Harnessing analogies for creativity and problem solving

By Christian Schunn

Christian Schunn (biography)

What is an analogy? How can analogies be used to work productively across disciplines in teams?

We know from the pioneering work of Kevin Dunbar (1995), in studying molecular biology labs, that analogies were a key factor in why multidisciplinary labs were much more successful than labs composed of many researchers from the same backgrounds. What is it about analogies that assists multi- and interdisciplinary work?

The advice that follows comes from a decade of research involving intensive analyses of hundreds of hours of interdisciplinary science and engineering teams, following the minute-by-minute processes of the teams, and using advanced statistical techniques to look for robust patterns in behavior over time and across teams. In general, our research has shown that creative teams generate twice as many ideas per unit time when they use analogies than when they do not.

So, what is an analogy? It is the accessing and transferring of elements from familiar categories or situations to the current problem. We all use analogies such as: ‘works like magic’, ‘stinks like a rotten egg’, or ‘hit-or-miss like a risqué joke’.

These prior situations might be a comparison to a relatively similar past experience of the team member, or might involve referencing experiences from a vastly different technical or everyday situation.
 As an example in the context of engineering, a team member raised the following analogy to an everyday experience in designing an unsupported tube to transport liquid: “the stuff you make Venetian blinds of for example… they can be bent.”

Note that analogy is a cognitive process in which the problem solvers reason through the relationship between the prior experience and the current problem. The reasoning process produces a number of inferences for the team that can serve many purposes. Analogies can help brainstorming sessions by:

  1. Generating new solutions
  2. Identifying likely problems that weren’t yet raised
  3. Finding additional functions that can be exploited, and
  4. Explaining ideas across the team

When a team needs to resolve uncertainties without a clear way forward, raising new analogies can bring possible solutions.

Productive use of analogies is useful for stimulating both divergent and convergent thinking processes and can combat two major team problems, those of group think and confirmation bias. On the divergent side, group think is getting stuck on a shared mediocre idea, instead of harnessing the breadth of knowledge in the team. On the convergent side, confirmation bias is preferring evidence in favor of the current plan, and failing to seek evidence that different team members have access to that points to flaws in the current favored plan.

Analogies can also reduce the fixation effects of starting with a non-functional prior solution. Even expert designers will fixate when given an example of what a solution should not do, but the harm is erased when designers are given a range of analogies to consider.

Analogies can produce disagreements among team members in several productive ways, for example about the aptness of the analogy at all, the way the analogy applies to the current situation, and whether inferences can be used. In our research this effect was found for productive kinds of disagreements (disagreements about the task and processes to be used), and not for the unproductive kind of disagreement (conflicts of personality or personal attacks).

Interestingly, the physical environment surrounding the team can also influence how often teams use analogies. Looking at highly detailed artifacts (e.g., prototypes or models) appears to suppress the rate of analogizing relative to having an open conversation or looking at more abstract sketches. Of course, getting very concrete and detail-oriented has an important place in teamwork, especially multidisciplinary teamwork. But there is a time and a place for it, and it can be helpful to put away the physical artifacts from time to time.

What have your experiences been with analogies? Can you provide additional examples of how they have been helpful?

Reference and related papers

Ball, L. J. and Christensen, B. T. (2009). Analogical reasoning and mental simulation in design: Two strategies linked to uncertainty resolution. Design Studies, 30, 2: 169–186.

Chan, J. and Schunn, C. D. (2015). The impact of analogies on creative concept generation: Lessons from an in-vivo study in engineering design. Cognitive Science, 39, 1: 126-155.

Chan, J., Dow, S. P. and Schunn, C. D. (2015). Do the best design ideas (really) come from conceptually distant sources of inspiration? Design Studies, 36: 31-58.

Christensen, B. T. and  Schunn, C. D. (2007). The relationship of analogical distance to analogical function and pre-inventive structure: The case of engineering design. Memory & Cognition, 35, 1: 29-38.

Dunbar, K. (1995). How scientists really reason: Scientific reasoning in real-world laboratories. In R. J. Sternberg and J. Davidson (Eds.). Mechanisms of insight. MIT Press: Cambridge, Massachusetts, Untied States of America, pp: 365-395.

Linsey, J., Tseng, I., Fu, K., Cagan, J., Wood, K. and Schunn, C. D. (2010). A study of design fixation, its mitigation and perception in engineering design faculty. Journal of Mechanical Design, 132, 4: 041003.

