Joint Modeling Framework Research Articles

A tutorial for joint modeling of longitudinal and time-to-event data in R

In biostatistics and medical research, longitudinal data are often composed of repeated assessments of a variable and dichotomous indicators to mark an event of interest. Consequently, joint modeling of longitudinal and time-to-event data has generated much interest in these disciplines over the previous decade. In behavioural sciences, too, often we are interested in relating individual trajectories and discrete events. Yet, joint modeling is rarely applied in behavioural sciences more generally. This tutorial presents an overview and general framework for joint modeling of longitudinal and time-to-event data, and fully illustrates its application in the context of a behavioral study with the JMbayes R package. In particular, the tutorial discusses practical topics, such as model selection and comparison, choice of joint modeling parameterization and interpretation of model parameters. In the end, this tutorial aims at introducing didactically the theory related to joint modeling and to introduce novice analysts to the use of the JMbayes package.

A discrete-continuous multi-vehicle anticipation model of driving behaviour in heterogeneous disordered traffic conditions

This study proposes a multi-vehicle anticipation-based discrete-continuous choice modelling framework for describing driver behaviour in heterogeneous disordered traffic (HDT) conditions. To incorporate multi-vehicle anticipation, the concept of an influence zone around a vehicle (subject vehicle) is introduced. Vehicles within the influence zone can potentially influence the subject vehicle’s driving behaviour. Further, driving decisions are characterized as combination of discrete and continuous components. The discrete component involves the decision to accelerate, decelerate, or maintain constant speed and the continuous component involves the decision of how much to accelerate or decelerate. A copula-based joint modelling framework that allows dependencies between discrete and continuous components is proposed. Such a joint modelling framework recognizes that the discrete and continuous decisions are made simultaneously, and common unobserved factors influence both decisions. Additionally, truncated distributions are employed for the continuous model components to avoid the prediction of unrealistically high acceleration or deceleration values. The parameters of the proposed model are estimated using a trajectory dataset from Chennai, India. The empirical results underscore (a) the importance of considering multi-vehicle anticipation for describing driving behaviour in HDT conditions, and (b) the efficacy of the joint discrete-continuous system for modelling driving behaviour. Further, not all traffic environment variables found to influence the discrete decisions were found influential on continuous decisions and vice versa. Moreover, the influence of several variables was found to be stronger on the decision to accelerate or decelerate than on the decision of how much to accelerate or decelerate.

Interdependence in active mobility adoption: Joint modeling and motivational spillover in walking, cycling and bike-sharing

Active mobility offers an array of physical, emotional, and social well-being benefits. However, with the proliferation of the sharing economy, new nonmotorized means of transport are entering the fold, complementing some existing mobility options while competing with others. The purpose of this research study is to investigate the adoption of three active travel modes—namely walking, cycling, and bikesharing—in a joint modeling framework. The analysis is based on an adaptation of the stages of change framework, which originates from the health behavior sciences. Multivariate ordered probit modeling drawing on U.S. survey data provides well-needed insights into individuals’ preparedness to adopt multiple active modes as a function of personal, neighborhood, and psychosocial factors. The research suggests three important findings. (1) The joint model structure confirms interdependence among different active mobility choices. The strongest complementarity is found for walking and cycling adoption. (2) Each mode has a distinctive adoption path with either three or four separate stages. We discuss the implications of derived stage-thresholds and plot adoption contours for selected scenarios. (3) Psychological and neighborhood variables generate more coupling among active modes than individual and household factors. Specifically, identifying strongly with active mobility aspirations, experiences with multimodal travel, possessing better navigational skills, along with supportive local community norms are the factors that appear to drive the joint adoption decisions. This study contributes to the understanding of how decisions within the same functional domain are related and help to design policies that promote active mobility by identifying positive spillovers and joint determinants.

Change in Estimated GFR and Risk of Allograft Failure in Patients Diagnosed With Late Active Antibody-mediated Rejection Following Kidney Transplantation.

