The neural correlates of reinforcement sensitivity theory: A systematic review of the frontal asymmetry and spectral power literature.

The original Reinforcement Sensitivity Theory (oRST) proposes two systems of approach (BAS) and avoidance (BIS) motivation to underpin personality and behavior. The revised-RST (rRST) model separates avoidance motivation into passive (BIS; anxiety) and active (FFFS; fear) systems. Prior research has attempted to map RST onto lateralized frontal asymmetry to provide a neurophysiological marker of RST. The main aim is to examine the relationships of the o/rRST scales with trait (baseline) and state (manipulated through experimental paradigms) frontal asymmetry. A systematic review was conducted, resulting in 158 studies designated to neuroimaging research. In total, 54 studies were included in this review using either frontal asymmetry or spectral power. The results were split into three main categories: resting frontal alpha asymmetry (N = 23), emotional induction and state-related frontal alpha asymmetry (N = 20), and spectral analysis (N = 16). Findings indicated that BAS was associated with enhanced left frontal asymmetry at baseline and during state-related paradigms. Findings for BIS were more inconsistent, especially at rest, suggesting that BIS, in particular, may require active engagement with the environment. Only 9 of the 54 papers included used the revised RST model, highlighting the need for more rRST research.

Domain specificity of error monitoring: An ERP study in young and older adults.

Metacognition refers to the ability to monitor and control one's cognitive processes, which plays an important role in decision-making throughout the lifespan. It is still debated whether metacognitive abilities decline with age. Neuroimaging evidence suggests that metacognition is served by domain-specific mechanisms. These domains may differentially decline with increasing age. The current investigates whether the error-related negativity (ERN) and the error positivity (Pe) which reflect error detection and error awareness, respectively, differ across perceptual and memory domains in young and older adults. In total, 38 young adults and 37 older adults completed a classic Flanker Task (perceptual) and an adapted memory-based version. No difference in ERN amplitude was found between young and older adults and across domains. Perceptual ERN peaked earlier than Memory ERN. Memory ΔERN was larger than Perceptual ΔERN. Pe was smaller in older adults and ΔPe was larger for perceptual than memory flanker. Memory Pe peaked earlier in young as compared to older adults. Multivariate analyses of whole scalp data supported cross-domain differences. During the task, ERN decreased in young but not in older adults. Memory Pe decreased in young adults but increased in older adults while no significant change in perceptual Pe was found. The study's findings suggest that neural correlates of error monitoring differ across cognitive domains. Moreover, it was shown that error awareness declines in old age but its within-task dynamics vary across cognitive domains. Possible mechanisms underlying metacognition impairments in aging are discussed.

RNA-sequencing of peripheral whole blood of individuals at ultra-high-risk for psychosis - A longitudinal perspective.

BACKGROUND: The peripheral blood is an attractive source of prognostic biomarkers for psychosis conversion. There is limited research on the transcriptomic changes associated with psychosis conversion in the peripheral whole blood. STUDY DESIGN: We performed RNA-sequencing of peripheral whole blood from 65 ultra-high-risk (UHR) participants and 70 healthy control participants recruited in the Longitudinal Youth-at-Risk Study (LYRIKS) cohort. 13 UHR participants converted in the study duration. Samples were collected at 3 timepoints, at 12-months interval across a 2-year period. We examined whether the genes differential with psychosis conversion contain schizophrenia risk loci. We then examined the functional ontologies and GWAS associations of the differential genes. We also identified the overlap between differentially expressed genes across different comparisons. STUDY RESULTS: Genes containing schizophrenia risk loci were not differentially expressed in the peripheral whole blood in psychosis conversion. The differentially expressed genes in psychosis conversion are enriched for ontologies associated with cellular replication. The differentially expressed genes in psychosis conversion are associated with non-neurological GWAS phenotypes reported to be perturbed in schizophrenia and psychosis but not schizophrenia and psychosis phenotypes themselves. We found minimal overlap between the genes differential with psychosis conversion and the genes that are differential between pre-conversion and non-conversion samples. CONCLUSION: The associations between psychosis conversion and peripheral blood-based biomarkers are likely to be indirect. Further studies to elucidate the mechanism behind potential indirect associations are needed.

Graph-based analysis of EEG for schizotypy classification applying flicker Ganzfeld stimulation.

