Word count: Front matter 233, text 10154, back matter 2626. Sustainable product development and tricks on the mind: Formulating conceptual models of cognitive illusions and mitigating actions

Rachael K. Gould and Martin Svensson

Blekinge Institute of Technology, Sweden

Abstract: Similar to visual illusions playing tricks on your eyes, cognitive illusions cause you to misjudge reality and therefore potentially make biased decisions. You are especially vulnerable when starting to introduce sustainability into concept development decision-making since this setting offers unfamiliarity, and complexity. Given a shortage of theories regarding which cognitive illusions product developers are susceptible to, we formulated a conceptual model. This model is based on the decision-making tasks that product developers undertake when they are developing concepts and the conditions that they experience when integrating sustainability into this decision-making. From decision-making literature, we identified the following cognitive illusions as occurring when undertaking those tasks under those conditions: availability, anchoring and confirmation bias when generating ideas; illusory correlation and validity effect when selecting between ideas; evaluability bias and status quo bias when gate reviewing. Based on the model, we synthesised literature on how to mitigate for the identified illusions and organised this synthesis according to when (during which task type) a product developer should perform the mitigating actions. These mitigating actions vary according to task type and focus on the quality of the decision-making process.

Keywords: the soft side of ecodesign; sustainable product development; sustainable design; decision-making; cognitive illusions; fallacies and biases in decision-making.

Submitted manuscript conditions (Arvai et al., 2012), which can trigger 1. Introduction inaction even when there is clear benefit from acting, such as failing to act on climate change (Gifford, 2011; For inexperienced individuals, starting to include Sunstein, 2006; Weber, 2006). We therefore wanted to sustainability in product development make the understand what decision-making challenges – decision-making environment tricky. This challenging cognitive illusions – product developers face when environment might lead to cognitive illusions (tricks bringing sustainability into early-phase (concept on the mind) and thus decision-making errors. development) decision-making. Cognitive illusions are defined as cognitive processes Examples of errors in decision-making that can that lead to a perception, judgement or memory that arise as a result of cognitive illusions include peoples’ deviates from ‘reality’; this deviation is systematic belief in climate disruption being biased by films that (reliable and in a predictable direction), appears they have recently seen (Arvai et al., 2012) and by involuntarily and is hard to avoid; such processes are higher predictions of the rise in the earth’s also distinct from normal forgetting and temperature (Joireman et al., 2010). Other examples misunderstanding (Pohl, 2004). Many cognitive include judges’ sentencing decisions being affected by illusions can be seen as “the flipside of otherwise the outcome of a dice throw (Englich et al., 2006) and adaptive and prudent algorithms” (Fiedler, 2004, pg customers paying more for less ice cream because it is 111). These algorithms are referred to by some as presented in a relatively smaller cup (Hsee, 1998). The heuristics, which are rules of thumb that may bias repercussions of paying more for less ice cream are decision-making (Tversky and Kahneman, 1974). perhaps trivial, but if we move into the arena of It is important to be knowledgeable about product development, imagine what might result from potential cognitive illusions when product developers errors in decision-making. Failure on the market, integrate sustainability into their work since product demotivated employees and damaged brand image lifecycle impacts are a major contributor to many of might ensue. society’s environmental and social challenges Over several decades, many researchers have (Kaebernick et al., 2003; Pujari, 2006; Tukker and studied cognitive illusions, their process and what Jansen, 2006). Sustainable product development is triggers them. These studies have largely been also a means to become and remain competitive experiments to isolate and characterise the illusions, (Dangelico and Pujari, 2010; Pujari et al., 2003). not to describe which illusions strike in the particular The integration of sustainability into product context that we want to study. There are many studies development is not only a technical challenge, but it identifying which illusions arise in other contexts; for also requires consideration of human (socio- example, in climate prediction (e.g. Nicholls, 1999), in psychological) elements (Boks and McAloone, 2009; medicine (e.g. Saposnik et al., 2016) and in Post and Altman, 1994), including individual aspects negotiations (e.g. Caputo, 2013). Some work identifies (Brones et al., 2014) such as behaviour and attitude which illusions occur during decision-making for (Hassi et al., 2009) individual judgement and decision- sustainability in general, for example, Arvai et al. making bias (Brones et al., 2017). Barriers to the (2012) and Gifford (2011), and very little identifying implementation of sustainableSubmitted product design are which occur during product development decision- related to the individual behavioural aspects in the making, for example, Antioco et al. (2008). In a product development project teams (Gould et al., perspective piece, Liedtka (2014) lists illusions that n.d.). might be addressed by design-thinking practices. In a The early phases of product developmentmanuscript are a review and research agenda piece, Despeisse and leverage point for integrating sustainability since the Vladimirova (2014) identified the need to understand early-phase decision-making influences the “the specific challenges when bringing sustainability sustainability impacts of manufacturing and other into manufacturing decision-making” [pg 148] lifecycle phases (Bhamra et al., 1999). Decisions in the (although not early phase product development early phases of the process also determine, to a large decision-making). At the juncture of decision-making extent, the success of a project (Martinsuo and for sustainability and product development decision- Poskela, 2011) and significantly impact development making, we found no literature on cognitive illusions. costs (Ulrich and Pearson, 1993). Indeed, individual behavioural aspects, including When making early-phase decisions, product individual decision-making and biases, have scarcely developers know little about the design problem been studied for the implementation of sustainable (Ullman, 2003). The earliest decision-making is product design (Brones et al., 2017). We therefore set sometimes even referred to as the fuzzy front end out to tackle this gap for product developers who are (Boks and Wever; Reid and de Brentani, 2004). starting out on their journey to implement sustainable Knowing only a little can cause product developers to product development by addressing the following judge inaccurately and evaluate incompletely (Chin, research questions: 1999). Furthermore, the novelty and complexity of sustainability results in tough decision-making

RQ1: What cognitive illusions may hinder Building on the descriptive perspective, another product developers when starting to include type of model, prescriptive models of decision- sustainability in concept development? making, strive to help decision-makers to close the gap RQ2: How might product developers be between how they make decisions (described by supported to mitigate the effects of the cognitive descriptive models) and an ideal (defined by illusions when starting to include sustainability in normative models) (Arvai et al., 2012; Bell et al., 1988). concept development? Sustainable development research is intentional, solutions orientated research – endeavouring to To address these questions, we used academic suggest how society can transition towards a literature to construct a conceptual model that sustainable future (Miller, 2013; Miller et al., 2014). We describes the relevance of specific cognitive illusions therefore considered it important to also include how and how to mitigate them. We present this model both product developers might be able to mitigate the with propositions and in a figure. The uniqueness of effects of cognitive illusions when they are this study arises from theorising at the (under- incorporating sustainability in concept development. investigated) intersection of three fields – sustainable development, product development and cognitive 2.2. Types of concept development decision-making illusions. We also contribute by synthesising relevant literature that can be used by designers of support for In this subsection, we describe the types of sustainable product development. Guidance on decision-making used in concept development. Since mitigating for cognitive illusions seems to be sustainability should be integrated into existing otherwise lacking in the sustainable product product design processes rather than creating new development field. processes for sustainable product design (Baumann et al., 2002; Brones and de Carvalho, 2015), these types 2. Background: Decision-making when starting are not specific to the integration of sustainability but to include sustainability in product development are generic for new product development. Product development decisions do not vary 2.1. A descriptive view of decision-making widely, even though the products can differ greatly. There are certain decisions that need to be made in There are two main perspectives on people as order to get to the point of having a product, and so decision-makers – normative and descriptive- these decisions end up being made, deliberately or prescriptive. For instance, in economic theory, otherwise (Krishnan and Ulrich, 2001). In a review of humans are modelled as perfect decision-makers, that the literature, Krishnan and Ulrich (2001) collected is, rational decision-makers who have access to and ‘generic’ decisions from across academic disciplines use all relevant information to form stable preferences and define concept development decisions as being – known as normative models of decision-making about specifying what the product will achieve and be, (Arvai et al., 2012; Bell et al., 1988). Such assumptions including generating and selecting product concepts. are considerably difficult to accept for sustainable Concept development decision-making involves concept development as there are many unknowns. Submittedproduct developers pursuing two types of activities - Over the past sixty years or so, behavioural decision generating options (divergent decision-making) and research has sought to understand how people selecting between them (convergent decision- actually make decisions, flaws and all – known as making) (Krishnan and Ulrich, 2001). Please note that descriptive models of decision-making (Arvai et al., manuscriptdeveloping concepts is not a two-step divergent- 2012; Bell et al., 1988). Behavioural decision research convergent process, but rather product developers includes empirical-based models about how people iterate through divergent and convergent decision- can, at best, be bounded in their rationality (see making. However, product developers and their (Simon, 1957)), the decision-making shortcuts and managers may have different views on how iterative rules of thumb that people use, and the ensuing the process is (Kihlander and Ritzén, 2012). systematic errors in decision-making, including Companies often have a gate review between cognitive illusions (Bell et al., 1988). It has been shown the concept development and product design phases. that people construct their preferences based on cues, The purpose is to manage risk by reviewing the rather than revealing some pre-existing, well-defined readiness of the new product project to proceed to the preferences (Slovic, 1995). Descriptive models help us next phase. Often, a single product concept has been to understand why, for example, you might not always selected and now it is time to review the readiness of maintain a balanced diet, especially when you are the whole project, with some authors advocating a stressed, even though you know that it would be wise. go/no-go type review. For example, Cooper (2008) Since we wanted very much to engage with the gives the reviewers’ four options – go, kill, hold or realities of product developers in order to find ways to recycle. The review is achieved by assessing the project help them, we chose to draw on the descriptive, rather against criteria, which usually cover four dimensions - than normative, models of decision-making. technical, strategic, customer acceptance, financial, have low knowledge and high design freedom. It is and market-related (Carbonell-Foulquié et al., 2004). difficult to even state when decisions are made – the The ‘gate’ nature of these reviews differs in decisions are embedded in the development of practice, at least in some industries, from Coopers’ concepts and they cannot easily be teased out, even if prescription. In some industries, such as aerospace, the company’s process prescribes making them at contracts have often already been signed and so a ‘kill’ certain points in time (Kihlander et al.). When making decision becomes highly unlikely and certainly these decisions, it is difficult to know how much undesirable (Christiansen and Varnes, 2008; information is needed to avoid major re-work later Johansson, 2014). This leads to a review that generates (Kihlander and Ritzén, 2012), that is, it is challenging an action plan comprising conditions for progression to know how good a decision is needed. to the next phase (Johansson, 2014). Product developers may also experience time- pressure. Perceived time-pressure have been seen to 2.3. Decision-making conditions when including negatively affect decision-making performance in lab- sustainability in concept development based settings (DeDonno and Demaree). Outside the lab, time pressure has been demonstrated to lead to In this subsection, we describe the decision- both positive and detrimental effects on quality for making conditions when integrating sustainability individuals (Lepine et al., 2005). Team decision- into concept development. making also suffered, particularly when the pressure did not lead to a satisfying, enjoyable and stimulating 2.3.1. Low knowledge, high uncertainty and time experience (Chong et al., 2011). Even when perceived pressure as a challenge that provides opportunities to problem- Product development decision-making is solve and learn, experiencing time pressure can add up characterised by uncertainty and long feedback loops, over time and lead to strain and negative impacts on commitment of large amounts of resources over long performance (Lepine et al., 2005). periods of time before the outcome and consequences are understood (Matheson, 1998). Including a large 2.3.2. Ad-hoc gate reviews and quality not explicit amount of uncertainty is something that product Although product developers’ knowledge about developers themselves have expressed experiencing the problem space and solution space has increased (Kihlander and Ritzén, 2012). Early in the process, since the beginning of their project, reviews are still product developers know less about the design made under significant uncertainty. However, the problem. As product developers work on the project, (low) quality of information available is not always they increase their knowledge of the problem and explicitly discussed (Johansson et al., 2009). Decision- possible solutions. Simultaneously, as product makers can mistake precise requirements for developers develop, select and detail attributes and uncertain placeholder values, and vice-versa (Stacey concepts, they reduce their design freedom. The lack and Eckert, 2003). The reviews can be ad-hoc in of knowledge while design freedom is high and lack of nature, depending on the people involved and the freedom when knowledge is highSubmitted is called the ‘design culture (Johansson, 2014). What's more, the reviewers process paradox’ by Ullman (2003). Figure 1 shows how likely feel pressured to make the decision quickly, Ullman represents the paradox. since it is important to avoid delay, especially in a situation where the company is already contracted to manuscriptdeliver by a given deadline (Christiansen and Varnes, 2008).

