Federico Mancinelli
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Federico Mancinelli has a total of 80 co-authors. Here are the top co-authors by number of shared publications:
- Dominik R. Bach
Department of Imaging Neuroscience, UCL Queen Square Institute of Neurology, University College London, London, UK
10 shared publications Huaiyu Liu
6 shared publications- Juliana Katharina Sporrer
Wellcome Centre for Human Neuroimaging, University College London, London, UK
4 shared publications Filip Melinscak
Department of Psychiatry, Psychotherapy, and Psychosomatics, University of Zurich, Zurich, Switzerland
3 shared publicationsJosua Zimmermann
Department of Psychiatry, Psychotherapy, and Psychosomatics, University of Zurich, Zurich, Switzerland
3 shared publicationsBrooks King-Casas
2 shared publications
Latest Publications
- Open Access
- 2 September 2026
- Published by Center for Open Science in PsyArXiv
Benchmarks for Associative Learning Models
Associative learning, like many mature fields of behavioural science, is endowed with a nearly unfathomable wealth of empirical data. These arise from multiple experimental procedures and animal species. To assess any proposed theory of associative learning, decisions must be made about which behavioral phenomena to account for. However, there is currently no systematic approach to selecting observations within this body of evidence. Here, we propose benchmarks for theories of associative learning by organizing phenomena according to their robustness across experiments, their generality across procedural variations, domains, and species, and the degree to which they occur under broad but clearly specified circumstances. We queried the expert community for phenomena to be included, reviewed the relevant literature, classified the phenomena into four categories of robustness and generality, and compared our classification with a second survey of the expert community. We identified 92 benchmark phenomena, of which 21 are domain- and species-general, and 24 are species- and/or domain-specific but highly robust across procedures. Strikingly, some textbook phenomena that have motivated influential and general associative learning theories are not among these. The remaining 47 phenomena were either constrained to particular experimental procedures or of inconclusive generality. Some of their implementational boundary conditions appear to be informally known but were rarely investigated systematically. In summary, our review provides a comprehensive evaluation of the current evidence, which will support theory building while also re-focusing research into areas in which evidence is lacking.
- 2 September 2026
- Published by Center for Open Science in PsyArXiv
Open AccessBenchmarks for Associative Learning Models
Associative learning, like many mature fields of behavioural science, is endowed with a nearly unfathomable wealth of empirical data. These arise from multiple experimental procedures and animal species. To assess any proposed theory of associative learning, decisions must be made about which behavioral phenomena to account for. However, there is currently no systematic approach to selecting observations within this body of evidence. Here, we propose benchmarks for theories of associative learning by organizing phenomena according to their robustness across experiments, their generality across procedural variations, domains, and species, and the degree to which they occur under broad but clearly specified circumstances. We queried the expert community for phenomena to be included, reviewed the relevant literature, classified the phenomena into four categories of robustness and generality, and compared our classification with a second survey of the expert community. We identified 92 benchmark phenomena, of which 21 are domain- and species-general, and 24 are species- and/or domain-specific but highly robust across procedures. Strikingly, some textbook phenomena that have motivated influential and general associative learning theories are not among these. The remaining 47 phenomena were either constrained to particular experimental procedures or of inconclusive generality. Some of their implementational boundary conditions appear to be informally known but were rarely investigated systematically. In summary, our review provides a comprehensive evaluation of the current evidence, which will support theory building while also re-focusing research into areas in which evidence is lacking.
- 20 August 2026
- Published by Elsevier in Neuroscience & Biobehavioral Reviews
- Vol. 190
A systematic review of human avoidance learning: Cognition, computation, and methods
- 20 August 2026
- Published by Elsevier in Neuroscience & Biobehavioral Reviews
- Vol. 190
A systematic review of human avoidance learning: Cognition, computation, and methods
- Open Access
- 3 August 2026
- Published by Center for Open Science in PsyArXiv
Triangulating the Cognitive Mechanisms of Naturalistic Human Avoidance Learning
Learning to avoid harm is essential for wellbeing, and its dysregulation is thought to underlie anxiety-related disorders. To date, there is no generally accepted computational theory of human avoidance learning. A potential reason is the absence of systematic empirical benchmarks capable of distinguishing competing theories. Here, we test 12 canonical and novel associative learning phenomena in a naturalistic avoidance paradigm across six preregistered virtual reality experiments in healthy humans (total N = 599). Our findings converge on three computational principles. First, avoidance learning is predominantly inference-based: participants updated avoidance following signalled US omission and transition structure changes. Second, it reflects cost-benefit integration: spontaneous extinction and retrospective revaluation emerged under naturalistic motor costs, but gave way to residual avoidance when costs were removed. Third, learning about compound cues suggests configural rather than linear feature integration. These principles, and the detailed pattern of findings, can guide the development of candidate computational theories.
