Sage Journals: Discover world-class research

Abstract

Organizations and universities use Design Thinking (DT) to facilitate team innovation. However, few empirical DT studies have quantified it. Across two experiments, each based on semester-long DT projects to generate innovative solutions to sustainability problems, several different DT strategies were compared. Experiment 1 (n = 145) compared team performance using a DT-Baseline process to a DT-Expanded process. Results showed that teams in DT-Expanded condition using resource constraint and “Yes and” prompts generated 64% more ideas overall than the DT-Baseline condition, but not more original ideas. Experiment 2 (n = 140) found that teams using the DT-Reverse Ideas strategy (adding a lateral thinking prompt) generated similar numbers of ideas overall, however, these teams generated 2x more original ideas than the DT-Expanded teams. This is one of the first large-scale, semester-long DT experiments measuring performance and suggests that small DT adjustments have potentially large effects on innovation.

Keywords

design thinking brainstorming team innovation creativity

Introduction

Innovation is a skill needed to solve the 21st century’s most pressing challenges. Companies often succeed at addressing customer needs by designing novel and value-added solutions. Universities build their reputations on graduating professionals who can innovate. One method organizations use to drive team innovation is Design Thinking (Liedtka, 2018). Design Thinking (DT) is a five-phase design process (Lewrick et al., 2018). One goal of DT is to leverage team cognition to support the generation of innovative, unique ideas as team cognition theory suggests that teams are more than the sum of individual contributions (Cooke et al., 2013). Meta-analyses demonstrate that positive team dynamics factors such as collaboration, safety, and communication are correlated with higher team innovation outcomes (Drach-Zahavy & Somech, 2001; Hülsheger et al., 2009). Teams may shift to be user centered (empathy), to solve a different part of the problem, or to encourage innovative solutions (Gerber, 2009; Lee et al., 2023; Moreau & Dahl, 2005).

What is Design Thinking?

DT is a team constructivist process with five phases: empathy, define, ideate, prototype, and solution test (Plattner, 2016). DT practitioners understand that the DT process is not a cookbook. Each team brings its own dynamics as well as unique skill sets and mindsets. During the DT ideate phase, teams brainstorm solutions to a given challenge and uses brainstorming to inform later prototyping. Understanding how to leverage the creative process is key to achieving innovative outcomes (Lewrick et al., 2018).

Research in Design Thinking

While DT empirical research is limited, one recent study measured participants evaluations during a large-scale, semester-long team DT exercise (Guaman-Quintanilla et al., 2023). Participants’ problem-solving skills were measured at three different times during the semester: before the DT exercise, middle of the term, and at the end of the DT procedure. Results indicated that participants self-rated and other-rated problem-solving skills and creativity skills improved over the DT exercise. However, this study did not vary the type of DT prompts, used subjective ratings, and did not measure team performance. We found no data comparing different DT methods on the innovativeness of solutions generated. The current study bridges this gap. Brainstorming is core to the DT ideate phase and has the potential to improve innovative outcomes. Research has shown that brainstorming activities affect the creativity and range of ideas generated (Putman & Paulus, 2009). DT uses different cognitive strategies to support team brainstorming.

Cognitive Prompts in Design Thinking

DT cognitive strategies that support innovation during brainstorming include constraining a team’s resources such as time, money, or resources (Moreau & Dahl, 2005), reverse ideas, reframing solutions to be the opposite of those generated, (de Bono, 1990; Lee et al., 2023) and team improv (Gerber, 2009). Resource constraints stimulate exploration and idea generation as teams are encouraged to construct solutions different from the norm (Moreau & Dahl, 2005). Reverse ideas help teams to see things from a fresh perspective (Lee et al., 2023) when they use previously brainstormed ideas as a jumping off point to generate ideas that are opposite of those provided. Improv is thought to stimulate a higher volume idea generation by establishes common ground (Clark & Brennan, 1991), facilitates collaboration and communication (Cooke et al., 2013), and creates a supportive environment for creative safety by reducing inhibitions during ideation (Gerber, 2009).