2.3.3. Sustainability is unfamiliar and adds complexity

The nature of sustainability is complex; it relates to social and ecological systems (Broman and Robert, 2017), and includes considering not only current global issues, but also potential future issues. Sustainability challenges can therefore be experienced as distant, in time and in space (Stoknes, 2014; Weber, 2015), which can lead to lower levels of concern (Spence et al., 2012). The sustainability consequences of a decision can be Figure 1 - The design process paradox. wide-ranging (for example, increased allergies and Figure reproduced from Ullman (2003) fig. child labour) and distributed (for example, mining in 1.11. far-flung lands), yet also interconnected (for example, deforestation and ocean acidification). Sustainability Concept development lies very early in the time- has therefore been named a ’wicked problem’ line of Ullman’s paradox, when product developers requiring a change of the whole system (Schäpke et al., sustainability and concept development. The field of 2013). cognitive illusions is vast, with a well-developed core Wrestling with the complexity of sustainability is and emerging edges. (The illusions range in credibility relatively new for product developers who are just and depth of description – see, for example, starting out with implementing sustainable product Wikipedia.) We therefore used compilations of development, compared to what they usually include illusions that were written and edited by experts in in their decision-making. This means that they likely cognitive illusions and decision-making for have relatively lower knowledge of sustainability, even sustainability in order to start from the core of the field in the beginning when (according to the design with well-established illusions. process paradox) they have generally low knowledge. To identify which compilations to use, we According to Slovic (1995), the complexity and searched for books and journal articles in a multi- unfamiliarity of environmental choices makes them “a database service with the following search terms and likely candidate for constructive processes” [pg 369- got 107 hits: cognitive illusions; biases. We 370]. The emotive and political nature of sustainable supplemented the list by also searching for books in development can even lead to individuals the library database Libris. From the resulting list, we experiencing discomforting dissonance (Stoknes, selected those sources that are scholarly, cover 2014). multiple illusions (that is, a compilation), are The novelty and complexity of sustainability published in English and are general or specific to gives challenging decision-making conditions (Arvai sustainable/product development. Some sources were et al., 2012). Responses to the challenge of climate specific to other fields, such as medicine (e.g. Saposnik change are not solely guided by rational processes et al., 2016; Yarritu et al., 2015), and were thus rejected (Kunreuther et al., 2014). In particular, bringing a from our selection. After reading the sources, we sustainability perspective to new product selected the 5 sources shown in Table 1. These sources development “complicates an already complex are also highly cited, and even cited by some of the process” (Goffin, 2012, pg 106) and the transformation specific studies that we excluded (e.g. Caputo, 2013). involved in sustainable production and consumption During our study, Pohl released a second edition of the is recognised as fundamentally complex (Tukker et al., handbook on cognitive illusions and so we included 2008). both the original and the second edition. In Table 1, we list the 29 illusions that we found in the selected 3. Research design - theorising literature.

To address the research questions, we employed existing literature (from various fields) to theorise – to 3.2. Identifying which cognitive illusions may arise formulate and express conceptual systems – in when starting to include sustainability accordance with Jaccard and Jacoby (2010). A Much of the research on cognitive illusions conceptual system is concepts placed in relationship demonstrates under which circumstances which with each other to facilitate understanding, a illusions arise and therefore, in order to use this conceptual model. In order toSubmitted empirically test for the literature, we needed to first understand the decision- presence of a cognitive illusion, you must isolate the making circumstances when product developers are individual in a controlled situation; you cannot to starting to include sustainability in concept detect the consequences of multiple illusions by development. The circumstances include the (a) observing product development. Indeed,manuscript these nature of the decision-making task (type of decision- experiments that are performed under controlled making) and the (b) conditions under which the conditions are exactly the type of research that is decisions are made. Parts a and b are described in the described in the literature that we have used in order background section (section 2). Once we understood to understand which conditions cause which illusions the circumstances, we could then (c) identify which and therefore identify which illusions are relevant to cognitive illusions arise under these circumstances. the context of integrating sustainability and product This is indicated in Figure 2. Our approach is also in development. line with how Montibeller and Winterfeldt (2015) identified biases in decision and risk analysis – by first 3.1. Identifying established cognitive illusions via understanding the task and then identifying the biases key sources relevant to the nature of the task. Our work is also of a similar design to De Treville et al. (2009) who discuss Our first step was to understand what cognitive cognitive illusions and mitigating actions for illusions are out there, in order to then be able to select operations management. the ones that are relevant when integrating Table 1 - The cognitive illusions (column 1) included in each of the five compilations (row 1) and therefore considered for inclusion in our model.

Arvai, J.; Pohl, R.F. Pohl, R.F. Campbell-Arvai, Cognitive Cognitive V.; Steel, P. Bazerman, Evans, J.S.B.T. illusions: A illusions: Decision-making M.H.; Moore, Hypothetical handbook on Intriguing for sustainability: D.A. Judgment thinking: Dual fallacies and biases phenomena in A systematic in managerial processes in in thinking, judgement, review of the body decision reasoning and judgement and thinking and of knowledge. making. 2012. judgement. memory. 2004. memory. 2017. 2012. (Arvai et al., (Bazerman and 2007. (Evans, (Pohl, 2004) (Pohl, 2017) 2012) Moore, 2012) 2007) anchoring     associative memory illusions   availability      base rate neglect     belief bias    confirmation bias     conjunction fallacy     effects of labelling   evaluability bias   halo effect  hindsight bias     illusion of control   illusions of change  illusory correlation   matching bias effect    mere exposure effect   misinformation effect   Moses illusion  orientation illusions  overconfidence    Pollyanna principle   probability matching  pseudodiagnosticity  representativeness     retrieval-induced forgetting  revelation effect  status quo bias    survival processing effect Submitted  validity effect   manuscript

Figure 2 - Which cognitive illusions arise when starting to include sustainability in concept development decision-making is determined by (a) the type of decision being made and (b) the conditions under which these decisions are made. See section 2 for details of a and b.

