Organizational Behavior and Human Decision Processes 164 (2021) 116–135
Available online 8 March 2021
0749-5978/© 2021 Elsevier Inc. All rights reserved.
Emotional acknowledgment: How verbalizing others’ emotions fosters
interpersonal trust
Alisa Yu
a
,
*
, Justin M. Berg
a
, Julian J. Zlatev
b
a
Stanford Graduate School of Business, 655 Knight Way, Stanford, CA 94305, United States
b
Harvard Business School, Wyss House, Soldiers Field, Boston, MA 02163, United States
ARTICLE INFO
Keywords:
Emotion
Costly signaling
Interpersonal trust
Emotional valence
Interpersonal relationships
Empathic accuracy
ABSTRACT
People often respond to others’ emotions using verbal acknowledgment (e.g., “You seem upset”). Yet, little is
known about the relational benets and risks of acknowledging others’ emotions in the workplace. We draw
upon Costly Signaling Theory to posit how emotional acknowledgment inuences interpersonal trust. We hy-
pothesize that emotional acknowledgment acts as a costly signal of the perceiver’s willingness to expend personal
resources to meet the needs of the expresser. Across six studies, we found convergent evidence that emotional
acknowledgment led to greater perceptions of costliness, and in turn, to higher evaluations of trust. These effects
were stronger for negative than positive emotions because acknowledging negative emotions involved a greater
perceived cost. Moreover, inaccurate acknowledgment fostered greater trust than not acknowledging when
positive emotions were mislabeled as negative, but not when negative emotions were mislabeled as positive.
These ndings advance theory on key dynamics between emotion and language in work-related relationships.
1. Introduction
Emotions are integral to the development of social relationships and
thus, to the functioning of organizations (Keltner & Haidt, 1999; van
Kleef, 2016). Social theories of emotions suggest that emotions provide
rapid insight into the goals and needs of others, thereby informing how
perceivers should respond (Keltner, Haidt, & Shiota, 2006, van Kleef,
2009). In short, emotions are powerful communication devices. Given
our capacity to glean how others may be feeling based on their
nonverbal emotional displays (Ekman & Keltner, 1997; Fridlund, 1994),
people often face important choices about whether or not to explicitly
acknowledge the emotions they see others express. Consider the
following example. An employee has a meeting with her manager, in
which she is berated for her slow progress on a project. She is feeling
distraught. When the employee returns to her desk, she takes a deep
sigh. Two coworkers notice her emotional expression. One pauses for a
moment, and then goes back to work. The other acknowledges her
emotions by saying “Hey, you seem upset.” We call this ubiquitous, but
potentially meaningful, response emotional acknowledgment—verbal
communication by which a perceiver signals the recognition of an ex-
presser’s emotional display.
Despite the role of emotional sharing and responsiveness in
strengthening interpersonal relationships (Greenberg, 2004; Reis &
Shaver, 1988; Rim
´
e, Finkenauer, Luminet, Zech, & Philippot, 1998),
surprisingly little is known about how verbally acknowledging others’
emotions affects relationships. The scant attention paid to emotional
acknowledgment is particularly noticeable when juxtaposed with the
wealth of research on the psychological and physiological benets of
using language for processing one’s own emotions (e.g., Brooks, 2014;
Pennebaker, 1997, 2018; Torre & Lieberman, 2018; Wolf, Lee, Sah, &
Brooks, 2016). Furthermore, research on active listening and partner
responsiveness typically focuses on responses to someone’s verbal ac-
counts of their personal events (Jones, 2011; Maisel, Gable, & Strach-
man, 2008; Weger, Bell, Minei, & Robinson, 2014), yet responses to
their nonverbal expressions have largely been ignored. What are the risks
and rewards of explicitly surfacing our interpretations of others’
nonverbal emotional expressions using language? In this paper, we
advance theory on this largely overlooked question by examining how
emotional acknowledgment shapes interpersonal trust—a universal and
core pillar of well-functioning social relationships (Rempel, Holmes, &
Zanna, 1985; Simpson, 2007).
We adopt the framework of Costly Signaling Theory (Zahavi, 1995)
to suggest that emotional acknowledgment is a costly signal that reveals
the willingness of a perceiver to allocate personal resources (time,
* Corresponding author at: Stanford Graduate School of Business, 655 Knight Way, Stanford, CA 94305, United States.
E-mail address: [email protected] (A. Yu).
Contents lists available at ScienceDirect
Organizational Behavior and Human Decision Processes
journal homepage: www.elsevier.com/locate/obhdp
https://doi.org/10.1016/j.obhdp.2021.02.002
Received 21 May 2020; Received in revised form 2 February 2021; Accepted 4 February 2021
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
117
energy, and attention) to the emotional needs of an expresser (Dovidio,
Piliavin, Shroeder, & Penner, 2006). Specically, we examine contexts
in which the emotions being expressed are neither caused by nor
directed at the perceiver, and in which the two parties already have an
established relationship. We hypothesize that individuals who
acknowledge others’ emotions will be seen as more trustworthy than
those who do not, because acknowledgment may be costly and conict
with their own interests to protect and conserve resources (Murray &
Holmes, 2009). Furthermore, we hypothesize that acknowledging
negative emotions will be a stronger driver of trust than positive emo-
tions, as acknowledging negative emotions will be seen as costlier.
Across six studies examining the perspectives of both expressers and
third-party observers, we found support for our hypotheses.
The present research offers important theoretical contributions.
First, whereas social theories of emotions predominantly focus on in-
ferences drawn about the expresser (van Kleef, 2009; Keltner & Haidt,
1999; Frijda & Mesquita, 1994), we focus instead on inferences drawn
about the perceiver, based on their response to the expresser’s emotions.
We suggest that representing the inuence of emotions as bidirectional,
rather than unidirectional, provides a more accurate depiction of
emotional sensemaking and communication. Second, by integrating
social theories of emotions with Costly Signaling Theory, we introduce a
novel theoretical perspective for understanding the role of emotions in
interpersonal relationships. Third, we complement research on labelling
one’s own emotions (e.g., Brooks, 2014; Wolf et al., 2016) by exploring
the consequences of labeling the emotions of others. Finally, we add to
research on active listening and responsiveness, which focuses on re-
sponses to what a speaker states verbally in a conversation. Supple-
menting this perspective, the current research examines responses to
what an expresser displays nonverbally (Jones, 2011; Maisel, Gable, &
Strachman, 2008; Weger, Bell, Minei, & Robinson, 2014), which may be
more ambiguous and therefore, potentially riskier to acknowledge.
Overall, our research sheds light on how the explicit acknowledgment of
emotions can be used as a powerful tool for shaping and understanding
social relationships.
1.1. Conceptualizing emotional acknowledgment in organizations
Whereas past research has examined emotional labelling, or verbal
appraisals of one’s own emotions (e.g., “I feel excited”, Brooks, 2014),
emotional acknowledgment refers to verbal appraisals of others’ emo-
tions. Emotional acknowledgment is directed at the expresser and con-
veys information about the perceiver’s interpretation of the emotional
display. For example, in response to noticing a coworker’s enthusiastic
grin, an employee might acknowledge this coworker’s emotion by
saying, “You seem excited.” Emotional acknowledgment communicates
both (1) the perception of an emotional expression (“I picked up on an
emotion”), and (2) the content inferred from the expression (“The
emotion I perceived was excitement”).
Conceptually, emotional acknowledgment belongs to a larger cate-
gory of responses that convey active listening and partner
responsiveness. For example, active listening includes displaying
nonverbal involvement (e.g., nodding, eye contact), paraphrasing the
speaker’s message without judgment, and asking questions prompting
the speaker to elaborate on their story (Bodie, 2011; Weger, Bell, Minei,
& Robinson, 2014). Similarly, responsiveness can be conveyed by
providing perspective on the situation, offering encouragement and
reassurance, or validating the person’s effort or identity (Maisel et al.,
2008). Generally, these responses lead to positive interpersonal evalu-
ations (e.g., asking follow-up questions increases liking; Huang et al.,
2017). However, research in these domains often makes an implicit
assumption that the listener responds only to what the speaker com-
municates explicitly and verbally in the conversation. In reality,
nonverbal channels also play a critical role in interpersonal communi-
cation. Emotional acknowledgment complements this body of work by
focusing on responsiveness to nonverbal displays. Because nonverbal
displays tend to be more ambiguous and difcult to decipher than verbal
remarks, perceivers may have greater latitude in how they choose to
respond. For example, it may be easier to ignore a coworker wearing a
frown, than a coworker who says, “I’m feeling really frustrated right
now.” Thus, the decision to acknowledge others’ emotions may be
perceived as riskier, but also more discretionary and volitional than
other potential responses.
Emotional acknowledgment is related to, but conceptually distinct
from, other emotion-relevant constructs, such as empathic accuracy
(Ickes, 1993) and emotional understanding (Mayer, Salovey, Caruso, &
Sitarenios, 2001). Table 1 provides two dimensions by which to compare
these constructs: (1) whether the construct refers to an intrapersonal or
interpersonal phenomenon, and (2) whether the construct refers to an
ability or behavior. Critically, many of the constructs arising from the
literature on emotional intelligence focus on personal abilities (Mayer &
Salovey, 1997). However, possessing the ability to decode and regulate
emotions is conceptually distinct from the decision to act on these
abilities. For example, a perceiver may notice when an expresser starts
crying, and may accurately infer that the expresser is experiencing
emotional distress, yet choose not to intervene. Indeed, research sug-
gests that empathic accuracy alone does not predict whether a partner
will be perceived as responsive (Winczewski, Bowen, & Collins, 2016).
