Human-data interaction and a spatial metaphor for epistemic action

02/21/2018 - 06:00 - 07:30

Add to calendar

iCal calendar

Abstract:


The presentation approaches computer-mediated explorative analysis of data as a chain of epistemic actions. Leaning on the perspectivist interpretation of G盲rdenfors鈥檚 theory of conceptual spaces (2000) suggested by Kaipainen & Hautam盲ki (2011, 2015, 2018), it presents concepts as 1) dynamically evolving perspective-relative spatial clusters and 2) compares cognitive perspectives to operative positions in concrete world 3) at which one arrives through temporally consequent epistemic actions. Further, it is proposed that concepts as grounded in that way allow relating to actionability in terms of 鈥済rasps鈥, in harmony with the Lakoff and Johnson鈥檚 metaphor theory (1980) among other relations of cognition with the embodied and situated mind. The approach implies applications in data analysis, dynamical models of learning and human-data interaction design, among others.