There are challenges in designing adequate, well-controlled studies of patients with active antibody-mediated rejection (AMR) after kidney transplantation (KTx). We assessed the functional relationship between change in estimated glomerular filtration rate (eGFR) following the diagnosis of AMR and the risk of subsequent death-censored graft failure using the joint modeling framework. We included recipients of solitary KTx between 1995 and 2013 at 4 transplant centers diagnosed with biopsy-proven active AMR at least 1 year post-KTx, who had a minimum of 3-year follow-up. A total of 91 patients across participating centers were included in the analysis. Of the 91 patients, n = 54 patients (59%) met the death-censored graft failure endpoint and n = 62 patients (68%) met the all-cause graft failure composite endpoint. Kaplan-Meier death-censored graft survival rates at 12, 36, and 60 months postdiagnosis of AMR pooled across centers were 88.9%, 58.9%, and 36.4%, respectively. Spaghetti plots indicated a linear trend in the change in eGFR, especially in the first 12 months postdiagnosis of active AMR. A significant change in eGFR was observed within the first 12 months postdiagnosis of active AMR, getting worse by a factor of -0.757 mL/min/1.73 m2 per month during the 12-month analysis period (a delta of -9.084 mL/min/1.73 m2 at 1 y). Notably, an extrapolated 30% improvement in the slope of eGFR in the first 12 months was associated with a 10% improvement in death-censored graft failure at 5 years. If prospectively validated, this study may inform the design of pivotal clinical trials for therapies for late AMR.

Learning niche features to improve image-based species identification

Species identification is a critical task of ecological research. Having accurate and intelligent methods of species identification would improve our ability to study and conserve biodiversity with saving much time and effort. But image-based deep learning methods have rarely taken account of domain knowledge, and perform poorly on imbalanced training dataset and similar species. Here, we propose NicheNet which combines ecological niche model and image-based deep learning model together expanding from the joint model framework invented by previous research. We incorporate an optimization process for NicheNet, and examine its performance on identifying Chinese Galliformes comparing with image-only model and Geo-prior-based model. For assessing the ability of both NicheNet and image-only models on distinguishing similar species, we initiate a new criterion named Near Error Rate in this study which decomposes identification error rate on each species along its classification. We show that the joint models gain significantly improvement on identifying our study species compared with image-only model. Against on an image-only baseline 82.5%, we observe 7.73% improvement in top-1 accuracy for NicheNet, and a 6.21% increase for Geo-prior-based model. NicheNet gains a 15.3% increment of average F1-Score and gets a − 8.5% mean decrement in Near Species Error Rate while −4.4% in Near Genus Error Rate. Further cases analysis shows the background mechanism of NicheNet to improve image-only models. We demonstrate that NicheNet can learn the features from images and niche-prior, which is generated by niche model and finer than geo-prior by Geo-prior-based model, together to promote its ability on identifying species, especially on those with small training dataset and similar outlook. It is a flexible model framework, and we can introduce more biological features and domain models to strengthen its accuracy and robust in the future work. Our study shows that NicheNet can be widely adopted and will accurately accelerate automatic species identification task for biodiversity research and conservation.

HetGP: Heteroskedastic Gaussian Process Modeling and Sequential Design in R

An increasing number of time-consuming simulators exhibit a complex noise structure that depends on the inputs. For conducting studies with limited budgets of evaluations, new surrogate methods are required in order to simultaneously model the mean and variance fields. To this end, we present the hetGP package, implementing many recent advances in Gaussian process modeling with input-dependent noise. First, we describe a simple, yet efficient, joint modeling framework that relies on replication for both speed and accuracy. Then we tackle the issue of data acquisition leveraging replication and exploration in a sequential manner for various goals, such as for obtaining a globally accurate model, for optimization, or for contour finding. Reproducible illustrations are provided throughout.

Prognostic Value of Tricuspid Valve Regurgitation in Patients With Pulmonary Arterial Hypertension and CTEPH: A Longitudinal Study