Ganzfeld conditions induce alterations in brain function and pseudo-hallucinatory experiences, particularly in people with high positive schizotypy. The current study uses graph-based parameters to investigate and classify brain networks under Ganzfeld conditions as a function of positive schizotypy. Participants from the general population (14 high schizotypy (HS), 29 low schizotypy (LS)) had an electroencephalography assessment during Ganzfeld conditions, with varying visual activation (8 frequencies of random light flicker) and soundscape-induced mood (neutral, serenity, and anxiety). Weighted functional networks were computed in six frequency sub-bands (delta, theta, alpha-low, alpha-high, beta, and gamma) as a function of light-flicker frequency and mood. The brain network was analyzed using graph theory parameters, including clustering coefficient (CC), strength, and global efficiency (GE). It was found that the LS groups had higher CC and strength than the HS groups, especially in bilateral temporal and frontotemporal brain regions. Moreover, some decreases in CC and strength measures were found in LS groups among occipital and parieto-occipital brain regions. LS groups also had significantly higher GE in all Ganzfeld conditions compared to the HS groups. The random under-sampling boosting (RUSBoost) algorithm achieved the best classification performance with an accuracy of 95.34%, specificity of 96.55%, and sensitivity of 92.85% during an anxiety-induction Ganzfeld condition. This is the first exploration of the relationship between brain functional state changes under Ganzfeld conditions in individuals who vary in positive schizotypy. The accuracy of graph-based parameters in classifying brain states as a function of schizotypy is shown, particularly for brain activity during anxiety induction, and should be investigated in psychosis.

Understanding suicidality in Pacific adolescents in New Zealand using network analysis.

INTRODUCTION: Pacific adolescents in New Zealand (NZ) are three to four times more likely than NZ European adolescents to report suicide attempts and have higher rates of suicidal plans. Suicidal thoughts, plans, and attempts, termed suicidality in this study, result from a complex dynamic interplay of factors, which emerging methodologies like network analysis aim to capture. METHODS: This study used cross-sectional network analysis to model the relationships between suicidality, self-harm, and individual depression symptoms, whilst conditioning on a multi-dimensional set of variables relevant to suicidality. A series of network models were fitted to data from a community sample of New Zealand-born Pacific adolescents (n = 550; 51% male; Mean age (SD) = 17 (0.35)). RESULTS: Self-harm and the depression symptom measuring pessimism had the strongest associations with suicidality, followed by symptoms related to having a negative self-image about looks and sadness. Nonsymptom risk factors for self-harm and suicidality differed markedly. CONCLUSIONS: Depression symptoms varied widely in terms of their contribution to suicidality, highlighting the valuable information gained from analysing depression at the symptom-item level. Reducing the sources of pessimism and building self-esteem presented as potential targets for alleviating suicidality amongst Pacific adolescents in New Zealand. Suicide prevention strategies need to include risk factors for self-harm.

The relationships between expressed emotion, cortisol, and EEG alpha asymmetry.

Families can express high criticism, hostility and emotional over-involvement towards a person with or at risk of mental health problems. Perceiving such high expressed emotion (EE) can be a major psychological stressor for individuals, especially those at risk of mental health problems. To reveal the biological mechanisms underlying the effect of EE on health, this study investigated physiological response (salivary cortisol, frontal alpha asymmetry (FAA)) to verbal criticism and their relationship to anxiety and perceived EE. Using a repeated-measures design, healthy participants attended three testing sessions on non-consecutive days. On each day, participants listened to one of three types of auditory stimuli, namely criticism, neutral or praise, and Electroencephalography (EEG) and salivary cortisol were measured. Results showed a reduction in cortisol following criticism but there was no significant change in FAA. Post-criticism cortisol concentration negatively correlated with perceived EE after controlling for baseline mood. Our findings suggest that salivary cortisol change responds to criticism in non-clinical populations and this response might be largely driven by individual differences in the perception of criticism (e.g., arousal and relevance). Criticisms expressed by audio comments may not be explicitly perceived as an acute emotional stressor, and thus, physiological response to criticisms could be minimum.

Enhancing biofeedback-driven self-guided virtual reality exposure therapy through arousal detection from multimodal data using machine learning.

