant details. The observation of potential useful coaching information generated by ChatGPT underscores the critical need for test security among clinical neuropsychologists, as nonadherence can facilitate coaching (e.g., AI-generated) and compromise test validity. In their position paper, the American Academy of Clinical Neuropsychology (AACN; Boone et al., 2022) emphasizes that objective neuropsychological assessment requires examinees to have no prior access to test questions and answers, while simultaneously ensuring adherence to legal and ethical obligations of disclosure to clients. Effects of AI-Generated Coaching in the Analogue Study The threat posed by AI-generated information to the validity of ADHD assessments is reinforced by the preregistered analyses of this analogue study. Individuals attempting to feign cognitive dysfunction were often shown to excessively overreport symptoms on SVTs and perform markedly poorly on PVTs. This effect is particularly pronounced in analogue studies compared to criterion group design studies in clinical settings. In criterion group designs, credible clinical groups are distinguished from noncredible clinical groups based on established criteria, for example, a number of established validity measures (Rogers, 2018a). The reason for excessive poor performance in analogue studies may be explained by the preoccupation of participants with creating a genuine disorder, rather than with focusing on the avoidance of detection (Rogers, 2018a). This pattern has been demonstrated with the same measures used in the current study, including ADHD symptom severity assessed with the CAARS in general and its infrequency index in particular (e.g., Wallace et al., 2019; Walls et al., 2017, for a meta-analysis), functional impairment evaluated with the WFIRS (Fuermaier et al., 2017a), selective attention measured with the WAFS (Becke et al., 2023), and working memory assessed by the RDS (Bing-Canar et al., 2022; Harrison et al., 2010). Our findings demonstrated that the AI-coached simulation group consistently moderated their symptom overreporting and cognitive underperformance compared to the symptom-coached group, as evidenced by group effects in mostly small to medium size (though nonsignificant in underpowered Bonferroni-corrected pairwise comparisons). This effect is also reflected in lower sensitivity rates for detecting individuals in the AI-coached simulation group compared to the symptom-coached group. Because SVTs and PVTs are conceptualized to detect symptom overreporting and cognitive underperformance, respectively, lower sensitivity rates of the AI-SG indicate a more nuanced symptom report and less excessively poor test performance after AI-coaching. The observed advantage of AI-coaching is particularly noteworthy, as prior research has consistently failed to demonstrate significant effects of coaching individuals before feigning in simulation studies (Boskovic et al., 2022; Dunn et al., 2003; Fuermaier et al., 2017a, b). AI-coaching appears to have a distinct impact compared to traditional coaching, raising significant concerns about the integrity and security of validity assessments when AI chatbots are used effectively. Limitations Several limitations must be considered when interpreting the findings of this study. First, the strict exclusion criteria applied during the post-experimental check resulted in a slightly smaller sample size in the simulation groups than initially planned, which consequently impacted statistical power. This is evident in the pairwise group comparisons between the simulation groups, where meaningful effect sizes were observed; however, the majority of statistical tests did not withstand Bonferroni-adjusted significance thresholds. Post hoc analyses further highlighted the limited statistical power, based on the sample sizes of our study, to reliably detect medium or smaller effects (44%), with sufficient power (85%) achieved only for large effects. Second, the observed