Rather, empathic accuracy only leads to perceptions of responsiveness
when coupled with high motivation to act compassionately toward
others. As such, studying responses to others’ emotions may shed unique
insight into the process of emotional communication that examining
personal abilities alone cannot.
1.2. Perception of emotional acknowledgment as a costly signal for
trustworthiness
We draw upon Costly Signaling Theory to explore the signaling value
of emotional acknowledgment. Costly Signaling Theory (Zahavi, 1995)
originated from the eld of evolutionary biology, but has since been
adapted by psychologists to explain how people interpret costly be-
haviors that seemingly violate norms of self-interest. The theory asserts
Table 1
Constructs that relate to emotional acknowledgment.
Ability Behavior
Interpersonal Emotional management:
The ability to manage emotions and emotional relationships for personal and interpersonal
growth (Mayer, Salovey, Caruso, & Sitarenios, 2001).
Empathic accuracy:
The ability to accurately infer the thoughts and feelings of another person (Ickes, 1993).
Emotional acknowledgment:
An externally-directed communication by which the perceiver
signals the recognition of an expresser’s emotions.
Intrapersonal Emotional Perception:
The ability to identify emotions through nonverbal emotional expressions (Mayer, Salovey,
Caruso, & Sitarenios, 2001).
Emotional Understanding:
The ability to comprehend emotional information about relationships, transitions from one
emotion to another, and linguistic information about emotions (Mayer, Salovey, Caruso, &
Sitarenios, 2001).
Emotional labeling:
A statement about one’s own experience of emotion (e.g., Brooks,
2014; Pennebaker, 1997, 2018; Wolf et al., 2016).
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
118
that it is difcult for people to ascertain whether someone has potential
as an ally, mate, or competitor because their underlying qualities are
often hidden (Smith & Bird, 2000). Consequently, people rely on
behavioral signals to infer these underlying qualities indirectly. The
theory suggests that the costliness of the signal helps distinguish those
who possess desirable qualities from those who lie about them because
presumably only individuals who can afford to bear the burden of the
signal would choose to engage in it. For example, adhering to strict
religious rituals is perceived as a costly signal and, thus, an indicator of
trustworthiness because presumably only the truly devout would be
willing to make the costly sacrices (Hall, Cohen, Meyer, Varley, &
Brewer, 2015). To this end, costly signals should convey higher-quality
information about the underlying attributes of the person sending the
signal.
Viewed through this lens, emotional acknowledgment may be
perceived as a costly signal that reveals the trustworthiness of the
perceiver. To be considered a costly signal, the behavior must be seen as
“expensive” for the signaler in terms of personal resources (Smith &
Bird, 2000; Hardy & Van Vugt, 2006). Responding to others’ emotions,
especially in organizational contexts, may be viewed as costly. Per-
ceivers have a nite pool of personal resources (time, energy, and
attention; Hobfoll, 2002). When perceivers direct their scarce personal
resources toward the emotional needs of others through emotional
acknowledgment, others are likely to recognize that this effort may
come at the expense of perceivers’ own needs or goals (e.g., meeting an
approaching deadline). Specically, perceivers may face tradeoffs be-
tween expending their personal reserve of cognitive, psychological, and
emotional resources to advance their own goals, or attend to the needs of
the expresser (Bono & Vey, 2005; Morris & Feldman, 1996; Zaki, 2014).
Furthermore, compared to everyday life, emotions in organizations can
be more difcult to interpret (DePaulo, 1992; Porter & ten Brinke, 2008)
because they are not always authentically expressed. Organizational
display rules prescribe which emotions employees should display for
their roles (Grandey, 2000; Hochschild, 1979; Van Maanen & Kunda,
1989), but these emotions may not reect their internal affective states.
For this reason, verbally acknowledging these “noisy” emotional signals
may be seen as reputationally risky because it could expose a norm
violation or faulty interpretation of an expresser’s emotion (e.g., mis-
labeling an expresser’s emotion). To this end, we expect expressers (and
third-party observers bearing witness to the interaction) to view those
who acknowledge emotions as possessing a higher capacity to respond
to others’ emotional needs.
By verbally calling on others’ emotions, emotional acknowledgment
can facilitate an open dialogue, providing permission for the expresser to
engage in subsequent self-disclosure (Antaki, 1988; Rempel, Ross, &
Holmes, 2001). When a perceiver says, “You look distressed”, not only
might this act imply a momentary resource investment (i.e., noticing the
emotion), it also implies the perceiver’s willingness to invest additional
resources in the future (e.g., listening to the expresser’s explanation or
providing social support). Because nonverbal expressions tend to be
more ambiguous and more easily ignored, the decision to engage in
acknowledgment may be perceived as more volitional, conveying the
perceiver’s voluntary desire to understand and construct a shared
meaning of the expresser’s experience (Rossignac-Milon & Higgins,
2018; Weick, 1995). To this end, emotional acknowledgment may act as
a catalyst for opening the emotional “oodgates”, licensing the
expresser to share their underlying needs. Because this process may
require the expenditure of scarce personal resources, emotional
acknowledgment should signal that the perceiver is willing to incur costs
for the sake of the expresser.
Insofar as emotional acknowledgment signals a readiness to accom-
modate and sacrice personal resources, emotional acknowledgment
should advance the formation of interpersonal trust, a cornerstone of
high-functioning and high-quality relationships (Balliet & Van Lange,
2013; Rempel, Holmes, & Zanna, 1985; Simpson, 2007). Although many
conceptualizations of interpersonal trust exist, we focus on a
benevolence-based denition: the expectation that an individual will
demonstrate care for the interests and welfare of the trustor (Holmes &
Rempel, 1989; Levine & Schweitzer, 2015; Mayer, Davis, & Schoorman,
1995). Past research suggests that trust often emerges when a conict
arises between the partner’s personal interests and the interests of the
relationship, putting the partner’s true goals and motives to the test
(Balliet & Van Lange, 2013; Holmes and Rempel, 1989). When a partner
voluntarily sacrices their own interests to prioritize the relationship,
this behavior provides unambiguous evidence of the partner’s benevo-
lent intent (Wieselquist, Rusbult, Foster, & Agnew, 1999), thereby
facilitating trust. Similarly, emotional acknowledgment may demon-
strate a perceiver’s costly willingness to prioritize an expresser’s needs
over their own. In turn, emotional acknowledgment should increase
perceptions of trustworthiness (see Fig. 1 for visual of hypotheses).
Hypothesis 1. When individuals engage in emotional acknowledg-
ment, they are seen as more trustworthy than if they do not engage in
emotional acknowledgment.
Hypothesis 2. The relationship between emotional acknowledgment
and interpersonal trust is mediated by perceptions of costliness.
1.3. Valence as a moderator: the costliness of acknowledging negative vs.
positive emotions
According to Costly Signaling Theory, the costliness of the signal
should enhance the credibility of the signaler’s desirable qualities, and
subsequently incur greater benets for the signaler. For example, Hardy
and Van Vugt (2006) showed that the most altruistic members in a group
achieve the highest levels of social status because they are seen as
bearing the greatest costs for their generosity to the group. Applying this
same logic, we expect seemingly costlier forms of emotional acknowl-
edgment to be rewarded with greater interpersonal trust.
Specically, we suggest that the valence of the expresser’s emotions
may inuence the link between emotional acknowledgment and trust
because negative emotions often signal greater underlying needs, which
subsequently places higher social, cognitive, and affective demands on
the perceiver than positive emotions (Fredrickson, 2001; Staw & Bar-
sade, 1993; Zaki, 2014). Consequently, the decision of whether to
engage in emotional acknowledgment may be more inuential on per-
ceptions of trust in the wake of negative than positive emotions. To
unpack this logic, we start by reviewing the evolutionary functions of
negative emotions.
The expression of negative emotions alerts others to problems that
require attention and potentially threaten survival (Frijda & Mesquita,
1994). As such, negative emotions should subsequently recruit valuable
resources and responses from the community (Baker, McNulty, &
Overall, 2014). Indeed, negative emotions communicate the unmet and
heightened needs of the expresser and compel perceivers to engage in
actions that alleviate or solve the problem (e.g., provide social support
or compassion; Keltner & Haidt, 1999). However, for perceivers,
responding to negative emotions may come at a price. Because negative
emotions typically signal a greater underlying need, addressing negative
emotions may be more demanding and “expensive” for the perceiver.
For example, attending to the negative emotions of others may require
the perceiver to engage in emotional regulation, which over time can
become depleting and harmful to the perceiver’s own well-being
(Grandey & Gabriel, 2015). Additionally, confronting others’ negative
emotions potentially obligates perceivers to help expressers, which may
require subsequent expenditure of personal resources (Zaki, 2014).
Furthermore, studies show that through contagion, perceivers often
“catch” the negative emotions of others (Barsade, 2002; Hateld,
Cacioppo, & Rapson, 1993), which can even increase the perceiver’s
own risk for depression (Rosenquist, Fowler, & Christakis, 2011).
