Journal of Neuropsychology

We aimed at validating the Mini Social Cognition and Emotional Assessment (Mini‐SEA) in a German cohort of mildly impaired behavioural‐variant frontotemporal dementia (bvFTD) patients and healthy controls. The Mini‐SEA comprises the Facial Emotion Recognition Test (FERT) and the Faux Pas Test (FPT) measuring Theory of Mind (ToM) abilities in social norm‐related real‐life stories. We examined the diagnostic performance of the Mini‐SEA alongside other neuropsychological assessments and investigated its structural neural correlates. We included 32 bvFTD patients and 54 controls in logistic regression models with forward‐stepwise selection containing demographics, standard neuropsychological battery (CERAD‐NAB+) and the Mini‐SEA scores to identify the most relevant variables. Demographic, neuropsychological and daily‐life activity associations were explored. Voxel‐based morphometry analysis was conducted in a subsample (14 bvFTD and 14 controls) on regions previously linked to emotion processing and ToM functions. The Mini‐SEA yielded a very good performance, being in the best‐fitting model with a high odds ratio alongside the executive‐language and memory measures. Specifically, the FERT indicated the strongest effect in the group differentiation. Mini‐SEA showed significant associations with executive‐language tests and daily‐life activities. In canonical emotion processing brain regions, we found associations of the Mini‐SEA composite and the FERT with grey matter volumes in the left insula and lentiform nucleus of putamen. Within ToM regions, associations were found for the Mini‐SEA composite and the FPT in cerebellar regions. The German Mini‐SEA discriminates well between mildly impaired bvFTD patients and controls. We also demonstrated its significant value for neuropsychological assessment and neuro‐behavioural associations in regions underlying emotion processing and ToM.

Harmonized neuropsychological assessment for neurocognitive disorders (NCDs) is an urgent priority in clinics. Neuropsychology assessments in NCDs seldom include tests exploring social cognitive skills. In 2022, we launched the SIGNATURE initiative to optimize socio-cognitive assessment in NCDs. Here, we report findings from the first initiative phase, including consortium creation and evaluation of the state of the art in socio-cognitive assessment in memory clinics. We developed an ad hoc online survey to explore practices and measures, relevance, and obstacles preventing the use of socio-cognitive testing in clinics. The survey was distributed within the SIGNATURE network. National coordinators were identified to disseminate the survey to local collaborators and scientific societies active in the field of dementia and/or neuropsychology. Data were analysed in aggregate form and stratified by geographical area and variables of interest. Four hundred and thirteen (413) responses from 10 European and Latin American geographical regions were recorded. Responders were balanced between physicians and psychologists. Seventy-eight (78) % of respondents reported no/limited experience with socio-cognitive measures; more than 85% agreed on their relevance in clinics. Ekman-60 faces was the most well-known and/or used task, followed by the Faux-Pas and Reading-the-Mind-in-the-Eyes tests. Lack of clinical measures, assessment time, guidelines, and education/training were reported as main obstacles. Real-life barriers prevent the adoption of socio-cognitive testing in clinics. Bidirectional collaboration between clinicians and researchers is required to address clinical needs and constraints and facilitate consistent socio-cognitive assessment.

Faced with increasing diversity in Europe, a large body of research in neuropsychology has emerged to develop tools for the reliable detection of cognitive disorders in diverse older adults. Following this perspective, few tools have been validated, particularly for assessing episodic memory, such as the Nine Images Test (TNI-93). The aim of the present study was to test the interest of adding a delayed recall to the classic TNI-93 procedure, for the detection of clinical Alzheimer’s disease (AD) in diverse populations. A French retrospective analysis was conducted based on the clinical and neuropsychological data of 281 patients (111 who received a clinical diagnosis of AD; 101 with a cognitive profile not suggestive of AD and 70 patients with a subjective cognitive decline). The sample was mostly composed by people with a low level of education and non-French speakers. The TNI-93 data from the neuropsychological assessment, including a free and a cued recall after a 20 min delay, were analyzed. First, AD patients performed more poorly than both other groups on all scores. However, performance decreased more significantly after a 20min delay in AD patients than in other groups. Second, the Receiver-Operating-Characteristic analysis showed that the higher diagnostic accuracy for detection of AD patients was obtained for the 20min free recall. These results confirm the relevance of the TNI-93 for the clinical detection of AD in diverse populations. The added delayed recall condition seems relevant to highlight the accelerated forgetting of AD patients in a 20min time window following learning.