effects of AI-coaching are limited to the specific assessment battery used in this study. Notably, the instruments in our assessment battery were explicitly included in the ChatGPT query. Future research should explore the effectiveness of AI-generated coaching across a broader range of assessment batteries and investigate its utility without prior specification of the instruments used. Third, it is important to emphasize that AI is a rapidly evolving field, and the quality and usefulness of AI-generated information must be continuously monitored and evaluated. The results of the present study are based on the free-to-use version of ChatGPT-4, and it is likely that subsequent, more advanced versions of ChatGPT or other AI chatbots, particularly those available through paid subscriptions, will provide more detailed and nuanced coaching information. Conclusions This study raises important concerns regarding the integrity of validity assessments in clinical, neuropsychological, and forensic contexts. We demonstrated that the current iteration of a widely used AI chatbot, ChatGPT-4, can offer clear and effective strategies for presenting oneself during clinical neuropsychological assessments in a manner consistent with the responses of someone genuinely suffering from ADHD. First empirical evidence supports this concern, showing that AI-generated coaching information surpassed symptom-coaching with meaningful effect size. Future research should aim to support these findings using study designs that include clinical populations. From our research, we conclude that researchers and clinicians should exercise caution when disclosing their assessment materials and be hesitant in sharing example items and scoring methodologies. Furthermore, we emphasize that AI chatbots are rapidly evolving and becoming increasingly sophisticated, which poses ongoing challenges for professionals involved in validity assessments. Data Availability Data can be accessed from the corresponding author upon reasonable request. References * American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM-5; 5th ed.). American Psychiatric Publishing. * Becke, M., Tucha, L., Weisbrod, M., Aschenbrenner, S., Tucha, O., & Fuermaier, A. B. M. (2021). Non-credible symptom report in the clinical evaluation of adult ADHD: Development and initial validation of a new validity index embedded in the Conners' adult ADHD rating scales. Journal of Neural Transmission, 128(7), 1045-1063. https://doi.org/10.1007/s00702-021-02318-y Article PubMed PubMed Central Google Scholar * Becke, M., Tucha, L., Weisbrod, M., Aschenbrenner, S., Tucha, O., & Fuermaier, A. B. M. (2022). Joint consideration of validity indicators embedded in Conners' adult ADHD rating scales (CAARS). Psychological Injury and Law, 15(2), 172-188. https://doi.org/ 10.1007/s12207-022-09445-1 Article Google Scholar * Becke, M., Tucha, L., Butzbach, M., Aschenbrenner, S., Weisbrod, M., Tucha, O., & Fuermaier, A. B. M. (2023). Feigning adult ADHD on a comprehensive neuropsychological test battery: An analogue study. International Journal of Environmental Research and Public Health, 20(5), 4070. https://doi.org/10.3390/ijerph20054070 Article PubMed PubMed Central Google Scholar * Bing-Canar, H., Phillips, M. S., Shields, A. N., OgramBuckley, C. M., Chang, F., Khan, H., Skymba, H. V., Ovsiew, G. P., Resch, Z. J., Jennette, K. J., & Soble, J. R. (2022). Cross-validation of multiple WAIS-IV digit spanembedded performance validity indices among a large sam-ple of adult attention deficit/hyperactivity disorder clinicalreferrals. Journal of Psychoeducational Assessment, 40(5), 678-688. https://doi.org/10.1177/ 07342829221081921 Article Google Scholar * Booksh, R. L., Pella, R. D., Singh, A. N., & Gouvier, W. D. (2010). Ability of college students to simulate ADHD on objective measures of attention. Journal of Attention Disorders, 13(4), 325-338. https://doi.org/10.1177/1087054708329927 Article Google Scholar * Boone, K. B., Sweet, J. J., Byrd, D. A., Denney, R. L., Hanks, R. A., Kaufmann, P. M., Kirkwood, M. W., Larrabee, G. J., Marcopulos, B. A., Morgan, J. E., Paltzer, J. Y., Mindt, M. R., Schroeder, R. W., Sim, A. H., & Suhr, J. A. (2022). Official position of the American Academy of Clinical Neuropsychology on test security. The Clinical Neuropsychologist, 36(3), 523-545. https://doi.org/10.1080/13854046.2021.2022214 Article PubMed Google Scholar * Boskovic, I., Akca, A.Y.E., & Giromini, L. (2022). Symptom coaching and symptom validity tests: An analog study using the structured inventory of malingered symptomatology, Self-report symptom inventory, and inventory of problems-29. Applied Neuropsychology: Adult, 1-13. https://doi.org/10.1080/ 23279095.2022.2057856 * Canadian ADHD Resource Alliance (CADDRA). (2017). Canadian ADHD practice guidelines. Retrieved from http://www.caddra.ca/ . Accessed Mar 2025 * Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum Associates. * Conners, C.K., Erhardt, D., & Sparrow, M.A. (1999). Conners' adult ADHD rating scales (CAARS). Multi- Health Systems. * Cook, C. M., Bolinger, E., & Suhr, J. (2016). Further validation of the Conner's adult attention deficit/hyperactivity rating scale infrequency index (CII) for detection of non-credible report of attention deficit/hyperactivity disorder symptoms. Archives of Clinical Neuropsychology, 31(4), 358-364. https:// doi.org/10.1093/arclin/acw015 Article PubMed Google Scholar * Cook, C., Buelow, M. T., Lee, E., Howell, A., Morgan, B., Patel, K., Bryant, A. M., Menatti, A., & Suhr, J. (2018). Malingered attention deficit/hyperactivity disorder on the Conners' adult ADHD rating scales: Do reasons for malingering matter? Journal of Psychoeducational Assessment, 36(6), 552-561. https://doi.org/ 10.1177/0734282917696934 Article Google Scholar * Crisan, I., Sava, F. A., & Maricutoiu, L. P. (2023). Strategies of feigning mild head injuries related to validity indicators and types of coaching: Results of two experimental studies. Applied Neuropsychology: Adult, 30(6), 705-715. https://doi.org/10.1080/ 23279095.2021.1973004 Article PubMed Google Scholar * Dandachi-FitzGerald, B., Merckelbach, H., Boskovic, I., & Jelicic, M. (2020). Do you know people who feign? Proxy respondents about feigned symptoms. Psychological Injury and Law, 13, 225-234. https://doi.org/10.1007/s12207-020-09387-6 Article Google Scholar * Dandachi-FitzGerald, B., Merckelbach, H., & Merten, T. (2024). Cry for help as a root cause of poor symptom validity: A critical note. Applied Neuropsychology: Adult, 31(4), 527-532. https:// doi.org/10.1080/23279095.2022.2040025 Article PubMed Google Scholar * Dong, H., Koerts, J., Pijnenborg, G. H. M., Scherbaum, N., Muller, B. W., & Fuermaier, A. B. M. (2023). Cognitive underperformance in a mixed neuropsychiatric sample at diagnostic evaluation of adult ADHD. Journal of Clinical Medicine, 12, 6926. https://doi.org/10.3390/jcm12216926 Article PubMed PubMed Central Google Scholar * Dunn, T. M., Shear, P. K., Howe, S., & Ris, M. D. (2003). Detecting neuropsychological malingering: Effects of coaching and information. Archives of Clinical Neuropsychology, 18, 121-134. https://doi.org/10.1093/arclin/18.2.121 Article PubMed Google Scholar * Fuermaier, A. B. M., Tucha, L., Koerts, J., Weisbrod, M., Grabemann, M., Zimmermann, M., Mette, C., Aschenbrenner, S., & Tucha, O. (2016). Evaluation of the CAARS infrequency index for the detection of noncredible ADHD symptom report in adulthood. Journal of Psychoeducational Assessment, 34(8), 739-750. Article Google Scholar * Fuermaier, A. B. M., Tucha, O., Koerts, J., Butzbach, M., Weisbrod, M., Aschenbrenner, S., & Tucha, L. (2017a). Susceptibility of functional impairment scales to noncredible responses in the clinical evaluation of adult ADHD. Clinical Neuropsychologist, 32(4), 671-680. https://doi.org/10.1080/ 13854046.2017.1406143 Article PubMed Google Scholar * Fuermaier, A. B. M., Tucha, O., Koerts, J., Lange, K. W., Weisbrod, M., Aschenbrenner, S., & Tucha, L. (2017b). Noncredible