Paletz, S. B. F., Schunn, C. D. and Kim, K. (2013). The interplay of conflict and analogy in multidisciplinary teams. Cognition, 126, 1: 1-19.

Biography: Christian Schunn is a Senior Scientist at the Learning Research and Development Center and a Professor of Psychology, Learning Sciences and Policy, and Intelligent Systems at the University of Pittsburgh. He directs a number of research projects studying expert engineering and science teams and the effects of new tools designed to increase innovation. He also leads research projects that apply this knowledge to building innovative technology-supported STEM curricula, and studying factors that influence student and teacher STEM learning.

13 thoughts on “Harnessing analogies for creativity and problem solving”

  1. Hello, and thank you for this post which is very relevant to my research.

    I am currently pursuing a PhD on the topic of pattern literacy in support of systemic intervention, and in particular developing a working theory of the pattern developing a working theory of the pattern as mediator and connector, as well as tools and methods for trans-disciplinary/systemic interventions.

    I make the hypothesis that the pattern is a shareable unit of meaning-making which can play the role of boundary object, i.e. object of shared inquiry and mediation.

    I am very interested in the cognitive and psychological processes involved. Analogy and other types of semantic connections are key in this context, because they can help connect different ‘facets’ of similar patterns.

    I am therefore glad to find here many references that can expand my view and approach, and looking forward to exchange more on the topic.

    In particular, and in relation to Joseph Guillaume’s question and other post on the blog, I would like to identify the different types of patterns, and look into frameworks that can help me do so, such as the OODA loop*, the pragmatic cycle and semiotic approach of Peirce, Luhmann’s systems theory.

    *Editor: This refers to the decision cycle of observe, orient, decide, act.

    • A few thoughts.

      1) one might think about a pattern as a kind of schema for a situation. Analogies are helpful for creating such schemata (the pattern that is shared by the two situations)
      2) I am reminded of Design Patterns in the software engineering literature. They have made a science of thinking of patterns in a way that guides problem solving.

  2. Discussion copied from the Linked In group Systems Thinking Network:

    Francisco Barreto: Thinking of Douglas Hofstadter… His works on the role of Analogy-Making in human thought..

    Christian Schunn: Yes, a number of scholars over the years have written about the role of analogies in creativity. Hofstadter’s main contribution was in providing some early thinking on the mechanisms by which the mind could do that work.

    William Donaldson, PhD: There is also a caution that needs to be added/considered. Problems and limitations can be carried over from the analogous root to the new idea

    Christian Schunn: Analogy is a kind of induction, not a kind of deduction. There is no guarantee of a workable solution and thus better thought of as a divergent thinking tool rather than a decision making tool.

    Jay Sorenson: Can brainstorming arrive at conclusions in excess of the intellectual capabilities of the individual participants? Are the conclusions limited to some relevant range. That is to say can a group of plumbers prove or disprove the correctness of string theory?

    Christian Schunn: On the one hand, Wisdom of the Crowd research suggests that even moderate collections of individuals can be remarkably smart (e.g., suggest which stocks to select more effectively than top stock brokers). On the other hand, most groups tend to use poor information sharing processes and so act less smart than expected. In other words, the process used by the group matters a lot.

    Jean-Louis Baudoin Why always talk about problem solving? What about developing the Thinking skills that AVOID problems upstream?

    Christian Schunn: Great point! In fact, expert engineers use analogies to anticipate problems that a particular possible solution might produce.

    S Anders Christensson Are analogies to be considered as puzzle-bits? First you seek bits that maybe corners, then add-ons to build frames. Then understand all red parts, blue parts and so on. Or the analogy can work as a envisioned pattern where ever it maybe in the puzzle. Kind of induction to use bits to contribute and build a larger hole?

    Christian Schunn: That is a lovely analogy about analogy!

    Ray Gallon I agree, but discovered that some people (includes one of my daughters) are are actually incapable of learning by analogy. I never knew such a condition existed.

    Christian Schunn: I would be amazed if there existed a human who could not learn by analogy—this would be worthy of publishing in Nature or Science, similar to finding a human who didn’t need oxygen or water. There has been extensive investigation of learning by analogy. Of the many variations on how analogies can be presented, sometimes there are individual differences in which ways of presenting analogies are best. For example, some learners learn more when the analogy is spelled out in greater detail and other learners learn more when they have to do more of the mapping work themselves. Interestingly, US math teachers and their Asian counterparts both use analogies equally often, but the US teachers frequently use analogies in unhelpful ways (e.g., by comparing a math concept to another topic the students also don’t understand).

