Page 119 - DP-2-2

Design+ Transposing human action research to design

well-known empirical studies whose experimental designs game speeds up. This results in an increasing portion of
rested on obviously well-defined, closed-ended tasks. The pieces being misplaced and lines not being dissolved. Once
generalizability of findings implied in either study may the full height of the interface is filled with undissolved
not apply to design, given that design was characterized rows, no more pieces can enter from the top, ultimately
as a distinct regime of human action with ill-structured, ending the game.
wicked, and messy qualities. To test the generalization of Employing an implementation of the game that was
their findings, we transposed and adapted both studies for modified to record player actions during game play, Kirsh
re-evaluation in design contexts. and Maglio observed that players did not always move
42
3.1. Epistemic and pragmatic actions pieces directly toward their preferred positions. Instead,
players frequently moved pieces temporarily away from
42
Kirsh and Maglio published a series of studies in the their preferred positions. Kirsh and Maglio divided these
42
1990s and early 2000s, in which they put forward their player actions into two categories, which they labeled as
distinction between epistemic and pragmatic actions. This pragmatic actions and epistemic actions. They described
distinction challenged earlier conceptions of purposeful pragmatic actions as directly pursuing clearly determined
human action as linearly goal-directed, with planning (i.e., goals (Figure 2). In contrast, Kirsh and Maglio as well as
42
“problem exploration and analysis,” according to Cross ) Loader described epistemic actions as probing the game
16
50
being a prerequisite to subsequent execution (i.e., “solution environment with a view to improving understanding of
synthesis,” according to Cross ). An example of this linear possible goals and actions to achieve them, as shown in
43
view on purposeful human action is Archer’s description Figure 3.
of design as a process consisting of an analytical, a
creative, and an executive phase. Schön, Suchman, This categorization of human action reflects
44
45
46
47
42
Kirsh and Maglio, Glanville, and, to some extent, Rittel Glanville’s later distinction between “understanding in
47
and Webber noted that this linear and comparatively order to act” and “acting in order to understand.” Kirsh
15
reductionist view was prevalent across early theories of and Maglio noted that typical Tetris play was characterized
design, reflecting attempts to scientize and prescribe design by an interplay of the two kinds of action and that earlier
practice and research. In contrast, more recent theories linear and reductionist theories failed to account for the
16
described human action in general and designed activity differences or the interplay between both categories. Kirsh
42
in particular, as circularly “argumentative,” “reflective,” or and Maglio extended their findings beyond problem-
“conversational.” 15,43,45,47-49 Kirsh and Maglio challenged solving in Tetris play, stating that “if epistemic are found
42
the above-described earlier theories of purposeful human in the time-limited context of Tetris, they are likely to be
action by testing the assumption that “the point of action found almost everywhere” and that their distinction “holds
is always pragmatic” and that “understanding” necessarily generally throughout all of human activity.” 42(p548)
precedes “acting.” To this end, they conducted an empirical The ongoing PhD project described here tested this
51
study of players tackling the well-structured classic video generalization. Designing, with its ill-defined goals and
game Tetris. We outlined our transposition of Kirsh and absence of predefined and predeterminable units and
Maglio’s experiment from the well-structured context into criteria of analysis, differs notably from Tetris play.
15
the ill-structured context. This allowed transposing Kirsh and Maglio’s study from
In Tetris play, various shaped pieces fall from the top the domain of well-structured problems to that of ill-
and stack up at the bottom of the game’s visual interface structured problems with the following research-about-
as illustrated in Figures 2 and 3. As it falls, the player design prompt: If observations of Tetris game play expose
42
can rotate and move each piece horizontally. The goal is epistemic and pragmatic actions, then what kinds of action
to fill horizontal rows without gaps, thereby dissolving can be discerned in the ill-structured context of design?
them. The process of a piece coming into play at the top, Figure 4 illustrates our extension of Kirsh and Maglio’s
traveling downward while being manipulated by the player, theory formulated in the well-structured context of Tetris
coming to rest at the bottom, and potentially resulting in to the ill-structured context of design.
the dissolution of rows is called a Tetris episode. A game We conducted an empirical lab study of six designers.
consists of consecutive episodes during which the speed Of these, two were digital design novices, two had
of the falling pieces increases gradually. During each intermediate skills, whereas the remaining two had
episode, the preferable target position (or one of multiple advanced skills (according to their self-categorization in a
equally preferable positions), and the shortest path toward pre-observation questionnaire). We asked the subjects to
it is unambiguously clear in principle. Their identification, design a space partitioning and shelving system for a retail
however, becomes cognitively more and more taxing as the space using Grasshopper in Rhino3D. The design briefly

Volume 2 Issue 2 (2025) 5 doi: 10.36922/dp.4875