Submitted manuscript An example of including an illusion in our model 4. Cognitive illusions that arise when starting to is: the validity effect occurs when decision-makers include sustainability in concept development are more familiar with some information than other In this subsection, we contribute by building information being used to inform a decision. Relative theory about what cognitive illusions that occur when unfamiliarity with sustainability aspects compared to performing the types of decisions outlined in section other aspects was one of the conditions that was 2.2 (generating, selecting, gate reviewing) under the identified and thus validity effect was included in our conditions outlined in section 2.3 (low knowledge, model. For details of the relationships between time pressure, unfamiliarity with sustainability, decision type, conditions and illusion, see section 3 for previously established hypothesis, judging an cognitive illusions that were identified as relevant. An alternative in isolation, unsure how much is good example of excluding an illusion is: the conjunction enough, feeling committed to a concept). We thus fallacy can be exaggerated by certain priming (Bakhti, address research question 1. We define each of the 2018) and is found to be higher in people who more selected cognitive illusions and justify their inclusion. strongly believe in conspiracy theories and the The section is divided into subsections according to paranormal (Brotherton and French, 2014). Since these type of decision-making. are not conditions that are specific to product developers starting to work with sustainability, this illusion was excluded from the model. 4.1. Cognitive illusions when generating alternatives 3.3. Prescribing how to mitigate In this subsection, we describe the cognitive illusions relevant when generating alternatives, and Once we had identified the cognitive illusions, associated implications when integrating we used decision-making literature to understand how sustainability into product development. to mitigate the effects of these illusions. On top of the literature that we had previously gathered, we also 4.1.1 Availability searched for scholarly sources (in Summon) on how to mitigate each of the selected illusions. From this, we Availability occurs when people think that the constructed a model of what designers of support can ease of recall indicates a high frequency (Schwarz et do to help product developers to start to include al., 1991) - it’s easy for me to remember times when I sustainability in concept development, thus have been assertive, and therefore it must happen often. addressing research question 2. This can arise irrespective of how many times the event has actually occurred (Reber, 2004; Tversky and 3.4. Quality of research design Kahneman, 1973). The illusion is particularly pervasive for vivid or sensational events (Reber, 2004). For Quality criteria for a theory are that the models example, exposure to films of extreme weather events are logically consistent internally and in agreement have been associated with increases in people’s belief with existing knowledge (Shaw, 1982). Our research that climate disruption is actually occurring (Arvai et was designed to do exactly this by using existing Submittedal., 2012). Otherwise people may not be impelled to knowledge of under what conditions cognitive respond to sustainability issues such as climate change illusions arise as strict criteria for inclusion in our because they can only bring to mind a few instances of model and similarly for inclusion of mitigating actions climate and extreme weather issues (Sunstein, 2006). in our suggestions for practice. manuscriptThis illusion occurs when people do not have the We did not set out to critique the literature, but information, or when they are not sufficiently rather to search for key texts and generally accepted motivated to systematically process the information knowledge, which of course has its limitations. We (Reber, 2004). Low motivation to consider chose to use systematic reviews and handbooks that sustainability occurs due to doomsday framing of assemble well-established knowledge from peer climate change issues (Stoknes, 2014). The combined reviewed sources compiled by experts in the cognitive potential lack of motivation to consider sustainability illusions and decision-making for sustainability fields. and general lack of information on during concept By using these collections of empirically tested theory, development provide ripe conditions for availability. robustness was added to our work, more so than if Proposition 1.1.1: When generating ideas while trying to cover all work on the edges of a vast and fast- starting to integrate sustainability into concept moving field. The list of cognitive illusions (Table 1) is development decision-making, the time pressure, also consistent with example illusions listed by which can cause lack of motivation to Krueger and Funder (2004) and Stanovich (2009). systematically process information, and the lack of information may lead to product developers basing expectations of frequency on how easy it is for them to recall related information, that is, being hindered by availability.

4.1.2. Anchoring decision-making (Oswald and Grosjean, 2004), and therefore also of product development decision- Anchoring is when decision-makers ‘anchor’ making. However, decision-makers often favour their judgement on some initial reference point information that supports their hypothesis over (anchor) when, for example, estimating a value. This information that does not fit well with the hypothesis. illusion appears even when the anchor is clearly This favouring of supporting information is known as irrelevant to the decision at hand (Arvai et al., 2012) or confirmation bias and occurs primarily when the an extreme value, and occurs even for experienced decision-maker has already established a hypothesis decision-makers (Mussweiler et al., 2004). For or when the decision-maker wants the hypothesis to example, judges’ sentencing decisions were biased by be true (Oswald and Grosjean, 2004). For example, sentencing demands (working as anchors) and this when Lord et al. (1979) presented two scientific studies bias occurred even when the demand (i) came from an − one supporting the effectiveness of the death penalty irrelevant source, (ii) was randomly determined, or and one non-supporting − to people opposed to the (iii) was randomly determined by the judges death penalty and people supporting it, participants themselves throwing dice (Englich et al., 2006). gave greater weight to the study that supported their Similarly, people who got a higher prediction of the own stance and more easily pointed out weak points increase in the earth’s temperature were more likely to in the other study. Even more alarming is that both believe that global warming is occurring and more sets of participants left feeling even more convinced of willing to pay to reduce it than those who received a their original view than before they read the studies lower prediction of temperature (Joireman et al., 2010). (Lord et al., 1979). Lovallo, a decision researcher, said There are two stages to anchoring - first selecting “Confirmation bias is probably the single biggest the anchor and then comparing the target value with problem in business, because even the most the anchor (Mussweiler et al., 2004). Product sophisticated people get it wrong” (in an interview developers could therefore base their judgement more quoted in Heath and Heath (2013, pg 12)). on initial financial costs than longer-term and During early concept development, particularly sustainability considerations because decision-makers when generating ideas, product developers form can be more heavily influenced by decision factors that hypotheses. Examples could include hypotheses about first come to mind (anchors) (Arvai et al., 2001). what attributes or which markets are best. Decision- When a decision-maker has an anchor, they then makers test their hypothesis by searching for new check the possibility that the target value is equal to information, interpreting existing information and the anchor value (the second stage). When they do remembering information. Confirmation bias occurs that check, they selectively retrieve knowledge from because the decision-makers unwittingly (i) are more memory that is consistent with the anchor. Then, likely to access memories that align with their when the decision-maker subsequently generates an hypothesis than memories that do not; (ii) weigh the estimate, they do so relying on easily accessible importance of supporting information more than the knowledge (Mussweiler et al., 2004). importance of information that does not fit well with The tendency to compare to an anchor occurs the hypothesis; and (iii) avoid sources that they know when there is little knowledgeSubmitted available (Mussweiler et could have information that would lead to rejecting al., 2004) and the magnitude of the anchoring effect the hypothesis (Oswald and Grosjean, 2004). Early in seems to be largest when faced with problems that the concept development phase and particularly when they have not thought much about in the past – starting to work with sustainable design, product emerging sustainability issues (Arvai etmanuscript al., 2012). developers have little knowledge (information) about Product developers are therefore prone to anchoring the problem space and there is therefore a potentially when integrating sustainability considerations into overwhelming huge range of information sources to be concept development, since they then have little searched. The lack of existing information on knowledge about the problem. For example, product sustainability and other consequences, as well as the developers may anchor their idea generation in expanse of potential sources to investigate, make it aspects other than sustainability. more likely that the product developers rely on Proposition 1.1.2: When generating ideas while memory, search for supporting information, and starting to integrate sustainability into concept development decision-making, little knowledge and ignore sources that they know to provide ill-fitting unfamiliarity of product developers with information, and therefore be prone to impeding a sustainability may lead to them basing judgements rejection of the existing hypothesis. on some (unrelated) value, that is, being hindered Proposition 1.1.3: When generating ideas while by anchoring. starting to integrate sustainability into concept development decision-making, the lack of 4.1.3. Confirmation Bias information and expanse of potential information sources may lead to product developers favouring Having a hypothesis (something you think or information that supports their existing believe to be true) and testing it is a part of everyday hypotheses, that is, being hindered by performance is likely to be much bigger than the other confirmation bias. combinations, since sustainability is fairly new. This means that, irrespective of the actual correlations (or 4.2. Cognitive illusions when selecting during lack of) between the variable pairs, product developers concept development may be prone to judging the combination of good functional performance and bad sustainability In this subsection, we describe the cognitive performance as more likely. illusions relevant when selecting between alternatives, Proposition 1.2.1: When selecting while starting to and associated implications when integrating integrate sustainability into concept development sustainability into product development. decision-making, uncertainty may lead to product developers seeing what they expect to see. In 4.2.1. Illusory Correlation addition, the high number of examples of products with a combination of poor sustainability People are great at detecting relationships performance and high functional performance may between variables, but sometimes this capability goes lead to product developers perceiving low- into overdrive such that decision-makers perceive sustainability-high-functional performance as the relationships that do not actually exist. These falsely most likely combination of these variables, that is, perceived relationships are known as illusory being hindered by illusory correlation. correlations (Fiedler, 2004). For example, Kao and Wasserman (1993) fed participants information about 4.2.2. Validity Effect the frequency with which an exotic plant flowered with and without fertiliser. When the rate of flowering People can perceive statements as truer just was equal for both with and without fertiliser, if the because the content is familiar to them, such as when number of plants tested with fertiliser was higher than they have heard a similar statement. This is known as tested without, participants thought that the fertiliser the validity effect and occurs both for fact and was successful. In other words, they perceived a opinion, both for statements that are generally correlation that was not there, just because the perceived as true or as false, and even for statements numbers in one column were higher than the numbers that are falsely perceived as true or false (Hackett in another column. Renner, 2004). The validity effect is strongest for the There are different types of illusory correlation, first repetition of information (Hackett Renner, 2004). such as illusions based on expectancies or sample size. However, decision-makers do not continue to increase These types differ in their underlying cognitive their perceived validity when, due to frequent process, which can operate simultaneously (Fiedler, repetition, they perceive an intent to persuade them 2004). Illusions based on expectancies are when you (Koch and Zerback, 2013). see what you expect to see, such as by following a It is when decision-makers are more familiar stereotype (Fiedler, 2004). Decision-makers are with information, such as when they believe that they especially susceptible to seeing what they expect when have heard it before) that they perceive it as truer they are making judgementsSubmitted under uncertainty (Boehm, 1994). Specifically, experts suffer worst from (Fiedler, 2004), a condition familiar in concept the validity effect in their area of expertise; they rate development, particularly for beginners in sustainable statements in their own area as being truer, even when product development. Product developers may the statements are factually false (Arkes et al., 1989). therefore see ‘sustainable’ concepts as moremanuscript expensive Due to the newness of integrating sustainability if they come into the decision-making believing that considerations in product development, product more sustainable concepts are more expensive. developers are likely less familiar with sustainability Another type of illusory correlation is related statements than many other product correlations based on sample size. When the sample development considerations and therefore prone to size of one pair of variables (such as flowering and viewing other statements as more valid than fertiliser used) is bigger than the other pairs of sustainability related statements. This would be variables, this can lead to decision-makers perceiving accentuated for statements related to their own area of the correlation between that variable pair even when expertise. it does not exist (Fiedler, 2004). This results in Proposition 1.2.2: When selecting while starting to minority groups having a potential disadvantage. integrate sustainability into concept development decision-making, product developers’ relative Sustainability attributes are in the minority, on the unfamiliarity with sustainability related market and in firms just starting to work with statements than with many other product ecodesign. This means that sustainable products may development considerations may lead to them be seen as less likely by product developers. For viewing other statements as more valid than example, for the variables ‘sustainability performance sustainability related statements, that is, being of product’ (good/bad) and ‘functional performance of hindered by validity effect. product’ (good/bad), the combination of good functional performance and bad sustainability