- 3 August 2026
- Published by Center for Open Science in PsyArXiv
Open AccessTriangulating the Cognitive Mechanisms of Naturalistic Human Avoidance Learning
Learning to avoid harm is essential for wellbeing, and its dysregulation is thought to underlie anxiety-related disorders. To date, there is no generally accepted computational theory of human avoidance learning. A potential reason is the absence of systematic empirical benchmarks capable of distinguishing competing theories. Here, we test 12 canonical and novel associative learning phenomena in a naturalistic avoidance paradigm across six preregistered virtual reality experiments in healthy humans (total N = 599). Our findings converge on three computational principles. First, avoidance learning is predominantly inference-based: participants updated avoidance following signalled US omission and transition structure changes. Second, it reflects cost-benefit integration: spontaneous extinction and retrospective revaluation emerged under naturalistic motor costs, but gave way to residual avoidance when costs were removed. Third, learning about compound cues suggests configural rather than linear feature integration. These principles, and the detailed pattern of findings, can guide the development of candidate computational theories.
- Open Access
- 24 July 2026
- Published by Center for Open Science in PsyArXiv
PupilFear: benchmarking pupillometry methods for fear conditioning on a distributed dynamical database
Behavioral science increasingly relies on complex, computational pipelines to infer latent psychological variables from time-series data. Because the same data can often be processed using multiple plausible approaches, analytic choices may substantially affect results, yet systematic method comparisons remain uncommon. Here, we adress this problem for pupillometric measurement of fear conditioning. We introduce M-BIDS, a constrained, machine-interpretable extension of the Brain Imaging Data Structure, together with a distributed dynamic database (DDDB) architecture for harmonizing source data stored across multiple repositories. Using this framework, we assembled PupilFear, comprising 361 pupillometry datasets from fear-acquisition experiments. We benchmarked more than 700 candidate measurement methods spanning alternative preprocessing and response-quantification choices. Performance was evaluated using retrodictive validity, and decision tree regression was used to identify the analytic decisions most strongly associated with performance. The distinction between condition- and trial-wise modelign was the dominant determinant of retrodictive validity. Condition-wise model-based approaches showed relatively stable performance across parameter settings, whereas trial-wise approaches were substantially more sensitive to preprocessing choices. Nevertheless, the best condition-wise and trial-wise methods achieved comparable retrodictive validity (d = 0.65). Methods also differed in their sensitivity to reinforced trials (US bias). Peak-scoring approaches showed lower retrodictive validity overall (d = 0.45) but were less influenced by reinforcement-related responses. These finding provide practical guidance for selecting pupillometry pipelines according to the research question. The publicly available PupilFear corpus and benchmarking workflow also enable researchers to reproduce the analyses and evaluate new methods against a common reference.
- 24 July 2026
- Published by Center for Open Science in PsyArXiv
Open AccessPupilFear: benchmarking pupillometry methods for fear conditioning on a distributed dynamical database
Behavioral science increasingly relies on complex, computational pipelines to infer latent psychological variables from time-series data. Because the same data can often be processed using multiple plausible approaches, analytic choices may substantially affect results, yet systematic method comparisons remain uncommon. Here, we adress this problem for pupillometric measurement of fear conditioning. We introduce M-BIDS, a constrained, machine-interpretable extension of the Brain Imaging Data Structure, together with a distributed dynamic database (DDDB) architecture for harmonizing source data stored across multiple repositories. Using this framework, we assembled PupilFear, comprising 361 pupillometry datasets from fear-acquisition experiments. We benchmarked more than 700 candidate measurement methods spanning alternative preprocessing and response-quantification choices. Performance was evaluated using retrodictive validity, and decision tree regression was used to identify the analytic decisions most strongly associated with performance. The distinction between condition- and trial-wise modelign was the dominant determinant of retrodictive validity. Condition-wise model-based approaches showed relatively stable performance across parameter settings, whereas trial-wise approaches were substantially more sensitive to preprocessing choices. Nevertheless, the best condition-wise and trial-wise methods achieved comparable retrodictive validity (d = 0.65). Methods also differed in their sensitivity to reinforced trials (US bias). Peak-scoring approaches showed lower retrodictive validity overall (d = 0.45) but were less influenced by reinforcement-related responses. These finding provide practical guidance for selecting pupillometry pipelines according to the research question. The publicly available PupilFear corpus and benchmarking workflow also enable researchers to reproduce the analyses and evaluate new methods against a common reference.