Virtual reality exposure therapy (VRET) is a novel intervention technique that allows individuals to experience anxiety-evoking stimuli in a safe environment, recognise specific triggers and gradually increase their exposure to perceived threats. Public-speaking anxiety (PSA) is a prevalent form of social anxiety, characterised by stressful arousal and anxiety generated when presenting to an audience. In self-guided VRET, participants can gradually increase their tolerance to exposure and reduce anxiety-induced arousal and PSA over time. However, creating such a VR environment and determining physiological indices of anxiety-induced arousal or distress is an open challenge. Environment modelling, character creation and animation, psychological state determination and the use of machine learning (ML) models for anxiety or stress detection are equally important, and multi-disciplinary expertise is required. In this work, we have explored a series of ML models with publicly available data sets (using electroencephalogram and heart rate variability) to predict arousal states. If we can detect anxiety-induced arousal, we can trigger calming activities to allow individuals to cope with and overcome distress. Here, we discuss the means of effective selection of ML models and parameters in arousal detection. We propose a pipeline to overcome the model selection problem with different parameter settings in the context of virtual reality exposure therapy. This pipeline can be extended to other domains of interest where arousal detection is crucial. Finally, we have implemented a biofeedback framework for VRET where we successfully provided feedback as a form of heart rate and brain laterality index from our acquired multimodal data for psychological intervention to overcome anxiety.

Investigation of social and cognitive predictors in non-transition ultra-high-risk' individuals for psychosis using spiking neural networks.

Finding predictors of social and cognitive impairment in non-transition Ultra-High-Risk individuals (UHR) is critical in prognosis and planning of potential personalised intervention strategies. Social and cognitive functioning observed in youth at UHR for psychosis may be protective against transition to clinically relevant illness. The current study used a computational method known as Spiking Neural Network (SNN) to identify the cognitive and social predictors of transitioning outcome. Participants (90 UHR, 81 Healthy Control (HC)) completed batteries of neuropsychological tests in the domains of verbal memory, working memory, processing speed, attention, executive function along with social skills-based performance at baseline and 4 × 6-month follow-up intervals. The UHR status was recorded as Remitters, Converters or Maintained. SNN were used to model interactions between variables across groups over time and classify UHR status. The performance of SNN was examined relative to other machine learning methods. Higher interaction between social and cognitive variables was seen for the Maintained, than Remitter subgroup. Findings identified the most important cognitive and social variables (particularly verbal memory, processing speed, attention, affect and interpersonal social functioning) that showed discriminative patterns in the SNN models of HC vs UHR subgroups, with accuracies up to 80%; outperforming other machine learning models (56-64% based on 18 months data). This finding is indicative of a promising direction for early detection of social and cognitive impairment in UHR individuals that may not anticipate transition to psychosis and implicate early initiated interventions to stem the impact of clinical symptoms of psychosis.

Constrained neuro fuzzy inference methodology for explainable personalised modelling with applications on gene expression data.

Interpretable machine learning models for gene expression datasets are important for understanding the decision-making process of a classifier and gaining insights on the underlying molecular processes of genetic conditions. Interpretable models can potentially support early diagnosis before full disease manifestation. This is particularly important yet, challenging for mental health. We hypothesise this is due to extreme heterogeneity issues which may be overcome and explained by personalised modelling techniques. Thus far, most machine learning methods applied to gene expression datasets, including deep neural networks, lack personalised interpretability. This paper proposes a new methodology named personalised constrained neuro fuzzy inference (PCNFI) for learning personalised rules from high dimensional datasets which are structurally and semantically interpretable. Case studies on two mental health related datasets (schizophrenia and bipolar disorders) have shown that the relatively short and simple personalised fuzzy rules provided enhanced interpretability as well as better classification performance compared to other commonly used machine learning methods. Performance test on a cancer dataset also showed that PCNFI matches previous benchmarks. Insights from our approach also indicated the importance of two genes (ATRX and TSPAN2) as possible biomarkers for early differentiation of ultra-high risk, bipolar and healthy individuals. These genes are linked to cognitive ability and impulsive behaviour. Our findings suggest a significant starting point for further research into the biological role of cognitive and impulsivity-related differences. With potential applications across bio-medical research, the proposed PCNFI method is promising for diagnosis, prognosis, and the design of personalised treatment plans for better outcomes in the future.

Prediction of Tinnitus Treatment Outcomes Based on EEG Sensors and TFI Score Using Deep Learning.

