Soutenance de thèse – Flora Gautheron

Quand :
27 septembre 2023 @ 9 h 00 min – 12 h 00 min
Où :
Amphi de la MSH Alpes
1251 rue des universités - 38400 Saint Martin d'Hères

Titre de la thèseÉtude empirique et modélisation computationnelle des dynamiques de prise de décision en cognition morale

Composition du Jury :

  • Mme Annique SMEDING, PR, Université Savoie Mont Blanc, Direction de thèse
  • M. Wim DE NEYS, DR, CNRS Délégation Paris Centre, Rapporteur
  • M. Giovanni PEZZULO, DR, National Research Council of Italy, Rapporteur
  • Mme Gaëlle VALLEE-TOURANGEAU, FULL PROFESSOR Kingston Business School, UK, Examinateur
  • M. Laurent BEGUE-SHANKLAND, PR, Université Grenoble Alpes, Examinateur
  • M. QUINTON Jean-Charles, Université Grenoble Alpes, Invité

Abstract :

Everyday decisions often involve moral considerations, encompassing a broad range of situations. Despite extensive research on the different psychological factors involved in moral decision-making, few studies have specifically examined the influence of moral factors on the decision-making process and its underlying mechanisms. The present thesis defends the pluralism of moral factors and their substantial varying influence on the decision-making process depending on the situation and the type of decisions involved. By integrating preregistered empirical studies and computational modeling of a wide range of everyday decisions, this thesis aims to unravel some of the underlying mechanisms that drive moral decisions. After introducing the scientific context with an overview of relevant literature, we will present a core model based on differential equations, employing a sequential sampling model coupled with sensorimotor control. This model covers dynamic neural fields model and extends classical drift diffusion models, by being suitable for continuous decision spaces. By successfully reproducing non-linear dynamics of human decision-making, as observed through mouse tracking techniques, the model demonstrates its utility in investigating moral and nonmoral decision processes. We then focus on the influence of decision situations, examining how paradigms affect moral decision-making. Through two experiments (N = 104 and N = 65) associated with computational modeling, this research uncovers the constraint induced by a binary (compared to a continuous) response mode, while proposing an adapted mouse-tracking paradigm to investigate moral decision-making in our subsequent studies. To examine the impact of decision factors, we then investigate conflicting moral factors in moral judgments. Through two empirical studies (N = 80 and N = 84), this research investigates how conflicting moral considerations, such as intent, action outcome, and causality, influence decisions and their underlying cognitive mechanisms. The findings reveal diverse dynamic influences of these moral factors, demonstrating their different impacts on decision processes.

We finally expand the investigation to scenarios where moral factors conflict with nonmoral factors, such as self-interest in two empirical (N = 52 and N = 42) and computational studies; as well as taste and health in dietary decisions in two other empirical studies (N = 77 and N = 92) and one associated computational model. The first study indicated that moral considerations tend to extremize decisions, additionally impacting the decision-making dynamics. However, when moral factors conflict with self-interest, everyday moral decisions exhibit greater stability, while still seeking compromise.

This pattern of compromise between moral and nonmoral factors was also observed in the context of dietary decisions. Throughout these studies, the computational model serves as a valuable tool to test hypotheses regarding the impact and dynamics of moral factors in decision-making processes, including reduced or enhanced impact and early or delayed influence, depending on the situation. This research thus highlights the intricate dynamics of conflicting decision factors and their role in shaping distinct underlying moral compared to nonmoral decision processes. By integrating empirical studies and computational modeling, this comprehensive approach offers complementary perspectives, enriching our knowledge of decision-making phenomena at both the behavioral and cognitive levels. Overall, this thesis advances the understanding of moral decision-making processes and offers new avenues for future research in the field.