Explanation in Human Thinking. Cassens, J., Habenicht, L., Blohm, J., Wegener, R., Korman, J., Khemlani, S., Gronchi, G., Byrne, R., Warren, G., Quinn, M., & Keane, M. T. In Proceedings of the 43rd Annual Meeting of the Cognitive Science Society, Vienna, Austria, July, 2021. Curran Associates.
abstract   bibtex   
Jörg Cassens, Rebekah Wegener, Lorenz Habenicht, and Julian Blohm discuss the dialogic form of explanations. Explanations are a long established research topic in a wide variety of disciplines, ranging from philosophy (van Fraassen, 1980; Achinstein, 1983) over the cognitive sciences and psychology (Lalljee et al., 1983; Keil and Wilson, 2000; Lombrozo, 2006) to computer science in general and artificial intelligence in particular (Schank, 1986; Leake, 1992; Leake (1995); Sørmo et al., 2005). However, while there is compelling research supporting the value, structure and function of explanation, as Edwards et al. (2019) argue, “accounts of explanation typically define explanation (the product) rather than explaining (the process)”. By contrast, we aim at an understanding of explanation as a functional variety of language behaviour that treats explanations as being - Contextualised, which itself is comprised of a) Context Awareness (knowing the situation the system is in) and b) Context Sensitivity (acting according to such situation), - Construed by user interest, - Multimodal, and - Dialogic. In this talk, we will focus on the latter two aspects, the dialogic form of explanations and its representation in different modalities (and codalities). We will report on our recent empirical work where we have been looking at explanatory situations, firstly the differences between human to human explanations and machine to human explanations (using of-the-shelf speech dialogue systems) and secondly multimodal human to human explanations.
@inproceedings{Cassens_EA-CogSci-2021,
  keywords =  {paper},
  author=     {Jörg Cassens and Lorenz Habenicht and Julian Blohm and Rebekah Wegener and Joanna Korman and Sangeet Khemlani and Giorgio Gronchi and Ruth Byrne and Greta Warren and Molly Quinn and Mark T. Keane},
  title=      {Explanation in Human Thinking},
  booktitle=  {Proceedings of the 43rd Annual Meeting of the Cognitive Science Society},
  month =     {July},
  year=       {2021},
  address =   {Vienna, Austria},
  isbn =      {9781713835257},
  publisher = {Curran Associates},
  abstract =  {Jörg Cassens, Rebekah Wegener, Lorenz Habenicht, and Julian Blohm discuss the dialogic form of explanations. Explanations are a long established research topic in a wide variety of disciplines, ranging from philosophy (van Fraassen, 1980; Achinstein, 1983) over the cognitive sciences and psychology (Lalljee et al., 1983; Keil and Wilson, 2000; Lombrozo, 2006) to computer science in general and artificial intelligence in particular (Schank, 1986; Leake, 1992; Leake (1995); Sørmo et al., 2005). However, while there is compelling research supporting the value, structure and function of explanation, as Edwards et al. (2019) argue, “accounts of explanation typically define explanation (the product) rather than explaining (the process)”. By contrast, we aim at an understanding of explanation as a functional variety of language behaviour that treats explanations as being - Contextualised, which itself is comprised of a) Context Awareness (knowing the situation the system is in) and b) Context Sensitivity (acting according to such situation), - Construed by user interest, - Multimodal, and - Dialogic. In this talk, we will focus on the latter two aspects, the dialogic form of explanations and its representation in different modalities (and codalities). We will report on our recent empirical work where we have been looking at explanatory situations, firstly the differences between human to human explanations and machine to human explanations (using of-the-shelf speech dialogue systems) and secondly multimodal human to human explanations.}
}

Downloads: 0