Reconciling Function- and Affordance-Based Design

ion for their problem definitions and avoid over-constraining the solution space (Ulrich, 2011). If designers analyze the end-user’s goals independent of any particular artifact-to-be-designed, they can begin the design process with the maximum breadth of solution space, increasing flexibility and delaying costly decisions. Explicitly linking required affordances to high-level goals builds system-level verification and validation opportunities into a design scenario, especially if those goals require execution of a plan involving many behaviors. This can help avoid situations in which individual affordance-level tests (e.g.,to 87 determine if a particular low-level action or behavior can be performed) fail to capture behavior that emerges or fails to emerge after system integration. Goal-level needs specifications are more general than behavior-level requirements because unlike behaviors, goals are purely intentional concepts and therefore more solution-agnostic. It is possible that early-stage design requirements specified at the behavioral level are more likely to act as a cognitive anchor during solution ideation. This is particularly true if the behavioral specifications are linked to the physical structure of existing solutions, such as defining the need for a more efficient car when the user’s real desire was a less expensive commute to work. Developing requirements in terms of goals allows designers to specify constraints on candidate designs with a minimum of information about the form of the solution. For example, the solution to a particular design problem might be a product to be sold, a product-service system, or simply training users to exercise new abilities. Lastly, goal-level specifications are necessary from the perspective of design-as-nichemodification. Because a niche is the set of affordances that a type of organism uses, modifying those affordances changes the niche. Without a goal-level requirement specification we are left without a means of identifying or comparing niches. If a fundamental niche consists of the variety of ways in which a particular goal can be achieved by a particular type of organism, and a realized niche is the specific set of affordances and affordance carriers used by that organism (Hutchinson, 1978), then Houkes and Vermass’ plan-based model of artifact use provides a graduated framework for describing niche alteration (Table 1). By altering Realized niche Fundamental Niche Goals Different niche type Different niche type Only actions (same goals) Different niche type Unaltered Only artifacts used (same actions, goals) Different niche type Unaltered Actions and artifacts (same goals) Different niche type Unaltered Table 1: Altering different elements of a use-plan results in different types of niche changes Our approach to a niche-based analysis of design is therefore based on decomposing use scenarios to understand how individuals currently achieve particular goals, followed by an analysis of how and why niche parameters can be altered by designers. While our immediate goal is the development of a framework 88 for understanding the design process, we also recognize the utility of this approach for designers themselves. For a designer, niche analysis represents a structured, holistic means for identifying and describing design opportunities. The core of our approach involves collecting sufficient information to specify the relevant niches for a particular artifact. Because of artifact non-exclusivity and multiplicity, artifacts can be part of multiple niches throughout their life-cycle because they provide affordances to multiple types of individuals. Because of boundedness, these niches can be identified and evaluated in terms of a single circumscribing niche. For an ecological approach to design to be complete, it is necessary to be able to define this circumscribing niche and account for all sub-niches relevant to the design, creation, and use of the artifact. For designers engaged in the design process, it may only be necessary to consider particular niches, such as end-use or a specific manufacturing process. Niche definitions should begin by identifying the niche inhabitants, such as an artifact’s user or intended type of user. For the analysis of existing artifacts one has the opportunity to directly identify the artifact’s users. For the identification of design opportunities, one must specify an intented type of user. Users can be usefully described in both functional and parametric terms. Functional user descriptions focus on the user’s behavioral abilities, which determine the affordances available to them in a particular environment. Parametric user descriptions may be affordance-defining ergonomic data like leg length (Warren, 1984) or demographic data such as socioeconomic class, geographical location, and cultural background. As Gibson (1979) noted, social and interpersonal affordances are extremely important for humans, and play a large role in day-to-day goal achievement. Social relationships, attitudes, and expectations may differ significantly between different cultures (Gaver, 1991), which may alter the existence of interpersonal affordances. Specifying the user’s goals determines the type of niche. For design purposes, goals should be described in observer-independent terms about the user’s existing or desired situation. Defining goals as minimally as possible without sacrificing specificity will help avoid over-constraining the design space. For a given overall goal there may exist subgoals, and the principle of artifact non-exclusivity implies that these goals may be used to define sub-niches.