The ‘KeyStrokes’ test (KS) was created as an online and computerized neuropsychological assessment to assess simple attention, processing speed, and executive function. This pilot study aims to show proof of concept of the KS test as a computerized assessment. Building on a previous feasibility study, we assessed the KS test's internal consistency and correlations to other neurocognitive assessments. Participants were recruited from a clinical sample of patients referred for standard neuropsychological evaluation and were asked to perform several standard neurocognitive tests and six subtests of the KS: two response time trials (arrows, words), three inhibition trials (arrows, words, arrows/words) and one inhibition/switching trial (arrows/words). We assessed internal consistency; conducted correlation analyses between each KS subtest, standard neuropsychological tests, and demographic characteristics (age, education, ethnicity, and gender); and conducted multiple regression analyses to assess the relationship between test performance and age and education. We assessed 87 individuals with a mean age of 54.09 years. Correlations between KS subtests were positive and strong (all above ρ > .72, p < .001). Subtests were generally positively correlated with select WAIS‐IV and Reynolds Interference Task subtests, and negatively correlated with trail making tests, the grooved pegboard test, and age. Age significantly predicted performance ( p < .001), whereas education did not. Ethnicity appeared to correlate with certain subtests, whereas gender did not. Analysis of correlations between the KS subtests and multiple well‐established neuropsychological tests showed the possible viability of the KS as a new neurocognitive measure assessing areas of attention, processing speed, and executive function. Additional study of the KS can provide more evidence for its use as a new computerized, and possible online neuropsychological assessment.

Transient global amnesia (TGA), a transient memory disorder in clinical neurology, is a unique clinical model for the study of hippocampal dysfunction and its implications for memory processes. While data are rather unequivocal concerning the relevance of the hippocampus for episodic memory, there is considerable dispute about its role for semantic memory. This study aimed at exploring how hippocampal impairment, which underlies the clinical presentation of TGA, affects semantic memory retrieval, particularly with regard to different retrieval strategies. Data from the acute and post‐acute phase of 17 TGA patients and 17 healthy controls matched on socio‐demographic factors were collected. Categorical word fluency tasks were differentiated into three retrieval strategies: first, with activation of episodic‐spatial memory content; second, with novel and flexible linking of semantic memory content and third, with activation of overlearned semantic memory content. We find that hippocampal impairment during TGA significantly restricts semantic word fluency performance, with the degree of impairment depending on the retrieval strategy used and most pronounced when flexible relinking of semantic content is required. Our results suggest an important hippocampal contribution to semantic retrieval, especially in connection with novel and flexible linking of semantic content. They may furthermore be practically relevant for the early differential diagnosis and therapy of memory disorders.

Updated criteria pertaining to the diagnosis of Alzheimer's disease (AD) have sparked debate over the reliance on biomarkers—particularly amyloid‐β and phosphorylated tau. While biomarkers promise earlier detection and standardized criteria, the potential extension and interpretation of their use in asymptomatic individuals remains controversial. Many individuals with abnormal biomarker profiles never experience cognitive decline, raising concerns about overdiagnosis, unintended negative psychosocial consequences and the blurring line between risk and definitive diagnosis. We, and others, argue that biomarker positivity should be reframed not as a definitive diagnosis but rather as an indicator of elevated risk, particularly in the absence of cognitive symptoms. Doing so better aligns with current evidence, preserves clarity in diagnosis, and avoids unintended psychosocial consequences. Crucially, the role of cognitive reserve—influenced by education, other life experiences and structural inequities—must be considered, particularly among racially and ethnically diverse populations historically underrepresented in AD research. Biomarker thresholds (as well as neuropsychological tools) derived from predominantly non‐Hispanic white cohorts may not generalize across groups, risking misclassification and inequity. As the field moves towards precision medicine and AI‐driven risk models, inclusive data and culturally valid frameworks are essential. Ultimately, embracing a risk‐based, multifactorial approach respects the complexity of AD and promotes equitable care. This perspective calls for interdisciplinary collaboration to refine diagnostic strategies that are scientifically grounded, socially conscious and responsive to the lived realities of diverse populations. Only then can we responsibly integrate biomarkers into practice without sacrificing nuance.

