XXIX.1 January - February 2022
Page: 84
Digital Citation

Nonlinear design thinking: The limitations of awareness in the search for climate care

Shunying Blevis, Eli Blevis

back to top 

The "A causes B" way of thinking is one-dimensional and linear whereas reality is multi-dimensional and non-linear. One has only to think of one's own life to see how absurd it is to think everything can be explained as a simple linear process of cause and effect.

— James Lovelock [1]


back to top  Linear and Nonlinear Design Thinking

From marketing research, we learn that people often expect linear relationships when nonlinear relationships actually hold. These expectations can lead to poor business decisions. These same principles apply to the notion of climate care.

back to top  Insights


This principle is elegantly expressed by De Langhe et al. in their article "Linear Thinking in a Nonlinear World" [2]. Writing in the Harvard Business Review, the authors provide a simple whiteboard graph to illustrate the disconnect between individual concern for the environment and purchase decisions that take environmental effects into account. They point out it would be natural to expect that the more individuals are environmentally concerned, the more likely they will be to make purchase decisions that take environmental effects into account. They further observe that this expected relationship does not hold. In fact, only the most environmentally concerned individuals actually make purchase decisions that take environmental effects into account. They give several other examples of how actual behaviors are not linearly related to attitudes, some with more explicit data sources than others.

Inspired by De Langhe et al.'s account, we consider the relationship between environmental awareness and environmental actions. These are more general notions than the environmental concerns and purchase decisions discussed in De Langhe et al. In the linear case (Figure 1), we may be tempted to assume that environmental awareness leads to environmental actions in a linear way.

ins03.gif Figure 1. Linear relationship between awareness and actions.

Following the reasoning of De Langhe et al., Figure 2 illustrates that the relationship between awareness and actions may be nonlinear. That is, only the most aware individuals take sound environmental actions. To the degree that this holds true for general environmentally motivated actions in the same way as it does for environmentally sound purchasing decisions, prompting environmental actions among those who are a little less aware presents the greater challenge, compared with focusing on those who are already very aware.

ins04.gif Figure 2. Nonlinear relationship between awareness and actions.

What is needed as a design goal in line with climate care is for our designs to prompt the majority of people to take environmentally sound actions, even those with only a modicum of awareness. This is illustrated by the nonlinear graph of Figure 3, which inverts the curve of Figure 2 to denote the greatest opportunity space for impact.

ins05.gif Figure 3. Needed/imagined relationship between awareness and actions for greatest impact.

How can we expect to make design decisions that prompt the larger effects the left side of Figure 3 suggests? To do this, imagine an ideal that seems even more unattainable than what is needed. That would be a straight line (Figure 4) denoting that people always take appropriate actions for climate care regardless of their levels of awareness. In other words, abandon altogether the possibility of a relationship between actions and awareness. Stated another way, can we design to prompt sound environmental actions regardless of awareness?

ins06.gif Figure 4. Ideal decoupling between awareness and actions.

back to top  The Novacene

Figure 4 appears at first to be an unattainable goal. Can such a state exist?

In Novacene: The Coming Age of Hyperintelligence, centenarian and visionary James Lovelock speculates that human awareness will not be a factor in a sustainable future, in keeping with Figure 4. He writes:

I am pretty sure that only Earth has incubated a creature capable of knowing the cosmos. But I am equally sure that the existence of that creature is imperiled. We are unique, privileged beings and, for that reason, we should cherish every moment of our awareness. We should now be cherishing those moments even more because our supremacy as the prime understanders of the cosmos is rapidly coming to end [1].

In Lovelock's vision, the Anthropocene—the age of human industrialization over the past 300 years—is coming to an end and will give way to the Novacene, an age of intelligent electronic agents. He sees such electronic forms as becoming fully autonomous and intuitive. AI (or AIs) will solve the issues of climate change more easily and more quickly than we can, essentially removing the need for human awareness from the climate care equation. According to Lovelock, this is good news, even if it is not under human control. The intelligent electronic agents of the Novacene will have an interest in keeping the planet to manageable temperatures, just as organic life does. To these agents, relatively slow-thinking humans will form part of the biosphere just like frogs or trees—helping keep Earth cool in keeping with the needs of intelligent electronic agents and organic life-forms alike. These agents will develop intuition and learn on their own at superhuman rates.

By understanding that relationships are nonlinear, designers can design systems that reward climate-careful actions over climate-careless actions.

If this seems far-fetched, we recommend reading Lovelock's visionary text firsthand to make your own judgments. Writing in the London Review of Books, Meehan Crist [3] offers a critical review that we also recommend. Regardless of how Lovelock's vision is received, it is easy to see that nowadays AI pervades everyday life in ways that are not easily predicted by its original human designers and that are increasingly autonomous.

back to top  The Role of Interaction Design

Without needing to be able to see as far ahead as Lovelock is uniquely qualified to see, interaction designers may realize they have more power than they think. By understanding that relationships are nonlinear, designers can design systems that reward climate-careful actions over climate-careless actions, regardless of whether and how awareness and actions align, or other salient relationships to actions such as influence, empowerment, universality, or resilience. We can propose a reasoning framework that some may find useful in their design work as a means of applying this discussion. Let's call it a nonlinear design-thinking framework (Table 1).

ins07.gif Table 1. Nonlinear design-thinking framework.

Using this framework to summarize our discussion, we have the information shown in Table 2.

ins08.gif Table 2. Climate care actions and awareness.