- Open Access
- 4 June 2026
- Published by Springer Nature in Behavior Research Methods
- Vol. 58 (7)
Experiment-based calibration: Inference and decision-making
Experiment-based calibration is an emerging approach for measurement validation in the behavioural sciences. It allows comparing multiple measurement methods by how well they reproduce a known experimental manipulation, providing insight into their measurement accuracy. Calibration entails questions unparalleled in classical validation approaches. The first is about inference: when should we conclude that one measurement method is truly more accurate than another? The second is about decisions: when should we decide that a method merits the investment of changing a measurement system? In this note, we review these questions in the context of the statistical challenges that arise in a calibration process: a potentially large and a priori unknown number of measurement methods; a requirement to integrate evidence across multiple calibration samples; and a possibility that some methods may not be available for all samples. We show that Bayesian meta-analytic model comparison is a suitable framework for inference in calibration, and propose a decision-theoretic approach to calculate immediate economic gain garnered through reduced sample sizes. In order to overcome the practical hurdles associated with the analysis of calibration experiments, we present CalibR , an R package for calibration inference.
- 4 June 2026
- Published by Springer Nature in Behavior Research Methods
- Vol. 58 (7)
Open AccessExperiment-based calibration: Inference and decision-making
Experiment-based calibration is an emerging approach for measurement validation in the behavioural sciences. It allows comparing multiple measurement methods by how well they reproduce a known experimental manipulation, providing insight into their measurement accuracy. Calibration entails questions unparalleled in classical validation approaches. The first is about inference: when should we conclude that one measurement method is truly more accurate than another? The second is about decisions: when should we decide that a method merits the investment of changing a measurement system? In this note, we review these questions in the context of the statistical challenges that arise in a calibration process: a potentially large and a priori unknown number of measurement methods; a requirement to integrate evidence across multiple calibration samples; and a possibility that some methods may not be available for all samples. We show that Bayesian meta-analytic model comparison is a suitable framework for inference in calibration, and propose a decision-theoretic approach to calculate immediate economic gain garnered through reduced sample sizes. In order to overcome the practical hurdles associated with the analysis of calibration experiments, we present CalibR , an R package for calibration inference.
- Open Access
- 2 March 2026
- Published by Center for Open Science in PsyArXiv
Optimal quantification of fear conditioning from skin conductance data
Accurate quantification of fear-conditioned skin conductance responses (SCR) depends critically on the validity of the underlying model of the psychophysiological process. In the framework of Psychophysiological Modelling (PsPM), the peripheral response model linking sudomotor nerve activity (SNA) to observed skin conductance changes has been extensively studied. In contrast, the neural model linking conditioned stimuli (CS) to SNA has not been systematically evaluated. Here, we assessed the retrodictive validity of a comprehensive set of model specifications, preprocessing choices, and parameter mappings using seventeen publicly available fear-conditioning datasets (N = 543). We examined short (≤4 s) and long (>6 s) CS–US intervals separately, and for long intervals tested both single- and dual-component neural response models. For both interval types, several optimized PsPM methods substantially outperformed the current PsPM default and all traditional peak scoring approaches. Bootstrap analyses showed that at least 100-200 participants are required to reliably distinguish method performance. These findings provide empirically validated recommendations for SCR quantification and highlight the importance of calibration-based optimization in psychophysiological measurement.
- 2 March 2026
- Published by Center for Open Science in PsyArXiv
Open AccessOptimal quantification of fear conditioning from skin conductance data
Accurate quantification of fear-conditioned skin conductance responses (SCR) depends critically on the validity of the underlying model of the psychophysiological process. In the framework of Psychophysiological Modelling (PsPM), the peripheral response model linking sudomotor nerve activity (SNA) to observed skin conductance changes has been extensively studied. In contrast, the neural model linking conditioned stimuli (CS) to SNA has not been systematically evaluated. Here, we assessed the retrodictive validity of a comprehensive set of model specifications, preprocessing choices, and parameter mappings using seventeen publicly available fear-conditioning datasets (N = 543). We examined short (≤4 s) and long (>6 s) CS–US intervals separately, and for long intervals tested both single- and dual-component neural response models. For both interval types, several optimized PsPM methods substantially outperformed the current PsPM default and all traditional peak scoring approaches. Bootstrap analyses showed that at least 100-200 participants are required to reliably distinguish method performance. These findings provide empirically validated recommendations for SCR quantification and highlight the importance of calibration-based optimization in psychophysiological measurement.