4.3. Cognitive illusions when gate reviewing overweighing easy-to-evaluate attributes. Similarly, relative unfamiliarity with sustainability In this subsection, we describe the cognitive can cause product developers to know with more illusions relevant when gate reviewing, as a special certainty about how they feel about factors other case of selecting decision-making, and associated than sustainability, as well as the complexity of implications when integrating sustainability into sustainability, can result in them experiencing product development. sustainability attributes as difficult-to-evaluate, which can lead to them overweighing other 4.3.1. Evaluability Bias attributes, that is, being hindered by evaluability bias. When judging an alternative in isolation (for example, not comparing multiple alternatives at a gate 4.2.2. Status Quo Bias and other reasons for sticking review), people tend to focus on attributes that are with the present easy-to-evaluate more than attributes that are difficult to evaluate. This is evaluability bias and it occurs The status quo bias surfaces when even when the difficult-to-evaluate attribute is more disproportional weight is assigned to present important to the decision at hand (Hsee, 1996). For conditions and when changes from this baseline are example, respondents in Hsee’s (1998) experiment perceived as losses. In order to side-step tackling tricky said, as you might expect, that they would pay more trade-offs or handling complex information, decision- for the larger amount of ice-cream when asked to makers often stick with the status quo when facing jointly evaluate how much they would pay for 8 ounces difficult sustainability-related decisions (Arvai et al., of ice-cream in a ten-ounce tub and for 7 ounces of ice- 2012). Resistance to new sustainability-related cream in a five-ounce tub. However, those who were business opportunities is a barrier to the asked to separately evaluate one tub (that is, without implementation of sustainable product development comparing) on average said that they would pay more (O’Rafferty and O’Connor, 2010). Explanations for why for the smaller amount of ice-cream (in the over-filled the bias occurs are found in loss aversion and tub). The relative fullness of the tub was easy-to- endowment effects. Although changes from a status evaluate and the absolute amount of ice-cream was quo typically can be gains as well as losses, losses are only evaluable via comparison. weighted more heavily than equivalent gains (Tversky How easy an attribute is to evaluate is also and Kahneman, 1981). A tendency to overemphasise related to how the decision-maker feels. If a decision- the avoidance of losses will thus favour retaining the maker knows exactly how they feel about a certain status quo, bringing about a status quo bias (Despeisse attribute, then that attribute likely carries more and Vladimirova, 2014). meaning for them and is therefore over-weighted in Endowment effect refers to that individuals decision-making (Slovic et al., 2007). When reviewing devote more value to things they possess even if it is a project at a gate, do product developers know how arbitrarily or recently acquired (Kahneman et al., they feel about each sustainability attribute? 1991). In a simple experiment, Kahneman et al. (1990) Difficult-to-evaluate attributesSubmitted are those that showed that individuals required twice as much require the decision-maker to undertake “effortful compensation for a trivial item (a mug) once their analysis” and have “contextual knowledge” (Arvai et ownership of the mug had been established than they al., 2012, pg 31) such as amount of child labour in the did when they were about to acquire the mug. Thus, value chain and associated importancemanuscript for the the endowment effect causes a lock-in, where not company. It is not only ice-cream quantity and letting go of current state of affairs is a stronger force amount of child labour that can be difficult to evaluate than changing them. When facing the challenge of without a comparative framework. No matter how integrating sustainability into concept development important or familiar, quantities of anything (even decision-making, gate reviewers may therefore favour number of lives saved or dollars) are sometimes not projects that they are involved in or concepts that are evaluable (Fetherstonhaugh et al., 1997; Slovic et al., similar to how things have been done in the past. 2007). In contrast, proportions and probabilities are Proposition 1.3.2: When selecting or gate reviewing often easy-to-evaluate and therefore over-weighted in while starting to integrate sustainability into decision-making, particularly when the decision concept development decision-making, the complexity of information and challenging trade- involves ambiguity about how much is good enough offs from including sustainability may lead to (Slovic et al., 2007), such as when considering newly product developers overweighing the loss of their introduced sustainability aspects. current concept, project or way of doing things Proposition 1.3.1: When selecting or gate reviewing compared to the gain of pursuing another concept, while starting to integrate sustainability into project or way of doing things, that is, being concept development decision-making, hindered by status quo bias. considering a single option (rather than comparing alternatives) and ambiguity about how good is good enough may lead to product developers 5. Mitigating for cognitive illusions when starting the support to vary according to activity being to include sustainability undertaken. Please note that the support should vary with type of activity, not phase. Within concept In line with Lovallo and Sibony (2010), we development, there are many iterations between idea selected practices and tools that are appropriate for generating and selecting activities, and it is these overcoming the identified illusions. In this section, we activities that need different support from one synthesise literature on decision-making and decision- another. making for sustainability to give advice on how to support product developers to mitigate for the 5.1.2. Pay attention to when to mitigate – perhaps in identified cognitive illusions. Even if we cannot say preceding step that every product developer will be thwarted by each of the illusions, we urge taking a precautionary In addition to needing different support in order approach in order to help product developers to to mitigate for different cognitive illusions, support achieve good decision-making when they are setting may need to encourage a mitigating action in the out on their journey to work with sustainable product activity that precedes the one where the illusion development. occurs. For example, to mitigate the effects of Providing support for product developers is confirmation bias (that might occur during selecting), important because (i) it is very challenging for product product developers need meaningful alternatives to developers themselves to be aware of these barriers select between. In other words, the support should and (ii) forewarned does not equate to forearmed. One encourage mitigating action (generating meaningful way in which our brain works, known as system 1, is alternatives) during the preceding idea generating automatic and we are unaware of its operations, and it step. Similarly, how information is selected influences can therefore send us off-track (Kahneman, 2003; the extent of anchoring and availability while Stanovich and West, 2000). Often the picture that generating. In the following subsections, we therefore system 1 builds for us is true, but cognitive illusions is describe the mitigating measures under the heading of a well-examined exception. Due to the nature of this the activity during which they need to be taken, rather system, we rarely catch ourselves in the act of making than the biased decision-making activity. errors, but rather simply accept the intuitive and effortless thinking. Even if we know which cognitive 5.2. Mitigating action to take when generating illusions we are facing, forewarned is not forearmed and decision-makers therefore need support for 5.2.1. Generate meaningful alternatives mitigating the cognitive illusions (Kahneman et al., Support should help product developers to seek 2011). For sustainability related decisions, using out alternatives that are meaningfully different from decision support techniques to mitigate for illusions each other. This can help reduce the effects of can lead to changes in preferences and even the confirmation bias (Kahneman et al., 2011) and is stabilising of preferences (which is a mark of good important even when one option/reasoning seems decision-making) (Hayashi et al., 2016). obvious (Oswald and Grosjean, 2004). Being able to It will vary how the followingSubmitted mitigating actions compare multiple alternatives, rather than trying to are integrated into sustainable product development judge a single option by itself, also helps prevent processes and practice (as indeed these processes and evaluability bias (Arvai et al., 2012). practices vary). Designers of support should therefore integrate these mitigating actions into themanuscript design of 5.2.2. Understand stakeholder and own perspectives the support. Support should help product developers to 5.1. Tailored support include different stakeholder perspectives in ideation in order to reduce availability (particularly when due 5.1.1. Vary type of support according to type of to a homogenous group) (Arvai et al., 2012). For decision-making activity example, through multi-functional teams (Antioco, M. et al., 2008), consumer co-creation, or supply chain The model in section 3 shows that product engagement. As well as searching out different developers may face different cognitive illusions for perspectives, support should encourage product different types of concept development activities. They developers to try to understand their own opinions therefore may need different types of support for and assumptions and then actively seek out integrating sustainability in the different types of information that contradicts, as well as information decision-making activity. For supporting decision- that reinforces, these assumptions. This is known as making for sustainability (in general), Arvai et al. using a diagnostic strategy and can help avoid (2012) mapped decision support techniques (from the confirmation bias (Oswald and Grosjean, 2004). literature) into two distinct groups depending on the type of decision being supported. We therefore reject one-size-fits-all solutions and highlight the need for

