In daily life, individuals constantly face decisions. Goal achievement and environmental adaptation often demand deductive and complex reasoning. The complexity of this task increases within interactive settings, that is, contexts where the outcome of an agent's choice also depends on others’ decisions. Individuals' limitations in performing multi-step, high-quality strategic reasoning have been intensively documented, fostering detrimental outcomes in multiple domains, such as social and economic contexts. In this regard, extensive research has reported different types of reasoners, based on their ability to make optimal decisions. Nonetheless, little is known about the neurocognitive factors driving human heterogeneity and the possibility of enhancing agents’ ability to make rational choices in complex reasoning settings. The present work aims to fill this gap through three experimental studies. Study 1 pursues to deepen the current theoretical knowledge of the cognitive mechanisms driving suboptimal strategic decisions, whereas Study 2 explores the feasibility of modeling individuals’ ability to perform optimal choices within interactive settings. The eye-tracking and mouse-tracking techniques have been employed in Study 1 and Study 2, respectively, to disclose individuals’ information search patterns and infer the decision processes underlying choices. Study 3 employs functional magnetic resonance imaging (fMRI) to explore the neural representation of deductive reasoning, specifically aiming to understand whether the neural underpinnings of deduction follow the principle of compositionality. Globally, our results enrich the body of knowledge of humans' reasoning mechanisms and computational abilities. Study 1 suggests that variability in strategic sophistication is driven by unsophisticated representations of the decision-making problem, namely bounded rationality, whereas Study 2 shows that strategic sophistication can be enhanced through training based on feedback exposure, leading to different learning outcomes depending on exogenous and endogenous factors. Eventually, Study 3 sheds light on how, our brain, represents deductive processes, suggesting that, at least in certain phases of deduction, the process follows the principle of compositionality. Altogether, these findings deepen our understanding of complex human reasoning, potentially informing interventions for improving individuals’ deductive and strategic abilities in multiple educational and professional real-life contexts.

Nella vita quotidiana, gli individui si trovano costantemente ad affrontare decisioni. Raggiungimento di obiettivi e adattamento all'ambiente spesso richiedono ragionamenti deduttivi e complessi. La complessità di questo compito aumenta in contesti interattivi, ovvero situazioni in cui l’esito della scelta di un agente dipende anche dalle decisioni di altre persone. Limiti degli individui nell’eseguire ragionamenti strategici di alta qualità, che richiedono più passaggi, sono stati ampiamente documentati, generando esiti negativi in molteplici contesti, come quelli sociali ed economici. In questo contesto, numerose ricerche hanno identificato diversi tipi di ragionatori, in base alla loro capacità di prendere decisioni ottimali. Tuttavia, si sa poco riguardo ai fattori neurocognitivi che determinano l’eterogeneità umana e alla possibilità di potenziare la capacità degli agenti di prendere decisioni razionali in contesti di ragionamento complesso. Il presente lavoro mira a colmare questa lacuna attraverso tre studi sperimentali. Lo Studio 1 cerca di approfondire l’attuale conoscenza teorica circa i meccanismi cognitivi che sottostanno decisioni strategiche non ottimali, mentre lo Studio 2 esplora la possibilità di modellare la capacità degli individui di prendere decisioni ottimali in contesti interattivi. Le tecniche di eye-tracking e mouse-tracking sono state impiegate nello Studio 1 e nello Studio 2, rispettivamente, per indagare il modo in cui gli individui cercano le informazioni e dedurre i processi decisionali alla base delle scelte. Lo Studio 3 utilizza la risonanza magnetica funzionale (fMRI) per esplorare la rappresentazione neurale del ragionamento deduttivo, mirando specificamente a comprendere se le basi neurali della deduzione seguano il principio di composizionalità. Globalmente, i nostri risultati arricchiscono il corpus di conoscenze sui meccanismi di ragionamento e sulle capacità computazionali umane. Lo Studio 1 suggerisce che la variabilità a livello di sofisticatezza strategica sia guidata da rappresentazioni non sofisticate del problema decisionale, ovvero limitata razionalità, mentre lo Studio 2 mostra che la sofisticatezza strategica può essere potenziata attraverso un training basato sull’esposizione a un feedback, portando a diversi risultati di apprendimento a seconda di fattori sia esogeni che endogeni. Infine, lo Studio 3 fa luce su come il nostro cervello rappresenti processi deduttivi, suggerendo che, almeno in certe fasi della deduzione, il processo segua il principio di composizionalità. Nel complesso, questi risultati approfondiscono la nostra comprensione del ragionamento umano complesso, potenzialmente informando interventi per migliorare le capacità deduttive e strategiche degli individui in molteplici contesti di vita reale sia educativi che professionali.

(2024). Unveiling the hidden threads: on the neurocognitive processes of reasoning. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2024).

Unveiling the hidden threads: on the neurocognitive processes of reasoning

DEL MAURO, LILIA
2024

Abstract

In daily life, individuals constantly face decisions. Goal achievement and environmental adaptation often demand deductive and complex reasoning. The complexity of this task increases within interactive settings, that is, contexts where the outcome of an agent's choice also depends on others’ decisions. Individuals' limitations in performing multi-step, high-quality strategic reasoning have been intensively documented, fostering detrimental outcomes in multiple domains, such as social and economic contexts. In this regard, extensive research has reported different types of reasoners, based on their ability to make optimal decisions. Nonetheless, little is known about the neurocognitive factors driving human heterogeneity and the possibility of enhancing agents’ ability to make rational choices in complex reasoning settings. The present work aims to fill this gap through three experimental studies. Study 1 pursues to deepen the current theoretical knowledge of the cognitive mechanisms driving suboptimal strategic decisions, whereas Study 2 explores the feasibility of modeling individuals’ ability to perform optimal choices within interactive settings. The eye-tracking and mouse-tracking techniques have been employed in Study 1 and Study 2, respectively, to disclose individuals’ information search patterns and infer the decision processes underlying choices. Study 3 employs functional magnetic resonance imaging (fMRI) to explore the neural representation of deductive reasoning, specifically aiming to understand whether the neural underpinnings of deduction follow the principle of compositionality. Globally, our results enrich the body of knowledge of humans' reasoning mechanisms and computational abilities. Study 1 suggests that variability in strategic sophistication is driven by unsophisticated representations of the decision-making problem, namely bounded rationality, whereas Study 2 shows that strategic sophistication can be enhanced through training based on feedback exposure, leading to different learning outcomes depending on exogenous and endogenous factors. Eventually, Study 3 sheds light on how, our brain, represents deductive processes, suggesting that, at least in certain phases of deduction, the process follows the principle of compositionality. Altogether, these findings deepen our understanding of complex human reasoning, potentially informing interventions for improving individuals’ deductive and strategic abilities in multiple educational and professional real-life contexts.
REVERBERI, FRANCO
Ragionamento; Decision-making; Process data; Pensiero strategico; Deduzione
Reasoning; Decision-making; Process data; Strategic thinking; Deduction
M-PSI/01 - PSICOLOGIA GENERALE
Italian
13-mag-2024
36
2022/2023
embargoed_20270513
(2024). Unveiling the hidden threads: on the neurocognitive processes of reasoning. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2024).
File in questo prodotto:
File Dimensione Formato  
phd_unimib_770341.pdf

embargo fino al 13/05/2027

Descrizione: Documento di tesi
Tipologia di allegato: Doctoral thesis
Dimensione 4.95 MB
Formato Adobe PDF
4.95 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/478661
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
Social impact