- Open Access
- 11 June 2025
- Published by Center for Open Science in PsyArXiv
Experiment-based calibration: inference and decision-making
Experiment-based calibration is an emerging approach for measurement validation. It allows comparing multiple measurement methods against each other by how well they reproduce a known experimental effect, which is informative about measurement accuracy. Calibration entails statistical questions unparal- leled in classical validation approaches. The first question is about inference: when should we conclude that one measurement method is truly more accurate than another? The second is about decisions: when should we decide that a method merits the investment of changing a measurement system? In this note, we review the particular challenges that arise in the context of a calibration process: a potentially large and a priori unknown number of measurement methods, a requirement to integrate evidence across multiple calibration samples, and a possibility that some methods are not available for all samples. We show that Bayesian meta-analytic model comparison is a suitable framework for inference in calibration, and propose a decision-theoretic approach to calculate immediate economic gain garnered through reduced sample sizes. In order to overcome the practical hurdles associated with the analysis of calibration experiments, we furnish calibr, an R package for calibration inference.
- 11 June 2025
- Published by Center for Open Science in PsyArXiv
Open AccessExperiment-based calibration: inference and decision-making
Experiment-based calibration is an emerging approach for measurement validation. It allows comparing multiple measurement methods against each other by how well they reproduce a known experimental effect, which is informative about measurement accuracy. Calibration entails statistical questions unparal- leled in classical validation approaches. The first question is about inference: when should we conclude that one measurement method is truly more accurate than another? The second is about decisions: when should we decide that a method merits the investment of changing a measurement system? In this note, we review the particular challenges that arise in the context of a calibration process: a potentially large and a priori unknown number of measurement methods, a requirement to integrate evidence across multiple calibration samples, and a possibility that some methods are not available for all samples. We show that Bayesian meta-analytic model comparison is a suitable framework for inference in calibration, and propose a decision-theoretic approach to calculate immediate economic gain garnered through reduced sample sizes. In order to overcome the practical hurdles associated with the analysis of calibration experiments, we furnish calibr, an R package for calibration inference.
- Open Access
- 11 June 2025
- Published by Center for Open Science in PsyArXiv
A systematic review of human avoidance learning: Cognition, computation, and methods
Avoidance behaviour is fundamental for survival but can become maladaptive in clinical conditions. A large literature has accumulated on the dynamics of human avoidance learning, but current theories do not provide an exhaustive account of the evidence. In this systematic review, we identify N = 116 studies on human avoidance learning. We analyse them with the goal of establishing empirical phenomena to include in theory-building, and examine their diagnostic value in differentiating between competing theories. We find that the evidence consistently contradicts theories based on associative learning, and favours inference-based accounts. Existing theories within this framework, however, do not predict several of the identified phenomena, calling for further theoretical development. Methodologically, we observe that the problem setting in the most common experimental paradigms is structurally less complex than real-world settings, and unlikely to expose the limitations of inference-based mechanisms. In consequence, we argue that paradigms with larger computational demands and a wider range of action options are required to expose the mechanisms underlying avoidance learning. Collectively, these insights provide a foundation for theory development and methodological innovation, with significant implications for advancing clinical interventions aimed at maladaptive avoidance behaviour.
- 11 June 2025
- Published by Center for Open Science in PsyArXiv
Open AccessA systematic review of human avoidance learning: Cognition, computation, and methods
Avoidance behaviour is fundamental for survival but can become maladaptive in clinical conditions. A large literature has accumulated on the dynamics of human avoidance learning, but current theories do not provide an exhaustive account of the evidence. In this systematic review, we identify N = 116 studies on human avoidance learning. We analyse them with the goal of establishing empirical phenomena to include in theory-building, and examine their diagnostic value in differentiating between competing theories. We find that the evidence consistently contradicts theories based on associative learning, and favours inference-based accounts. Existing theories within this framework, however, do not predict several of the identified phenomena, calling for further theoretical development. Methodologically, we observe that the problem setting in the most common experimental paradigms is structurally less complex than real-world settings, and unlikely to expose the limitations of inference-based mechanisms. In consequence, we argue that paradigms with larger computational demands and a wider range of action options are required to expose the mechanisms underlying avoidance learning. Collectively, these insights provide a foundation for theory development and methodological innovation, with significant implications for advancing clinical interventions aimed at maladaptive avoidance behaviour.