Aims: The prognostic value of functional tricuspid valve regurgitation (TR) in patients with pulmonary arterial hypertension and chronic thromboembolic pulmonary hypertension (CTEPH) remains undetermined. This study aims to quantify the prognostic role of TR during follow-up in relation to right ventricle (RV) dysfunction on clinical outcomes. Methods and Results: Patients from our prospective cohort study with pulmonary arterial hypertension and chronic thromboembolic pulmonary hypertension were included. The primary endpoint was a composite of death or lung transplantation. The longitudinal evolution of TR and RV dysfunction were modelled with generalized mixed-effect models, which were inserted in a cox model under the joint-modelling framework in order to investigate the association of TR and RV dysfunction with the endpoint. In total, 120 patients were included (median age:59, females:62%), with a mean follow-up of 3.2±2.1 years. 31 patients reached the endpoint (2 transplant, 29 mortality). On average the probability of moderate-to-severe TR decreased during follow-up, whereas the probability of moderate-to-severe RV dysfunction remained stable over time. The cumulative effect of moderate-to-severe TR (hazard ratio: 1.01 95%CI[1.00-1.01],P<0.001) and moderate-to-severe RV dysfunction (hazard ratio: 1.01 95%CI[1.00-1.01],P<0.001) was associated with the composite endpoint in univariable joint-models. In a multivariable joint-model with both the evolutions of TR and RV dysfunction only TR remained a significant predictor (hazard ratio: 1.01 95%CI[1.00-1.01],P<0.001). Conclusion: Persistent moderate-to-severe tricuspid valve regurgitation during follow-up predicts adverse outcomes and might be a better predictor of lung transplantation and mortality compared to right ventricle dysfunction.

A joint modeling approach for longitudinal microbiome data improves ability to detect microbiome associations with disease.

Changes in the composition of the microbiome over time are associated with myriad human illnesses. Unfortunately, the lack of analytic techniques has hindered researchers' ability to quantify the association between longitudinal microbial composition and time-to-event outcomes. Prior methodological work developed the joint model for longitudinal and time-to-event data to incorporate time-dependent biomarker covariates into the hazard regression approach to disease outcomes. The original implementation of this joint modeling approach employed a linear mixed effects model to represent the time-dependent covariates. However, when the distribution of the time-dependent covariate is non-Gaussian, as is the case with microbial abundances, researchers require different statistical methodology. We present a joint modeling framework that uses a negative binomial mixed effects model to determine longitudinal taxon abundances. We incorporate these modeled microbial abundances into a hazard function with a parameterization that not only accounts for the proportional nature of microbiome data, but also generates biologically interpretable results. Herein we demonstrate the performance improvements of our approach over existing alternatives via simulation as well as a previously published longitudinal dataset studying the microbiome during pregnancy. The results demonstrate that our joint modeling framework for longitudinal microbiome count data provides a powerful methodology to uncover associations between changes in microbial abundances over time and the onset of disease. This method offers the potential to equip researchers with a deeper understanding of the associations between longitudinal microbial composition changes and disease outcomes. This new approach could potentially lead to new diagnostic biomarkers or inform clinical interventions to help prevent or treat disease.

Joint modelling of longitudinal lipids and time to coronary heart disease in the Jackson Heart Study

BackgroundMultiple longitudinal responses together with time-to-event outcome are common in biomedical studies. There are several instances where the longitudinal responses are correlated with each other and at the same time each longitudinal response is associated with the survival outcome. The main purpose of this study is to present and explore a joint modeling approach for multiple correlated longitudinal responses and a survival outcome. The method will be illustrated using the Jackson Heart Study (JHS), which is one of the largest cardiovascular studies among African Americans.MethodsFour longitudinal responses, i.e., total cholesterol (TC), high density lipoprotein (HDL) cholesterol, triglyceride (TG) and inflammation measured by high-sensitivity C-reactive protein (hsCRP); and time-to-coronary heart disease (CHD) were considered from the JHS. The repeated lipid and hsCRP measurements from a given subject overtime are likely correlated with each other and could influence the subject’s risk for CHD. A joint modeling framework is considered. To deal with the high dimensionality due to the multiple longitudinal profiles, we use a pairwise bivariate model fitting approach that was developed in the context of multivariate Gaussian random effects models. The method is further explored through simulations.ResultsThe proposed model performed well in terms of bias and relative efficiency. The JHS data analysis showed that lipid and hsCRP trajectories could exhibit interdependence in their joint evolution and have impact on CHD risk.ConclusionsWe applied a unified and flexible joint modeling approach to analyze multiple correlated longitudinal responses and survival outcome. The method accounts for the correlation among the longitudinal responses as well as the association between each longitudinal response and the survival outcome at once. This helps to explore how the combination of multiple longitudinal trajectories could be related to the survival process.