Because negative emotions convey the heightened needs of the
expresser, and therefore make them potentially more challenging and
personally costly for the perceiver, acknowledging negative emotions
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
119
should be seen as highly diagnostic of the perceiver’s capacity to nurture
the expresser’s needs. Thus, the effect of acknowledging emotions on
trust should be stronger when responding to negative emotions.
In contrast, the acknowledgment of positive emotions may be
perceived as less costly. Positive emotions are used to build and broaden
personal resources for the long term (Fredrickson, 2001). They
encourage expressers to engage in approach behaviors toward others,
which procure social resources for the future. However, the call to action
for perceivers is less immediate and clear for positive emotions (Fre-
drickson & Levenson, 1998). Positive emotions largely arise in the
absence of problems, and therefore do not necessarily compel perceivers
to deliver a specic momentary response. For positive emotions, because
there are no pressing needs requiring immediate assistance, the decision
to respond may be lower stakes and therefore, less resource-depleting or
demanding. Furthermore, even though responding to positive emotions
may still require an expenditure of one’s own personal resources, doing
so may reap some personal benets. For example, perceivers may
“catch” the positive emotions of others, thereby elevating their own
affective experiences (Barsade, 2002; Morelli, Ong, Makati, Jackson, &
Zaki, 2017). Research on capitalization suggests that the sharing of
positive emotions can strengthen existing social bonds (Gable & Reis,
2010). Because responding to positive emotions can potentially provide
benets to the perceiver, the act may be perceived as more compatible
with self-interest than responding to negative emotions. Consequently,
the allocation of resources required to acknowledge positive emotions
may be interpreted as relatively “inexpensive”. Thus, although
acknowledging positive emotions should increase trust more than not
acknowledging, we expect a weaker effect for positive, relative to
negative, emotions.
Hypothesis 3a. The relationship between emotional acknowledg-
ment and interpersonal trust is moderated by the valence of the ex-
presser’s emotion, such that the relationship is stronger when the
emotions expressed are negative as opposed to positive.
Hypothesis 3b. This moderation occurs because negative emotions
are seen as costlier to acknowledge than positive emotions.
2. Overview of studies
Employing a diverse set of methodologies, we examine the rela-
tionship between emotional acknowledgment and interpersonal trust
from the perspective of the expresser (Studies 1–4) and third-party ob-
servers (Studies 5–6). In Study 1, we surveyed a eld sample of hospital
employees, providing preliminary support for the link between
emotional acknowledgment and trust in a high-stakes organizational
context. In Study 2, we provide causal evidence for our hypotheses using
an experiment involving work-related scenarios. In Study 3, we show
that emotional acknowledgment led to greater trust than non-emotional
acknowledgment. In Study 4, we show that the relationship between
emotional acknowledgment and trust is attenuated when perceivers are
seen as acknowledging out of concern for their own reputation. In Study
5, we examine our hypotheses from a third-party perspective using
videos of two coworkers interacting in a breakroom. In Study 6, we
examine situations in which perceivers acknowledge an expresser’s
emotions inaccurately and test empathic accuracy as an alternative
explanation. See Table 2 for an overview of each study (i.e., study
purpose, sample, design, perspective, and hypotheses tested).
For the ve experiments (Studies 2–6), we set a xed rule prior to
data collection to collect approximately 100 participants per cell and
then exclude any participants who failed the comprehension checks that
were included in each experiment (Simmons, Nelson, & Simonsohn,
2011). See the Online Supplement for details on the exclusions and
comprehension checks, which showed that the manipulations were
effective in each of the ve experiments (Studies 2–6). Data and mate-
rials for all studies are available at: https://osf.io/dn3wh/
3. Study 1: Examining emotional acknowledgment and trust in
the eld
This study examined the link between emotional acknowledgment
and trust in a eld sample of healthcare providers working in two
intensive care units of a children’s hospital. As initial tests, we explored
the association between emotional acknowledgment and trust (H1) and
investigated whether the relationship depended on the valence of the
expresser’s emotions (H3a).
3.1. Participants and procedures
As part of an organizational development initiative, all healthcare
providers in the Pediatric Intensive Care Unit (PICU) and the Cardio-
vascular Intensive Care Unit (CVICU) of a large children’s hospital in
California were invited to participate in this study. The study was con-
ducted in two phases. The hospital sent the survey to a subset of em-
ployees in the PICU during the rst phase (n = 92; response rate of
25.3%), and to all CVICU and remaining PICU employees during the
second phase (n = 154; response rate of 42.3%). We capitalized on the
staggered rollout by administering a general measure of emotional
acknowledgment during the rst phase and measures of negative and
positive emotional acknowledgment during the second phase. This
allowed us to examine the main effect of emotional acknowledgment
and the differential effects of acknowledging positive and negative
emotions, respectively.
Overall, we surveyed 246 healthcare employees. The employees in
the sample occupied several different roles: 54.5% were nurses, 14.6%
H1
H2
H3a
Perceptions of Costliness
Emotional
Acknowledgment
Perceptions of
Interpersonal Trust
Valence of Expresser’s
Emotions
H3b
Fig. 1. Theoretical model of hypotheses.
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
120
were respiratory therapists, 12.2% were physicians, 2.8% were nurse
practitioners, 2.4% were social workers, 3.7% were rehabilitation
therapists, and 9.8% occupied other roles. Most (79.3%) employees
worked in full-time positions, 18.3% worked part-time, and 2.4% had
other working arrangements. The majority (65.9%) worked day shifts,
27.2% worked night shifts, and 6.9% worked approximately equal
numbers of night and day shifts. Lastly, 82.9% worked weekdays and
weekends, while 16.7% worked weekdays only. In line with the hospi-
tal’s wishes to increase participation by protecting the employee’s
identifying information, we did not collect additional demographic
information.
3.2. Measures
Participants were asked to answer all survey items based on their
attitudes and behaviors during the past two weeks. The survey included
measures of interpersonal trust, emotional acknowledgment, and addi-
tional control variables. All items were rated on a 5-point Likert-type
scale (1 = “strongly disagree” to 5 = “strongly agree”).
3.2.1. Interpersonal trust
Interpersonal trust (
α
= 0.78) was measured using a four-item scale
adapted from Mayer and Davis (1999): “My coworkers look out for what
is important to me”, “My coworkers would not knowingly do anything to
hurt me”, “My coworkers are concerned about my welfare”, “My co-
workers think my needs and preferences are important”.
3.2.2. General emotional acknowledgment
During the rst phase of the study, general emotional acknowledg-
ment (
α
= 0.88) was measured using a four-item scale: “When my co-
workers notice my emotions, they will bring them up”, “When my
coworkers see that I am experiencing an emotion, they will mention it.”,
“My coworkers do NOT refer to my emotions” (reverse-scored), “When
coworkers see that I am feeling an emotion, they do NOT say anything
about it” (reverse-scored). The pre-registered validation of this scale can
be found in Appendix A.
3.2.3. Negative/positive emotional acknowledgment
During the second phase of the study, negative (M = 3.24, SD = 0.71,
α
= 0.85) and positive emotional acknowledgment (M = 3.41, SD =
0.69,
α
= 0.87) were measured using the following eight items: “When
my coworkers notice my negative [positive] emotions, they will bring
them up”, “When my coworkers see that I am experiencing a negative
[positive] emotion, they will mention it.”, “My coworkers do NOT refer
to my negative [positive] emotions” (reverse-scored), and “When co-
workers see that I am feeling negatively [positively], they do NOT say
anything about it” (reverse-scored).
3.2.4. Control variables
Based on recommendations provided by Becker et al. (2016), we
included three control variables that were empirically and theoretically
related to interpersonal trust, our main dependent variable (De Jong,
Dirks, & Gillespie, 2016; Guinot, Chiva, & Roca-Puig, 2014): job satis-
faction, perceived stress, and perceived team functioning. Job
satisfaction (
α
= 0.91) was measured using three items adapted from
Cammann, Fichman, Jenkins, and Klesh (1983; e.g., “I have enjoyed
working as a member of this care team”). Perceived stress (
α
= 0.59) was
measured using four items adapted from Cohen, Kamarck, and Mer-
melstein (1983; e.g., “I felt difculties were piling up so high at work
that I could not overcome them”). Perceived team functioning (
α
= 0.71)
was measured using three items adapted from the Agency for Healthcare
Research and Quality (2006; e.g., “My unit operates at a high level of
efciency”).
3.3. Results and discussion
To examine the association between emotional acknowledgment and
interpersonal trust, we used our sample from the rst phase and con-
ducted a series of linear regression models (Table 3). We found that
participants were more likely to trust coworkers when their emotions
were acknowledged, β = 0.59, SE = 0.09, p < .001. This relationship
persisted when we controlled for demographic variables, β = 0.63, SE =
0.10, p < .001, and when additional controls were included for job
satisfaction, perceived stress, and perceived team functioning, β = 0.30,
SE = 0.11, p = .006. Consistent with H1, these results provide pre-
liminary evidence of a relationship between general emotional
acknowledgment and interpersonal trust in an organizational context.
To examine the differential effects of positive and negative emotional
acknowledgment on interpersonal trust (H3a), we used our sample from
the second phase. Using a series of linear regression models (Table 4),
we found that participants were more likely to trust coworkers when
their positive emotions were acknowledged, even after controlling for
demographic variables, β = 0.23, SE = 0.08, p = .004. However, the
relationship did not persist when we included additional controls for job
satisfaction, perceived stress, and perceived team functioning, β = 0.12,
SE = 0.08, p = .12. Likewise, we found that participants were more likely
to trust coworkers when their negative emotions were acknowledged,
even after controlling for demographic variables, β = 0.34, SE = 0.08, p
< .001. Moreover, this relationship persisted even when we included the
additional controls, β = 0.23, SE = 0.08, p = .004. When we added
positive and negative emotional acknowledgment to the same model
with all covariates, negative emotional acknowledgment persisted as a
signicant predictor of trust, β = 0.20, SE = 0.08, p = .01, but positive
emotional acknowledgment did not, β = 0.07, SE = 0.08, p = .36,
indicating that only negative emotional acknowledgment accounted for
unique variance in trust. These ndings provide suggestive evidence in
support of H3a that acknowledging negative emotions has a stronger
effect on trust than acknowledging positive emotions.
This rst study provides initial evidence in support of H1 and H3a in
a high-stakes organizational context. However, the cross-sectional na-
ture of this study precluded us from drawing inferences about causality,
and we did not test our proposed mechanism of perceived costliness
directly (H2 and H3b). We address these limitations in Study 2.
4. Study 2: Experimental tests of H1-H3 from the expresser’s
perspective
In Study 2, we build on our cross-sectional eld study using an
Table 2
Summary table of all studies.
Study Main Purpose Sample Design Perspective Hypotheses Tested
1 To establish external validity N = 246 Field survey with hospital employees First-party (expresser) H1, H3a
2 To nd causal evidence N = 374 Between-subject experiment using vignettes First-party (expresser) H1, H2, H3a, H3b
3 To compare emotional vs. non-emotional
acknowledgment
N = 195 Mixed-design experiment using vignettes First-party (expresser) H1, H2, H3a
4 To examine moderation by reputational concern N = 395 Between-subject experiment using vignettes First-party (expresser) H1, H2
5 To test hypotheses from third-party perspective N = 541 Between-subject experiment using video stimuli Third-party H1, H2, H3a, H3b
6 To examine the role of acknowledgment accuracy N = 581 Between-subject experiment using video stimuli Third-party H4, H5, H6a, H6b
(outlined on pg. 33–34)
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
121
experimental design. Participants imagined themselves in a hypothetical
scenario with a coworker. We varied the description of how the
participant was feeling at the time (either positively or negatively) and
how the coworker responded (either by acknowledging their emotions
or not). In addition to interpersonal trust and emotional acknowledg-
ment, we also measured perceived costliness, thereby allowing a test of
our full theoretical model (H1-H3b).
4.1. Participants and procedures
Participants included 374 individuals recruited through Mechanical
Turk (35.3% female; M
age
= 37.08, SD
age
= 12.71). The sample was
12.0% African-American, 7.5% Asian, 71.1% Caucasian, 5.3% Hispanic,
and 4.0% other. The experiment employed a 2 (valence of expressed
emotion: positive, negative) × 2 (emotional acknowledgment: yes, no)
between-subjects design. Participants were randomly assigned to one of
four scenarios, in which they imagined themselves interacting with a
coworker. The scenarios were described as follows:
You just nished a meeting with your boss, and it went very poorly [very
well].
You are feeling upset [happy].
Your coworker nearby is ipping through a report and then says to you, “I
just talked to the team about the new deadline and they’re okay with it.”
Your coworker looks up at you and pauses for a moment.
In the acknowledgment condition, participants then read: “Then,
your coworker says, “You seem upset [happy].” In the no acknowledgment
condition, participants instead read: “Then, your coworker continues
talking about the report.” Next, participants answered questions about
their perceptions of this coworker and lled out a set of demographic
questions.
4.2. Measures
4.2.1. Interpersonal trust
Interpersonal trust (
α
= 0.88) was measured using adapted versions
of the four items from Study 1 (e.g., “This coworker is concerned about
my welfare”).
4.2.2. Perceived costliness
Perceived costliness was measured using a ve-item scale that
captured potential costs associated with emotional acknowledgment in
the focal context (
α
= 0.96): “This coworker is willing to spend [atten-
tion]/[effort]/[time]/[energy]/[personal resources] on the relation-
ship.” Specically, we constructed these items based on previous
conceptual denitions of costliness in Costly Signaling Theory (Smith &
Bird, 2000; Hardy & Van Vugt, 2006), focusing on the forms of costliness
that are relevant in workplace relationships.
4.3. Results and discussion
4.3.1. Interpersonal trust
In support of H1, we found a main effect of emotional acknowledg-
ment on interpersonal trust. Specically, participants trusted the
perceiver signicantly more in the acknowledgment (M = 5.09, SD =
1.06) than the no acknowledgment condition (M = 4.44, SD = 1.30), F
(1, 370) = 29.27, p < .001,
η
2
= 0.07. Consistent with previous research
Table 3
Standardized OLS regression models for rst phase of Study 1.
Dependent variable:
Interpersonal Trust
(1) (2) (3)
Emotional Acknowledgment 0.59
***
0.63
***
0.30
**
(0.09) (0.10) (0.11)
Role: Physician −0.01 −0.26
(0.36) (0.30)
Role: Rehab. Therapist 0.57 0.42
(0.54) (0.45)
Role: Resp. Therapist 0.06 0.24
(0.31) (0.27)
Role: Other 0.44 0.04
(0.38) (0.32)
Shift: Night 0.36 0.21
(0.24) (0.20)
Shift: Both Day and Night −0.22 −0.14
(0.52) (0.44)
Days on Duty: Weekdays only −0.27 −0.35
(0.33) (0.28)
Days on Duty: Weekend only −1.24 −1.00
(0.88) (0.75)
Full-Time Status: Part-Time −0.19 −0.18
(0.26) (0.22)
Full-Time Status: Other −0.54 −0.44
(0.54) (0.45)
Job Satisfaction 0.33
**
(0.12)
Stress −0.16
(0.12)
Team Functioning 0.15
(0.10)
Constant 0.05 0.01 0.09
(0.09) (0.15) (0.13)
Observations 82 82 82
R
2
0.34 0.41 0.61
Adjusted R
2
0.33 0.31 0.52
Residual Std. Error 0.82 (df = 80) 0.83 (df = 70) 0.69 (df = 67)
F Statistic 41.65
***
(df = 1; 80) 4.38
***
(df = 11; 70) 7.39
***
(df = 14; 67)
Note: *p < .05, **p < .01, ***p < .001.
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
122
Table 4
Standardized OLS regression models for second phase of Study 1.
Dependent variable:
Interpersonal Trust
(1) (2) (3) (4) (5)
Positive Emotional Acknowledgment 0.23
**
0.12 0.07
(0.08) (0.08) (0.08)
Negative Emotional Acknowledgment 0.34
***
0.23
**
0.20*
(0.08) (0.08) (0.08)
Role: Nurse Practitioner 0.63 0.59 0.85 0.74 0.71
(0.44) (0.40) (0.44) (0.40) (0.40)
Role: Physician 0.65* 0.41 0.73
**
0.48 0.51
(0.28) (0.26) (0.28) (0.26) (0.27)
Role: Rehab. Therapist 0.67 0.41 0.76 0.51 0.55
(0.53) (0.49) (0.53) (0.49) (0.49)
Role: Resp. Therapist 0.26 0.06 0.36 0.15 0.16
(0.23) (0.22) (0.24) (0.22) (0.22)
Role: Social Worker 0.96* 1.05* 0.89 1.01* 1.01*
(0.46) (0.42) (0.46) (0.42) (0.42)
Role: Other 0.30 0.12 0.24 0.07 0.08
(0.30) (0.28) (0.29) (0.28) (0.28)
Shift: Night 0.42* 0.27 0.49* 0.34 0.35
(0.19) (0.18) (0.20) (0.18) (0.18)
Shift: Both Day and Night 0.04 0.10 −0.11 −0.0005 0.01
(0.37) (0.33) (0.36) (0.33) (0.33)
Days on Duty: Weekdays only −0.14 −0.25 −0.04 −0.18 −0.18
(0.27) (0.25) (0.27) (0.25) (0.25)
Full-Time Status: Part-Time −0.14 −0.23 −0.16 −0.23 −0.20
(0.22) (0.20) (0.22) (0.20) (0.20)
Full-Time Status: Other 0.79 0.73 0.29 0.33 0.33
(0.57) (0.52) (0.69) (0.63) (0.63)
Job Satisfaction 0.30
**
0.26
**
0.25
**
(0.09) (0.09) (0.09)
Stress −0.03 −0.06 −0.05
(0.08) (0.08) (0.08)
Team Functioning 0.20* 0.21* 0.20*
(0.08) (0.08) (0.08)
Constant −0.34* −0.18 −0.39* −0.22 −0.23
(0.16) (0.15) (0.16) (0.15) (0.15)
Observations 152 151 146 145 145
R
2
0.14 0.31 0.19 0.35 0.35
Adjusted R
2
0.07 0.24 0.12 0.27 0.27
Residual Std. Error 0.96 (df = 139) 0.88 (df = 135) 0.95 (df = 133) 0.87 (df = 129) 0.87 (df = 128)
F Statistic 1.93* (df = 12; 139) 4.10
***
(df = 15; 135) 2.63
**
(df = 12; 133) 4.54
***
(df = 15; 129) 4.31
***
(df = 16; 128)
Note: *p < .05, **p < .01, ***p < .001.
Fig. 2. Interactive effects of emotional acknowledgment and emotional valence on rst-party trust perceptions (Study 2). Note: The depicted bars represent mean
values of interpersonal trust, and error bars represent 95% condence intervals.
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
123
showing that positive emotions make people more trusting than nega-
tive emotions (Dunn & Schweitzer, 2005), we also found a main effect of
emotional valence—participants trusted the perceiver more in the pos-
itive (M = 4.99, SD = 1.13) than negative emotion condition (M = 4.51,
SD = 1.29), F(1, 370) = 16.73, p < .001,
η
2
= 0.04.
In support of H3a, we also found a signicant interaction between
emotional acknowledgment and valence on trust, F(1, 370) = 8.96, p =
.003,
η
2
= 0.02, indicating that the effect of emotional acknowledgment
depended on the valence of the simulated emotion—see Fig. 2. Partici-
pants trusted the perceiver more when the perceiver acknowledged their
positive emotion (M = 5.15, SD = 1.05) than when they did not (M =
4.83, SD = 1.19), t(2 00) = 2.00, p = .047, d = 0.28. However, in the
wake of experiencing negative emotions, differences in trust evaluations
were even more pronounced. That is, participants trusted the perceiver
to a greater extent when the perceiver acknowledged their negative
emotions (M = 5.03, SD = 1.07) than when they did not (M = 4.00, SD =
1.29), t(170) = 5.70, p < .001, d = 0.87.
4.3.2. Mediation and moderated mediation
Next, we conducted mediation analyses (Hayes, 2013) to test
whether perceived costliness mediated the relationship between
emotional acknowledgment and interpersonal trust (H2), and whether
this mediation was moderated by the valence of the expressed emotions
(H3b). Using 5000 bootstrapped samples, we found a signicant medi-
ation pathway from emotional acknowledgment to interpersonal trust
through perceived costliness (indirect effect = 0.86, SE = 0.11, 95% CI
= [0.66, 1.08]), which supports H2. We also found evidence in support
of H3b (index of moderated mediation = 0.28, SE = 0.10, 95% CI =
[0.08, 0.48]). Specically, the effect of emotional acknowledgment on
interpersonal trust through perceived costliness was stronger for nega-
tive (indirect effect = 0.54, SE = 0.08, 95% CI = [0.38, 0.69]) than
positive emotions (indirect effect = 0.25, SE = 0.07, 95% CI = [0.13,
0.38]).
In sum, Study 2 provides causal evidence in support of H1-H3. Ex-
pressers deemed perceivers as more trustworthy when the perceived
costs of acknowledging were heightened (i.e., responding to negative
rather than positive emotions). Correspondingly, the effect of emotional
acknowledgment on trust through perceived costliness was signicantly
stronger in response to an expresser’s negative than positive emotions.
5. Study 3: Comparison of emotional and non-emotional forms
of acknowledgment
Studies 1 and 2 leave open the possibility that the results could be
attributed to the general act of acknowledgment, as opposed to
acknowledgment of emotions in particular. In Study 3, we modied our
design to compare emotional and non-emotional acknowledgment. That
is, we used non-emotional acknowledgment, rather than no acknowl-
edgment, as our control condition. We predicted that perceivers who
engage in emotional acknowledgment, which signals openness to
emotional self-disclosures from expressers, would be seen as more
trustworthy than those who engage in non-emotional acknowledgment.
Past research suggests that when people reveal information about their
emotions (as opposed to factual or descriptive information), they feel
greater intimacy and closeness to their listening partners (Laurenceau,
Barrett, & Pietromonaco, 1998). Indeed, emotional self-disclosures may
foster more social closeness than factual self-disclosures because emo-
tions and feelings tend to be more central to one’s own self-denition
(Bosma & Kunnen, 2001; Kristjansson, 2010). As such, to the extent
that emotional acknowledgment prompts and licenses emotional self-
disclosures, we expect emotional acknowledgment to be seen as cost-
lier—and thus increase trust more—than non-emotional acknowledg-
ment. Furthermore, to complement our measure of perceived costliness,
which captures the general willingness to expend future resources into
the relationship, we explored the specic resource costs that participants
anticipate will be expended in the interaction. The design, analyses, and
hypotheses for Study 3 were pre-registered on AsPredicted.org (https://
aspredicted.org/i6tz7.pdf).
5.1. Participants, procedures, and measures
The sample included 195 participants recruited through Prolic
(53.8% female; M
age
= 31.47, SD
age
= 9.78). The sample was 7.2%
African-American, 14.5% Asian, 67.1% Caucasian, 6.2% Hispanic, and
5.1% other.
This experiment used a mixed design, with valence of expressed
emotion (positive, negative) manipulated between-subjects and
emotional acknowledgment (yes, no) varied within-subjects. Partici-
pants read the same scenario stimuli as in Study 2. However, at the end
of the scenario, each participant read about two possible responses (i.e.,
emotional or a non-emotional acknowledgment) from the coworker,
presented in random order:
Now, we’d like for you to consider two possible responses from your
coworker:
Response 1: Then your coworker says, “You looked upset [happy] after
the meeting. How are you feeling about it?”
Response 2: Then your coworker says, “It looked like the meeting went
poorly [well]. How are you thinking about it?”
Next, participants answered questions about their perceptions of the
coworker based on each of these two responses and then completed a set
of demographic questions. We used the same items as Study 2 to mea-
sure interpersonal trust (
α
non-emotional
= 0.90;
α
emotional
= 0.88) and
perceived costliness (
α
non-emotional
= 0.95;
α
emotional
= 0.92).
Finally, to complement our general measure of perceived costliness,
we also included three items capturing the specic resource costs that
participants may anticipate in the interaction. These items asked par-
ticipants to compare how they thought the conversation would unfold if
their coworker engaged in emotional vs. non-emotional acknowledg-
ment. First, we presented participants with a sliding scale ranging from
0 to 30 min and asked, “How much longer (in minutes) would this
coworker stay in a conversation with you?” Second, we presented par-
ticipants with a sliding scale from 0% (minimal attention) to 100%
(maximal attention) and asked, “How attentive would this coworker be
in the rest of this conversation?” Third, we presented participants with a
sliding scale from 0% (minimal effort) to 100% (maximal effort) and
asked, “How much effort would this coworker put into the rest of this
conversation?”
5.2. Results and discussion
5.2.1. Interpersonal trust
To analyze these data, we conducted a 2 (Acknowledgment:
Emotional, Non-Emotional) × 2 (Emotional Valence: Positive, Negative)
mixed ANOVA with repeated measures on the rst factor. As predicted
in H1, we found that participants trusted the perceiver signicantly
more in the emotional acknowledgment (M = 5.32, SD = 0.95) than the
non-emotional acknowledgment condition (M = 4.48, SD = 1.03), F(1,
193) = 117.45, p < .001,
η
2
= 0.38. We did not nd a main effect of
emotional valence—participants did not trust the perceiver more in the
positive (M = 4.93, SD = 0.08) than negative emotion condition (M =
4.91, SD = 0.08), F(1, 193) = 0.01, p = .91,
η
2
= 0.00.
As predicted in H3a, we found a signicant interaction between
emotional acknowledgment and valence on trust, F(1, 193) = 31.23, p =
.00,
η
2
= 0.14, indicating that the effect of acknowledgment depended
on the valence of the simulated emotion—see Fig. 3. For the positive
emotion condition, participants trusted the perceiver more when the
perceiver engaged in emotional (M = 5.13, SD = 1.09) than non-
emotional acknowledgment (M = 4.70, SD = 0.96), paired-t(96) =
4.23, p < .001, d = 0.43. However, for the negative emotion condition,
differences in trust evaluations were even more
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
124
pronounced—participants trusted the perceiver to a greater extent when
the perceiver engaged in emotional (M = 5.59, SD = 0.74) than non-
emotional acknowledgment (M = 4.26, SD = 1.05), paired-t(97) =
10.50, p < .001, d = 1.06.
5.2.2. Mediation
Next, we conducted mediation analyses (Hayes, 2013) to test
whether perceived costliness mediated the relationship between
emotional acknowledgment and interpersonal trust (H2). Using 5000
bootstrapped samples, we found a signicant mediation pathway from
emotional acknowledgment to interpersonal trust through perceived
costliness (indirect effect = 0.67, SE = 0.08, 95% CI = [0.52, 0.84]),
which supports H2.
5.2.3. Anticipated costs in the interaction
As an exploratory analysis, we examined the specic resource costs
participants anticipated from coworkers who engaged in emotional vs.
non-emotional acknowledgment. Participants expected the coworker to
continue the conversation longer when they engaged in emotional
acknowledgment (M = 10.12 min, SD = 6.17) than non-emotional
acknowledgment (M = 6.96 min, SD = 5.15), paired-t(194) = 8.48, p
< .001, d = .61
1
. Similarly, participants expected the coworker to
expend more attention when they engaged in emotional acknowledg-
ment (M = 67.98%, SD = 24.06) than non-emotional acknowledgment
(M = 50.46%, SD = 23.33), paired-t(194) = 10.01, p < .001, d = 0.72.