cognitive performance at clinical evaluation of adult ADHD: An embedded validity indicator in a visuospatial working memory test. Psychological Assessment, 29(12), 1466-1479. https:// doi.org/10.1037/pas0000534 Article PubMed Google Scholar * Fuermaier, A. B. M., Tucha, O., Koerts, J., Tucha, L., Thome, J., & Faltraco, F. (2021). Feigning ADHD and stimulant misuse among Dutch university students. Journal of Neural Transmission, 128, 1079-1084. https://doi.org/10.1007/s00702-020-02296-7 Article PubMed PubMed Central Google Scholar * Fuermaier, A. B. M., Gontijo-Santos Lima, C., & Tucha, O. (2024a). Impairment assessment in adult ADHD and related disorders: Current opinions from clinic and research. Journal of Attention Disorders, 28(12), 1529-1541. https://doi.org/10.1177/ 10870547241261598 Article PubMed PubMed Central Google Scholar * Fuermaier, A. B. M., Tucha, L., Merten, T., Fathollah, Gol, M., & Tucha, O. (2024b). Symptom validity testing in adults with clinically diagnosed ADHD: Comparison of the Conner's Adult ADHD Rating Scale (CAARS) and the Self-Report Symptom Inventory (SRSI) . Journal of Clinical and Experimental Neuropsychology, 46(8), 693-706. https://doi.org/10.1080/13803395.2024.2411365 Article PubMed Google Scholar * Garcia-Willingham, N.E., Bosch, C.M., & Walls, B.D. (2018). Assessment of feigned cognitive impairment using standard neuropsychological tests. In R. Rogers & S.D. Bender (Eds.), Clinical assessment of malingering and deception (4th ed., pp. 329-358). Guilford Press. * Greiffenstein, M. F., Baker, W. J., & Gola, T. (1994). Validation of malingered amnesia measures with a large clinical sample. Psychological Assessment, 6, 218-224. Article Google Scholar * Harrison, A. G. (2015). Child and adolescent psychoeducational evaluations. In M. W. Kirkwood (Ed.), Validity assessment in child and adolescent assessment. Guilford Press. Google Scholar * Harrison, A. G. (2017). Clinical, ethical, and forensic implications of a flexible threshold for LD and ADHD in postsecondary settings. Psychological Injury and Law, 10, 138-150. https://doi.org/10.1177/108705472210920 Article Google Scholar * Harrison, A. G., & Edwards, M. J. (2010). Symptom exaggeration in postsecondary students: Preliminary base rates in a Canadian sample. Applied Neuropsychology, 17(2), 135-143. https://doi.org/ 10.1080/09084281003715642 Article PubMed Google Scholar * Harrison, A. G., Rosenblum, Y., & Currie, S. (2010). Examining unusual digit span performance in a population of postsecondary students assessed for academic difficulties. Assessment, 17(3), 283-293. https://doi.org/10.1177/1073191109348590 Article PubMed Google Scholar * Hirsch, O., Fuermaier, A. B. M., Tucha, O., Albrecht, B., Chavanon, M. L., & Christiansen, H. (2022). Symptom and performance validity in samples of adults at clinical evaluation of ADHD: A replication study using machine learning algorithms. Journal of Clinical and Experimental Neuropsychology, 44(3), 171-184. https://doi.org/10.1080/13803395.2022.2105821 Article PubMed Google Scholar * Larrabee, G. J. (2012). Performance validity and symptom validity in neuropsychological assessment. Journal of the International Neuropsychological Society, 18(4), 625-30630. https://doi.org/ 10.1017/s1355617712000240 Article PubMed Google Scholar * Lavigne, S., Rios, A., & Davis, J.J. (2024). Does generative artificial intelligence pose a risk to performance validity test security? The Clinical Neuropsychologist, 1-14. https://doi.org/ 10.1080/13854046.2024.2379023 * Mascarenhas, M. A., Cocunato, J. L., Armstrong, I. T., Harrison, A. G., & Zakzanis, K. K. (2023). Base rates of non-credible performance in a post-secondary student sample seeking accessibility accommodations. The Clinical Neuropsychologist, 37 (8), 1608-1628. https://doi.org/10.1080/13854046.2023.2167737 Article PubMed Google Scholar * Nelson, J. M., Whipple, B., Lindstrom, W., & Foels, P. A. (2014). How is ADHD