Neurocomputational modelling of human decision making and volition : from neural mechanisms to behavioral outcomes
Year: 2024
Type: dissertation
Abstract: In today's world, amidst the bombardment of information, the mental health and well-being of a society with higher level of responsibility hinge on a central issue related to decision making and human agency. The essence of studying these issues lies in unraveling the role of individual and social life experiences in evolving human choices and intentional actions. Researchers aim to unravel the foundational principles shaping individual behavior by comprehending how the brain, as a complex system, can be conceptualized, understood psychologically, and addressed socially. \n \n \n \nIn this thesis, neurocomputational models are developed to bridge the gaps between micro (neuronal), meso (brain areas), and macro (cognition/behavior) levels. These models primarily focus on the mesoscale neurodynamics of cortical structures, with the goal of linking neural structures, functions, and the influences of internal and environmental factors on decision-making processes and volitional action control. These are neurally-inspired models, providing insight into the dynamics of neural oscillations through attractor networks, recognized as distinctive markers of various cognitive functions. My research is divided into two parts based on studying decision making without and with considering human agency, respectively. \n \n \n \nIn the first part, the focus is on studying the neural mechanisms underlying decision making without explicitly considering human agency. Two slightly different neurocomputational models are developed based on the sources of input information, i.e. either internal or external. In both these models, an integration of rational and emotional processes is the essence of decision-making process. This interplay results in different model behaviors due to the sources of information: 1) the subjective values and attitudes and 2) the observational-based perceived behavior of others. The former model type addresses how an individual's behavior dynamically changes in response to the interplay between rational and emotional factors regarding internal signals while the latter one deals with the social adaptive characteristic of an individual, where dynamic changes in her behaviors are connected with the impact of trust on rational-emotional interactions. This part was part of the EU-funded project COMPLEX - Knowledge Based Climate Mitigation Systems for a Low Carbon Economy. \n \n \n \nIn the second part, the central premise is understanding volitional decision-making process while individual’s actions are goal-directed guided by either internal or predicted external triggers. In this context, fundamental questions concerning the causal efficacious of intentions in decision making, as well as distinctions between 1) self-initiated and 2) externally-triggered actions, were explored. In this regard, a neurocomputational model has been developed to study the neurodynamics of structures involved in the intentional preparatory process of these two volitional processes, while shedding light on the dynamics of attractor networks and neurodynamical changes. Furthermore, the obtained results in both volitional contexts have been compared qualitatively with real EEG data to validate the models as well as explain the observed real neural behavior regarding this descriptive comparison. This research was conducted as a part of the Neurophilosophy of Free Will project funded by a joint Templeton/Fetzer grant. \n \n \n \nRegarding the fact that the simulation results mimic EEG and MEG readouts, further qualitative comparisons with experimental and clinical data should be conducted in the future. However, due to limitations in the available data, the next step can involve providing data with higher resolution using source-localized electrodes. Additionally, I intend to disentangle a meta-model to investigate how two models of self-initiated and externally-triggered actions can be transformed to each other. This exploration can provide insights into the behavioral adaptations of human beings in everyday life.
Author: Azadeh Hassannejad Nazir
Institution: Karolinska Institutet
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: green
Funder: Fetzer Institute
OpenAlex ID: W6993000458
