Insights on the self from a cognitive robotics perspective. A theoretical framework for the design of social robots

Despite the high level of performance robots are able to achieve in task-specific contexts, they are still far from reaching intuitiveness in social interactions and versatility in making use of their skills in different contexts. Cognitive robotics is trying to address such issues by providing robots with cognitive architectures that integrate and dynamically coordinate the core cognitive abilities, enabling the agent to show flexible and context-sensitive behavior during interaction in order to cope with unpredictable environments. Inspired by the functioning of human cognition, these models are being studied with the aim of improving autonomy in robotic platforms. Furthermore, the use of humanoid robots extends this objective toward the perspective of embodied cognition. My project aligns with this approach with a specific focus on the concept of self. Conceived as an operational definition to describe the subjective character of human experience, the self is, in my opinion, a suitable concept to address the problem of autonomy in robots. Indeed, it is precisely the unity and consistency of the self that constitute a reference point to generalize previous experience and to engage in a meaningful and effective contact with the world. Such a subjective continuity, analyzed in some of its main aspects in order to outline a theoretical framework, is the object of my dissertation. Within the first section, an examination of the prevalent terminology in discussions surrounding the self across various disciplines highlights the importance of selecting appropriate terms based on context and intended emphasis. In particular, the focus is on last century’s research and theories, whose contribution is more visible in the contemporary discourse. The resultant taxonomy and the proposed selection of four core elements of the self, emerged from a thorough examination of how the concept of self and its related notions have been employed in philosophical and scientific discussions. In the effort not to lose its richness and complexity, the synthesis proposed aims to serve also as a practical framework for understanding the self in the field of cognitive robotics. In section II, the interdisciplinary perspective is prominent in the dialogue between philosophy and robotics. The concept of experience emerges as a pivotal point in fostering collaboration between the two disciplines: the idea is to pass from the integration of information to the interpretation of information on the basis of the embodied self. This cognitive-experiential structure is proposed to allow robots to be more autonomous in learning from interaction with the environment and others. Adaptivity proves to be another crucial element to integrate, within an ever-changing context, invariable features and evolving components. Therefore, a proposal is put forward to foster a natural and adaptive social interaction that relies on a hermeneutical circle between context and experience for robots collaborating with humans. A synthesis between the phenomenological approach in philosophy and the Predictive Processing (PP) approaches in neuroscience is attempted in section III. With the aim of a concrete proposal to integrate the tripartite conception of temporality by Edmund Husserl and the one by PP approaches, with the idea of a minimal self, the pipeline for a possible cognitive architecture is put forward. The coordination between different modules of the architecture combines top-down and bottom-up strategies to account for the temporal unfolding of the perceptual process, in a dynamic that integrates and interprets real-time perception with predictions and prior experience. Such a structure is conceived for an embodied artificial agent and could be of inspiration for future implementations.