Maura Mezzetti
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Maura Mezzetti has a total of 141 co-authors. Here are the top co-authors by number of shared publications:
- Carlo La Vecchia
University of Milan
4 shared publications - Adriano Decarli
Department of Clinical Sciences and Community Health, University of Milan, Milan, Italy
3 shared publications - Anna Gigli
Institute for Research on Population and Social Policies, National Research Council, Rome, Italy
3 shared publications Chris Robertson
Division of Epidemiology and Biostatistics, European Institute of Oncology, Milan, Italy
3 shared publications- Frederic Yves Bois
Certara UK Limited, Simcyp Division, Level 2-Acero, 1 Concourse Way, Sheffield, S1 2BJ, United Kingdom
3 shared publications - Ilia Negri
Università della Calabria
3 shared publications
Latest Publications
- 17 July 2026
- Published by Springer Nature
- p. 63-68
A Bayesian Framework for Estimating the True Incidence of Gender-Based Violence in Italy
- 17 July 2026
- Published by Springer Nature
- p. 63-68
A Bayesian Framework for Estimating the True Incidence of Gender-Based Violence in Italy
- Open Access
Modeling Psychophysical Data in R: A Comparative Study of Four Model Frameworks
Inferential models in psychophysics are essential for quantifying the relation between physical properties of the stimulus and their perceptual representations. The psychometric function is typically used to model the responses of individual participants in forced-choice experiments. The accuracy and the noise of the response can be estimated from the Point of Subjective Equality (PSE) and the Just Noticeable Difference (JND) of the function, respectively. Traditionally, a two-level approach is used to model the behavior of a group of participants, where psychometric functions are first fitted to individual participant data, followed by hypothesis testing across participants on parameters of interest. Recent studies have introduced alternative approaches based on hierarchical models, such as Generalized Linear Mixed Models (GLMM) and Models within the Bayesian Hierarchical Framework (BHF), to analyze in a single framework the responses from multiple participants. These two approaches can be effectively implemented in R, thanks to its flexibility and robust statistical capabilities. Here, we provide a tutorial on how to model and analyze data from psychophysical experiments in R, using both two-level and hierarchical frameworks. Our goal is to provide researchers with a practical guide for building a complete and reproducible analysis pipeline, using core R functionalities together with custom packages, and facilitate rigorous and efficient data analysis in psychophysics.
Open AccessModeling Psychophysical Data in R: A Comparative Study of Four Model Frameworks
Inferential models in psychophysics are essential for quantifying the relation between physical properties of the stimulus and their perceptual representations. The psychometric function is typically used to model the responses of individual participants in forced-choice experiments. The accuracy and the noise of the response can be estimated from the Point of Subjective Equality (PSE) and the Just Noticeable Difference (JND) of the function, respectively. Traditionally, a two-level approach is used to model the behavior of a group of participants, where psychometric functions are first fitted to individual participant data, followed by hypothesis testing across participants on parameters of interest. Recent studies have introduced alternative approaches based on hierarchical models, such as Generalized Linear Mixed Models (GLMM) and Models within the Bayesian Hierarchical Framework (BHF), to analyze in a single framework the responses from multiple participants. These two approaches can be effectively implemented in R, thanks to its flexibility and robust statistical capabilities. Here, we provide a tutorial on how to model and analyze data from psychophysical experiments in R, using both two-level and hierarchical frameworks. Our goal is to provide researchers with a practical guide for building a complete and reproducible analysis pipeline, using core R functionalities together with custom packages, and facilitate rigorous and efficient data analysis in psychophysics.
- 7 November 2025
- Published by Springer Nature
- p. 169-183
Assessing Academic Career Progression: A Cox Proportional Hazard Model for Stagnant Levels
- 7 November 2025
- Published by Springer Nature
- p. 169-183
Assessing Academic Career Progression: A Cox Proportional Hazard Model for Stagnant Levels
- 17 June 2025
- Published by Springer Nature
- p. 102-107
A Hierarchical Bayesian Model for Combining Multiple Data Sources to Estimate Under-Reporting in Count Data
- 17 June 2025
- Published by Springer Nature
- p. 102-107
A Hierarchical Bayesian Model for Combining Multiple Data Sources to Estimate Under-Reporting in Count Data
- 21 January 2025
- Published by Springer Nature
- p. 330-336
The Efficacy of Social Distancing Measures in Controlling the Spread of SARS-CoV-2: A Retrospective Study in Italy
- 21 January 2025
- Published by Springer Nature
- p. 330-336
The Efficacy of Social Distancing Measures in Controlling the Spread of SARS-CoV-2: A Retrospective Study in Italy
- 4 December 2024
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 34 (3) , 984-993
Latent Markov Time-Interaction Processes
- 4 December 2024
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 34 (3) , 984-993
Latent Markov Time-Interaction Processes
- 27 November 2024
- Published by Springer Nature in Scientometrics
- Vol. 129 (12) , 7443-7474
Hierarchical Bayesian model to estimate and compare research productivity of Italian academic statisticians
- 27 November 2024
- Published by Springer Nature in Scientometrics
- Vol. 129 (12) , 7443-7474
Hierarchical Bayesian model to estimate and compare research productivity of Italian academic statisticians
- Open Access
- 2 March 2023
- Published by Frontiers Media SA in Frontiers in Computational Neuroscience
- Vol. 17, 1108311
Bayesian hierarchical models and prior elicitation for fitting psychometric functions
Our previous articles demonstrated how to analyze psychophysical data from a group of participants using generalized linear mixed models (GLMM) and two-level methods. The aim of this article is to revisit hierarchical models in a Bayesian framework. Bayesian models have been previously discussed for the analysis of psychometric functions although this approach is still seldom applied. The main advantage of using Bayesian models is that if the prior is informative, the uncertainty of the parameters is reduced through the combination of prior knowledge and the experimental data. Here, we evaluate uncertainties between and within participants through posterior distributions. To demonstrate the Bayesian approach, we re-analyzed data from two of our previous studies on the tactile discrimination of speed. We considered different methods to include a priori knowledge in the prior distribution, not only from the literature but also from previous experiments. A special type of Bayesian model, the power prior distribution, allowed us to modulate the weight of the prior, constructed from a first set of data, and use it to fit a second one. Bayesian models estimated the probability distributions of the parameters of interest that convey information about the effects of the experimental variables, their uncertainty, and the reliability of individual participants. We implemented these models using the software Just Another Gibbs Sampler (JAGS) that we interfaced with R with the package rjags. The Bayesian hierarchical model will provide a promising and powerful method for the analysis of psychometric functions in psychophysical experiments.
