Acting to Understand and Understanding to Act in Design Processes

Description

This study investigates a distinction between two kinds of purposeful human action: 'understanding in order to act' and 'acting in order to understand.' The former describes purposeful human action as linearly goal-directed, with planning (i.e., understanding) being a prerequisite to subsequent execution (i.e., acting). This notion has been prevalent in early design theories and is still encountered frequently today. The more recent view of understanding preceding action and action preceding understanding in conversational and reflective cycles is increasingly recognised as offering a more appropriate theory of purposeful human action in general and design in particular. While the distinction between 'understanding in order to act' and 'acting in order to understand' appears in various guises across several theories of human action, it has yet to be tested empirically in the design context. One significant effort, tangential to design, was documented in a series of publications by David Kirsh and Paul Maglio. Based on their observations of problem-solving in Tetris play, the team describes purposeful human action as a combination of pragmatic action (action that follows understanding) and epistemic action (action that leads to new understanding). Occasionally, Kirsh and Maglio acknowledge that particular actions may serve both pragmatic and epistemic purposes; however, the extent to which the two actions overlap and how they interplay is not fully clarified in their work. Kirsh and Maglio generalise their findings, arguing that the presence of epistemic actions in a pragmatically oriented game suggests that the two kinds of action are widespread throughout – and that the distinction between them is applicable to – all human activity. This generalisation of the binary pragmatic-epistemic distinction from the rule-based and goal-oriented environment of Tetris to all human activity remains empirically untested. Its applicability to the "wicked," open-ended problems characteristic of design practice in particular appears doubtful and is in need of testing. So far, this issue has received little empirical attention. To address this gap, this study offers an empirical test of the main research questions: Is Kirsh and Maglio's distinction between pragmatic and epistemic actions sufficiently differentiated to account for design activities? And: How do pragmatic and epistemic activities interplay in design processes? For the purpose of theory-testing, this study hypothesises that all actions in design can be categorised as either pragmatic or epistemic, and that this distinction enables adequate representations and analyses of design processes. It does so by examining protocols of six digital design processes in an empirical study of six individual designers with varying levels of expertise. It employs concurrent think-aloud protocol analysis, primarily collecting data through audio and video recordings. This qualitative data is transcribed, segmented, and coded using a scheme developed to investigate instances of pragmatic or epistemic objective-setting and their possible pragmatic or epistemic fulfilments. The coded data is then linked to discern pragmatic and epistemic episodes using a linkography-inspired notation system. The analysis led to several insights into the interplay of both kinds of action, considering the ill-defined and subjective nature of design and the absence of clear temporal markers that allow the identification of unambiguous units of analysis discernible in Tetris play. In multiple instances, pragmatic design objectives are observed to yield epistemic fulfilments, and vice versa. Occasionally, design episodes also result in dead-ends, yielding neither pragmatic nor epistemic fulfilments. The analysed design episodes fall into six possible relationships, beginning with pragmatic or epistemic objectives and ending with pragmatic or epistemic fulfilments or dead-ends. Additionally, numerous design episodes best described by one category of action at some given time frame are found to be better described by other categories of action over different time frames. To some extent, the findings align with Kirsh and Maglio's pragmatic-epistemic distinction, reaffirming its value in extending linear design process theories to circular ones. However, they also show that the distinction offers limited descriptive value in the context of design. This research makes three main contributions to the field: (1) theoretically, it extends Kirsh and Maglio's framework from tame problem-solving contexts to the wicked, ill-defined realm of design, demonstrating the need for more nuanced categorisations of purposeful human actions; (2) methodologically, it introduces a novel analytical framework combining extended protocol analysis with a linkography-inspired notation system specifically developed for investigating the temporal nature of designerly actions; and (3) empirically, it identifies six distinct episodic relationships that provide a more refined vocabulary for describing human actions in open-ended design contexts. These contributions offer substantial utility for design education, practice, management, and empirical research, particularly in understanding how actions initially categorised in one way may transform over different time frames – a finding that illuminates 'trial-and-error' operations in design processes.

Period	18 Sept 2024 → 30 Sept 2025
Examinee	Chitraj Bissoonauth
Examination held at
Degree of Recognition	International