- Open Access
- 25 June 2024
- Published by Center for Open Science in PsyArXiv
The CogLearn Toolkit for Unity: Validating A Virtual Reality Paradigm for Human Avoidance Learning
Avoidance learning encompasses the acquisition of behaviours that enable individuals to evade or withdraw from potentially harmful stimuli, prior to their occurrence. Maladaptive avoidance is a crucial feature of anxiety and trauma-related disorders. In biological and clinical settings, avoidance behaviours usually involve uninstructed, idiosyncratic and complex motor actions. However, there is a lack of laboratory paradigms that allow investigating how such actions are acquired. To fill this gap, we developed a wireless virtual reality platform to investigate avoidance learning in naturalistic settings, with an uncomfortable sound as unconditioned stimulus (US), a physically plausible avoidance action, and allowing for unconstrained movements. This platform, the CogLearn Toolkit for Unity, is publicly available and allows conducting various types of learning experiments with only text files as input. We validated this platform in an exploration-confirmation approach with 5 independent experiments. Overall, participants showed successful acquisition of avoidance behaviour in all experiments. In three exploration experiments, we refined the paradigm and identified mean distance from sound source during conditioned stimulus (CS) presentation (before US occurs) as sensitive measure of avoidance among several dependent variables. Two confirmation experiments revealed stronger avoidance for CS+ than CS- during avoidance learning, whether or not this phase was preceded by Pavlovian acquisition. Furthermore, we demonstrated reduced avoidance during extinction with instruction to approach CS, but persistent residual avoidance during this phase. We found evidence of reinstatement in one of two confirmation experiments. Overall, our study provides robust evidence supporting the efficacy of our paradigm to study avoidance learning in conditions of high ecological relevance.
- 25 June 2024
- Published by Center for Open Science in PsyArXiv
Open AccessThe CogLearn Toolkit for Unity: Validating A Virtual Reality Paradigm for Human Avoidance Learning
Avoidance learning encompasses the acquisition of behaviours that enable individuals to evade or withdraw from potentially harmful stimuli, prior to their occurrence. Maladaptive avoidance is a crucial feature of anxiety and trauma-related disorders. In biological and clinical settings, avoidance behaviours usually involve uninstructed, idiosyncratic and complex motor actions. However, there is a lack of laboratory paradigms that allow investigating how such actions are acquired. To fill this gap, we developed a wireless virtual reality platform to investigate avoidance learning in naturalistic settings, with an uncomfortable sound as unconditioned stimulus (US), a physically plausible avoidance action, and allowing for unconstrained movements. This platform, the CogLearn Toolkit for Unity, is publicly available and allows conducting various types of learning experiments with only text files as input. We validated this platform in an exploration-confirmation approach with 5 independent experiments. Overall, participants showed successful acquisition of avoidance behaviour in all experiments. In three exploration experiments, we refined the paradigm and identified mean distance from sound source during conditioned stimulus (CS) presentation (before US occurs) as sensitive measure of avoidance among several dependent variables. Two confirmation experiments revealed stronger avoidance for CS+ than CS- during avoidance learning, whether or not this phase was preceded by Pavlovian acquisition. Furthermore, we demonstrated reduced avoidance during extinction with instruction to approach CS, but persistent residual avoidance during this phase. We found evidence of reinstatement in one of two confirmation experiments. Overall, our study provides robust evidence supporting the efficacy of our paradigm to study avoidance learning in conditions of high ecological relevance.
- Open Access
- 17 April 2024
- Published by Elsevier in Journal of Psychiatric Research
- Vol. 175, 470-478
Attachment and borderline personality disorder as the dance unfolds: A quantitative analysis of a novel paradigm
- 17 April 2024
- Published by Elsevier in Journal of Psychiatric Research
- Vol. 175, 470-478
Open AccessAttachment and borderline personality disorder as the dance unfolds: A quantitative analysis of a novel paradigm