- 2 March 2023
- Published by Frontiers Media SA in Frontiers in Computational Neuroscience
- Vol. 17, 1108311
Open AccessBayesian hierarchical models and prior elicitation for fitting psychometric functions
Our previous articles demonstrated how to analyze psychophysical data from a group of participants using generalized linear mixed models (GLMM) and two-level methods. The aim of this article is to revisit hierarchical models in a Bayesian framework. Bayesian models have been previously discussed for the analysis of psychometric functions although this approach is still seldom applied. The main advantage of using Bayesian models is that if the prior is informative, the uncertainty of the parameters is reduced through the combination of prior knowledge and the experimental data. Here, we evaluate uncertainties between and within participants through posterior distributions. To demonstrate the Bayesian approach, we re-analyzed data from two of our previous studies on the tactile discrimination of speed. We considered different methods to include a priori knowledge in the prior distribution, not only from the literature but also from previous experiments. A special type of Bayesian model, the power prior distribution, allowed us to modulate the weight of the prior, constructed from a first set of data, and use it to fit a second one. Bayesian models estimated the probability distributions of the parameters of interest that convey information about the effects of the experimental variables, their uncertainty, and the reliability of individual participants. We implemented these models using the software Just Another Gibbs Sampler (JAGS) that we interfaced with R with the package rjags. The Bayesian hierarchical model will provide a promising and powerful method for the analysis of psychometric functions in psychophysical experiments.
- Open Access
- 7 March 2022
- Published by Springer Nature in Statistical Methods & Applications
- Vol. 31 (5) , 1245-1271
A Bayesian approach to model individual differences and to partition individuals: case studies in growth and learning curves
The first objective of the paper is to implement a two stage Bayesian hierarchical nonlinear model for growth and learning curves, particular cases of longitudinal data with an underlying nonlinear time dependence. The aim is to model simultaneously individual trajectories over time, each with specific and potentially different characteristics, and a time-dependent behavior shared among individuals, including eventual effect of covariates. At the first stage inter-individual differences are taken into account, while, at the second stage, we search for an average model. The second objective is to partition individuals into homogeneous groups, when inter individual parameters present high level of heterogeneity. A new multivariate partitioning approach is proposed to cluster individuals according to the posterior distributions of the parameters describing the individual time-dependent behaviour. To assess the proposed methods, we present simulated data and two applications to real data, one related to growth curve modeling in agriculture and one related to learning curves for motor skills. Furthermore a comparison with finite mixture analysis is shown.
- 7 March 2022
- Published by Springer Nature in Statistical Methods & Applications
- Vol. 31 (5) , 1245-1271
Open AccessA Bayesian approach to model individual differences and to partition individuals: case studies in growth and learning curves
The first objective of the paper is to implement a two stage Bayesian hierarchical nonlinear model for growth and learning curves, particular cases of longitudinal data with an underlying nonlinear time dependence. The aim is to model simultaneously individual trajectories over time, each with specific and potentially different characteristics, and a time-dependent behavior shared among individuals, including eventual effect of covariates. At the first stage inter-individual differences are taken into account, while, at the second stage, we search for an average model. The second objective is to partition individuals into homogeneous groups, when inter individual parameters present high level of heterogeneity. A new multivariate partitioning approach is proposed to cluster individuals according to the posterior distributions of the parameters describing the individual time-dependent behaviour. To assess the proposed methods, we present simulated data and two applications to real data, one related to growth curve modeling in agriculture and one related to learning curves for motor skills. Furthermore a comparison with finite mixture analysis is shown.
- 20 November 2019
- Published by Wiley in Statistics in Medicine
- Vol. 39 (1) , 26-44
Combining individual and aggregated data to investigate the role of socioeconomic disparities on cancer burden in Italy
Quantifying socioeconomic disparities and understanding the roots of inequalities are growing topics in cancer research. However, socioeconomic differences are challenging to investigate mainly due to the lack of accurate data at individual‐level, while aggregate indicators are only partially informative. We implemented a multiple imputation algorithm within a statistical matching framework that combines diverse sources of data to estimate individual‐level associations between income and risk of breast and lung cancer, adjusting for potential confounding factors in Italy. The framework is computationally flexible and can be adapted to similar contexts.
- 20 November 2019
- Published by Wiley in Statistics in Medicine
- Vol. 39 (1) , 26-44
Combining individual and aggregated data to investigate the role of socioeconomic disparities on cancer burden in Italy
Quantifying socioeconomic disparities and understanding the roots of inequalities are growing topics in cancer research. However, socioeconomic differences are challenging to investigate mainly due to the lack of accurate data at individual‐level, while aggregate indicators are only partially informative. We implemented a multiple imputation algorithm within a statistical matching framework that combines diverse sources of data to estimate individual‐level associations between income and risk of breast and lung cancer, adjusting for potential confounding factors in Italy. The framework is computationally flexible and can be adapted to similar contexts.