Research in Sloyd Education and Crafts Science A:2/1997

Composition and Construction
in Experts´ and Novices´ Weaving Design

Pirita Seitamaa-Hakkarainen


The purpose of this study was to analyze expertise in weaving design from the viewpoint of cognitive science. Weaving design was seen as a complex problem-solving process requiring domain-specific knowledge of the principles of weaving as well as more domain-general knowledge of visual design. The study was aimed at identifying basic elements of weaving design, analyzing qualitative differences between novices´ and experts´ design processes and investigating the relationships between visual and technical designing in depth in order to develop theoretical models for expert and novice design processes.

The process of weaving design was characterized as a dual-space search like many other areas of problem-solving. Thus, weaving design was divided into two problem spaces: composition space, which represented visual design; and construction space, which represented technical design. The former space represented a more domain-general knowledge of production of woven textiles. A necessary prerequisite for successful weaving design was the existence of a close interaction between the spaces. In accord with the author´s earlier study (Seitamaa-Hakkarainen 1989), the main hypothesis was as follows: The basic difference between experts and novices in their approach to weaving design is the novices´ tendency to structure the process in a singular fashion around the visual elements of the task whereas experts were able to integrate both the visual and technical elements of weaving and consider them in a parallel way during the design process.

In the study, four subjects, two professional weaving designers, and two graduate students of textiles completed two design tasks each. The novices in this study had specialized in textiles in general, whereas the experts were specialists in weaving design in particular. The purpose of the first task, an open-ended weaving-design assignment, was to identify the basic, comprehensive elements of weaving design as utilized by the subjects and the general qualitative differences between professional and non-professional methods of design. The second task was a clearly defined format of weaving-draft problems which required domain specific knowledge and application of weaves and drafting. Protocol analysis was used for collection of data. To aid in the documentation of these processes, sessions with the subjects were recorded on videotape to enable a comparision of the verbal report with observed activity. Protocol data were analyzed qualitatively and problem-behaviour graphs (PBG) were constructed.

The main hypothesis, was supported by the data, including the problem-behaviour graphs. The experts´ process of integration tended to start from the construction space and move between visual and technical design. The experts could process many aspects of the design in parallel fashion within and between the two design spaces. The problem-bahaviour graphs suggest that it was typical of the novices to handle the elements of the design serially. The novices organized their weaving process around composition (visual) design, only in the end did they jump to the construction space to explore how visual ideas could be realized. Thus, toward the end of the design process there was an increased novitiate concern about construction problems, though technical design remained a segregated aspect of the process. The non-professionals´ occasionally considered design elements parallelly, but interactions between different design elements were almost exclusively within the composition space. In general, analysis of weaving design process revealed cyclical (iterative) nature of designing. Both the experts and the novices processed design elements several times, "jumping", in many cases, from one design element to another. Results of the study showed that the novices utilized iterative designing only in the context of visual composition.

In the second task of weaving draft, the subjects were asked to apply the principles of weaving by analyzing how a given textile could be reproduced by weaving. Experts were able to produce from 7 to 13 different correct solutions in a shorter time whereas novices produced the relatively low number of 2 to 4 partially correct solutions in more time. Experts were apparently able to rely on fundamental principles of weaving in their solutions, whereas the novices were bound to the surface features of the task.

Keywords:Weaving design, protocol analysis, problem-behaviour graph, dual-space search

Obtainable from:
University of Helsinki
Faculty of Education
Department of  Teacher Education
Section of Textiles, Clothing and Craft Design Studies
P.O. Box 32
FIN-00014 University of Helsinki, Finland

Phone +358-(0)9-191 7065
+358-(0)9-191 7043

NordFo Techne serien Redaktionsrådet
Juha Hartvik, Mia Porko-Hudd