Patients with dural arteriovenous fistulas (DAVF) suffer from cognitive impairments that go often unrecognized. This study aimed to explore the severity of cognitive impairment as well as specific cognitive domains affected in DAVF patients and to track its evolution over long‐term follow‐up after embolization. Consecutive DAVF patients (and an equal number of healthy controls) were prospectively enrolled and underwent a comprehensive baseline neuropsychological (NP) assessment. These patients were re‐evaluated postembolization at a short‐term follow‐up of 1 month and long‐term follow‐up of 8–12 months. Thirty‐one patients were included, with a male‐to‐female ratio of 5.2:1 and an average age of 45.1 years. NP assessments revealed significantly impaired cognitive scores across all domains (Addenbrooke's Cognitive Examination [m‐ACE], the Rey Auditory Verbal Learning Test [RAVLT], the Wechsler Memory Scale, digit span forward and backward tests, and the Trail Making Test Parts A and B) in the DAVF group compared to healthy controls ( p < .001). Post embolization, the m‐ACE ( p < .001), RAVLT ( p = .04), WMS‐verbal delay ( p = .002) and Trail making test B ( p = .019) scores showed statistically significant improvement compared to healthy controls at 1 month. However, the cognitive scores did not fully recover to the level of healthy controls at long‐term follow‐up. Though treatment leads to significant cognitive recovery, lasting residual cognitive deficits are persistent in DAVF patients compared to healthy controls. Inclusion of comprehensive NP evaluation in work up can unmask subtle cognitive deficits that may guide in therapeutic decision making especially in ‘benign’ DAVFs.

We introduce a special issue of the Journal of Neuropsychology dedicated to a recent paradigm shift in Alzheimer's disease diagnosis. Joint workgroups from the (US) National Institute on Aging and the Alzheimer's Association (NIA‐AA) recently issued policy guidelines reclassifying Alzheimer's disease as a biological entity. These guidelines shift the onus of diagnosis in favour of protein biomarkers, relegating cognitive symptoms (e.g. subjective memory and language disorders) as supportive rather than core features. We invited experts in the study of Alzheimer's disease and Related Disorders (ADRDs) to express their views on this paradigmatic shift in dementia management. In this editorial, we synthesize some of the main points advanced in the commentaries. Contributors identified the promise of blood‐based biomarker testing for improving equitable detection of dementia in large swathes of the world population. This enthusiasm was tempered by concerns about the biomarker‐only diagnostic approach, including the potential for significant harm (e.g. stigma, depression, suicide) caused by labelling asymptomatic older adults who might otherwise never behaviourally express the underlying disease pathology.

This study used fMRI to examine the effect that the abstract‐concrete dimension may have on the thematic‐taxonomic distinction. The dual hub theory (DHT) posits that left angular gyrus supports thematic relationships and left temporal pole supports taxonomic relationships; however, the DHT is largely based on evidence from concrete words. It is important that theories of semantic organization include abstract words as they are ubiquitous in everyday discourse (Lupyan & Winter, Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences , 2018, 373, 20170137). Additionally, there is reason to believe that there may bean interaction between the abstract‐concrete dimension and the thematic‐taxonomic distinction, based on the different representational frameworks (DRF) hypothesis, which posits that abstract concepts are primarily organized by association/theme, and that concrete concepts are primarily organized by similarity/taxonomy. However, there appears to be a mismatch among the DHT, predictions of the DRF hypothesis for brain activation, and existing neuroimaging data for the concreteness effect. Thus, we sought to include abstract words in a test of the DHT and determine whether any interactions exist between the abstract‐concrete dimension and the thematic‐taxonomic distinction. While we replicated the localization of abstract and concrete word processing, we did not extend the DHT to include abstract words, nor did we find an interaction that would support the DRF hypothesis. Rather, our results align more closely with the hub‐and‐spoke model (Patterson et al., Nature Reviews Neuroscience , 2007, 8, 976).