assessed and documented? Examination of psychological reports submitted to determine eligibility for postsecondary disability. Journal of Attention Disorders, 23(14), 1780-1791. https://doi.org/10.1177/1087054714561860 Article PubMed Google Scholar * OpenAI. (2023). ChatGPT-4 (Mar 14 version) [Large language model]. https://chat.openai.com/chat. Accessed Oct 2024 * Ovsiew, G. P., Cerny, B. M., Boer, A. B., Petry, L. G., Resch, Z. J., Durkin, N. M., & Soble, J. R. (2023). Performance and symptom validity assessment in attention deficit/hyperactivity disorder: Base rates of invalidity, concordance, and relative impact on cognitive performance. The Clinical Neuropsychologist, 37(7), 1498-1515. https://doi.org/10.1080/13854046.2022.2162440 Article PubMed Google Scholar * Phillips, M. S., Wisinger, A. M., Lapitan-Moore, F. T., Ausloos-Lozano, J. E., Bing-Canar, H., Durkin, N. M., Ovsiew, G. P., Resch, Z. J., Jennette, K. J., & Soble, J. R. (2023). Cross-validation of multiple embedded performance validity indices in the Rey Auditory Verbal Learning test and Brief Visuospatial Memory test-revised in an adult attention deficit/ hyperactivity disorder clinical sample. Psychological Injury and Law, 16(1), 27-35. https://doi.org/10.1007/s12207-022-09443-3 Article Google Scholar * Rabiner, D. L. (2013). Stimulant prescription cautions: Addressing misuse, diversion and malingering. Current Psychiatry Reports, 15, 375. https://doi.org/10.1007/s11920-013-0375-2 Article PubMed Google Scholar * Robinson, E. V., & Rogers, R. (2018). Detection of feigned ADHD across two domains: The MMPI-2-RF and CAARS for faked symptoms and TOVA for simulated attention deficits. Journal of Psychopathology & Behavioral Assessment, 40(3), 376-385. https:// doi.org/10.1007/s10862-017-9640-8 Article Google Scholar * Rogers, R., Bagby, R. M., & Chakraborty, D. (1993). Feigning schizophrenic disorders on the MMPI-2: Detection of coached simulators. Journal of Personality Assessment, 60(2), 215-226. https://doi.org/10.1207/s15327752ipa6002_1 Article PubMed Google Scholar * Rogers, R. (2018a). Researching response styles. In R. Rogers & S.D. Bender (Eds.), Clinical assessment of malingering and deception (4th ed., pp. 592-614). Guilford Press. * Rogers, R. (2018b). Detection strategies for malingering and defensiveness. In R. Rogers & S.D. Bender (Eds.), Clinical assessment of malingering and deception (4th ed., pp. 18-41). Guilford Press * Schuhfried, G. (2013). Vienna test system (VTS) 8 (Version 8.2.00). [Computer software] Vienna, Austria: Schuhfried * Sibley, M. H. (2021). Empirically-informed guidelines for first-time adult ADHD diagnosis. Journal of Clinical and Experimental Neuropsychology, 43(4), 340-351. https://doi.org/ 10.1080/13803395.2021.1923665 Article PubMed Google Scholar * Song, Q., Zhang, Y., Li, X., & Rudan, I. (2021). The prevalence of adult attention-deficit hyperactivity disorder: A global systematic review and meta-analysis. Journal of Global Health, 11 , 04009. https://doi.org/10.7189/jogh.11.04009 Article PubMed PubMed Central Google Scholar * Sturm, W. (2006). Perception and attention functions (WAF). Vienna, Austria: Schuhfried. Google Scholar * Suhr, J. A., Buelow, M., & Riddle, T. (2011). Development of an infrequency index for the CAARS. Journal of Psychoeducational Assessment, 29(2), 160-170. https://doi.org/10.1177/ 0734282910380190 Article Google Scholar * Wallace, E. R., Garcia-Willingham, N. E., Walls, B. D., Bosch, C. M., Balthrop, K. C., & Berry, D. T. R. (2019). A meta-analysis of malingering detection measures for attention-deficit/ hyperactivity disorder. Psychological Assessment, 31(2), 265-270. https://doi.org/10.1037/pas0000659 Article PubMed Google Scholar * Walls, B. D., Wallace, E. R., Brothers, S. L., & Berry, D. T. R. (2017). Utility of the Conners' adult ADHD rating scale validity scales in identifying simulated attention-deficit hyperactivity disorder and random responding. Psychological Assessment, 29(12), 