Perspective Shifting Strategies for Design Thinking

Teams who engage in perspective shift strategies are shown to generate more original solutions (de Bono, 1990; Hoever et al., 2012). Ideation in DT relates to perspective shifting. Perspective shifting (Damnik et al., 2017; Mezirow, 1978) is a key constructivist cognitive strategy in DT that is leveraged in several different phases to help designers to see challenges, problems, and solutions through new paradigms. Perspective shifting can occur at the individual level and the team level (Klein et al., 2006). Consequently, it is a motivation for cultivating diverse teams as diverse team members from different disciplines contribute their own framing lens. Diverse teams support perspective shift (Drach-Zahavy & Somech, 2001; Hoever et al., 2012). The ideate phase of DT presents an opportunity to test cognitive strategies supporting perspective shift. Empirical studies demonstrate the success of individual factors supporting team innovation, yet few studies have empirically examined these processes in a DT context or with a large project.

Current Study Contributions

Our research goals were to quantify the effect of different DT strategies designed to encourage team innovation. We designed two experiments to test different DT strategies to improve ideation outcomes. We started with a baseline DT strategy, then compared two different sets of cognitive prompts each encouraging original idea generation in a different way. These DT strategies were applied to a semester-long project. Student teams were challenged to innovate around one of the UNESCO 17 real-world sustainable development goals examples include; clean water, sustainable cities, and zero hunger (United Nations, 2015). Outcome measures included the number of common and innovative ideas generated (Hülsheger et al., 2009; Kohn et al., 2011). Our research questions were:

Do resource constraints and a “Yes and” improv prompt increase the number of ideas and the number of original ideas teams generate? (Exp 1)

Does adding the lateral thinking prompt “reverse ideas” increase the number of ideas and the number of original ideas? (Exp 2)

Experiment 1

The purpose of this experiment was to compare the baseline DT process to an expanded DT process using two popular DT prompts, while controlling for ideation time. Everything was the same for the teams except the instructions in the DT ideate phase. This will provide quantitative evidence related to different DT strategies designed to support innovation compared to a baseline DT process. Our hypotheses were:

H1: Teams using an expanded DT process (additional prompts) will generate more ideas than teams using the baseline DT process.

H1a: Teams using additional DT prompts to think creatively (DT-Expanded) will generate more innovative ideas than teams using the baseline DT process (DT-Baseline)

Methods

Participants

One hundred and forty-five students (45 teams) in a large undergraduate Introduction to Psychology course worked on a semester-long project to design a solution to 1 of the 17 United Nations sustainability challenges (United Nations, 2015). About 70% of students were from Engineering and Computer Science majors. Team sizes were three to four people. Participants signed a consent form to share their data for this study, and they received research credit for participating. All team members needed to agree for the data to be included and all did.

Experimental Design

Data were analyzed using a 2 DT Condition (DT-Baseline vs. DT-Expanded) ×2 Idea Type (Common, Innovative Ideas) mixed-factorial ANOVA. DT condition was a between-subjects independent variable. Idea type was a nested factor within Teams. DT strategy was between subject factor and the Idea type and DT conditions were treated as fixed factors (using R). The dependent measure was the number of common and innovative ideas generated by each team during a 10-min DT ideate phase.

Originality Coding Scheme

Two independent coders rated the originality of each of the 865 idea solutions using the coding scheme in Table 1. The originality coding scheme distinguished between three types of innovative ideas: novel, divergent and extended ideas and was informed by creativity research (Lee et al., 2023; Moreau & Dahl, 2005). Coders achieved an inter-rater reliability (IRR) of .82 (Cohen, 2007).

Table 1.

Originality Coding Scheme.

Originality	Definition	Example solution
Common	A solution widely used	Promote charities around campus.
Novel	New and unique idea	Build a dome of mirrors to focus energy on one point
Divergent	Applies an existing solution in an unusual way	Sponsor exchange programs to immerse students in a (UN) challenge
Extended	By adding features, or changing an idea’s function	Use QR codes around campus that direct students to companies that plant trees

Procedure

All teams completed the semester-long DT project. The manipulation was focused on the 45 min in-class DT ideate phase during which teams spent 10 min brainstorming under different DT strategy conditions and generating ideas. One DT strategy was the DT-Baseline activity, the other included a DT-Expanded condition with a problem constraint and a “Yes and” prompt. Teams in the DT- Baseline condition chose a problem, defined it and then brainstormed for 10 min. Teams in the DT-Expanded condition, chose a problem, defined it, and then brainstormed for 10 min total with two manipulations added. After 4 min of baseline brainstorming, a resource constraint “A solution that costs at least $1M” was added, teams then brainstormed for 3 min. Next, teams brainstormed for another 3 min using the “Yes and” Improv step by building off any idea on the table. They read the idea aloud, then said, “Yes and” and added their own idea. See Figure 1.