Transcranial electrical stimulation (tES) holds promise for neuropsychological rehabilitation by leveraging the brain's inherent plasticity to enhance cognitive and motor functions. However, early results have been variable due to oversimplified approaches. This manuscript explores the potential and complexities of tES, particularly focusing on a protocol defined transcranial direct current stimulation as a reference model for all tES protocols, emphasising the need for precision in tailoring stimulation parameters to individual characteristics. By integrating intrinsic (i.e. the neuro‐physiological system state) and extrinsic factors (i.e. experimental set up), highlighting the critical role of state‐dependent effects and the synergy with cognitive tasks, we aim to refine tES protocols. This approach not only addresses the complexity of the brain system (as defined by its current state) but also highlights the importance of collaborative research and data sharing to understand the underlying mechanisms of tES‐induced changes and optimising therapeutic efficacy. Emphasising the integration of tES with targeted tasks and clearer hypotheses, this work underscores the potential for more effective neurorehabilitation strategies.

This case report presents new semantic learning and long‐term retention data collected over a 5‐year period from an amnesic adolescent boy, CJ. Compared to his younger sister, a novel abbreviation‐learning task captured CJ's slower semantic acquisition across three weekly training sessions. By contrast, his rate of forgetting between sessions was comparable to that of the control's and was slower over long delays of up to 5 years but recalled information without any reliable report of the original learning context.

We standardized a new version of the Digit span test and the first version of the Bisyllabic non‐word span test, both measuring the phonological loop, in an Italian sample of neurologically healthy adults ( n = 225). All stimuli were administered to the participants through a computerized procedure to avoid the influence of the examiner on participants' performance. We used a preliminary test to exclude the presence of sensory‐perceptual and articulatory‐motor difficulties that might have influenced the results. The results revealed that both Age and Education were significant predictors of participants' performance on the Digit span test. By contrast, only Age predicted significantly participants' performance on the Bisyllabic non‐word span test. The average Digit span was approximately twice as large as the average Bisyllabic non‐word span, suggesting that the latter might be a strategy‐free measure of the phonological loop. The Bisyllabic non‐word span is an innovative and specific measure for identifying phonological short‐term memory deficits. For all tests, adjusted and equivalent scores are provided to facilitate results interpretation and clinical applicability.

We present a patient with left mesial temporal medically refractory epilepsy who developed new onset abnormal eating behaviour following surgical ablation of the left hippocampus and amygdala. The patient underwent a second ablation due to seizure recurrence targeting a remnant of the amygdala, after which seizures resolved, but disordered eating behaviours continued with no appreciable change. We discuss the role of aberrant limbic signalling in mesial temporal epilepsy and its contribution towards eating behaviours in the context of surgical treatment.

The aim of this research was to analyse the reliability and validity of the Computerised Battery for Neuropsychological Evaluation of Children (BENCI) in a Cuban population of children and adolescents. The study involved 1714 Cuban students between the ages of 6 and 18 who were divided into three groups according to their level of education (Elementary: 6–11 years old; High School: 12–14 years old; and Pre‐University: 15–18 years old). All participants were evaluated using the BENCI with some also undergoing additional neuropsychological testing. The BENCI evaluates the following cognitive domains: processing speed, visuomotor coordination, attention, memory, language, and executive functions. The results showed that the BENCI has good test–retest reliability indices and high internal consistency values in Reasoning, Reaction Time, and Working Memory. In terms of validity, the data revealed significant correlations between the BENCI tests and other neuropsychological tests assessing similar cognitive functions. The BENCI also has discriminative validity, as it was found that performance on the tests varies according to the level of education of those being evaluated. In terms of construct validity, confirmatory factor analysis showed that the fit indices for the executive functions dimension of the BENCI are adequate. The data show that the BENCI is a reliable and valid instrument for assessing neurodevelopment in Cuban children and adolescents. Since this is the first neuropsychological test of its kind to be validated in the Cuban population, this finding is of particular importance.