5.2.3. Employ passive decision-support techniques Therefore, selecting and gate reviewing decision- making incorporating sustainability, likely require When developing support for generating ideas, active decision-support. consider passive decision-support techniques. Passive Active decision-support breaks down a decision-support takes advantage of known biases in complicated decision into manageable parts and uses decision-making to make the ‘right’ decision easier particular techniques to mitigate for cognitive (Arvai et al., 2012). Examples include setting goals, illusions – techniques such as those listed in the ‘nudging’ (see (Thaler and Sunstein, 2008)), giving following subsections. This type of support explicitly feedback and designing defaults (Arvai et al., 2012; addresses consequences and trade-offs, and provides a Brones et al., 2014). It is therefore worth investigating structured process for including objectives that are not how to use such techniques during ideation, such that easily defined or quantified and for including product developers also generate sustainable ideas, subjective values (Arvai et al., 2012). Emphasizing such as for product concepts that are highly values and value-trade-offs helps decision-makers to sustainable. navigate the complexity of the decision-making If part of a large ideation process, generating a context and focus on the sustainability considerations single idea is a decision that is repeated frequently and that matter for the given project (Arvai et al., 2001; quickly, perhaps hundreds of times. Generating Keeney, 1992), with empirical evidence from risk activities are important and the quality of the best communication (Arvai et al., 2001). ideas will influence success on the market, however the generation of a single idea (amongst hundreds) 5.3.2. Check for credible alternatives requires a low degree of accuracy since it contributes a small part to the overall quantity and quality. The For hurdling barriers due to confirmation bias, objectives have already been selected (such as desired Kahneman et al. (2011) suggest checking for the bias by attributes selected following needfinding) for asking yourself “Are credible alternatives included generating ideas. When generating ideas, particularly along with the recommendation?” (pg. 56) and for when considering sustainability, the consequences of decision-makers to request additional alternatives, or each idea are not readily apparent. Decision-making at least check that meaningful alternatives have been with these characteristics should be supported with considered. Directly comparing alternatives allows for passive decision-support techniques (Arvai et al., better understanding of important principles of the 2012), otherwise known as choice architectures. problem, as well as generating designs in parallel Proposition 2.1: If you are designing support for leading to better outcomes (Dow et al., 2010). sustainable product development, help product Furthermore, in order to reduce the potential for developers who are generating ideas during spurious confirmation, it is important to consider concept development to mitigate for cognitive alternatives even when one option/reasoning seems illusions by employing passive decision- obvious (Oswald and Grosjean, 2004). support techniques and facilitating them to …generate alternatives that are meaningfully 5.3.3. Use ends-objectives different from each otherSubmitted in order to mitigate for evaluability bias and confirmation bias, …include Double-check the requirements that you will use stakeholder perspectives in order to reduce to select between ideas and even your assumptions of availability, and …use a diagnostic strategy to what matters. Check that you are using ends- mitigate for confirmation bias. manuscriptobjectives (as opposed to means-objectives), such as values-based requirements, in order to reduce 5.3. Mitigating action to take when selecting and anchoring (Arvai et al., 2012). This can be achieved by gate reviewing asking ‘why is that important?’ until you find a reason that is important in its own right (Gregory et al., 2012). 5.3.1. Emphasise values and value trade-offs Focusing on ends-objectives helps you to avoid anchoring in a particular solution and to explore Arvai et al.’s (2012) description of characteristics multiple ways to achieve the ends-objective (Arvai et of decisions requiring active decision-support al., 2012; Gregory et al., 2012). techniques fits well to selecting decision-making. These decisions are frequently multi-stakeholder (for 5.3.4. Develop a comparative framework example, customer, producer, society) and can involve competing objectives (for example, reduce short-term Support should assist gate reviewers to compare costs versus increase sustainability of production alternatives, rather than trying understand their techniques). Selecting product concepts and projects preference for a single option by itself, in order to can determine a company’s success on the market and mitigate for evaluability bias (Arvai et al., 2012). so is a high-stakes decision. Decisions that will impact Research has shown that when multiple options were production over a long time-frame are also full of developed, success rates increased from 56% to 70% uncertainty with regards to risk and sustainability. (Nutt, 1999) and boards made 6 times as many ‘very good’ decisions compared to evaluating a single option that are regarded as more certain. To encourage more (Gemünden and Hauschildt, 1985). Some may think equal familiarity, product developers could also be that comparing alternatives takes more time than educated in the sustainability related statements and evaluating a single option, but it has been shown that assumptions. executives who weigh more options actually make decisions faster since it helps them better understand 5.3.9. Search for evidence of the unexpected the situation, what is possible and the factors involved, As stated earlier, illusory correlation is more which in turn gives them enough confidence to make prominent for decision-making under uncertainty. It a quicker decision (Eisenhardt, 1989). is perhaps therefore not so surprising that given sufficient information, illusory correlation disappears. 5.3.5. Make sustainability easy-to-evaluate However, what may be more worrying for concept When summarising and presenting the development decision-making, is that experiments (comparative) information that will be reviewed, have shown that little information can lead to the product developers should include sustainability strongest occurrence of this illusion (Murphy et al., considerations such that they are understandable and 2011). Given that product developers cannot do much as easy-to-evaluate as other considerations. This will about the low-knowledge decision-making conditions help mitigate for evaluability bias (Arvai et al., 2012). in concept development, it might be worth supporting them to pit one cause factor against another in order 5.3.6. Prioritise searching for key knowledge to reduce illusory correlation. For example, by encouraging product developers to test the hypothesis When selecting between ideas under the of an unexpected event (such as highly sustainable conditions of little knowledge, it is important to strive products achieving high functional performance) pits to have key knowledge. To mitigate for availability, sample-size against expectancy effects (Fiedler, 2004). Kahneman et al. (2011) suggest asking: “If you had to make this decision again in a year’s time, what 5.3.10. Frame your results in multiple ways information would you want, and can you get more of it now?” (pg. 56). It can also help to have checklists In general, a good practice for mitigating bias in that detail what kind of data is required to be gathered environmental selection decisions is to use multiple for different types of decision (Kahneman et al., 2011). framings – positive gains and negative losses, factual and emotional – to guard against some aversions 5.3.7. Use a diagnostic strategy (Kühberger, 2017), to encourage using a breadth of types of thinking and to draw on both thinking of When striving to obtain the key knowledge, it is individuals and of the group (Gregory et al., 2012). This also critical to reflect on how you search for that is particularly helpful in mitigating for illusory information. Use multiple models to generate data correlation (Fiedler, 2004), but is also useful against (Kahneman et al., 2011). To help avoid confirmation framing effects, which can show up in a whole range of bias, product developers should actively search for decision situations (Kühberger, 2017). For example, information that contradicts,Submitted as well as information describing the same programme as a tax or as an offset that supports, the current hypothesis, known as using can influence support for the programme (Hardisty et a diagnostic strategy (Oswald and Grosjean, 2004). al., 2010). Framing is a very important topic and we recommend the following reading for designers of 5.3.8. Scrutinise where the info came frommanuscript support: The psychology of climate change Unfortunately, ‘forewarned is not forearmed’ is communication (Shome and Marx, 2009) and Framing particularly pertinent for anchoring. Even when (Kühberger, 2016). forewarned, judges were still biased (Mussweiler et al., 2004). However, considering why an anchor is 5.3.11. Get unstuck from the status quo inappropriate increases the availability of anchor in- For mitigating effects of aversion to loss, consistent knowledge and therefore reduces the effect Kahneman et al. (2011) suggest checking for the bias by of the anchor (Mussweiler et al., 2000). Kahneman et asking if the decision-makers are being overly al. (2011) suggest asking yourself: “Do you know where cautious. Gate reviewers can encourage exploration of the numbers came from? Can there be more risky options by stating that they will share the …unsubstantiated numbers? …extrapolation from risk or framing the acceptance of a certain amount of history? …a motivation to use a certain anchor?” risk of failure, thus helping individuals to explore what Asking these questions should help decision-makers they perceive as a large individual risk. look out for when estimates have been made based on Proposition 2.2: If you are designing support for adjusting from another number and where that sustainable product development, help product number came from. Since bias due to the validity effect developers who are selecting between ideas during is evoked by familiar information, it may be important concept development to mitigate for cognitive to scrutinise even those statements and assumptions illusions by facilitating them to …emphasise

