Session 4
My Session Status
Sub Sessions
Reasoning about what to believe is often analyzed using probabilities, while reasoning about what to do combines probabilities with utilities. A systematic alternative is provided by coherence theories of belief and decision, in ways that are more psychologically plausible and computationally efficient than probabilities. Nevertheless, powerful new AI models seem to use neither probability or coherence. What does their success tell us about rational t...
Logical abduction has played an important role in the cognitive sciences since its origins. Abduction is a type of ampliative reasoning that together with induction, complement deduction, and altogether constitute the inquiry cycle, according to Charles Peirce, the XIXth century philosopher who characterized abduction.I will address two aspects of the role of logical abduction in the cognitive sciences. One of them concerns whe...
I will show how abduction represents the fundamental form of hypothetical reasoning that also underlies human creativity and scientific discovery. Adopting an eco-cognitive perspective, I will demonstrate that abduction is not a purely internal mental process, but a distributed, embodied, and environmentally embedded activity. Through external manipulations, models, and environmental affordances, humans generate creative hypotheses that turn the world into cogni...
Reasoning with conditionals: abduction and deduction by Serge RobertThis talk will present a model of causal cognition based on abductive reasoning. We will show how this kind of reasoning is systematically used in our cognition to establish causality, when we try to explain past events and predict future events. Contrary to what is usually held, our causal reasoning is not deductive, it is abductive and it is so because it relies mainly on our long-term mem...