Figure 1.

DT baseline and DT-expanded condition procedures.

Results

Table 2 breaks down the total number of ideas generated across all teams by prompt and DT condition. This provides a sense of how many more ideas are generated by prompt type. DT-Expanded teams generated 280 new ideas across the two, additional DT prompts and 189 more ideas overall than DT-Baseline teams in the same 10-min period.

Table 2.

Total Number of Ideas by DT Condition and Prompt.

Condition	Brainstorming	Constraint	Yes and	Total
DT Baseline	338	-	-	338
DT Expanded	247	135	145	527

To evaluate H1 and H1a, data were analyzed using a 2 DT Condition (DT-Baseline vs. DT-Expanded) ×2 Idea Type (Common, Innovative Ideas) mixed-factorial ANOVA using the lmer function of the R lme4 library. Means are provided in Table 3. In support of H1, DT-Expanded teams generated 64% more ideas on average than the DT-Baseline teams, but not more innovative ideas F(1,35) = 5.56, p = .024, η2 = .14, a large effect. There was also a statistically significant interaction between DT Condition and Idea Type. To evaluate H1a and the interaction, planned comparisons showed no statistically significant differences for DT Condition on the number of innovative ideas, t(31) = 1.329, p = .258. However, there was for common ideas, t(45) = 11.23, p = .0016, η2 = .2. Not only was this difference surprising, but it was also a large effect.

Table 3.

Mean Number of Ideas by Idea Type and DT Condition.

Condition	Common M (SD)	Innovative M (SD)	Total M (SD)
DT-Baseline	12.0 (5.59)	3.56 (2.92)	8.53 (6.26)
DT-Expanded	19.8 (9.77)	4.09 (2.07)	12.48 (10.78)

Discussion

The results of Experiment 1 showed support for H1, teams in the DT-Expanded condition generated more ideas overall (Gerber & Carroll, 2012). However, H1a was not supported, surprisingly there was no significant difference in the number of innovative ideas generated. Teams in the DT-Expanded condition using the “Yes and” prompt did build on each other’s ideas, but did not generate more original ideas as might be expected (c.f., Cooke et al., 2013; Hülsheger et al., 2009). As we could find no quantitative research with prompts, this research will need to be replicated. One explanation of the “Yes and” prompt, is that teams may have anchored (Kahneman, 2011) on an idea and built it out, rather than pivoting or broadening their ideas which negatively impacted innovation. Therefore, Experiment 2 was designed to replicate the DT-Expanded result and test a new lateral thinking strategy.

Experiment 2

While our “Yes and” prompt helped with team collaboration, it did not encourage perspective shift (de Bono, 1990; Hoever et al., 2012). This led to Experiment 2 which used a lateral thinking DT brainstorming prompt to generate opposite ideas or reverse ideas. Experiment 2 compared a new lateral thinking prompt condition (DT- Reverse ideas) to a DT-Expanded condition (Figure 2). This was a replication for the DT-Expanded condition and the new condition focused on two lateral thinking strategies and was hypothesized to help teams increase the number of ideas and original ideas. Our hypotheses were:

H2: Teams using the increased lateral thinking strategies with the “reverse ideas” prompt will generate more ideas overall than teams using the DT-Expanded strategies.

H2a: Teams using the lateral thinking strategies with the “reverse ideas” prompt will generate more innovative ideas than teams using the DT-Expanded strategies.

Figure 2.

DT-expanded and DT-reverse ideas procedures.

Methods

Participants

One hundred and forty students (42 teams) in a separate large undergraduate Introduction to Psychology course worked on the same semester-long design task and earned research credit for their participants. Again, about 70% of students were from Engineering and Computer Science majors. Team size was three to four people.

Experimental Design

Data were analyzed using a 2 DT strategy (DT-Expanded vs. DT-Reverse Ideas) ×2 Idea Type (Common, Innovative Ideas) mixed-factorial ANOVA. The experiment design, dependent measures, and coding scheme (Table 1) were the same as described in Experiment 1.

Procedure

Experiment 2 leveraged the same experiment protocol and dependent measures as Experiment 1. The key difference was a new condition to increase perspective shift. The DT-Reverse Ideas condition included a resource constraint and a reverse ideas strategy where students brainstormed solutions the opposite of solutions already shared. During the DT-Reverse Ideas (Figure 2), teams brainstormed for 4 min, then a resource constraint, “a solution that costs at least $1M dollars,” was added and they brainstormed for three more minutes. The next manipulation was to brainstorm a new solution that is the opposite of an idea already posted. They brainstormed opposite ideas for 3 min.