Finally, participants expected the coworker to expend more effort when
they engaged in emotional acknowledgment (M = 66.71%, SD = 22.46)
than non-emotional acknowledgment (M = 48.07%, SD = 23.27),
paired-t(194) = 9.75, p < .001, d = 0.70. In sum, when acknowledgment
was emotional in nature (as compared to non-emotional), participants
expected their coworker to expend signicantly more time, attention,
and effort.
Overall, these ndings replicate the results from Study 2 and provide
evidence that emotional acknowledgment leads to greater levels of trust
than non-emotional acknowledgment, because emotional acknowledg-
ment tends to be perceived as costlier for the perceiver.
6. Study 4: Moderation by reputational concern
In Studies 1–3, we found that emotional acknowledgment increased
trust because the perceiver was presumably seen as acting in the in-
terests of the expresser, rather than the self. However, reputational
concerns may also drive emotional acknowledgment, especially in
organizational contexts. In Study 4, we manipulate whether the
perceiver has high or low motivation to engage in impression manage-
ment (Leary & Kowalski, 1990). Prior research suggests that people
make attributions based on the perceived motives of the actor (Abele &
Wojciszke, 2014). When impression management motives are salient,
we expect that acknowledgment would still be seen as costlier than no
acknowledgment. However, the attribution that expressers make about
the costs may change. Specically, expressers may believe that per-
ceivers are willing to bear the costs of acknowledgment in part to serve
their own self-interest, not just to help the expresser, thereby weakening
the positive relationship between acknowledgment and trust. In sum, we
hypothesize that the relationship between emotional acknowledgment
and trust will be attenuated when the perceiver is highly motivated to
engage in impression management.
Furthermore, in addition to perceived costliness, we explore two
alternative mechanisms in this study: response importance and
empathic concern. We consider the possibility that emotional
acknowledgment increases trust because, in the eyes of the expresser,
this response is more useful for having their emotional needs met. For
example, to the extent that expressers experiencing negative emotions
require greater social support and intervention, emotional acknowl-
edgment should be a more meaningful response to the expresser because
it creates an impression that the perceiver is prepared to meet their
needs. We also consider the possibility that acknowledgment increases
trust because it signals that the perceiver demonstrates empathic
concern, or the tendency to respond to others with tender and
compassionate feelings (Batson, Fultz, & Schoenrade, 1987). Although
the pathways from emotional acknowledgment to trust are likely to be
multiply determined, we hypothesize that perceived costliness will be a
stronger mediator than response importance or empathic concern, and
that it will still remain a signicant mediator even after controlling for
these other two mechanisms. The design, analyses, and hypotheses for
Study 4 were pre-registered on AsPredicted.org (https://aspredicted.
org/if4ij.pdf).
Fig. 3. Interactive effects of type of acknowledgment and emotional valence on rst-party trust perceptions (Study 3). Note: The depicted bars represent mean values
of interpersonal trust, and error bars represent 95% condence intervals.
1
This analysis was conducted using a log-transformation of this variable to
account for positive skew.
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
125
6.1. Participants and procedures
Participants included 395 individuals recruited through Prolic
(49.6% female; M
age
= 34.58, SD
age
= 12.13). The sample was 7.8%
African-American, 13.7% Asian, 65.3% Caucasian, 7.8% Hispanic, and
5.3% other. This experiment employed a 2 (impression management
motive: high, low) × 2 (emotional acknowledgment: yes, no) between-
subjects design. Participants were randomly assigned to one of four
scenarios, in which they imagined themselves interacting with a
coworker. The scenarios were described as follows:
You were working on a presentation for work when your computer
crashes. The changes to your presentation were not saved, so you have to
start over.
You are feeling angry.
One of your coworkers is sitting nearby.
[In the high impression management condition: This coworker knows that
you need to ll out a performance review for them soon. Your perfor-
mance review will determine whether they get a job promotion. Because of
this, they have been acting in ways to get on your good side.]
In the acknowledgment condition, participants then read:
In this moment, your coworker sees you and says, “You look angry. Did
something happen?” This coworker has never acknowledged your emo-
tions before.
In the no acknowledgment condition, participants instead read:
In this moment, your coworker sees you, but then goes back to doing their
work. This coworker has never acknowledged your emotions.
Next, participants answered questions about their perceptions of this
coworker and lled out a set of demographic questions.
6.2. Measures
6.2.1. Interpersonal trust
Interpersonal trust (
α
= 0.87) was measured using the same items as
Studies 2 and 3.
6.2.2. Perceived costliness
Perceived costliness (
α
= 0.97) was measured using the same items as
Studies 2 and 3.
6.2.3. Response importance
Response importance (
α
= 0.90) was measured using a four-item
scale that captured how participants viewed the signicance of the
perceiver’s response (
α
= 0.96): “This coworker’s response was
[important]/[signicant]/[meaningful]/[useful]”.
6.2.4. Empathic concern
Empathic concern (
α
= 0.87) was measured using the seven-item
scale from Davis (1980; e.g., “This coworker often has tender, con-
cerned feelings for people less fortunate than him/her.”)
6.3. Results and discussion
6.3.1. Interpersonal trust
Replicating results from Studies 1–3, we found a main effect of
emotional acknowledgment on trust (H1). Participants trusted the
perceiver signicantly more in the acknowledgment (M = 4.26, SD =
1.10) than the no acknowledgment condition (M = 2.92, SD = 1.25), F
(1, 391) = 130.36, p < .001,
η
2
= 0.25. We also found a main effect of
impression management: participants trusted the perceiver more in the
low (M = 3.72, SD = 1.38) than high impression management condition
(M = 3.48, SD = 1.33), F(1, 391) = 4.24, p = .04,
η
2
= 0.009.
As hypothesized, we also found a signicant interaction between
emotional acknowledgment and impression management, F(1, 391) =
7.19, p = .008,
η
2
= 0.02, indicating that the effect of emotional
acknowledgment depended on whether the perceiver had high or low
impression management motives (Fig. 4). When the perceiver did not
have a salient motive to engage in impression management, participants
trusted the perceiver who acknowledged their emotion (M = 4.53, SD =
1.00) more than the one who did not (M = 2.87, SD = 1.20), t(193) =
10.52, p < .001, d = 1.51. However, when the perceiver was highly
motivated to engage in impression management, these differences were
attenuated. Participants still trusted the perceiver who acknowledged
their emotions (M = 4.00, SD = 1.14) more than the one who did not (M
= 2.97, SD = 1.31), but the effect was signicantly smaller, t(198) =
Fig. 4. Interactive effects of emotional acknowledgment and impression management motive on rst-party trust perceptions (Study 4). Note: The depicted bars
represent mean values of interpersonal trust, and error bars represent 95% condence intervals.
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
126
5.93, p < .001, d = 0.84.
6.3.2. Mediation including alternative mechanisms
As stated in our preregistration, we conducted our mediation ana-
lyses only within the low impression management condition, which
more closely resembled the context in our previous studies. Using 5000
bootstrapped samples, we found a signicant mediation pathway from
emotional acknowledgment to interpersonal trust through perceived
costliness (indirect effect = 1.47, SE = 0.16, 95% CI = [1.16, 1.79]),
supporting H2. Furthermore, as expected, we found that when entered
into the same model, perceived costliness (indirect effect = 1.19, SE =
0.17, 95% CI = [0.87, 1.56]) was a signicantly stronger mediator than
response importance (indirect effect = 0.08, SE = 0.08, 95% CI =
[−0.07, 0.25]) or empathic concern (indirect effect = 0.31, SE = 0.10,
95% CI = [0.13, 0.51]). Finally, we hypothesized and found that
perceived costliness persisted as a mediator, even after controlling for
response importance and empathic concern (indirect effect = 0.36, SE =
0.13, 95% CI = [0.14, 0.63]).
The ndings from Study 4 suggest that the perceiver’s motives in-
uence the interpretation of emotional acknowledgment. When the
perceiver was highly concerned with his/her own reputation, the effect
of emotional acknowledgment on trust was weakened, but did not fully
disappear. This was presumably because the act was perceived as more
self-, rather than other-oriented. Additionally, we found that empathic
concern, but not response importance, also mediated the relationship
between emotional acknowledgment and trust. However, perceived
costliness was a signicantly stronger mediator than these alternative
mechanisms.
7. Study 5: Experimental tests of H1–H3 from a third-party
perspective
Thus far, the studies have demonstrated the potential benets of
emotional acknowledgment on the expresser’s perceptions (i.e., rst-
party perceptions) of interpersonal trust. However, solely examining
rst-party perceptions of acknowledgment on interpersonal trust may
introduce confounds. Past research has documented the effect of various
emotional states on trust perceptions (Dunn & Schweitzer, 2005; Gino &
Schweitzer, 2008). These studies nd that emotions affect our cognition
(Schwarz & Clore, 2007)—negative emotions, such as anger, decrease
trust, whereas positive emotions, such as happiness and gratitude, in-
crease trust (Dunn & Schweitzer, 2005). Indeed, in Study 2, we found a
main effect of valence: participants trusted the perceiver signicantly
more in the positive (vs. negative) emotion condition. As such, the rst-
party perspective makes it difcult to disentangle whether the effects on
trust are due to emotional acknowledgment or the subjective experience
of the emotion. Additionally, rst-party expressers may be particularly
sensitive to the costliness of emotional acknowledgment because they
stand to receive direct benets from the act. However, according to
Costly Signaling Theory, because emotional acknowledgment is an
observable response that signals important underlying qualities, the
costs of the act should also be discernable by third-party observers.
Moreover, the fact that third-party observers have little or no “skin in the
game” may make tests of the hypotheses from a third-party perspective
more conservative than a rst-party perspective.
Thus, in Study 5, we investigated the relationship between emotional
acknowledgment and trust (H1- H3) from a third-party perspective to
mitigate these potential confounds and strengthen our empirical evi-
dence. In particular, we hired two professional actors and created videos
of two coworkers interacting in a workplace breakroom. In the videos,
one actor played the role of expresser, and the other the role of
perceiver. We created a separate video to represent each condition of the
experiment—the scene depicted in each video was altered to manipulate
emotional acknowledgment and valence. Participants were randomly
assigned to watch and respond to questions about one of the videos.
Furthermore, in addition to no acknowledgment, we included
another control to compare against emotional acknowledgment:
directing the conversation to a topic unrelated to emotions (i.e.,
distraction). Past work on interpersonal emotion regulation suggest that
people can manage an expresser’s emotions by either facilitating
engagement or disengagement from the emotional situation (Pauw,
Sauter, van Kleef, & Fischer, 2019). Whereas emotional acknowledg-
ment presumably facilitates engagement, distraction should facilitate
disengagement from the emotional situation. As such, we included a
distraction condition to compare the efcacy of an engagement and
disengagement approach. Additionally, the distraction condition facili-
tates a fairer comparison than the no acknowledgment condition, as the
amount of verbal interaction in the distraction and emotional
acknowledgment conditions are more similar (Cooper & Richardson,
1986). We hypothesized that emotional acknowledgment would in-
crease trustworthiness more than no acknowledgment or distraction,
because engaging with emotions directly should be a costlier signal of
the perceiver’s willingness to help the expresser.
7.1. Participants and procedures
The sample included 541 participants recruited through Mechanical
Turk (38.4% female; M
age
= 36.40, SD
age
= 10.24). Our sample was
25.0% African-American, 4.4% Asian, 63.5% Caucasian, 5.7% Hispanic,
and 2.4% other. This experiment employed a 2 (valence of expressed
emotions: positive, negative) × 3 (emotional acknowledgment:
acknowledgment, no acknowledgment, distraction) between-subjects
design.
All participants read the same background story about Daniel and
Brandon, who were purportedly two employees working at a publishing
company. Then, to provide context, participants were told: “You will see
a video of Daniel and Brandon’s interaction in the breakroom. Specif-
ically, Brandon receives an email with some good [bad] news when
Daniel walks into the breakroom”. Next, participants were randomly
assigned to watch one of six videos (see Online Supplement), each
representing one of the six conditions.
To create the six videos, we hired two professional actors to play two
colleagues interacting at work. The actors were instructed to dress in
business attire and to thoroughly memorize and rehearse the script
before arriving on set. We recorded the videos in a workplace break-
room. All six videos start with Brandon (the expresser) eating lunch and
checking his email. As he reads his email, he either displays a happy or
angry emotional expression. Then, Daniel (the perceiver), walks into the
breakroom, and grabs his lunch. He sits down next to Brandon, looks at
Brandon, and pauses for a moment. In the emotional acknowledgment
condition, Daniel acknowledges Brandon’s emotions by saying, “You
seem happy [upset]”. In the no acknowledgment condition, Daniel says
nothing. In the distraction condition, Daniel redirects attention away
from the emotional expression by saying “Did you catch the game last
night?” Each video was about 40 s in length. Following the video, par-
ticipants were asked about their perceptions of Daniel and to ll out
demographic questions.
7.2. Measures
7.2.1. Interpersonal trust
Similar to the prior studies, interpersonal trust (
α
= 0.89) was
measured using a four-item scale adapted from Mayer and Davis (1999):
e.g., “Daniel looks out for what is important to other people”.
7.2.2. Perceived costliness
Perceived costliness (
α
= 0.95) was measured using adapted versions
of the ve items used in Studies 2–4: “Daniel is willing to spend
[attention]/[effort]/[time]/[energy]/[personal resources] on the
relationship”.
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
127
7.3. Results and discussion
7.3.1. Interpersonal trust
First, we examined the effect of emotional acknowledgment and
valence on interpersonal trust. In line with H1, we found a main effect of
emotional acknowledgment on trust, F(2, 535) = 27.97, p < .001,
η
2
=
0.09. Specically, emotional acknowledgment (M = 5.39, SD = 1.01) led
to greater perceptions of trust than no acknowledgment (M = 4.47, SD =
1.34), t(36 3) = 7.51, p < .001, d = 0.79, and distraction (M = 4.95, SD
= 1.26), t(360) = 3.73, p < .001, d = 0.39. We also found a main effect
of emotional valence, F(1, 535) = 4.34, p = .04,
η
2
= 0.007. Participants
trusted the perceiver more in the positive (M = 5.04, SD = 1.19) than
negative emotion condition (M = 4.83, SD = 1.34). However, the size of
this main effect was relatively small (
η
2
= 0.007), considerably smaller
than the equivalent effect size from Study 2 (
η
2
= 0.04). This makes
sense, as we would expect the effects of positive and negative emotion
on trust found in past research (Dunn & Schweitzer, 2005) to be weaker
from a third-party than a rst-party perspective.
Furthermore, using a two-way ANOVA, we found a signicant
interaction between emotional acknowledgment and valence on trust
(Fig. 5), F(2, 535) = 7.85, p < .001,
η
2
= 0.03. Using contrast coding, we
probed the interaction by comparing the acknowledgment vs. no
acknowledgment, and acknowledgment vs. distraction for the negative
and positive emotion conditions. The effect of acknowledgment (vs. no
acknowledgment) on trust was signicantly greater for negative than
positive emotions, B = 0.59, SE = 0.25, t = 2.37, p = .02. Tukey-HSD
multiple comparisons revealed a signicant difference between
acknowledgment (M = 5.55, SD = 0.84) and no acknowledgment (M =
4.31, SD = 1.45) for the negative emotion condition, p < .0001. For the
positive emotion condition, the difference between acknowledgment (M
= 5.25, SD = 1.12) and no acknowledgment (M = 4.61, SD = 1.12) was
signicant but weaker than in the negative emotion condition, p =
.0005.
Likewise, the effect of acknowledgment (vs. distraction) on trust was
signicantly greater for negative than positive emotions, B = 0.98, SE =
0.25, t = 3.93, p < .001. Tukey-HSD multiple comparisons revealed a
signicant difference between acknowledgment (M = 5.55, SD = 0.84)
and distraction (M = 4.57, SD = 1.32) for the negative emotion condi-
tion, p < .0001. In contrast, there was virtually no difference between
acknowledgment (M = 5.25, SD = 1.12) and distraction (M = 5.26, SD =
1.12) for the positive emotion condition, p = 1.00. In sum, these results
support H3a, as acknowledging negative emotions was a more potent
driver of trust than acknowledging positive emotions.
7.3.2. Mediation and moderated mediation
Next, we conducted mediation analyses to test H2 and H3b. Using
5000 bootstrapped samples, we found a signicant mediation pathway
from emotional acknowledgment to interpersonal trust through
perceived costliness for acknowledgment vs. distraction (indirect effect
= 0.34, SE = 0.09, 95% CI = [0.16, 0.53]), as well as acknowledgment
vs. no acknowledgment (indirect effect = 0.77, SE = 0.10, 95% CI =
[0.57, 0.97]). These results provide additional support for H2.
Next, we used moderated mediation models to test H3b. When
comparing acknowledgment and no acknowledgment, moderated
mediation was supported (index of moderated mediation = 0.55, SE =
0.20, 95% CI = [0.17, 0.96]). The effect of emotional acknowledgment
(vs. no acknowledgment) on interpersonal trust through perceived
costliness was stronger for negative emotion (indirect effect = 1.06, SE
= 0.16, 95% CI = [0.76, 1.39]) than positive emotion (indirect effect =
0.51, SE = 0.13, 95% CI = [0.25, 0.77]). These results support H3b.
When comparing acknowledgment and distraction, moderated media-
tion was also supported (index of moderated mediation = 0.79, SE =
0.18, 95% CI = [0.43, 1.16]). The effect of emotional acknowledgment
(vs. distraction) on interpersonal trust through perceived costliness was
signicant for negative emotions (indirect effect = 0.77, SE = 0.14, 95%
CI = [0.49, 1.05]), but not positive emotions (indirect effect = −0.02,
SE = 0.12, 95% CI = [−0.25, 0.22]).
The results from Study 5 suggest that the effects of emotional
acknowledgment on trust posited in H1–H3 replicate when using eval-
uations of the perceiver from a third-party perspective. Participants
rated the perceiver as more trustworthy when they engaged in
acknowledgment rather than no acknowledgment or distraction (H1)
Fig. 5. Interactive effects of emotional acknowledgment and emotional valence on third-party trust perceptions (Study 5). Note: The depicted bars represent mean
values of interpersonal trust, and error bars represent 95% condence intervals.