values and value trade-offs in order to maintain implementing sustainable product development and focus on what matters, …check that the for the decision-making that takes place in concept alternatives are credible in order to mitigate for development. Confirmation bias occurs when there is confirmation bias, …use ends-objectives instead a lack of information and an expanse of sources; these of means-objectives in order to reduce anchoring, conditions probably occur often in product …compare alternatives rather than considering development in general (although likely even stronger each option in isolation in order to mitigate for for beginners starting to integrate sustainability). All evaluability bias, …make sustainability as easy- of the other illusions occur very much due to the to-evaluate as other aspects in order to mitigate introduction of sustainability considerations, for for evaluability bias, …prioritise searching for example, owing to the relative unfamiliarity and key knowledge in order to mitigate for availability, …use a diagnostic strategy to inexperience of product developers with mitigate for confirmation bias, …scrutinise where sustainability. However, if these conditions were information came from in order to mitigate for identified in another product development situation anchoring and validity bias, …search for evidence where the product developer is making the same type of the unexpected in order to mitigate for illusory of decision, then the listed cognitive illusions may also correlation, …frame analysis results in multiple pose a risk there. And, in that case, the same ways, and …ask themselves if they are being mitigating actions may be useful. overly cautious in order to mitigate for status quo bias. 6.2. Discussion of credibility Important to notice is that the above mitigating actions are process improvements. This is crucial The model of cognitive illusions and when they since, for achieving good decision-making, the quality may occur has been constructed from well-established of the decision-making process (including exploiting theories of how people make decisions. We cannot be analysis and reaching a decision) can be 6 times as certain that all the listed cognitive illusions will arise important as the quantity and detail of analysis for every product developer under the conditions performed, as well as a good process helps avoid poor identified. However, we are confident that this an analysis (Lovallo and Sibony, 2010). initial model of which cognitive illusions should be considered when developing support for including 6. Discussion and conclusions sustainability. The above suggestions for how to mitigate are Our contribution to theory is the developed based on empirical testing in a variety of contexts, but model about what cognitive illusions may hinder not as broad testing as for the cognitive illusions. product developers when they start integrating Given the strength of the evidence in the literature, we sustainability into their concept development are confident that (a) the focus should be on the decision-making. This model is communicated in the quality of the decision-making process (including form of the propositions in section 3 and summarised exploiting analysis and reaching a decision), more so in Figure 3. Our contribution to practice is a synthesis than the quantity and detail of the analysis, and (b) the of what the literature advisesSubmitted that support developers support for introducing sustainability into concept can do to help product developers to mitigate for the development should vary according to the type of illusions. This synthesis is communicated in the form activity undertaken. of the propositions in section 5 and summarised in Figure 4. Our suggestions address a gapmanuscript in help for 6.3. Comparison with other studies those developing support for sustainable product development. We did not find any studies that addressed our research question. There is, of course, the literature on 6.1. Models of cognitive illusions and mitigating decision-making for sustainability by Arvai et al. (2012), which we used when building our model. actions However, we found no model for which cognitive We found a number of relevant cognitive illusions may hinder product developers when starting illusions and therefore product developers may need to include sustainability in concept development, or help to mitigate for these cognitive illusions (shown in how to mitigate for these illusions. Figure 3) when they are starting to integrate sustainability in concept development. Furthermore, developers of support for sustainable product development could enable product develops to mitigate for those illusions by taking the actions shown in Figure 4. The decision-making conditions (left-hand side of Figure 3) were identified for the specific situation where individual product developers are beginners at

Figure 3 - Cognitive illusions for the three types of concept development decision-making with the conditions that lead to them. Submitted manuscript

Figure 4 - Mitigating for cognitive illusions when integrating sustainability into concept development.

As mentioned in our introduction, Liedtka (2014) While working on this study, some other has published a perspective piece on design thinking questions arose. For example, how does the extent to and reducing cognitive bias. It is difficult to compare which a product development process already includes the studies since the perspective piece does not mitigating actions effect the smoothness of the include a description of how the biases were selected. transition to sustainable product development? She identified 9 cognitive illusions (biases) related to product development, 2 of which are the same as ours. 6.6. Conclusions It would be interesting to know if Liedtka (2014) When developing support for product identified the biases by considering which ones would developers to include sustainability in concept likely be mitigated through design-thinking. This development decision-making, consider cognitive would give insight into whether integrating illusions. The illusions that product developers are sustainability into design-thinking is an especially vulnerable to depends on whether the specific promising approach to sustainable design. decision-making task is a generating, selecting or gate- With regards to comparing to studies on reviewing task. Mitigating actions should also vary mitigating bias, prescriptions for how to do product according to task type. These activities focus on the development aim to support good decision-making, quality of the decision-making process, more so than even if they do not actively consider cognitive on the quantity of analysis. Our model for the specific illusions. Since there is little literature on how cognitive illusions and the suggestions for mitigating prescribed product development processes mitigate actions provide a starting point for researchers and for cognitive illusions, they may have evolved through support developers. trial and error. Given the new challenge of integrating sustainability along with the time pressure to get it Acknowledgments: We gratefully acknowledge done, society does not have the luxury of slow financial support for the Model Driven Development evolution. A model for how to achieve good decision- and Decision-Support project from The Knowledge making when integrating sustainability is therefore Foundation (Stiftelsen för kunskaps- och needed, and as soon as possible. kompetensutveckling), Blekinge Institute of Technology, and the other project partners. 6.4. Discussion of implications Through developing theory about potentially References relevant cognitive illusions and then identifying some ways to mitigate for them, we hope to help those Antioco, M., Moenaert, R.K., Lindgreen, A., 2008. developing support for product developers starting to Reducing ongoing product design decision- include sustainability. In particular, those developing making bias. Journal of Product Innovation support can have an increased awareness of the cognitive illusions that product developers face and Management 25(6), 528-545. some ways in which to mitigate. With the cognitive illusions mitigated, product Submitteddevelopers will have a better chance of reducing errors in concept Antioco, M., Moenaert, R.K., Lindgreen, A., 2008. development decision-making. This is important since Reducing Ongoing Product Design Decision- decisions in the early phases of the process determine, Making Bias*. Journal of Product Innovation to a large extent, the success of a developmentmanuscript project (Martinsuo and Poskela, 2011), development costs Management 25(6), 528-545. (Ulrich and Pearson, 1993) and sustainability impacts of the product’s lifecycle (Bhamra et al., 1999). The end result is hopefully therefore competitive products that Arkes, H.R., Hackett, C., Boehm, L., 1989. The are more sustainable than today’s products. generality of the relation between familiarity and judged validity. Journal of Behavioral Decision 6.5. Recommendations for research Making 2(2), 81-94. Our model is in need of empirical testing and evolution. The suggestions that we make on how to mitigate need to be tested in a variety of contexts and Arvai, J., Campbell-Arvai, V., Steel, P., 2012. Decision- decision-making evaluated. Both short-term and long- making for sustainability: A systematic review of term tests are probably required. Processes related to cognitive illusions may the body of knowledge. Network for Business amplify each other. This is an emerging field of Sustainability, Ontario. research that could be used to further develop the work presented in this article.

Arvai, J.L., Gregory, R., McDaniels, T.L., 2001. Testing Boks, C., Wever, R., Design for Sustainability in the a Structured Decision Approach: Value-Focused Fuzzy Front End. The Centre for Sustainable Thinking for Deliberative Risk Communication. Design. Risk Analysis 21(6), 1065-1076.

Broman, G.I., Robert, K.H., 2017. A framework for Bakhti, R., 2018. Religious versus reflective priming strategic sustainable development. Journal of and susceptibility to the conjunction fallacy. Cleaner Production 140, 17-31. Applied Cognitive Psychology 32(2), 186-191.

Brones, F., de Carvalho, M.M., 2015. From 50 to 1: Baumann, H., Boons, F., Bragd, A., 2002. Mapping the integrating literature toward a systemic ecodesign green product development field: engineering, model. Journal of Cleaner Production 96, 44-57. policy and business perspectives. Journal of

Cleaner Production 10(5), 409-425. Brones, F., Gyldendal Melberg, M., Monteiro de Carvalho, M., Pigosso, D.C., Antelmi, McAloone, Bazerman, M.H., Moore, D.A., 2012. Judgment in T.C., 2014. Driving behavioural change towards Managerial Decision Making, 8th ed. JW Wiley, ecodesign integration: Nudging experiment in New York. industry, Sustainable Innovation. pp. 15-23.

Bell, D.E., Raiffa, H., Tversky, A., 1988. Decision Brones, F.A., de Carvalho, M.M., Zancul, E.D., 2017. Making: Descriptive, Normative and Prescriptive Reviews, action and learning on change Interactions. Cambridge University Press, New management for ecodesign transition. Journal of York. Cleaner Production 142, 8-22.

Bhamra, T.A., Evans, S., McAloone, T.C., Simon, M., Brotherton, R., French, C.C., 2014. Belief in Conspiracy Poole, S., Sweatman, A., 1999. Integrating Theories and Susceptibility to the Conjunction environmental decisions into the product Fallacy. Applied Cognitive Psychology 28(2), 238- development process. SubmittedI. The early stages, 248. Proceedings First International Symposium on

Environmentally Conscious Design and Inverse Caputo, A., 2013. A literature review of cognitive biases Manufacturing. IEEE. manuscript in negotiation processes. International Journal of Conflict Management 24(4), 374-398. Boehm, L.E., 1994. The Validity Effect: A Search for

Mediating Variables. Personality and Social Carbonell-Foulquié, P., Munuera-Alemán, J.L., Psychology Bulletin 20(3), 285-293. Rodrı́guez-Escudero, A.I., 2004. Criteria employed for go/no-go decisions when developing successful Boks, C., McAloone, T.C., 2009. Transitions in highly innovative products. Industrial Marketing Sustainable Product Design Research. Management 33(4), 307-316. International Journal of Product Development 9(4),

429-449. Chin, K.-S., 1999. Integrated product concepts development and evaluation. International Journal of Computer Integrated Manufacturing 12(2), 179- 190. Chong, D.S.F., Eerde, W.V., Chai, K.H., Rutte, C.G., leads to better design results, more divergence, 2011. A Double-Edged Sword: The Effects of and increased self-efficacy. ACM Transactions on Challenge and Hindrance Time Pressure on New Computer-Human Interaction 17(4), 1-24. Product Development Teams. IEEE Transactions

on Engineering Management 58(1), 71-86. Eisenhardt, K.M., 1989. MAKING FAST STRATEGIC DECISIONS IN HIGH-VELOCITY Christiansen, J.K., Varnes, C., 2008. From models to ENVIRONMENTS. Academy of Management practice: decision making at portfolio meetings. Journal 32(3), 543-576. International Journal of Quality & Reliability

Management 25(1), 87-101. Englich, B., Mussweiler, T., Strack, F., 2006. Playing Dice With Criminal Sentences: The Influence of Cooper, R.G., 2008. Perspective: The Stage-Gate®Idea- Irrelevant Anchors on Experts’ Judicial Decision to-Launch Process—Update, What's New, and Making. Personality and Social Psychology NexGen Systems. Journal of Product Innovation Bulletin 32(2), 188-200. Management 25(3), 213-232.

Evans, J.S.B.T., 2007. Hypothetical Thinking: Dual Dangelico, R.M., Pujari, D., 2010. Mainstreaming Processes in Reasoning and Judgement. Green Product Innovation: Why and How Psychology Press, UK, USA & Canada. Companies Integrate Environmental

Sustainability. Journal of Business Ethics 95(3), Fetherstonhaugh, D., Slovic, P., Johnson, S., Friedrich, 471-486. J., 1997. Journal of Risk and Uncertainty 14(3), 283- 300. De Treville, S., Hoffrage, U., Petty, J.S., 2009.

Managerial decision making and lead times: the Fiedler, K., 2004. Illusory correlation, in: Pohl, R.F. impact of cognitive illusions, in: Reiner , G. (Ed.) (Ed.) Cognitive illusions: A handbook on fallacies Rapid Modelling for Increasing Competitiveness. and biases in thinking, judgment and memory. Springer, London, pp. 3-14.Submitted Psychology Press, New York, pp. 97-114.

DeDonno, M.A., Demaree, H.A., Self-Efficacy: An Gemünden, H.G., Hauschildt, J., 1985. Number of Important Component in Decision Makingmanuscript under alternatives and efficiency in different types of top- Time Pressure. American Psychological management decisions. European Journal of Association (APA), Operational Research 22(2), 178-190. http://dx.doi.org/10.1037/e722352011-022.

Gifford, R., 2011. The dragons of inaction: Despeisse, M., Vladimirova, D., 2014. Decision Making Psychological barriers that limit climate change for Sustainability: Review and Research Agenda, mitigation and adaptation. American Psychologist IFIP Advances in Information and Communication 66(4), 290-302. Technology. Springer Berlin Heidelberg, pp. 146- 153. Goffin, K., 2012. Sustainability and new product development. Cranfield on Corporate Dow, S.P., Glassco, A., Kass, J., Schwarz, M., Schwartz, Sustainability, Sustainability and new product D.L., Klemmer, S.R., 2010. Parallel prototyping development, pp. 105-118.

Gould, R.K., Bratt, C., Svensson, M., Broman, G.I., n.d. Hsee, C.K., 1998. Less is better: when low‐value Shrinking and scaffolding: supporting individual options are valued more highly than high‐value behaviour change towards implementing options. Journal of Behavioral Decision Making sustainable product design. 11(2), 107-121.

Gregory, R., Failing, L., Harstone, M., Long, G., Jaccard, J., Jacoby, J., 2010. Theory construction and McDaniels, T., Ohlson, D., 2012. Structured model-building skills: a practical guide for social decision making: a practical guide to scientists. Guilford Press, New York. environmental management choices, 1. Aufl.;1; ed.

Wiley-Blackwell, Chichester, West Johansson, C., 2014. MANAGING UNCERTAINTY Sussex;Hoboken, N.J;. AND AMBIGUITY IN GATES: DECISION MAKING IN AEROSPACE PRODUCT DEVELOPMENT. Hackett Renner, C., 2004. Validity Effect, in: Pohl, R.F. International Journal of Innovation and (Ed.) Cognitive illusions: A handbook on fallacies Technology Management 11(02), 1450012. and biases in thinking, judgment and memory.

Psychology Press, New York, pp. 201-213. Johansson, C., Parida, V., Larsson, A., 2009. How are knowledge and information evaluated?: decision Hardisty, D.J., Johnson, E.J., Weber, E.U., 2010. A Dirty making in stage-gate processes. p. 195. Word or a Dirty World? Attribute Framing,

Political Affiliation, and Query Theory. Joireman, J., Barnes Truelove, H., Duell, B., 2010. Effect Psychological Science 21(1), 86-92. of outdoor temperature, heat primes and anchoring on belief in global warming. Journal of Hassi, L., Peck, D., Wever, R., DeWulf, K., 2009. Environmental Psychology 30(4), 358-367. Sustainable Innovation - organisation and goal

finding Joint Actions on Climate Change. Kaebernick, H., Kara, S., Sun, M., 2003. Sustainable product development and manufacturing by Hayashi, K., Hondo, H., SubmittedMoriizumi, Y., 2016. considering environmental requirements. Preference Construction Processes for Renewable Robotics and Computer-Integrated Manufacturing Energies: Assessing the Influence of Sustainability 19(6), 461-468. Information and Decision Supportmanuscript Methods.

Sustainability 8(11), 14. Kahneman, D., 2003. A perspective on judgment and choice: Mapping bounded rationality. American Heath, D., Heath, C., 2013. Decisive: how to make Psychologist 58(9), 697-720. better decisions in life and work. Cornerstone

Digital. Kahneman, D., Knetsch, J.L., Thaler, R.H., 1990. Experimental Tests of the Endowment Effect and Hsee, C.K., 1996. The Evaluability Hypothesis: An the Coase Theorem. Journal of Political Economy Explanation for Preference Reversals between Joint 98(6), 1325-1348. and Separate Evaluations of Alternatives.

Organizational Behavior and Human Decision Kahneman, D., Knetsch, J.L., Thaler, R.H., 1991. Processes 67(3), 247-257. Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias. Journal of Economic Kunreuther, H., Gupta, S., Bosetti, V., Cooke, R., Dutt, Perspectives 5(1), 193-206. V., Ha-Duong, M., Held, H., Llanes-Regueiro, J., Patt, A., Shittu, E., Weber, E., 2014. Integrated Risk

Kahneman, D., Lovallo, D., Sibony, O., 2011. Before you and Uncertainty Assessment of Climate Change make that big decision. Harvard Business School, Response Policies, in: Edenhofer, O., Pichs- Watertown, pp. 50-137. Madruga, R., Sokona, Y., Farahani, E., Kadner, S., Seyboth, K., Adler, A., Baum, I., Brunner, S., Eickemeier, P., Kriemann, B. (Eds.), Contribution Kao, S.-F., Wasserman, E.A., 1993. Assessment of an of Working Group III to the Fifth Assessment information integration account of contingency Report of the Intergovernmental Panel on Climate judgment with examination of subjective cell Change. Cambridge University Press, UK & USA. importance and method of information presentation. Journal of Experimental Psychology: Learning, Memory, and Cognition 19(6), 1363-1386. Kühberger, A., 2016. Framing, in: Pohl, R.F. (Ed.) Cognitive illusions: Intriguing phenomena in judgement, thinking and memory. Psychology Keeney, R.L., 1992. Value-focused thinking: a path to Press, London, pp. 79-98. creative decisionmaking. Harvard University Press, United States. Kühberger, A., 2017. Framing, in: Pohl, R.F. (Ed.) Cognitive illusions: Intriguing phenomena in Kihlander, I., Ritzén, S., 2012. Compatibility before judgement, thinking and memory. Psychology completeness—Identifying intrinsic conflicts in Press, London, pp. 79-98. concept decision making for technical systems. Technovation 32(2), 79-89. Lepine, J.A., Podsakoff, N.P., Lepine, M.A., 2005. A META-ANALYTIC TEST OF THE CHALLENGE Kihlander, I., Ritzén, S., Maskinkonstruktion, Skolan STRESSOR--HINDRANCE STRESSOR för industriell teknik och, m., Integrerad, p., Kth, FRAMEWORK: AN EXPLANATION FOR Deficiencies in Management of the Concept INCONSISTENT RELATIONSHIPS AMONG Development Process: TheorySubmitted and Practice. p. 267. STRESSORS AND PERFORMANCE. Academy of Management Journal 48(5), 764-775. Koch, T., Zerback, T., 2013. Helpful or Harmful?manuscript How Frequent Repetition Affects Perceived Statement Liedtka, J., 2014. Perspective: Linking Design Thinking Credibility. Journal of Communication 63(6), 993- with Innovation Outcomes through Cognitive Bias 1010. Reduction. Journal of Product Innovation Management 32(6), 925-938. Krishnan, V., Ulrich, K.T., 2001. Product development decisions: A review of the literature. Management Lord, C.G., Ross, L., Lepper, M.R., 1979. Biased Science 47(1), 1-21. assimilation and attitude polarization: The effects of prior theories on subsequently considered Krueger, J.I., Funder, D.C., 2004. Towards a balanced evidence. Journal of Personality and Social social psychology: Causes, consequences, and Psychology 37(11), 2098-2109. cures for the problem-seeking approach to social behavior and cognition. Behavioral and Brain Sciences 27(3), 313-+.