Tinnitus is a hearing disorder that is characterized by the perception of sounds in the absence of an external source. Currently, there is no pharmaceutical cure for tinnitus, however, multiple therapies and interventions have been developed that improve or control associated distress and anxiety. We propose a new Artificial Intelligence (AI) algorithm as a digital prognostic health system that models electroencephalographic (EEG) data in order to predict patients' responses to tinnitus therapies. The EEG data was collected from patients prior to treatment and 3-months following a sound-based therapy. Feature selection techniques were utilised to identify predictive EEG variables with the best accuracy. The patients' EEG features from both the frequency and functional connectivity domains were entered as inputs that carry knowledge extracted from EEG into AI algorithms for training and predicting therapy outcomes. The AI models differentiated the patients' outcomes into either therapy responder or non-responder, as defined by their Tinnitus Functional Index (TFI) scores, with accuracies ranging from 98%-100%. Our findings demonstrate the potential use of AI, including deep learning, for predicting therapy outcomes in tinnitus. The research suggests an optimal configuration of the EEG sensors that are involved in measuring brain functional changes in response to tinnitus treatments. It identified which EEG electrodes are the most informative sensors and how the EEG frequency and functional connectivity can better classify patients into the responder and non-responder groups. This has potential for real-time monitoring of patient therapy outcomes at home.

Depression, anxiety and worry in young Pacific adults in New Zealand during the COVID-19 pandemic.

OBJECTIVE: To measure symptoms of anxiety, depression and hopelessness in a sample of young Pacific adults living in Auckland, New Zealand during the 2020/2021 COVID-19 pandemic and identify protective factors. METHODS: Participants were 267 Pacific adults (58% female) who completed a survey online. Analyses included descriptive statistics, correlations, linear regression and symptom network analysis. RESULTS: Around 25% of the sample scored in the range for moderate to severe anxiety and 10% for moderate to severe depression on standard measures. Almost 40% indicated that they found the first lockdown very stressful and 55% noted that some members of their family found it stressful. Only 16% worried about COVID-19 and their future quite a bit or constantly, while another 25% worried sometimes. Self-compassion and Pacific Identity had moderate, negative correlations, and Worry about COVID-19 had weak positive correlations, with anxiety, depression, hopelessness and perceived stress. CONCLUSION: These results suggest that, while the prevalence of depression and anxiety are quite high among this population, fostering ethnic identity and self-compassion in Pacific children and adolescents might protect against developing depression and anxiety.

Coping and Co-Occurrence of Gaming Disorder and Substance Use in Recovering Substance Users.

BACKGROUND: There are a wide range of negative effects associated with both substance use disorders and behavioural disorders and their co-occurrence. Understanding the way in which at-risk populations (e.g., substance-abstinent users) interact with potentially addictive behaviours (e.g., gaming) and substance use-while navigating life stressors through differing coping styles-can inform preventative strategies. METHODS: Therefore, the present study investigated 64 clinical participants and 138 general population participants. Each cohort was required to complete a battery of psychometric scales exploring problematic behaviours, problematic substance use, co-occurrence, and coping styles. Additional exploratory direct comparisons of gamers in the clinical cohort and gamers in the general cohort were carried out. RESULTS: The study's findings suggest that gamers from different populations (i.e., general and clinical) share similar at-risk behaviours. These problematic behaviours were more pronounced among abstinent substance use gamers, and more specifically among poly-substance use gamers. CONCLUSIONS: The findings of the present study add to the literature and suggest that dysfunctional coping style and the co-occurrence of problematic behaviours may have an impact on the assessment and potential treatment of substance abstinent gamers. The findings offer support for an integrated treatment approach, wherein both substance use and the other problematic behaviours (e.g., gaming) are considered in tandem.

Co-Occurrence of Gaming Disorder and Other Potentially Addictive Behaviours between Australia, New Zealand, and the United Kingdom.

Background: Evidence suggests that gamers can have varying experiences of disordered gaming behaviours due to coping mechanisms and how they can act as risk or protective factor in the development and/or maintenance of disordered behaviours. A particular area of interest is how this may manifest across different countries. Understanding the interplay of these potential risk and protective factors within different countries will aid identifying and preventing disordered behaviours. Methods: Three cohorts were recruited from Australia, New Zealand, and the United Kingdom. Each cohort was required to complete a battery of psychometric scales exploring problematic behaviours, problematic substance use, co-occurrence, coping styles, and personality. A latent profile analysis was conducted to examine the differences between cohorts and further investigated with additional analyses. Results: The findings suggested that a minority of gamers were affected by gaming disorder, and there appeared an at-risk cohort who utilise gaming as a maladaptive coping strategy. Other accompanying potentially addictive behaviour or substance use may be exacerbated as a result, the manifestation of which can be influenced by cultural elements. Conclusions: When considering gamers from countries which hold similar views, it is important to be cognisant of the variations found in the manifestations of disordered gaming and accompanying potentially addictive behaviours. This will allow for a more precise identification of at-risk behaviours, which will result in more favourable treatment outcomes for those who are considered at-risk or high-risk individuals.