    Ray Gallon Well, I can only comment on both what I observed and what the school diagnosed when she was 16. Obviously she understands them on some level, but they don’t help her to learn. Paradoxicaly she is super smart at pattern recognition (visual) which you would think was connected.
    She might be able to learn maths by analogy, but more literary out philosophical notions need to be explicitly explained for her to get them. She’s now finishing university do it hasn’t hindered her. I should mention we are in Spain (Catalonia). Culture also plays a role.

  3. Hi Christian, thanks for the article.

    I think analogies have been with us for a long time and used in communicating, teaching and conveying concepts, ideas and meaning to others. There are numerous forms found in all sorts of literature and language.

    They are powerful for creative thinking if there is sufficient alignment between your problem and the outside or given concept – then its reasonable to port across useful elements to fashion an idea.

    I see people doing this all the time, we generally aren’t conscious of this or use it deliberately. The skill of the facilitator in the group is to apply techniques in a way to elicit the main elements of an analogy reliably and consistently to get predictable results. Ie if the team is employed to solve problems, it has to solve problems!

    Famous works from the folk that developed Synectics is interesting – quite hard to get through their work to fashion a method, but worth it. They had 4 types of analogous thinking. Have you seen this or come across it Christian?

    Regards, Nat.

    • Dear Nate,

      I have heard of Synetics although have not directly worked with someone specifically implementing that exact approach. There are a number of design consulting groups that make formal use of analogy or something like it. That it appears so regularly in their work suggests 1) a shared experience with the power of analogy AND 2) a shared experience that some facilitation is required (otherwise why need a consulting group to help with it).


  4. It may be worth observing that analogies also serve as a flexible means of building understanding and picking apart complex problems by revealing and illustrating patterns or schemas which can be mapped to other, perhaps simpler or already known situations. Obviously, along with all other strengths, analogy is just as practical in individual work as it is in team context discussed in the article.

    BTW. I completely agree with the broader and more encompassing meaning of analogy, inclusive of every type of correspondence in structures, behaviours and functions. Despite my interest in systems science,I ‘m not familiar with term homology in context of design.

    • The literature on schema learning has suggested that analogies are a powerful way of building new schemata (the category of things shared by the example).

      Analogies can be useful to individuals, but they are more powerful for groups. Individuals have trouble retrieving useful analogies on their own. In addition, analogies are an important part of interdisciplinary communication.

      • Does it follow that analogy used for problem solving is a form of slow, more deliberate thinking, and that its fast, instinctive equivalent is the application of a tacit pattern/schema?
        If so, then using analogies for schema learning involves distilling key patterns that can be internalised for everyday use?
        Do you have an opinion about value of directly use analogies in problem solving vs. using them to build schemas that are applied in a second step?

        • A number of theorists (e.g., John Anderson of CMU) have made exactly that proposals: analogies distill key patterns for more rapid later use. There is some disagreement about whether it is via building up schema (a way of categorizing) or creating of rules (a way of acting). I think there is good evidence for both.

          Anderson, J. R., Fincham, J. M., & Douglass, S. (1997). The role of examples and rules in the acquisition of a cognitive skill. Journal of experimental psychology: learning, memory, and cognition, 23(4), 932.

          Gick, M. L., & Holyoak, K. J. (1983). Schema induction and analogical transfer. Cognitive psychology, 15(1), 1-38.

  5. Terrific article Christian. Thank you for sharing this. I think you’re right that analogies are low-hanging fruit when it comes to creativity and problem definition/solving. Another related topic is the difference between an analogy and a homology. Where an analogy refers to a “correspondence in function,” a homology is “a likeness in structure” (or set of causal relationships). In systems thinking terms, we might call these “kernel structures” or archetypes.

    For more on the value of homologies, you might want to read this short article:

    • Interesting. I had not yet seen that distinction being made. Within the psychology literature, both correspondence of function and of structure would be labeled analogy, since both invoke similar psychological processes. There was a debate back in the 1990s about whether correspondence of causal structures had special status, but in the end that didn’t get much empirical support.

      You might also be interested in relational reasoning more generally, which includes analogy, anomaly, antimony, and antithesis. Sometimes how things are different is just as interesting as how things are the same. http://link.springer.com/article/10.1007%2Fs10648-016-9370-6


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: