Part I
Foundations of trust
The foundations part includes four chapters which examine six key topics that are important for the field. In the first of our foundation chapters, Daan van Knippenberg examines affect-based trust, a topic which is neglected in many studies, leading to it having an underestimated role in trust. He makes the case for it being distinct from cognition-based trust, considering its unique antecedents and consequences. Important here, and why we deliberately position this as our first foundation chapter, is the contention that affect-based trust has a stronger influence on trust than cognitive-based trust. This chapter is concerned not merely with emotions but with relationships.
Then in Chapter 2 M. Audrey Korsgaard reviews the interpersonal concept of reciprocal trust, a complex, bidirectional phenomenon which involves a self-reinforcing dynamic process. Korsgaard develops further the importance of relationships, which van Knippenberg puts central as a foundation, to consider three key facets of reciprocal trust: trust spirals and their directions, the dynamics of relationships and the shapes of the trust trajectory.
This is followed by Kirsimarja Blomqvist and Karen S. Cook who focus on swift trust, updating a concept which Meyerson, Weick and Kramer (1996) introduced. In their chapter, as well as testing some of the founding propositions, they look at new organizational phenomena and specifically at the role of technology and virtual interactions to explore knowledge-creation activities, in which trust might be required to form rapidly and has to deliver immediate benefits.
Sim B. Sitkin and Katinka M. Bijlsma-Frankema in Chapter 4 take us to the debate of trust versus distrust. Their chapter considers two distinct ways that distrust has been approached: as a subset of trust and in its own right as a distinct phenomenon. They review the evidence to identify further directions that arise from these different perspectives for the field.
1
Reconsidering Affect-Based Trust
A new research agenda
Daan van Knippenberg
It is hard to envision social relationships that would not require some level of trust between parties to function effectively. Regardless of whether these are relationships between individuals, group members, organizations or linkages across those levels (e.g. individualâemploying organization), regardless of whether these are work or non-work relationship, regardless of whether these are short-term or longer-term relationships, arguably all relationships build on trust between parties to be effective (cf. Fulmer & Gelfand, 2012). Not surprisingly then, trust is a concept studied widely in the social and behavioural sciences (Rousseau, Sitkin, Burt & Camerer, 1998). The jumping off point for the current discussion of trust is that in this impressive body of work there is such an emphasis on trust as a cognitive phenomenon that the role of affective processes in trust is under-investigated and, I would contend, underestimated. The idea that one would trust, or distrust, someone because one has âa good feelingâ about the person, or a âbad feelingâ about the person seems completely natural; yet, this idea is essentially absent from the empirical work on trust.
Following the widely used definition by Rousseau et al. (1998) I understand trust to be the willingness to accept vulnerability based upon positive expectations about anotherâs behaviour (key here is that vulnerability entails risk, and trust thus captures the willingness to expose oneself to that risk). Expectations are cognitions, and it is therefore probably not surprising that trust research has been strongly dominated by a cognitive approach to trust in which trust is understood as the outcome of a subjectively rational cognitive process. This is illustrated in influential conceptual treatments of trust (e.g. Mayer, Davis & Schoorman, 1995) and in the observation that most empirical work tends to focus on cognition-based trust (e.g. Dirks & Ferrin, 2002). Even so, affect â moods and emotions â may play a role in trust. In trust research, this is captured first and foremost by the concept of affect-based trust, trust based on the personal bond between two people (McAllister, 1995; Webber, 2008). Despite the fact that affect-based trust is in a sense a âminority perspectiveâ in trust research, there is enough work on affect-based trust to take stock of its contribution to the trust literature. The aim of this chapter is to do exactly that, and to review what we know from empirical research on affect-based trust to assess the case for a unique role of affect within the greater body of trust research â is there evidence of unique antecedents? Of unique consequences? The review aims to not merely âsum upâ the evidence, but to arrive at both integrative conclusions and identify a way forward for research in affect-based trust â a research agenda.
Presaging things to come, the first-blush conclusion would be that affect-based trust and cognition-based trust are moderately to strongly correlated, but do represent different constructs with unique antecedents and consequences. Moreover, there is evidence that affect-based trust tends to be the stronger influence and that cognition-based trust can be seen as a precursor to affect-based trust. On closer consideration, however, one may spot the elephant in the room: affect-based trust does not so much invoke affect as it invokes relationship; it is relationship-based trust much more than affect-based trust. From its definition as trust based on the personal bond between two people, onwards it is arguably better labelled relationship-based trust than affect-based trust in that the relationship is definitional but affect is not. In that sense the label affect-based trust is something of a misnomer, and the primary driver of the difference between affect-based and cognition-based trust may not be so much the contrast between affect and cog nition but the extent to which the trust is relationship-based or reflects an assessment of the other partyâs trustworthiness that is relatively independent of the relationship between parties. Recognizing this important alternative understanding of the affect-basedâcognition-based trust distinction, I propose that research in affect-based trust does not, or at the very least not un equivocally, address the role of affect in trust. The key conclusion I would advance, therefore, concerns the need for research on the one hand to clarify the relationship basis versus affective basis of affect-based trust, and on the other hand to unambiguously study the role of affect in trust.
Affect-based trust: state of the science
As outlined by Lewis and Weigart (1985) and Mayer et al. (1995), our assessment of the extent to which some other party (person, group, organization) can be trusted has a strong component of subjective rationality. Trust follows from what one sees as good reasons â indicators of a partyâs trustworthiness. Schoorman et al. (1995) for instance argue that the key indicators of trustworthiness concern the other partyâs competence, integrity and benevolence. As noted by Lewis and Weigart (1985) and McAllister (1995), however, the quality of the relationship between parties and its emotional connotations can also be a basis for trust, and this has led to the rise of the concept of affect-based trust (as I will argue below, a misnomer for what conceptually and empirically is relationship-based trust). In research in management, McAllister (1995) captured this probably in the most influential way by explicitly distinguishing and operationalizing cognition-based trust and affect-based trust. In this understanding of trust, the key antecedent of affect-based trust is the quality of the relationships between parties. Indeed, operationally McAllisterâs measure of affect-based trust reflects judgment of relationship quality as much as the willingness to be vulnerable in the relationship (i.e. only the latter matches the more common definition of trust) and does not have the notion of affect as a basis for trust as a main theme underlying the items. Cognition-based trust, in contrast, would be based primarily on more factual indicators of a partyâs competence, integrity and benevolence (i.e. feeding into expectations of what the party is able and motivated to do; cf. Mayer et al., 1995).
Affect is clearly implied by the quality of a relationship, but I argue that this is a limited and limiting view of the role affect may play in trust, and moreover one that does not unambiguously speak to the role of affect but rather puts the quality of the relationship and not affect centre-stage. Before we get to that argument, however, I first provide a concise review of research on affect-based trust. The review aims to be representative, not exhaustive. To reflect my misgivings about the construct naming, I will refer to the construct as âaffect-based trustâ, that is, within quotation marks to indicate I am quoting the label used but not sharing the conclusion that the label is correct.
âAffect-based trustâ: empirical evidence
âAffect-based trustâ is conceptualized as rooted in the quality of the relationship between parties. By implication, cognition-based trust would not be, or to a lesser degree, rooted in this relationship. When cognition-based trust is understood as rooted in assessment of the other partyâs competence, integrity and benevolence (Mayer et al., 1995), it seems reasonable to conclude that these are assessments that need not be based on the quality of the relationship with the other party (even when a relationship might give one access to relevant information pertaining to competence and integrity). For instance, we may assess someoneâs competence on the basis of their CV, or judge a personâs integrity based on information available through others without any relationship with that person. In that sense, cognition-based trust could be said to be person-based â in contrast to the relationship-based nature of the concept known as âaffect-based trustâ. Cognition-based trust can thus be a precursor to âaffect-based trustâ to the extent that cognitive judgments of the other partyâs trustworthiness may set the stage for the development of a high-quality relationship or for the absence of such a development.
Consistent with this logic, McAllister (1995) found support for a model in which relationship quality was a unique antecedent of âaffect-based trustâ in peers, and cognition-based trust too was an antecedent of âaffect-based trustâ (and âaffect-based trustâ predicted performance). Other studies also report such evidence of cognition-based trust as an antecedent of âaffect-based trustâ. Schaubroeck, Peng and Hannah (2013) for instance found that cognition-based trust in leader and peers predicted âaffect-based trustâ, which in turn predicted identification and performance. Newman, Kiazad, Miao and Cooper (2014) observed that ethical leadership predicted cognition-based trust, which in turn predicted âaffect-based trustâ, which predicted organizational citizenship behaviour. Other research shows that such findings generalize beyond inter personal relationships at work, such as for consumer trust (Johnson & Grayson, 2005), inter-organizational trust (Robson, Katsikeas & Bello, 2008; see also Dowell, Morrison & Heffernan, 2015) and trust in virtual communities (open source software development; Stewart & Gosain, 2006).
Research also corroborates McAllisterâs (1995) argument and findings that âaffect-based trustâ and cognition-based trust have unique antecedents and consequences. Chua, Ingram and Morris (2008) for instance found that âaffect-based trustâ and cognition-based trust have different types of social network ties as precursors, and Dunn, Ruedy and Schweitzer (2012) show that âaffect-based trustâ suffers from upward comparison to the other party (i.e. because it puts one in an unfavourable light) whereas downward comparison is bad for cognition-based trust (i.e. because it implies lower competence).
Other research shows that âaffect-based trustâ and cognition-based trust have (partially) different consequences. Ergeneli, Ari and Metin (2007) found for instance that cognition-based trust predicted the competence and meaning aspects of psychological empowerment, whereas âaffect-based trustâ predicted the impact aspect of empowerment. Ng and Chua (2006) found that âaffect-based trustâ was positively related to cooperation whereas cognition-based trust was not. Ha, Park and Cho (2011) found that âaffect-based trustâ was related to information sharing and benefit/risk sharing, whereas cognition-based trust was related to joint decision making and benefit/risk sharing. Chua, Morris and Mor (2012) found that cultural meta-cognition was positively related to âaffect-based trustâ (but not cognition-based trust) in cross-cultural relationships. Taking a moderator perspective Parayitam and Dooley (2009) studied conflict and trust in strategic decision-making teams and found that cognition-based trust was a moderator in the relationship between conflict and outcomes whereas âaffect-based trustâ did not moderate the relationship.
A series of studies also speaks to different roles for âaffect-based trustâ and cognition-based trust in leadership. Yang, Mossholder and Peng (2009) observed that cognition-based trust mediated the relationship of supervisory procedural justice with performance and job satisfaction, whereas âaffect-based trustâ mediated the relationship between supervisory procedural justice and helping behaviour at work. In a different spin on the mediating role of these two types of trust for justice effects, Colquitt, LePine, Piccolo, Zapata and Rich (2012) showed that âaffect-based trustâ mediated the social exchange-based (i.e. relationship-based) effects of organizational justice, whereas cognition-based trust mediated the uncertainty-reducing effects of justice. Miao, Newman, Schwarz and Xu (2013) reported that âaffect-based trustâ mediated the relationship between participative leadership and organizational commitment whereas cognition-based trust did not have a mediating role. Schaubroeck, Lam and Peng (2011) found that servant leadership influenced team performance through âaffect-based trustâ and team psychological safety. Transformational leadership (a concept that has been invalidated by conceptual and empirical problems; van Knippenberg & Sitkin, 2013) influenced team performance through cognitive trust. Zhu, Newman, Miao and Hooke (2013) found that âaffect-based trustâ mediated the relationship between transformational leadership and commitment, performance and organizational citizenship behaviour, whereas cognition-based...