Results

Table 4 shows the total number of ideas generated across all teams by prompt and DT condition to provide a sense of how many more ideas are generated by prompt type. With the additional prompts and no change in overall ideate time, DT-Reverse Ideas and DT-Expanded are similarly distributed. The DT-Expanded teams in Experiment 1 (Table 2) and Experiment 2 (Table 4) generated a similar number of total ideas and average number of ideas (Table 5).

Table 4.

Experiment 2 Total Ideas by DT Condition.

Condition	Brainstorming	Constraint	Prompt	Total
DT expanded	259	168	109	536
DT reverse ideas	232	205	135	573

Table 5.

Experiment 2 Mean Ideas by DT Condition.

	Common M (SD)	Innovative M (SD)	Total ideas M (SD)
DT- expanded	19.47 (8.99)	4.66 (SD = 3.13)	13 (6.06)
DT-reverse ideas	19.36 (3.59)	10.6 (7.12)	13.3 (5.35)

To evaluate H2 and H2a, data were again analyzed using a 2 DT Condition (DT-Baseline vs. DT-Expanded) ×2 Idea Type (Common, Innovative Ideas) mixed-factorial ANOVA using the lmer function of the R lme4 library. Means are provided in Table 5.

Teams in the DT-Reverse Ideas condition which used a resource constraint and then an opposite idea prompt generated 2x more innovative ideas on average than the DT- Expanded teams. There was a statistically significant interaction between DT Condition and Idea Type, F(1,37) = 4.97, p = .03, η2 = .12, indicating a large effect size. To evaluate the interaction, planned comparisons revealed no statistically significant difference of DT Strategy for common ideas, t(40) = 0.002, p = .96. However, DT-Reverse ideas teams generated statistically more innovative ideas than DT- Expanded teams, t(37) = 11.97, p = .014, η2 = .24, indicating a large effect size. This result supports H2a.

Discussion

To summarize, in Experiment 1, while the DT- Expanded teams generated 64% more ideas on average than the DT-Baseline teams, there was no difference in the number of innovative ideas. In contrast, for Experiment 2, DT-Reverse Ideas teams doubled the number of innovative ideas compared to the DT-Expanded teams. Not only was this effect statistically reliable, but it was important as the effect size was large. DT-Reverse Ideas was found to be an impactful lateral thinking strategy and seemed to shift perspective and spur creativity in support of past research (de Bono, 1990; Hoever et al., 2012; Lee et al., 2023).

These experiments were well-powered statistically, conducted with >40 teams each, and evaluated in a semester-long design project; it suggests that small changes to the DT ideate phase can have a big impact on team innovation performance. This is important for evidence-based research and interesting for team collaboration as it is one of the few experiments capturing DT outcome measures. It should be noted that all conditions were DT conditions, these experiments do not provide any data comparing DT strategies to other team innovation strategies. Furthermore, students had already started with a set of DT strategies to support innovation (i.e., empathy, define) before the ideate phase. Given this was only comparing DT strategies, not DT to another team innovation strategy, this result makes an important contribution to the DT literature (Liedtka, 2018), and building on previous recent studies by adding a DT performance measure (c.f., Guaman-Quintanilla et al., 2023).

As with any experiment, there are a few limitations. First, this study only compared DT conditions, it did not compare DT to another design method. Therefore, the results should be considered in that context. Another limitation of this study is that the outcomes may have been influenced by students’ experience with UNESCO sustainability design solutions. Basically, if students were more expert in sustainability, they might have generated more innovative ideas. Finally, idea quality was based on the number and types of ideas generated, not the quality of those ideas. It is possible that the DT-Reverse ideas teams generated more innovative ideas, but not that would be implementable. For the quality to be evaluated, a panel of experts in sustainability could have rated each idea generated.

For DT practitioners, these experiments suggest that not all lateral thinking strategies have a similar impact on innovation or DT outcomes. Practitioners may want to choose strategies and tailor them to the task to leverage the skills of the team (Lewrick et al., 2018). For example, if generating a wider range of ideas is important, then including a reverse ideas prompt would be useful. More research is needed, but these experiments suggest that increasing perspective shifting strategies will increase team innovation. This small change during the DT ideate instructions had a big impact on team innovation.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Lisa M. Casper

Elizabeth S. Veinott

References

Clark

H. H.

Brennan

S. E.

(1991). Grounding in communication. In Perspectives on socially shared cognition (pp. 127–149). American Psychological Association.