Lovallo, D., Sibony, O., 2010. The case for behavioral Selective Accessibility. Personality and Social strategy. McKinsey Quarterly(2), 30-40. Psychology Bulletin 26(9), 1142-1150.

Martinsuo, M., Poskela, J., 2011. Use of Evaluation Nicholls, N., 1999. Cognitive illusions, heuristics, and Criteria and Innovation Performance in the Front climate prediction. Bulletin of the American End of Innovation*. Journal of Product Innovation Meteorological Society 80(7), 1385-1397. Management 28(6), 896-914.

Nutt, P.C., 1999. Surprising but true: Half the decisions Matheson, D., 1998. The smart organization : creating in organizations fail. Academy of Management value through strategic R&D, United States. Perspectives 13(4), 75-90.

Miller, T.R., 2013. Constructing sustainability science: Oswald, M.E., Grosjean, S., 2004. Confirmation bias, emerging perspectives and research trajectories. in: Pohl, R.F. (Ed.) Cognitive illusions: A handbook Sustainability Science 8(2), 279-293. on fallacies and biases in thinking, judgment and memory. Psychology Press, New York, pp. 79-96.

Miller, T.R., Wiek, A., Sarewitz, D., Robinson, J., Olsson, L., Kriebel, D., Loorbach, D., 2014. The O’Rafferty, S., O’Connor, F., 2010. Regional future of sustainability science: a solutions- Perspectives on Capacity Building for Ecodesign – oriented research agenda. Sustainability Science Insights from Wales, in: Sarkis, J., Cordeiro, J.J., 9(2), 239-246. Vazquez Brust, D. (Eds.), Facilitating Sustainable Innovation through Collaboration: A Multi-

Montibeller, G., von Winterfeldt, D., 2015. Cognitive Stakeholder Perspective. Springer Netherlands, and Motivational Biases in Decision and Risk Dordrecht, pp. 159-183. Analysis. Risk Analysis 35(7), 1230-1251. Pohl, R.F., 2004. Cognitive Illusions: a handbook on

Murphy, R.A., Schmeer, S., Vallée-Tourangeau, F., fallacies and biases in thinking, judgement and Submittedmemory, 2012 ed. Psychology Press, USA & Canada, Mondragón, E., Hilton, D., 2011. Making the illusory correlation effect appear and then p. 438. disappear: The effects of increasedmanuscript learning. Quarterly Journal of Experimental Psychology Pohl, R.F., 2017. Cognitive Illusions: Intriguing 64(1), 24-40. Phenomena in Judgement, Thinking and Memory, 2nd ed. Psychology Press (Routledge), London &

Mussweiler, T., Englich, B., Strack, F., 2004. Anchoring New York. effect, in: F., P.R. (Ed.) Cognitive illusions: A handbook on fallacies and biases in thinking, Post, J.E., Altman, B.W., 1994. Managing the judgment and memory. Psychology Press, New environmental change process: barriers and York, pp. 183-200. opportunities. Journal of Organizational Change Management 7(4), 64-81.

Mussweiler, T., Strack, F., Pfeiffer, T., 2000. Overcoming the Inevitable Anchoring Effect: Pujari, D., 2006. Eco-innovation and new product Considering the Opposite Compensates for development: understanding the influences on market performance. Technovation 26(1), 76-85. Pujari, D., Wright, G., Peattie, K., 2003. Green and Simon, H.A., 1957. Models of man: social and rational; competitive. Journal of Business Research 56(8), mathematical essays on rational human behavior 657-671. in a social setting. Wiley, New York.

Reber, R., 2004. Availability, in: Pohl, R.F. (Ed.) Slovic, P., 1995. THE CONSTRUCTION OF Cognitive illusions: A handbook on fallacies and PREFERENCE. American Psychologist 50(5), 364- biases in thinking, judgment and memory. 371. Psychology Press, New York, pp. 147-163.

Slovic, P., Finucane, M.L., Peters, E., MacGregor, D.G., Reid, S.E., de Brentani, U., 2004. The Fuzzy Front End 2007. The affect heuristic. European Journal of of New Product Development for Discontinuous Operational Research 177(3), 1333-1352. Innovations: A Theoretical Model. Journal of

Product Innovation Management 21(3), 170-184. Spence, A., Poortinga, W., Pidgeon, N., 2012. The Psychological Distance of Climate Change. Risk Saposnik, G., Redelmeier, D., Ruff, C.C., Tobler, P.N., Analysis 32(6), 957-972. 2016. Cognitive biases associated with medical

decisions: a systematic review. Bmc Medical Stacey, M., Eckert, C., 2003. Against Ambiguity. Informatics and Decision Making 16, 14. Computer Supported Cooperative Work (CSCW) 12(2), 153-183. Schwarz, N., Bless, H., Strack, F., Klumpp, G., et al.,

1991. Ease of retrieval as information: Another look Stanovich, K.E., 2009. Distinguishing the reflective, at the availability heuristic. Journal of Personality algorithmic, and autonomous minds: Is it time for and Social Psychology 61(2), 195-202. a tri-process theory?, in: Evans, J.S.B.T.F., Keith (Ed.) In two minds: Dual processes and beyond . . Schäpke, N., Omann, I., Wittmayer, J., MariaMock, M., Oxford University Press, Oxford. von Raggamby, A., 2013. Supporting sustainability transitions by enhancing theSubmitted human dimension via Stanovich, K.E., West, R.F., 2000. Individual empowerment, social learning and social capital . differences in reasoning: Implications for the SCORAI Europe - Sustainable Consumption rationality debate? Behavioral and Brain Sciences Transitions Series. Rotterdam, Niederlande,manuscript pp. 23(5), 645-665. 277-293.

Stoknes, P.E., 2014. Rethinking climate Shaw, M.E., 1982. Theories of social psychology. communications and the “psychological climate McGraw-Hill, New York. paradox”. Energy Research & Social Science 1, 161- 170. Shome, D., Marx, S., 2009. The psychology of climate

change communication: A guide for scientists, Sunstein, C.R., 2006. The Availability Heuristic, journalists, educators, political aides, and the Intuitive Cost-Benefit Analysis, and Climate interested public. Center for research on Change. Climatic Change 77(1-2), 195-210. environmental decisions, New York.

Thaler, R.H., Sunstein, C.R., 2008. Nudge: improving Yarritu, I., Matute, H., Luque, D., 2015. The dark side decisions about health, wealth, and happiness. of cognitive illusions: When an illusory belief Yale University Press, New Haven. interferes with the acquisition of evidence‐based knowledge. British Journal of Psychology 106(4),

Tukker, A., Emmert, S., Charter, M., Vezzoli, C., Sto, 597-608. E., Andersen, M.M., Geerken, T., Tischner, U., Lahlou, S., 2008. Fostering change to sustainable consumption and production: an evidence based view. Journal of Cleaner Production 16(11), 1218- 1225.

Tukker, A., Jansen, B., 2006. Environment impacts of products - A detailed review of studies. Journal of Industrial Ecology 10(3), 159-182.

Tversky, A., Kahneman, D., 1973. Availability: A heuristic for judging frequency and probability. Cognitive Psychology 5(2), 207-232.

Tversky, A., Kahneman, D., 1974. Judgment under Uncertainty: Heuristics and Biases. Science 185(4157), 1124-1131.

Tversky, A., Kahneman, D., 1981. The framing of decisions and the psychology of choice. Science 211(4481), 453-458. Submitted

Ullman, D.G., 2003. The mechanical design process, 3. ed. McGraw-Hill, New York. manuscript

Ulrich, K.T., Pearson, S.A., 1993. Does product design really determine 80% of manufacturing cost? Sloan School of Management.

Weber, E.U., 2006. Experience-Based and Description- Based Perceptions of Long-Term Risk: Why Global Warming does not Scare us (Yet). Climatic Change 77(1-2), 103-120.

Weber, E.U., 2015. Climate Change Demands Behavioral Change: What Are the Challenges? SOCIAL RESEARCH 82(3), 561-580.