1437-1446. https://doi.org/10.1037/pas0000530 Article PubMed Google Scholar * Wechsler, D. (2008). Wechsler adult intelligence scale (4th ed.). Pearson Assessment. Google Scholar * Weis, R., Till, C. H., & Erickson, C. P. (2019). ADHD assessment in college students: Psychologists' adherence to DSM- 5 criteria and multi-method/multi-informant assessment. Journal of Psychoeducational Assessment, 37(2), 209-225. https://doi.org/ 10.1177/0734282917735152 Article Google Scholar * Wetter, M. W., & Corrigan, S. (1995). Providing information on clients about psychological tests: A survey of attorney's and law students' attitudes. Professional Psychology: Research and Practice, 26(5), 474-477. Article Google Scholar * Winter, D., & Braw, Y. (2024). Online search strategies utilized in feigning attention deficit/hyperactivity disorder (ADHD) while performing a continuous performance test (CPT). Applied Neuropsychology: Adult, 31(6), 1365-1374. https://doi.org/10.1080 /23279095.2022.2128356 Article PubMed Google Scholar * World Health Organization (2019). International statistical classification of diseases and related health problems (11th ed.). https://icd.who.int/browse11. Accessed Mar 2025 * Youngjohn, J. R. (1995). Confirmed attorney coaching prior to neuropsychological evaluations. Assessment, 2(3), 279-283. Article Google Scholar * Zhao, X., Coxe, S. J., Timmons, A. C., & Frazier, S. L. (2022). Mental health information seeking online: A Google trends analysis of ADHD. Administration and Policy in Mental Health and Mental Health Services Research, 49(3), 357-373. Article PubMed Google Scholar Download references Acknowledgements We thank Sebastiaan Bensink, Alicia Bigeng, and Hui Dong, for their valuable support in study preparation, assessments, and data processing. Author information Authors and Affiliations 1. Department of Clinical and Developmental Neuropsychology, Faculty of Behavioral and Social Sciences, University of Groningen, Grote Kruisstraat 2/1, 9712 TS, Groningen, The Netherlands Anselm B. M. Fuermaier 2. Clinical Psychology, Open University of the Netherlands, Heerlen, The Netherlands Isabella J. M. Niesten Authors 1. Anselm B. M. Fuermaier View author publications You can also search for this author inPubMed Google Scholar 2. Isabella J. M. Niesten View author publications You can also search for this author inPubMed Google Scholar Corresponding author Correspondence to Anselm B. M. Fuermaier. Additional information Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Rights and permissions Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Reprints and permissions About this article Check for updates. Verify currency and authenticity via CrossMark Cite this article Fuermaier, A.B.M., Niesten, I.J.M. ChatGPT Helps Students Feign ADHD: An Analogue Study on AI-Assisted Coaching. Psychol. Inj. and Law 18, 97-107 (2025). https://doi.org/10.1007/s12207-025-09538-7 Download citation * Received: 14 January 2025 * Accepted: 06 March 2025 * Published: 24 March 2025 * Issue Date: June 2025 * DOI: https://doi.org/10.1007/s12207-025-09538-7 Share this article Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative Keywords * Adult ADHD * Malingering * Coaching * Chat GPT * Artificial intelligence Use our pre-submission checklist Avoid common mistakes on your manuscript. Advertisement Search Search by keyword or author [ ] Search Navigation * Find a journal * Publish with us * Track your research Discover content * Journals A-Z * Books A-Z Publish with us * Journal finder * Publish your research * Open access publishing Products and services * Our products * Librarians * Societies * Partners and advertisers Our brands * Springer * Nature Portfolio * BMC * Palgrave Macmillan * Apress * Discover * Your privacy choices/Manage cookies * Your US state privacy rights * Accessibility statement * Terms and conditions * Privacy policy * Help and support * Legal notice * Cancel contracts here 216.82.21.147 Not affiliated Springer Nature (c) 2025 Springer Nature