Adjustment for the measurement error in evaluating biomarker performances at baseline for future survival outcomes: Time-dependent receiver operating characteristic curve within a joint modelling framework

The performance of a biomarker is defined by how well the biomarker is capable to distinguish between healthy and diseased individuals. This assessment is usually based on the baseline value of the biomarker; the value at the earliest time point of the patient follow-up, and quantified by ROC (receiver operating characteristic) curve analysis. However, the observed baseline value is often subjected to measurement error due to imperfect laboratory conditions and limited machine precision. Failing to adjust for measurement error may underestimate the true performance of the biomarker, and in a direct comparison, useful biomarkers could be overlooked. We develop a novel approach to account for measurement error when calculating the performance of the baseline biomarker value for future survival outcomes. We adopt a joint longitudinal and survival data modelling formulation and use the available longitudinally repeated values of the biomarker to make adjustment of the measurement error in time-dependent ROC curve analysis. Our simulation study shows that the proposed measurement error-adjusted estimator is more efficient for evaluating the performance of the biomarker than estimators ignoring the measurement error. The proposed method is illustrated using Mayo Clinic primary biliary cirrhosis (PBC) study.

Bayesian design of clinical trials using joint models for longitudinal and time-to-event data.

Joint models for longitudinal and time-to-event data are increasingly used for the analysis of clinical trial data. However, few methods have been proposed for designing clinical trials using these models. In this article, we develop a Bayesian clinical trial design methodology focused on evaluating the treatment's effect on the time-to-event endpoint using a flexible trajectory joint model. By incorporating the longitudinal outcome trajectory into the hazard model for the time-to-event endpoint, the joint modeling framework allows for non-proportional hazards (e.g., an increasing hazard ratio over time). Inference for the time-to-event endpoint is based on an average of a time-varying hazard ratio which can be decomposed according to the treatment's direct effect on the time-to-event endpoint and its indirect effect, mediated through the longitudinal outcome. We propose an approach for sample size determination for a trial such that the design has high power and a well-controlled type I error rate with both operating characteristics defined from a Bayesian perspective. We demonstrate the methodology by designing a breast cancer clinical trial with a primary time-to-event endpoint and where predictive longitudinal outcome measures are also collected periodically during follow-up.

Dynamic predictive models for ridesourcing services in New York City using daily compositional data

In recent years, there has been a tremendous change in the landscape of mobility offerings with the introduction with the introduction of Uber and Lyft, companies collectively known as Transportation Network Companies (TNCs), otherwise known as ridesourcing companies. Due to the nascency of these companies, there is a small but rapidly growing body of literature on the impacts these TNCs are having on traditional forms of shared ride modes, however, most of the emerging literature centers on the relationships between TNC and only one other shared mode of transportation (e.g. TNCs and Public Transit). This research attempts to contribute to the literature by examining the relationship between TNCs and multiple shared ride modes simultaneously. To this end, a joint modeling framework was used to study New York City ridership data for TNCs, taxi services, bikesharing, and the subway from January 2015 to June 2017. The goal of the research presented is two-fold: explore the dynamic relationships between TNCs and other modes of shared ridership, as well as to build a predictive model for the daily ridership usage for each modal offering and total daily ridership usage. To accomplish these tasks, we first used a compositional time series approach in which the four series are modeled as proportions of total daily demand and then, after a suitable transformation, jointly modeled them via a vector autoregression with exogenous predictors (VARX) to account for trend, a weekly seasonal structure, and exogenous predictors. The second part of our analysis involved modeling the daily total usage of the four modes using a dynamic linear model (DLM) and then using that model to draw inferences about the total ridership in NYC. Results of the models were then combined to produce medium-term forecasts for each modal ridership. Our findings corroborate those of others in investigating correlations over time between usage of TNCs and taxis in servicing consumers. In keeping with our second goal, this analysis demonstrates that our modeling framework may be useful for forecasting several competing types of shared ride modes in New York City.

A marginal estimate for the overall treatment effect on a survival outcome within the joint modeling framework.

