Research shows that humans extract information from various sources, directly or indirectly, to perceive and reason about the cause-effect relations in both everyday lives and in more formal contexts such as science.
Work on causal cognition focuses on mapping and understanding the cognitive processes that are involved in formulating these judgments, with particular regard to thinking about evidence; when and how causal relations are identified; how different sources and types of information are construed and communicated, and with what degree of accuracy; and how the ability to make causal inferences develops.
Over the past years, research in computer science and AI has moved much closer to the modeling of human cognition, aiming to capture a variety of cognitive modes by taking inspiration from some leading psychologists. For example, work has been done on a compositional model of Gardenfors' conceptual spaces, and its relation to Smolensky's architectures.
There is indeed a great drive for making machines mirror human reasoning and make their behaviors more human-like. At the core of these formulations lies the role of causality, inspired both by human-perceived causality and causality in the physical world.
Although investigations in both disciplines have focused on similar territory, the links regarding causal cognition have received less attention, and the implications and potential of the two fields to inform each other are largely unexplored.
This is a particularly relevant moment for researchers coming from these different backgrounds to share their different theoretical frameworks and methodologies.
A thorough understanding of causality is argued to be indispensable for making informed decisions and assuming responsibilities, in both artificial and real-world contexts. Applications include fostering reasoning and learning, developing the ability to identify cause-effect relationships, enhancing problem-solving skills, augmenting predictive power, and improving the accuracy of causal inferences. Such competencies are no longer exclusive to humans, but are drawn upon by the artificial intelligent systems that increasingly permeate our daily lives.
At the same time, the apparent successes and limitations of recent large language models in AI have raised vital questions about the necessity and role of causal reasoning in artificial systems.
Following the first workshop in 2019, the second Interdisciplinary Conference on Causal Cognition in Humans and Machines*, held in Oxford in January 2024, aimed to assemble researchers, industry pioneers, engineers, and students from various disciplines to engage in comprehensive discussions on causal intelligence. The goal was to stimulate an interdisciplinary exploration of causality across psychology, computer science, AI, and real-world applications. The key targets were:
· Exploring the connections between intelligence and causal thinking in both humans and machines
· Establishing an interdisciplinary platform for researchers in causal intelligence from cognitive science, computer science, and AI
· Examining the necessity of causal reasoning for AI in light of current developments (e.g. large language models) and in comparison to human cognition
· Developing new methodologies for analyzing causal cognition and intelligent systems
The topic remains highly relevant four years after Volume I** was launched due to continuous advancements and unresolved challenges in understanding and modelling causal cognition. Despite significant progress, we are still far from even mimicking human causal reasoning within AI systems or, conversely, enhancing human causal thinking using AI systems. This second volume aims to contribute to this reciprocal relationship by highlighting the crucial need for interdisciplinary research and introducing new perspectives and methodologies that have emerged since the first workshop.
We welcome original Research manuscripts, reviews, hypothesis/theory manuscripts, mini-reviews, perspectives, and Brief Research reports.
* https://amcs-community.org/events/causal-cognition-humans-machines/
** Volume I - https://www.frontiersin.org/research-topics/9874/causal-cognition-in-humans-and-machines/overview
Keywords:
causal cognition, computer science, machines
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
Research shows that humans extract information from various sources, directly or indirectly, to perceive and reason about the cause-effect relations in both everyday lives and in more formal contexts such as science.
Work on causal cognition focuses on mapping and understanding the cognitive processes that are involved in formulating these judgments, with particular regard to thinking about evidence; when and how causal relations are identified; how different sources and types of information are construed and communicated, and with what degree of accuracy; and how the ability to make causal inferences develops.
Over the past years, research in computer science and AI has moved much closer to the modeling of human cognition, aiming to capture a variety of cognitive modes by taking inspiration from some leading psychologists. For example, work has been done on a compositional model of Gardenfors' conceptual spaces, and its relation to Smolensky's architectures.
There is indeed a great drive for making machines mirror human reasoning and make their behaviors more human-like. At the core of these formulations lies the role of causality, inspired both by human-perceived causality and causality in the physical world.
Although investigations in both disciplines have focused on similar territory, the links regarding causal cognition have received less attention, and the implications and potential of the two fields to inform each other are largely unexplored.
This is a particularly relevant moment for researchers coming from these different backgrounds to share their different theoretical frameworks and methodologies.
A thorough understanding of causality is argued to be indispensable for making informed decisions and assuming responsibilities, in both artificial and real-world contexts. Applications include fostering reasoning and learning, developing the ability to identify cause-effect relationships, enhancing problem-solving skills, augmenting predictive power, and improving the accuracy of causal inferences. Such competencies are no longer exclusive to humans, but are drawn upon by the artificial intelligent systems that increasingly permeate our daily lives.
At the same time, the apparent successes and limitations of recent large language models in AI have raised vital questions about the necessity and role of causal reasoning in artificial systems.
Following the first workshop in 2019, the second Interdisciplinary Conference on Causal Cognition in Humans and Machines*, held in Oxford in January 2024, aimed to assemble researchers, industry pioneers, engineers, and students from various disciplines to engage in comprehensive discussions on causal intelligence. The goal was to stimulate an interdisciplinary exploration of causality across psychology, computer science, AI, and real-world applications. The key targets were:
· Exploring the connections between intelligence and causal thinking in both humans and machines
· Establishing an interdisciplinary platform for researchers in causal intelligence from cognitive science, computer science, and AI
· Examining the necessity of causal reasoning for AI in light of current developments (e.g. large language models) and in comparison to human cognition
· Developing new methodologies for analyzing causal cognition and intelligent systems
The topic remains highly relevant four years after Volume I** was launched due to continuous advancements and unresolved challenges in understanding and modelling causal cognition. Despite significant progress, we are still far from even mimicking human causal reasoning within AI systems or, conversely, enhancing human causal thinking using AI systems. This second volume aims to contribute to this reciprocal relationship by highlighting the crucial need for interdisciplinary research and introducing new perspectives and methodologies that have emerged since the first workshop.
We welcome original Research manuscripts, reviews, hypothesis/theory manuscripts, mini-reviews, perspectives, and Brief Research reports.
* https://amcs-community.org/events/causal-cognition-humans-machines/
** Volume I - https://www.frontiersin.org/research-topics/9874/causal-cognition-in-humans-and-machines/overview
Keywords:
causal cognition, computer science, machines
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.