Design creativity research: recent developments and future challenges

In 1983 Howard Gardner (1983/2011) first published ‘Frames of mind: The theory of multiple intelligences’. His view that intelligence can take many forms and is contingent on particular disciplines and on socio-cultural circumstances was revolutionary and had a lasting influence on the study of intelligence. This theory shuttered the former belief that intelligence is innate, and that it can be universally measured in the form of a standard IQ score. Gardner listed seven (later the list was extended to nine) basic types of intelligence, which translate into learning styles: visual-spatial, bodily-kinesthetic, musical, interpersonal (later dubbed emotional or social intelligence), intrapersonal, linguistic and logical-mathematical. Gardner, who is a prominent creativity researcher as well, did not come up with a similarly innovative theory of creativity; instead, he demonstrated extraordinary creativity in a range of different disciplines. A generation ago the study of creativity was largely descriptive and included much anecdotal material. In parallel, assessments of creativity continued to emphasize divergent thinking, as proposed by Guilford (1950) and evident in fluency, flexibility and originality (later elaboration was added), for which Torrance (1974) developed the most prevalent creativity tests. There are many offshoots of the original 1970s’ tests, but they continue to rest on the same principles.

Two main factors have changed creativity theories in more recent times. First, artificial intelligence (AI), which has become a strong simulation and development tool in studies of intelligence, has been instrumental also in creativity research. One of the first scholars to illuminate the role of AI in analyzing as well as generating creative ideas is Margaret Boden, who distinguishes between two levels of creativity. The first is connectionist, whereby familiar ideas are combined in novel ways. A higher level of creativity is achieved by “mapping, exploration and transformation of conceptual spaces” (Boden, 1991/2004). AI models can “help define a space”, and show how it is navigated when acts of mapping, exploration and transformation are performed. When this theory was first presented, AI tools could only explore spaces. Today we have tools that can also transform them, thus assisting in the production of creative ideas.

The second factor that has a profound impact on creativity research is neurological and neurocognitive sciences, which have, in recent years, been able to expand creativity research from the realm of cognition and the mind into the realm of the brain. In particular, the combinatorial theories of creativity which talk about associations of ideas can now be demonstrated in terms of patterns of neural connections in the brain, with the help of contemporary imaging technologies such as fMRI (e.g., Ellamil, Dobson, Beeman, & Christoff, 2012; Gabora, 2010; Goel, 2014; Lazar, 2018). This line of research has helped weaken the hitherto dominating view of the ‘exclusivity’ of divergent thinking in creativity. Even with simpler tools than imaging, it is now possible to show that convergent thinking is equally important in creative thinking, including in design. Furthermore, since we can now work with finer resolutions in research, we can also defy the paradigm that creative design consists of prolonged divergent phases followed by convergent ones. Goldschmidt (2016) has demonstrated that shifts between divergent and convergent segments of thought are very frequent and occur throughout the idea generation phase of creative design.

In conclusion, the important developments of the last decade or two resemble the progress that had transpired earlier in the study of intelligence. On the one hand, we realize that our research must be conducted at a much finer grain than had been done before, and we now have the tools to do so, in collaboration with other relevant scientific disciplines such as neuroscience. Anecdotal studies are no longer admissible, other than as illustrations. On the other hand, with the coming of age of the relevant computational tools, we can now apply AI to automate creative designing, at least to some degree. ‘Artificial Creativity’ is still met with some suspicion, but it is reasonable to predict that with progress in computational means it will become a mainstream assistant in both research and practice of creative design and designing.