There is emerging evidence of social cognitive impairments in bipolar disorders (BD). Less evident is the question if social cognitive impairments are predictive of psychosocial functioning, independently of neurocognitive impairment. The aims of the study were to investigate if patients with BD‐I showed impairments in facial emotion recognition and alterations in affective empathy, in relation to healthy controls, and if these impairments would predict psychosocial functioning, after accounting for neurocognitive impairments. Thirty‐seven patients diagnosed with BD‐I, in an euthymic state, and 37 matched healthy controls underwent an assessment including a facial recognition test (ERT) and a self‐report scale of affective empathy (BEES). Patients additionally underwent an extensive neuropsychological assessment consisting of traditional tests. Patients with BD‐I were significantly less able to recognize the emotion fear compared to healthy controls. However, the lower ability to recognize fear did not predict psychosocial functioning. In addition, it was not related to any of the other neuropsychological variables. The degree of self‐reported empathy did not differ between patients and healthy controls. The results add to the evidence of a specific deficit in recognizing fear in BD‐I; however, a link with psychosocial functioning was lacking. It is possible that the ability to recognize fear is related to a more narrow concept of interpersonal functioning than to the broad concept of psychosocial functioning. Future research should be directed towards aspects of social functioning in relation to social cognitive impairments, while taking account of subgroups of social cognitive functioning.

Fatigue is one of the most common symptoms following traumatic brain injury (TBI). Despite its prevalence, fatigue remains a challenging concept to define and measure. The aim of the present study was to explore potential relationships between self‐rated fatigue in patients with TBI and performance on several widely used neuropsychological tests. In a cross‐sectional design, patients with TBI ( n = 68) and healthy controls ( n = 27) underwent a comprehensive battery of commonly used neuropsychological tests and completed two self‐assessment fatigue scales, the Fatigue Severity Scale and the Mental Fatigue Scale. Patients with TBI performed worse on neuropsychological tests of short‐term memory, processing speed and executive functioning (inhibition) compared to healthy controls. Within the TBI group, only the Paced Auditory Serial Addition Test (PASAT) and the Stroop—Inhibition task showed significant correlations with measures of fatigue. However, after adjusting for relevant demographic variables, including age, gender, education and TBI severity, only PASAT remained significantly associated with the Mental Fatigue Scale ( r = .45, p = .005). Within the healthy control group, no such associations were found. These results highlight an interesting relationship between PASAT performance and self‐assessed fatigue. With further research, PASAT, modifications of it or similar measures could potentially help clinicians in evaluating fatigue in patients with TBI.

Attention deficit hyperactivity disorder (ADHD) is associated with diverse cognitive deficits of which problems related to memory and learning are well-established but poorly understood. In an online experiment, we studied whether verbal memory impairment in adult ADHD is related to differences in spontaneous use of memory strategies that hinge upon metacognitive and executive skills. Eighty-one ADHD adults and 209 neurotypical controls performed a Word List Learning task where the same words were presented three times, each time coupled with an open strategy report. Bayesian analyses indicated that the ADHD group recalled less words, did not differ from controls in efficacy of strategy use, but exhibited more limited strategy use than the controls. This suggests that less versatile strategy employment, possibly related to core executive deficits, may play a role in verbal memory impairment in adults with ADHD. Concerning treatment implications, future research could probe to what extent external strategy instruction might increase the versatility of mnemonic strategy use in adult ADHD, and thereby partly compensate for their verbal memory deficits.