Risk factors for depression in Pacific adolescents in New Zealand: A network analysis.

BACKGROUND: Network analysis provides opportunities to gain a greater understanding of the complex interplay of risk factors for depression and heterogeneous symptom presentations. This study used network analysis to discover risk factors associated with both depression severity and depression symptoms amongst Pacific adolescents in New Zealand. METHODS: Mixed graphical models with regularization were fitted to data from a community sample of New Zealand born, Pacific adolescents, (n = 561; 51% male; Mean age (SD) = 17 (0.35)) and associations between a wide range of potentially explanatory variables and depression severity and depression symptoms investigated. The associations identified were then tested for reliability, using resampling techniques and sensitivity analysis. RESULTS: In the networks, the explanatory variables associated with both depression severity and depression symptoms were those related to quality of the relationships with mother or friends, school connectedness, and self-assessed weight, but the symptoms they were associated with varied substantially. In the depression severity networks, impulsivity appeared to be a bridging node connecting depression severity with delinquency and negative peer influence. LIMITATIONS: The data were analysed cross-sectionally, so causal inferences about the directions of relationships could not be inferred and most of the data were self-reported. CONCLUSIONS: The results illustrate the varied way that adolescent depression can manifest itself in terms of symptoms and suggest specific items on the depression inventory that might be suitable targets for prevention strategies and interventions, based on the risk factor - depression symptom profiles of individuals or groups.

What has social neuroscience learned from hyperscanning studies of spoken communication? A systematic review.

A growing body of literature examining the neurocognitive processes of interpersonal linguistic interaction indicates the emergence of neural alignment as participants engage in oral communication. However, questions have arisen whether the study results can be interpreted beyond observations of cortical functionality and extended to the mutual understanding between communicators. This review presents evidence from electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) hyperscanning studies of interbrain synchrony (IBS) in which participants communicated via spoken language. The studies are classified into: knowledge sharing; turn-taking speech co-ordination; cooperation, problem-solving and creativity; and naturalistic discussion paradigms according to the type of interaction specified in each study. Alignment predominantly occurred in the frontal and temporo-parietal areas, which may reflect activation of the mirror and mentalizing systems. We argue that the literature presents a significant contribution to advancing our understanding of IBS and mutual understanding between communicators. We end with suggestions for future research, including analytical approaches and experimental conditions and hypothesize that brain-inspired neural networks are promising techniques for better understanding of IBS through hyperscanning.

New insights into neural networks of error monitoring and clinical implications: a systematic review of ERP studies in neurological diseases.

Error monitoring allows for the efficient performance of goal-directed behaviors and successful learning. Furthermore, error monitoring as a metacognitive ability may play a crucial role for neuropsychological interventions, such as rehabilitation. In the past decades, research has suggested two electrophysiological markers for error monitoring: the error-related negativity (ERN) and the error positivity (Pe), thought to reflect, respectively, error detection and error awareness. Studies on several neurological diseases have investigated the alteration of the ERN and the Pe, but these findings have not been summarized. Accordingly, a systematic review was conducted to understand what neurological conditions present alterations of error monitoring event-related potentials and their relation with clinical measures. Overall, ERN tended to be reduced in most neurological conditions while results related to Pe integrity are less clear. ERN and Pe were found to be associated with several measures of clinical severity. Additionally, we explored the contribution of different brain structures to neural networks underlying error monitoring, further elaborating on the domain-specificity of error processing and clinical implications of findings. In conclusion, electrophysiological signatures of error monitoring could be reliable measures of neurological dysfunction and a robust tool in neuropsychological rehabilitation.

Neurophysiological markers of prospective memory and working memory in typical ageing and mild cognitive impairment.

