Complex Problem Solving through Human-AI Collaboration: Literature Review on Research Contexts

Abstract

Solving complex problems has been proclaimed as one major challenge for hybrid teams of humans and artificial intelligence (AI) systems. Human-AI collaboration brings immense opportunities in these complex tasks, in which humans struggle, but full automation is also impossible. Understanding and designing human-AI collaboration for complex problem solving is a wicked and multifaceted research problem itself. We contribute to this emergent field by reviewing to what extent existing research on instantiated human-AI collaboration already addresses this challenge. After clarifying the two key concepts (complex problem solving and human-AI collaboration), we perform a systematic literature review. We extract research contexts and assess them considering different complexity features. We thereby provide an overview of existing and guidance for designing new, suitable research contexts for studying complex problem solving through human-AI collaboration and present an outlook for further work on this research challenge.

The relation of complex problem solving with reflective abstraction: a systematic literature review

“…Table 1. Summary of research findings of the included studies Type of content References Relation CPS to knowledge i) in the CPS process, knowledge is used to understand problems and is applied to achieve goals [20]- [22]; ii) knowledge development is required for complex problem solving [23]; iii) the knowledge possessed, and the environment can stimulate CPS abilities on cognitive and environmental factors [7], [24]; iv) knowledge is built through abstraction to complex problems solving [25]; v) acquiring knowledge of the problem becomes the relationship between intelligence and the application of knowledge in the CPS process [26]- [28]; vi) CPS includes a large number of human higher-order cognitive processes [29]; vii) CPS is related to knowledge and understanding [27], [30], [31]; viii) the importance of knowledge tracking in CPS [32]; ix) there is influence of knowledge and cognitive ability on CPS performance [33], [34]; x) there was no interaction between CPS ability and gender, even though there were differences in math ability [35]; xi) knowledge of the problem will have a positive impact on the success of CPS [1]; xii) through learning can develop skills and experience in dealing with CPS [36]; xiii) metacognition plays a role in increasing motivation in the CPS process [37]; xiv) the roadmap helps the cultivation of knowledge and skills in CPS [38]; xv) problem-solving-based mathematics textbooks can increase knowledge in CPS [39]; xvi) the use of technology in learning can encourage knowledge in CPS [40]; xvii) the contribution of knowledge-based students by making examples based on their own experiences can train students' CPS [3], [41]; xviii) intelligence is highly correlated with CPS performance if the level of knowledge is moderate and has a small correlation if knowledge is very little or very much [16], [42]; xix) CPS as the application of knowledge and skills in the real world [43]; xx) knowledge as a basis for decision-making on CPS [44]; ...…”

Human-AI collaboration by design

“…The nature of problems across many domains is evolving to become highly knowledge-intensive, interdisciplinary, and complex, surpassing the capabilities of individual humans and specialized AI (Memmert and Bittner, 2022). HAIC emerges as a promising paradigm to harness the complementary strengths of humans and AI for problem-solving, insight generation, and value creation.…”