Joint models for longitudinal and survival data are increasingly used and enjoy a wide range of application areas. In this article, we focus on the application of joint models on clinical trial data with special interest in the treatment effect on the survival outcome. Within a joint model, the estimated treatment effect on the survival outcome is an aggregate comprising the indirect treatment effect through the longitudinal outcome and the direct treatment effect on the survival outcome. This overall treatment effect is, however, conditional on random effects, and therefore has a subject‐specific interpretation. The conditional interpretation arises from the shared random effects between the longitudinal and survival process in combination with the nonlinear link function of the survival model. The overall treatment effect is, therefore, not valid for population‐based inference, which is the goal for most clinical trials. We propose a method to obtain a marginal estimate of the overall treatment effect on the survival outcome in a joint model. Additionally, we extend our proposal to allow for different parameterizations for the association between the longitudinal and survival outcome. The proposed method is demonstrated on data of a clinical study on the effect of synbiotic on the gut microbiota of cesarean delivered infants, where we estimate the marginal overall treatment effect on the risk of eczema or atopic dermatitis using longitudinal information on fecal bifidobacteria.

Dynamic predictions of kidney graft survival in the presence of longitudinal outliers.

In kidney transplantation, dynamic predictions of graft survival may be obtained from joint modelling of longitudinal and survival data for which a common assumption is that random-effects and error terms in the longitudinal sub-model are Gaussian. However, this assumption may be too restrictive, e.g. in the presence of outliers, and more flexible distributions would be required. In this study, we relax the Gaussian assumption by defining a robust joint modelling framework with t-distributed random-effects and error terms to obtain dynamic predictions of graft survival for kidney transplant patients. We take a Bayesian paradigm for inference and dynamic predictions and sample from the joint posterior densities. While previous research reported improved performances of robust joint models compared to the Gaussian version in terms of parameter estimation, dynamic prediction accuracy obtained from such approach has not been yet evaluated. Our results based on a training sample from the French DIVAT kidney transplantation cohort illustrate that estimates for the slope parameters in the longitudinal and survival sub-models are sensitive to the distributional assumptions. From both an internal validation sample from the DIVAT cohort and an external validation sample from the Lille (France) and Leuven (Belgium) transplantation centers, calibration and discrimination performances appeared to be better under the robust joint models compared to the Gaussian version, illustrating the need to accommodate outliers in the dynamic prediction context. Simulation results support the findings of the validation studies.

Getting the best of both worlds: a framework for combining disaggregate travel survey data and aggregate mobile phone data for trip generation modelling

Traditional approaches to travel behaviour modelling primarily rely on household travel survey data, which is expensive to collect, resulting in small sample sizes and infrequent updates. Furthermore, such data is prone to reporting errors which can lead to biased parameter estimates and subsequently incorrect predictions. On the other hand, mobile phone call detail records (CDRs), which report the timestamped locations of mobile communication events, have been successfully used in the context of generating travel patterns. However, due to their anonymous nature, such records have not been widely used in developing mathematical models establishing the relationship between the observed travel behaviour and influencing factors such as the attributes of the alternatives and the decision makers. In this paper, we propose a joint modelling framework that utilises the advantages offered by both travel survey data and low-cost CDR data to optimise the prediction capacity of traditional trip generation models. In this regard, we develop a model that jointly explains the reported trips for each individual in the household survey data and ensures that the aggregated zonal trip productions are close to those derived from CDR data. This framework is tested using data from Dhaka. Bangladesh consisting of household survey data (65,419 persons in 16,750 households), mobile phone CDR data (over 600 million records generated by 6.9 million users), and aggregate census data. The model results show that the proposed framework improves the spatial and temporal transferability of the joint models over the base model which relies on household travel survey data alone. This serves as a proof-of-concept that augmenting travel survey data with mobile phone data holds significant promise for the travel behaviour modelling community, not only by saving the cost of data collection, but also improving the prediction capability of the models.

Incorporating survival analysis into the safety effectiveness evaluation of treatments: Jointly modeling crash counts and time intervals between crashes

This study aims to incorporate survival theory into the estimation of crash modification factors (CMFs) and to examine the performance. A joint modeling framework, which considers crash counts and time intervals between crashes simultaneously, is proposed. To assess the performance of the joint model, this study collected roadway and crash data on 240 rural two-lane roadway segments in Texas, developed CMFs for a dummy treatment and further compared the results with those developed using the traditional empirical Bayes (EB) method. The main findings are summarized as follows: (1) The traditional EB method tends to overestimate the CMFs for the treatment, and underestimate the standard errors. In most cases, the results are biased; (2) The estimated CMF values with the joint model are closer to the true effect, and they have higher standard errors. The confidence intervals of the CMFs cover the CMF for the dummy treatment (i.e., 1.0), which is more realistic; (3) Temporal instability in traffic crashes are also observed in this study. Increasing the duration of the study period does not always increase the accuracy of CMF estimates. In addition to crash counts, safety analysts are encouraged to incorporate time intervals between crashes while estimating CMFs.