Up to 45% of patients with Parkinson's disease (PD) experience impulse control disorders (ICDs), characterized by a loss of voluntary control over impulses, drives or temptations. This study aimed to investigate whether previously identified genetic and psychiatric risk factors interact towards the development of ICDs in PD. A total of 278 de novo PD patients (ICD‐free at enrollment) were selected from the Parkinson's Progression Markers Initiative database. ICD presence at baseline and yearly follow‐up assessments were evaluated via the Questionnaire for Impulsive‐Compulsive Disorders. Symptoms of anxiety and depression at baseline were measured via the State–Trait‐Anxiety Inventory and Geriatric Depression Scale, respectively. Furthermore, an individual dopamine genetic risk score was calculated according to polymorphisms in genes coding for dopamine (D1, D2 and D3 receptors and catechol‐O‐methyltransferase), with higher scores reflecting higher central dopamine neurotransmission. In total, 146 participants (47.5%) developed an ICD with an average onset time of 36 months (range 3–96) from baseline. Results from a Cox proportional hazards model showed a trait anxiety × genetics interaction, suggesting that individuals with both higher baseline trait anxiety scores and higher dopamine genetic risk scores were at increased risk of ICD development. This interaction remained significant after controlling for age, sex and motor symptom severity. Our findings suggest that genetic and psychiatric predictors of impulsivity in PD interact and jointly yield increased ICD risk during the course of the disorder. This implies that early screening of anxiety symptoms in combination with genotyping can be useful to identify those at risk for ICD.

In 1941, André Rey published the Rey Complex Figure, a widely used test for assessing visual‐constructional ability and visual memory. It consists of two parts: copy and recall. Evaluating the copy portion presents challenges, as it requires the administrator to focus on both the process and outcome. The assessor must systematically track how the patient copies the figure in real‐time to evaluate their planning, organisation and executive abilities. This ‘clinician's copy’ serves as a record of the patient's approach, aiding later judgements about their cognitive skills. To ensure accuracy, clinicians need a method to record this process for later review or colleague consultation. This paper revisits Rey's suggestion of using different coloured pencils to observe the copy sequence, proposing an alternative. Instead of providing coloured pencils to the patient, we recommend that the administrator use them to record the copy sequence. This method aligns with test norms, reducing potential distractions for the patient while enabling both experienced and novice administrators to easily track and document the sequence of copying.

This study aims to demonstrate that children and adolescents diagnosed with Attention Deficit Hyperactivity Disorder (ADHD) who exhibit autism traits have a more severe clinical profile in terms of emotion regulation, clinical features related to ADHD, and functionality, compared to those diagnosed with ADHD without these traits. 50 patients with and 64 patients without autism traits between the ages of 8–16 were recruited for the study among the children and adolescents diagnosed with ADHD. The Schedule for Affective Disorders and Schizophrenia for School‐Age Children‐Present and Lifetime Version, DSM‐5‐2016‐Turkish Adaptation (K‐SADS‐PL‐DSM‐5‐T) was used to exclude the diagnosis of Autism Spectrum Disorder (ASD) and detect comorbid psychiatric diagnosis. The Social Reciprocity Scale (SRS) was completed by parents to determine groups based on autism traits. Children completed the Childhood Anxiety Sensitivity Index (CASI) and the Affective Reactivity Index (ARI). The Strengths and Difficulties Questionnaire (SDQ), Affective Reactivity Index‐Parent Report (ARI‐P) and Weiss Functional Impairment Rating Scale‐Parent Report (WFIRS‐P) were completed by the parents. We found that the group with autism traits had significantly more hyperactivity/inattention, conduct problems, emotional problems, and peer problems and significantly more irritability and frequent separation anxiety disorder. Although there was no significant impairment in functionality in either group, the level of impairment was significantly higher in the group with autism traits. Children diagnosed with Attention Deficit Hyperactivity Disorder who exhibit autism traits experience higher levels of irritability and separation anxiety disorder, as well as greater impairment in functionality, compared to those without these traits.

In recognising emotions expressed by others, one can make use of both embodied cognition and mechanisms that do not necessarily require activation of the limbic system, such as evoking from memory the meaning of morphological features of the observed face. Instead, we believe that the recognition of the authenticity of an emotional expression is primarily based on embodied cognition, for which the mirror system would play a significant role. To verify this hypothesis, we submitted 20 parkinsonian patients and 20 healthy control subjects to the Emotional Authenticity Recognition test, a novel test using dynamic stimuli to evaluate the ability to recognise emotions and their authenticity. Analysis of variance of the test scores shows that Parkinsonian patients perform worse than controls when they had to recognise the authenticity of emotions, although they are able to identify them. Our results confirm a deficit in the recognition of the authenticity of emotions in patients with Parkinson's disease attributable to the disruption of extrapiramidal limbic circuit between ventral striatum and orbitomesial-prefrontal cortex.

