This thesis examines inter-rater reliability and cognitive biases in the forensic assessment of criminal responsibility. The evaluation of the capacity to understand and to intend is a complex process in which the expert must integrate clinical data, legal criteria, and a retrospective reconstruction of the defendant’s mental functioning at the time of the offence. Drawing on Dror’s Hierarchy of Expert Performance model, the forensic decision is considered not as a single final outcome, but as a sequence of inferential steps: symptom observation, diagnostic formulation, and assessment of the relationship between mental disorder and criminal responsibility. The empirical study is based on the analysis of 105 Italian forensic cases, with the aim of identifying the stage of the evaluation process at which disagreement between experts emerges. The results show considerable variability: disagreement was found in 45.71% of cases at the symptom level, in 58.10% at the diagnostic level, and in 61.90% in the final assessment of criminal responsibility. These findings suggest that disagreement does not concern only the translation of diagnosis into legal categories, but may already arise during the selection and interpretation of clinical data. Overall, the study highlights the importance of more transparent assessment procedures, explicit criteria, specific training on cognitive biases, and greater awareness of uncertainty in forensic reasoning. From this perspective, the HEP model provides a useful framework for understanding and improving the quality of psychological-forensic assessment.
La presente tesi analizza il tema dell’inter-rater reliability e dei bias cognitivi nella valutazione dell’imputabilità in ambito psicologico-forense. La valutazione della capacità di intendere e di volere rappresenta un processo complesso, in cui l’esperto deve integrare dati clinici, elementi giuridici e ricostruzione retrospettiva del funzionamento mentale al momento del fatto-reato. Alla luce del modello Hierarchy of Expert Performance di Dror, lo studio considera la decisione peritale non come un esito unitario, ma come una sequenza di passaggi inferenziali: osservazione dei sintomi, formulazione diagnostica e valutazione del nesso tra disturbo mentale e imputabilità. La ricerca empirica si basa sull’analisi di 105 casi peritali italiani, con l’obiettivo di individuare il livello del processo valutativo in cui emerge il disaccordo tra consulenti. I risultati mostrano una significativa variabilità: la discordanza compare nel 45,71% dei casi a livello sintomatologico, nel 58,10% a livello diagnostico e nel 61,90% nel giudizio finale sull’imputabilità . Tali dati suggeriscono che il disagreement non riguarda soltanto la traduzione della diagnosi in categorie giuridiche, ma può emergere già nella selezione e interpretazione dei dati clinici . Lo studio evidenzia quindi l’importanza di procedure valutative più trasparenti, criteri espliciti, formazione specifica sui bias cognitivi e maggiore consapevolezza dell’incertezza nel ragionamento peritale. In questa prospettiva, il modello HEP offre una cornice utile per comprendere e migliorare la qualità della valutazione psicologico-forense.
Inter-rater reliability e bias cognitivi nella valutazione dell’imputabilitá: uno studio empirico secondo il modello Hierarchy Expert Performance
TOPPAO, ALICE
2025/2026
Abstract
This thesis examines inter-rater reliability and cognitive biases in the forensic assessment of criminal responsibility. The evaluation of the capacity to understand and to intend is a complex process in which the expert must integrate clinical data, legal criteria, and a retrospective reconstruction of the defendant’s mental functioning at the time of the offence. Drawing on Dror’s Hierarchy of Expert Performance model, the forensic decision is considered not as a single final outcome, but as a sequence of inferential steps: symptom observation, diagnostic formulation, and assessment of the relationship between mental disorder and criminal responsibility. The empirical study is based on the analysis of 105 Italian forensic cases, with the aim of identifying the stage of the evaluation process at which disagreement between experts emerges. The results show considerable variability: disagreement was found in 45.71% of cases at the symptom level, in 58.10% at the diagnostic level, and in 61.90% in the final assessment of criminal responsibility. These findings suggest that disagreement does not concern only the translation of diagnosis into legal categories, but may already arise during the selection and interpretation of clinical data. Overall, the study highlights the importance of more transparent assessment procedures, explicit criteria, specific training on cognitive biases, and greater awareness of uncertainty in forensic reasoning. From this perspective, the HEP model provides a useful framework for understanding and improving the quality of psychological-forensic assessment.| File | Dimensione | Formato | |
|---|---|---|---|
|
Toppao_Alice.pdf
Accesso riservato
Dimensione
3.03 MB
Formato
Adobe PDF
|
3.03 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/110650