Artificial Intelligence and High-Level Cognition

Ragni, Marco

doi:10.1007/978-3-030-06170-8_14

Cited by 5 publications

(4 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, we aim to investigate whether preferences over answer sets or weighted counting could allow for more detailed modeling of cognitive principles. Furthermore, inspired by our idea of employing existing techniques from AI, and as already mentioned in the introduction, the cognitive science community could discuss and design an event establishing benchmarks for human reasoning tasks as suggested in (Ragni, 2020) explaining different majority responses using one or many existing frameworks.…”

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…An additional challenge for cognitive reasoning is to identify the relevant problems that a model should account for (Ragni, 2020). Therefore, Ragni (2020) suggested establishing generally accepted benchmarks, similar to the PRECORE Challenge (Ragni, Riesterer, & Khemlani, 2019) for human reasoning tasks. The evaluation of this challenge was done with the benchmarking tool Cognitive COmputation for Behavioral Reasoning Analysis (CCOBRA) framework (Riesterer & Shadownox, 2021).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Quantitative Symbolic Approach to Individual Human Reasoning

Dietz¹,

Fichte²,

Hamiti³

2022

Preprint

View full text Add to dashboard Cite

Cognitive theories for reasoning are about understanding how humans come to conclusions from a set of premises. Starting from hypothetical thoughts, we are interested which are the implications behind basic everyday language and how do we reason with them. A widely studied topic is whether cognitive theories can account for typical reasoning tasks and be confirmed by own empirical experiments. This paper takes a different view and we do not propose a theory, but instead take findings from the literature and show how these, formalized as cognitive principles within a logical framework, can establish a quantitative notion of reasoning, which we call plausibility. For this purpose, we employ techniques from non-monotonic reasoning and computer science, namely, a solving paradigm called answer set programming (ASP). Finally, we can fruitfully use plausibility reasoning in ASP to test the effects of an existing experiment and explain different majority responses.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Quantitative Symbolic Approach to Individual Human Reasoning

Dietz¹,

Fichte²,

Hamiti³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Third, the literature often conveys the impression that most problem-solving areas can in principle be solved by machine learning. This overlooks the hierarchy of problem solving [63] where one must differentiate between at least permutation problems, insight problems, and complex problems. These differ in the underlying environment (i.e., if it is static or changes), observability (i.e., whether all relevant information is given), and operators (i.e., whether all possible actions are known) among others [64].…”

Section: Ai Perspectives On Computational Thinkingmentioning

confidence: 99%

Survey: Artificial Intelligence, Computational Thinking and Learning

et al. 2022

Self Cite

View full text Add to dashboard Cite

Learning is central to both artificial intelligence and human intelligence, the former focused on understanding how machines learn, the latter concerned with how humans learn. With the growing relevance of computational thinking, these two efforts have become more closely connected. This survey examines these connections and points to the need for educating the general public to understand the challenges which the increasing integration of AI in human lives pose. We describe three different framings of computational thinking: cognitive, situated, and critical. Each framing offers valuable, but different insights into what computational thinking can and should be. The differences between the three framings also concern the views of learning that they embody. We combine the three framings into one framework which emphasizes that 1) computational thinking activities involve engagement with algorithmic processes, and 2) the mere use of a digital artifact for an activity is not sufficient to count as computational thinking. We further present a set of approaches to learning computational thinking. We argue for the significance of computational thinking as regards artificial intelligence on three counts: (i) Human developers use computational thinking to create and develop artificial intelligence systems, (ii) understanding how humans learn can enrich artificial intelligence systems, and (iii) such enriched systems will be explainable. We conclude with an introduction of the articles included in the Special Issue, focusing on how they call upon and develop the themes of this survey.

show abstract