Cohen

(2007). A power primer. Tutorials in Quantitative Methods for Psychology, 3(2), 79. https://doi.org/10.20982/tqmp.03.2.p079

Cooke

N. J.

Gorman

J. C.

Myers

C. W.

Duran

J. L.

(2013). Interactive team cognition. Cognitive Science, 37(2), 255–285. https://doi.org/10.1111/cogs.12009

Damnik

Proske

Körndle

(2017). Designing a constructive learning activity with interactive elements: The effects of perspective-shifting and the quality of source material. Interactive Learning Environments, 25(5), 634–649.

de Bono

. (1990). Lateral thinking: A textbook of creativity. Penguin UK.

Drach-Zahavy

Somech

(2001). Understanding team innovation: The role of Team Processes and structures. Group Dynamics Theory Research and Practice, 5(2), 111–123. https://doi.org/10.1037//1089-2699.5.2.111

Gerber

(2009). Using improvisation to enhance the effectiveness of brainstorming. In Proceedings of the SIGCHI conference on human factors in computing systems. Boston, MA.

Gerber

Carroll

(2012). The psychological experience of prototyping. Design Studies, 33(1), 64–84.

Guaman-Quintanilla

Everaert

Chiluiza

Valcke

(2023). Impact of design thinking in higher education: A multi-actor perspective on problem solving and creativity. International Journal of Technology and Design Education, 33(1), 217–240. https://doi.org/10.1007/s10798-021-09724-z

10.

Hoever

I. J.

van Knippenberg

van Ginkel

W. P.

Barkema

H. G.

(2012). Fostering team creativity: Perspective taking as key to unlocking diversity's potential. Journal of Applied Psychology, 97(5), 982–996.

11.

Hülsheger

U. R.

Anderson

Salgado

J. F.

(2009). Team-level predictors of innovation at work: A comprehensive meta-analysis spanning three decades of research. Journal of Applied Psychology, 94(5), 1128–1145. https://doi.org/10.1037/a0015978

12.

Kahneman

(2011). Thinking, fast and slow. Macmillan.

13.

Klein

Moon

Hoffman

R. R.

(2006). Making sense of sensemaking 1: Alternative perspectives. IEEE Intelligent System, 21(4), 70–73.

14.

Kohn

N. W.

Paulus

P. B.

Choi

(2011). Building on the ideas of others: An examination of the idea combination process. Journal of Experimental Social Psychology, 47(3), 554–561. https://doi.org/10.1016/j.jesp.2011.01.004

15.

Lee

O’Mahony

Lebeck

(2023). Prompts for creativity. In Creativity and Innovation: Everyday Dynamics and Practice (pp. 85–115). Palgrave Macmillan.

16.

Lewrick

Link

Leifer

(2018). The design thinking playbook: Mindful digital transformation of teams, products, services, businesses and ecosystems. John Wiley & Sons.

17.

Liedtka

(2018). Why design thinking works. Harvard Business Review, 96(5), 72–79.

18.

Mezirow

(1978). Perspective transformation. Adult Education, 28(2), 100–110.

19.

Moreau

Dahl

(2005). Designing the solution: The impact of constraints on consumers’ creativity. Journal of Consumer Research, 32(1), 13–22. https://doi.org/10.1086/429597

20.

Plattner

(2016). An introduction to design thinking PROCESS GUIDE. Institute of Design at Stanford.

21.

Putman

V. L.

Paulus

P. B.

(2009). Brainstorming, brainstorming rules and decision making. Journal of Creative Behavior, 43(1), 29–40. https://doi.org/10.1002/j.2162-6057.2009.tb01304.x

22.

United Nations. (2015). The 17 goals| sustainable development. Sustainable Development. https://sdgs.un.org/goals

Design Thinking: Experimental Evidence of Ideation Strategies That Support Team Innovation

Abstract

Keywords

Introduction

What is Design Thinking?

Research in Design Thinking

Cognitive Prompts in Design Thinking

Perspective Shifting Strategies for Design Thinking

Current Study Contributions

Experiment 1

Methods

Participants

Experimental Design

Originality Coding Scheme

Procedure

Results

Discussion

Experiment 2

Methods

Participants

Experimental Design

Procedure

Results

Discussion

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iDs

References