Real-time Adaptive Design Optimization Within Functional MRI Experiments

Efficient data collection is an important goal in cognitive neuroimaging studies because of the high cost of data acquisition. One method of improving efficiency is to maximize the informativeness of the data collected on each trial. We propose an Adaptive Design Optimization (Cavagnaro et al. Neural Computation 22, 887–905 2010; Myung et al. Journal of Mathematical Psychology 57, 53–67 2013) procedure to optimize the sequencing of stimuli for model-based functional neuroimaging studies. Our method uses the Joint Modeling Framework (Turner et al. NeuroImage 72, 193–206 2013, 2019) to maximize the information learned about how the brain produces a behavior by integrating over neural and behavioral data simultaneously. We validate our method in simulation and real-world experiments by showing how Adaptive Design Optimization proposes the optimal stimulus sequence to reduce uncertainty and improve accuracy from a Bayesian perspective.

An introduction to joint models-applications in nephrology.

In nephrology, a great deal of information is measured repeatedly in patients over time, often alongside data on events of clinical interest. In this introductory article we discuss how these two types of data can be simultaneously analysed using the joint model (JM) framework, illustrated by clinical examples from nephrology. As classical survival analysis and linear mixed models form the two main components of the JM framework, we will also briefly revisit these techniques.

The Clinical Impact and Long-Term Outcome of Tricuspid Regurgitation in Patients with Orthotopic Heart Transplantation

Tricuspid regurgitation (TR) is common in patients with orthotopic heart transplantation (HTX). Nevertheless, the clinical impact and the long-term outcome of TR remains controversial. In this study we aim to elucidate the clinical impact and long-term outcome of TR, taking into account its dynamic nature. All consecutive adults patients undergoing biatrial HTX in our tertiary center with a follow-up echocardiogram were included in this study. Mixed-models were used to model the evolution of TR over time. Thereafter, the mixed-model was inserted into a Cox model, under the joint-model framework, in order to address the association of the dynamic TR with mortality. In total, 572 patients were included (median age: 50 years, males: 74.9%). Approximately 33% of patients had moderate-to-severe TR immediately after surgery. However, this declined to approximately 15% after 5 years and 10% after 10 years of surgery (Fig 1). Predictors of TR were: a higher operation urgency, higher donor age, no pre-implant mechanical assist device, and concurrent post-HTX LV dysfunction. Overall survival at 1, 5, 10, 20 years was 97±1%, 88±1%, 66±2% and 23±2%, respectively. The presence of moderate-to-severe TR during follow-up was associated with higher mortality (HR: 1.08, 95%CI[1.03-1.14], p<0.001). The course of TR was positively correlated with the course of creatinine, ASAT and ALAT. Moderate-to-severe TR after HTX is common and the highest immediately after HTX and decreases steadily over the years. TR during follow-up is significantly associated with higher morbidity and mortality. Therefore, further research is urgently needed for the prevention, appropriate medical treatment, and timely surgical intervention of TR post-HTX.

A Joint Modelling Approach to Analyze Risky Decisions by Means of Diffusion Tensor Imaging and Behavioural Data.

Understanding dependencies between brain functioning and cognition is a challenging task which might require more than applying standard statistical models to neural and behavioural measures to be accomplished. Recent developments in computational modelling have demonstrated the advantage to formally account for reciprocal relations between mathematical models of cognition and brain functional, or structural, characteristics to relate neural and cognitive parameters on a model-based perspective. This would allow to account for both neural and behavioural data simultaneously by providing a joint probabilistic model for the two sources of information. In the present work we proposed an architecture for jointly modelling the reciprocal relation between behavioural and neural information in the context of risky decision-making. More precisely, we offered a way to relate Diffusion Tensor Imaging data to cognitive parameters of a computational model accounting for behavioural outcomes in the popular Balloon Analogue Risk Task (BART). Results show that the proposed architecture has the potential to account for individual differences in task performances and brain structural features by letting individual-level parameters to be modelled by a joint distribution connecting both sources of information. Such a joint modelling framework can offer interesting insights in the development of computational models able to investigate correspondence between decision-making and brain structural connectivity.