We are not arguing that these steps are necessarily easy to implement. There may be some relationships that are in fact linear. It may be hard to know what the actual nonlinear relationships are. It is easier to create designs for people who are already aware than for those who are either unaware or reluctant to prioritize climate care actions over their own pressing everyday concerns. There may not always be a way to eliminate the need for a factor like human awareness, as in Lovelock's account. Nonetheless, we are arguing that this framework is one way to understand and augment design possibilities for greater impact.

Thinking about smartphones, tablets, computers, and other digital devices, Table 3 shows a less abstract instantiation to illustrate the utility of our nonlinear design-thinking framework for interaction designers.

ins09.gif Table 3. Energy-aware settings and personal choice.

The framework may also be applied inwardly to the research on and practice of sustainable HCI (SHCI), shown in Table 4.

ins10.gif Table 4. Climate care and SHCI.

back to top  Furthermore

In this short article, we have privileged the analysis of De Langhe et al. [2] and the vision of Lovelock [1], focusing on the distinction between linear and nonlinear thinking. There are many related sources that should be mentioned. We discuss just a few here to round out our discussion. One thing that we have left out of our discussion is the notion of multidimensionality. Recall from the opening quote Lovelock's claim that reality is multidimensional and nonlinear. Our graphs are too simple for a complete analysis. Climate care actions are not related to just, or possibly decoupled from, awareness but also other dimensions such as influence, empowerment, universality, and resilience, as stated earlier. A more ambitious framework than the one we propose here would take multidimensional relationships into account.

Our discussion is also related to work in behavioral economics about notions of nudge theory [4] and human decision making [5]. Kahneman et al. [5] distinguish between noise (random outliers that are easy to ignore in a rule-based algorithmic sense) and bias (systematic misjudgments that are hard to identify as a matter of human cognition and deep-learning algorithmic bias). This distinction between noise and bias is germane to this discussion as forces that can further complicate understanding the nonlinear relationships between awareness and actions. Cass Sunstein and Lucia Reisch [6] take up the implications of nudge theory for environmentally sound actions. Their work aligns with the energy-aware settings and personal choice instantiation of our framework in Table 3.

In design outside of the arena of sustainable interaction design and SHCI, arguments for climate-aware design have a longer history. Victor Papanek's Design for the Real World [7] is often cited as the seminal work linking industrial design to the possibility of doing more harm than good in the perspective of environmental and ontological sustainability.

In an earlier article for ACM Interactions [8], we describe the connection of interaction design to the tipping point for climate change, inspired in part by Lovelock's earlier writing. Our 2010 article led to collaborations with others about the limits of mitigation, which in turn led to notions of collapse informatics (e.g., [9]), later integrating with the workshop series Computing within Limits (https://computingwithinlimits.org/) that began in 2015 and continues to this day. SHCI has archived several literature survey articles over years that are not listed here owing to the limits of the Interactions feature article format.

In future work, we plan to more fully describe these related arenas of research with respect to nonlinear design thinking and climate care. In particular, we plan to augment our explanation and diagrams here to include notions of multidimensional, nonlinear design thinking.

back to top  Acknowledgments

We gratefully acknowledge the many helpful comments and suggestions from Mikael Wiberg.

back to top  References

1. Lovelock, J. Novacene: The Coming Age of Hyperintelligence. MIT Press, 2019.

2. De Langhe, B., Puntoni, S., and Larrick, R.P. Linear thinking in a nonlinear world. Harvard Business Review (May–June 2017), 11.

3. Crist, M. Our cyborg progeny. London Review of Books 43, 1 (Jan. 7, 2021).

4. Thaler, R.H. and Sunstein, C.R. Nudge: Improving Decisions About Health, Wealth, and Happiness. Penguin, 2009.

5. Kahneman, D., Sibony, O., and Sunstein, C.R. Noise: A Flaw in Human Judgement. Little, Brown, 2021.

6. Sunstein, C.R. and Reisch, L.A. Automatically green: Behavioral economics and environmental protection. Harv. Envtl. L. Rev. 38 (2014), 127.

7. Papanek, V. Design for the Real World: Human Ecology and Social Change (2nd ed.). Academy Chicago, Chicago, 1985.

8. Blevis, E. and Blevis, S.A. Hope for the best and prepare for the worst: interaction design and the tipping point. Interactions 17, 5 (Sep.–Oct. 2010), 26–30; https://doi.org/10.1145/1836216.1836223

9. Tomlinson, B., Blevis, E., Nardi, B., Patterson, D.J., Silberman, M.S., and Pan, Y. Collapse informatics and practice: Theory, method, and design. ACM Trans. Comput.-Hum. Interact. 20, 4 (Sep. 2013), Article 24; https://doi.org/10.1145/2493431

back to top  Authors

Shunying An Blevis is a lecturer in informatics at the Luddy School of Informatics, Computing, and Engineering at Indiana University Bloomington. She holds a Ph.D. in informatics for her work on creativity, innovation, and entrepreneurship. san@indiana.edu

Eli Blevis is a professor of informatics in human-computer interaction design (HCI/d) at the Luddy School of Informatics, Computing, and Engineering at Indiana University Bloomington. His primary area of research, and the one for which he is best known, is sustainable interaction design. His research also engages visual thinking, including the photographic foundations of HCI and design theory, especially transdisciplinary design. eblevis@indiana.edu

back to top 

Copyright held by authors. Publication rights licensed to ACM.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2022 ACM, Inc.

Post Comment

No Comments Found