Understanding the relative contribution of various factors influencing initial severity of aphasia and recovery after a stroke is essential for optimising neurorehabilitation programmes. We investigated how various significant sociodemographic, cognitive, clinical, stroke‐related and rehabilitation‐related factors modulate aphasia severity and language recovery following left‐hemispheric stroke. Employing an innovative method, we conducted a retrospective analysis of 96 stroke participants to explore the combined impact of these factors. The initial severity of aphasia was categorised into severe, mild/moderate and no aphasia based on the severity of their language deficits in the subacute phase (Aphasia Severity Rating Scale, ASRS). To assess speech‐and‐language recovery, we classified 53/96 patients with aphasia into high and poor recovery categories using a gain score formula (ASRS_discharge—ASRS_admission)/ASRS_admission. Subsequently, we performed statistical analyses (univariate analyses and forward stepwise logistic regression combined with bootstrap) to identify the determinants of the initial severity of aphasia and the degree of recovery. Our analyses unveiled that more severe aphasia initially was correlated with a more severe stroke (Odds Ratio, OR = .90, p = .041), moderate/severe executive dysfunction (OR = .068, p < .001) and larger lesion size (OR = .068, p < .001). Furthermore, the degree of recovery was associated with the intensity of speech‐and‐language therapy (OR = 1.47, p = .043). These findings enrich our understanding of the determinants of aphasia severity and language recovery, employing an original methodology to scrutinise the collective effect of multiple variables in a retrospective analysis of stroke participants. A better knowledge of these factors may help implement personalised language rehabilitation programmes to maximise speech‐and‐language recovery.

Stroke causes severe long‐term disabilities with a significant reduction in quality of life. This study aims to explore the predictive value of cognitive screening in the acute phase of mild stroke on patients' functional outcome after discharge. A total of 110 patients with mild stroke were recruited. Patients were included in the study if they were discharged directly home from the acute units. The cognitive profile of patients was assessed with the Oxford Cognitive Screen (OCS). The OCS was administered 3–10 days after stroke, providing a five domain‐specific cognitive profile. Long‐term functional outcomes were evaluated by the Stroke Impact Scale 3.0 (SIS 3.0), a self‐reported questionnaire that includes physical, cognitive, emotional and social participation dimensions. All patients completed the survey online on average 10 months after stroke. Our results show that OCS is positively associated with physical and cognitive dimensions, after adjusting for age and stroke severity measured by NIHSS at admission. In conclusion, OCS in acute mild stroke seems to be an independent predictor of long‐term functional outcomes and could help clinicians in the long‐term management of patients.

Aims Predicting the diagnosis of an older adult solely based on their financial capacity performance or other neuropsychological test performance is still an open question. The aim of this study is to highlight which tests are of importance in diagnostic protocols by using recent advancements in machine learning. Methods For this reason, a neuropsychological battery was administered in 543 older Greek patients already diagnosed with different types of neurocognitive disorders along with a test specifically measuring financial capacity, that is, Legal Capacity for Property Law Transactions Assessment Scale (LCPLTAS). The battery was analysed using a random forest algorithm. The objective was to predict whether an older person suffers from dementia. The algorithm's performance was tested through cross‐validation. Results Machine learning was applied for the first time in data analysis regarding financial capacity and three factors‐tests were revealed as the best predictors: two subscales from the LCPLTAS measuring ‘financial decision making’ and ‘cash transactions’, and the widely used MMSE which measures general cognition. The algorithm demonstrated good performance as measured by the F1‐score, which is a measure of the harmonic mean of precision and recall. This evaluation metric in binary and multi‐class classification integrates precision and recall into a single metric to gain a better understanding of model performance. Conclusions These findings reveal the importance of focusing on these scales and tests in neuropsychological assessment protocols. Future research may clarify in other cultural settings if the same variables are of importance.