OBJECTIVE: Prospective memory (PM) -the memory of delayed intentions- is impacted by age-related cognitive decline. The current event-related potential study investigates neural mechanisms underpinning typical and atypical (Mild Cognitive Impairment, MCI) age-related decline in PM. METHODS: Young adults (YA, n = 30, age = 24.7, female n = 13), healthy older adults (OA, n = 39, age = 72.87, female n = 24) and older adults with MCI (n = 27, age = 77.54, female n = 12) performed two event-based PM tasks (perceptual, conceptual) superimposed on an ongoing working memory task. Electroencephalographic data was recorded from 128 electrodes. Groups were compared for P2 (higher order perceptual processing), N300/frontal positivity (cue detection), the parietal positivity (retrieval), reorienting negativity (RON; attention shifting). RESULTS: Participants with MCI had poorer performance (ongoing working memory task, conceptual PM), lower P2 amplitudes, and delayed RON (particularly for perceptual PM) than YA and OA. MCI had lower parietal positivity relative to YA only. YA had earlier latencies for the parietal positivity than MCI and OA, and lower amplitudes for N300 (than OA) and frontal positivity (than OA and MCI). CONCLUSIONS: Impaired attention and working memory may underpin PM deficits in MCI. SIGNIFICANCE: This is the first study to document the role of RON in PM and to investigate neurophysiological mechanisms underpinning PM in MCI.

Increased functional connectivity patterns in mild Alzheimer's disease: A rsfMRI study.

Background: Alzheimer's disease (AD) is the most common age-related neurodegenerative disorder. In view of our rapidly aging population, there is an urgent need to identify Alzheimer's disease (AD) at an early stage. A potential way to do so is by assessing the functional connectivity (FC), i.e., the statistical dependency between two or more brain regions, through novel analysis techniques. Methods: In the present study, we assessed the static and dynamic FC using different approaches. A resting state (rs)fMRI dataset from the Alzheimer's disease neuroimaging initiative (ADNI) was used (n = 128). The blood-oxygen-level-dependent (BOLD) signals from 116 regions of 4 groups of participants, i.e., healthy controls (HC; n = 35), early mild cognitive impairment (EMCI; n = 29), late mild cognitive impairment (LMCI; n = 30), and Alzheimer's disease (AD; n = 34) were extracted and analyzed. FC and dynamic FC were extracted using Pearson's correlation, sliding-windows correlation analysis (SWA), and the point process analysis (PPA). Additionally, graph theory measures to explore network segregation and integration were computed. Results: Our results showed a longer characteristic path length and a decreased degree of EMCI in comparison to the other groups. Additionally, an increased FC in several regions in LMCI and AD in contrast to HC and EMCI was detected. These results suggest a maladaptive short-term mechanism to maintain cognition. Conclusion: The increased pattern of FC in several regions in LMCI and AD is observable in all the analyses; however, the PPA enabled us to reduce the computational demands and offered new specific dynamic FC findings.

Personalised predictive modelling with brain-inspired spiking neural networks of longitudinal MRI neuroimaging data and the case study of dementia.

BACKGROUND: Longitudinal neuroimaging provides spatiotemporal brain data (STBD) measurement that can be utilised to understand dynamic changes in brain structure and/or function underpinning cognitive activities. Making sense of such highly interactive information is challenging, given that the features manifest intricate temporal, causal relations between the spatially distributed neural sources in the brain. METHODS: The current paper argues for the advancement of deep learning algorithms in brain-inspired spiking neural networks (SNN), capable of modelling structural data across time (longitudinal measurement) and space (anatomical components). The paper proposes a methodology and a computational architecture based on SNN for building personalised predictive models from longitudinal brain data to accurately detect, understand, and predict the dynamics of an individual's functional brain state. The methodology includes finding clusters of similar data to each individual, data interpolation, deep learning in a 3-dimensional brain-template structured SNN model, classification and prediction of individual outcome, visualisation of structural brain changes related to the predicted outcomes, interpretation of results, and individual and group predictive marker discovery. RESULTS: To demonstrate the functionality of the proposed methodology, the paper presents experimental results on a longitudinal magnetic resonance imaging (MRI) dataset derived from 175 older adults of the internationally recognised community-based cohort Sydney Memory and Ageing Study (MAS) spanning 6 years of follow-up. SIGNIFICANCE: The models were able to accurately classify and predict 2 years ahead of cognitive decline, such as mild cognitive impairment (MCI) and dementia with 95% and 91% accuracy, respectively. The proposed methodology also offers a 3-dimensional visualisation of the MRI models reflecting the dynamic patterns of regional changes in white matter hyperintensity (WMH) and brain volume over 6 years. CONCLUSION: The method is efficient for personalised predictive modelling on a wide range of neuroimaging longitudinal data, including also demographic, genetic, and clinical data. As a case study, it resulted in finding predictive markers for MCI and dementia as dynamic brain patterns using MRI data.