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
128
because acknowledgment was seen as costlier (H2). Additionally,
consistent with Studies 2 and 3, we found that the relationship between
emotional acknowledgment and trust was moderated by the valence of
the expresser’s emotions (H3a). The effect of emotional acknowledg-
ment was stronger when the expresser displayed negative than positive
emotion, because participants perceived acknowledging negative (vs.
positive) emotions as costlier (H3b). Interestingly, in the case of positive
emotions, perceivers who engaged in distraction were seen as equally
trustworthy as perceivers who acknowledged the emotion. However, in
the case of negative emotions, distraction was less effecti-
ve—participants rated perceivers who distracted as less trustworthy
than those who acknowledged the emotion directly. These results
highlight that third-party observers may be more vigilant of how per-
ceivers respond to emotions in the wake of witnessing negative, as
opposed to positive, emotional displays. This may be because responses
to negative emotions, which come at a greater cost to the perceiver’s
self-interest, are more diagnostic of the perceiver’s benevolent motives.
Until this point, the language and design of our studies have implied
that when perceivers acknowledged the expressers’ emotions, they are
accurate in their assessment of expressers’ emotions. However, people
differ widely in their abilities to accurately perceive emotions (Mayer &
Salovey, 1997), and even people relatively high in emotional intelli-
gence still make mistakes. Because we have only explored accurate
forms of emotional acknowledgment, it is unclear whether accuracy is a
necessary condition to reap the benets of emotional acknowledgment.
In Study 6, we explore this potential boundary condition by examining
how acknowledgment accuracy inuences perceptions of trust. In so
doing, we address the possibility that our results were driven solely by
the perceiver’s ability to accurately infer others’ emotional states (Ickes,
1993), rather than the perceiver’s decision to acknowledge the emotion.
8. Study 6: Examination of the role of accuracy
In Study 6, we compare the effects of accurate, inaccurate, and no
emotional acknowledgment. We used a similar paradigm and set of
video stimuli as Study 5, but we built on the prior studies with the
intention of accomplishing three goals.
First, whereas our prior studies examined the decision to acknowl-
edge (or not), here we also tested the accuracy (or inaccuracy) of
emotional acknowledgment. Overall, we predicted accurate acknowl-
edgment would increase trust more than inaccurate acknowledgment
because accuracy signals greater attentiveness, thereby indicating a
willingness to expend resources, as well as a greater capacity to provide
helpful support to the expresser. Following a similar logic, we also ex-
pected inaccurate acknowledgment to foster greater trust than no
acknowledgment, because inaccurate acknowledgment, although mis-
directed, would still signal greater effort by the perceiver (Goldstein,
Vezich, & Shapiro, 2014). Thus, we still expected perceived costliness to
mediate the relationship between inaccurate acknowledgment and trust,
but we also expected the effects to be stronger overall for accurate than
inaccurate acknowledgment.
Hypothesis 4. When individuals acknowledge others’ emotions
accurately, they are seen as more trustworthy than when they
acknowledge inaccurately.
Hypothesis 5. The relationship between emotional acknowledg-
ment accuracy and interpersonal trust is mediated by perceptions of
costliness.
Consistent with the logic of H3, we expected the advantage of ac-
curacy over inaccuracy to be especially strong when the emotion
expressed is negative (as opposed to positive). Inaccurately acknowl-
edging an expresser’s positive emotions (by verbally labeling them as
negative), signals an eagerness and readiness to provide resources to the
expresser, even when the expresser’s needs are relatively low. However,
inaccurately acknowledging an expresser’s negative emotions (by
labeling them as positive), should signal a lack of ability or will to
provide for the expresser, even when the expresser’s needs are relatively
high. As such, inaccurately acknowledging an expresser’s negative
emotions should receive a greater penalty than inaccurately acknowl-
edging an expresser’s positive emotions.
Hypothesis 6a. The relationship between emotional acknowledg-
ment accuracy and interpersonal trust is moderated by the valence of
the expresser’s emotion, such that the relationship is stronger when
the emotions expressed are negative as opposed to positive.
Hypothesis 6b. This moderation occurs because negative emotions
are seen as costlier to acknowledge than positive emotions.
Our second goal was to explore whether empathic accuracy—the
ability to infer the thoughts and feelings of others (Ickes, 1993)—could
serve as an alternative theoretical account for our ndings. Because
perceivers in our previous studies accurately acknowledged the ex-
presser’s emotions, it is plausible that the relationship between
emotional acknowledgment and trust was driven not by the costly act
itself, but by the perceiver’s accurate identication of the emotion.
However, if a perceiver acknowledges an expresser’s emotions inaccu-
rately, then the predictions made by costly signaling and empathic ac-
curacy diverge, thereby allowing us to more clearly delineate the
mechanism. By comparing the effects of inaccurate, accurate, and no
acknowledgment, we were able to compare costly signaling against
empathic accuracy.
Our third goal was to test whether results held across different
negative and positive emotions, complementing the focus on anger and
happiness in Studies 2–5. Thus, in Study 6, we used a different pair of
discrete emotions: anxiety and excitement. We selected these emotions
because anxiety and excitement are comparable on arousal level, but
opposing in valence (Russell, 1980). Furthermore, because the behav-
ioral and physiological manifestations of anxiety and excitement are
similar (Brooks, 2014), it would be more plausible for perceivers to
mistake one emotion for the other.
8.1. Participants, procedures, and measures
Participants included 581 individuals who were recruited through
Mechanical Turk (37.6% female; M
age
= 38.36, SD
age
= 11.92). The
sample was 10.0% African-American, 10.3% Asian, 69.7% Caucasian,
6.9% Hispanic, and 3.1% other.
This experiment employed a 2 (valence of expressed emotions: pos-
itive, negative) × 3 (emotional acknowledgment: accurate, inaccurate,
no acknowledgment) between-subjects design. As in Study 5, partici-
pants were randomly assigned to watch a video of two colleagues
interacting in a breakroom at work. However, in these videos, Brandon
either shows anxiety or excitement. Daniel responds by saying, “You
seem anxious [excited]” in the acknowledgment conditions, or nothing
at all in the no acknowledgment condition. Still shots of the videos are
shown in Fig. 6, and links to the full videos are in the Online Supple-
ment. We operationalized accurate acknowledgment as a match be-
tween the expresser’s emotions and the perceiver’s verbal
acknowledgment, and inaccurate acknowledgment as a mismatch.
Specically, we coded acknowledgment as inaccurate if Brandon looked
anxious, but Daniel said, “You seem excited”, or if Brandon looked
excited, but Daniel said, “You seem anxious”. We used the same items as
Study 5 to measure interpersonal trust (
α
= 0.89) and perceived costli-
ness (
α
= 0.96).
8.2. Results and discussion
8.2.1. Interpersonal trust
First, we examined how the accuracy of emotional acknowledgment
and emotional valence inuenced perceptions of trust. We found that the
three acknowledgment conditions (accurate, inaccurate, no) signicantly
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
129
differed in ratings of trust, F(2, 575) = 29.62, p < .001,
η
2
= 0.09.
Consistent with H1 and Study 5, accurate emotional acknowledgment (M
= 5.04, SD = 1.06) led to signicantly higher ratings of trust than no
acknowledgment (M = 4.16, SD = 1.22), t(3 91) = 7.61, p < .001, d =
0.77. In support of H4, we also found that accurate emotional acknowl-
edgment led to signicantly higher ratings of trust than inaccurate
emotional acknowledgment (M = 4.63, SD = 1.20), t(3 82) = 3.55, p <
.001, d = 0.36, and that inaccurate acknowledgment led to signicantly
higher ratings of trust than no acknowledgment, t(383) = 3.81, p < .001,
d = 0.39. In contrast to Study 5, we did not nd a signicant main effect
of emotional valence on trust, F(1, 575) = 1.79, p = .18,
η
2
= 0.003.
Participants did not trust the perceiver signicantly more in the positive
(M = 4.67, SD = 1.17) than negative emotion condition (M = 4.55, SD =
1.26), t(579) = 1.25, p = .21, d = 0.10. Given that the size of this main
effect was fairly small in Study 5 (
η
2
= 0.007), it is not surprising that the
equivalent main effect was not signicant in Study 6 (
η
2
= 0.003). As
expected, both of these main effects were much smaller compared to that
of the rst-party perspective in Study 2 (
η
2
= 0.04).
Using a two-way ANOVA, we found a signicant interaction between
emotional acknowledgment and valence on trust (Fig. 7), F(2, 575) =
13.44, p < .001,
η
2
= 0.04. To test H3a in the same fashion as Study 5, we
conducted planned contrasts. When comparing accurate vs. no
acknowledgment, the effect of acknowledgment on trust was signi-
cantly greater for negative than positive emotions, B = 0.73, SE = 0.23, t
= 3.19, p = .001. Tukey-HSD multiple comparisons revealed a signi-
cant difference between accurate acknowledgment (M = 5.29, SD =
1.00) and no acknowledgment (M = 4.04, SD = 1.21) for the negative
emotion condition, p < .001. The difference between accurate
acknowledgment (M = 4.79, SD = 1.07) and no acknowledgment (M =
4.27, SD = 1.24) was also signicant but weaker for the positive emotion
condition, p = .005. These results provide additional support for H3a.
Next, consistent with H6a, the advantage of accurate (vs. inaccurate)
acknowledgment in increasing trust was signicantly greater for nega-
tive than positive emotions, B = 1.19, SE = 0.23, t = 5.13, p < .001.
Tukey-HSD multiple comparisons revealed a signicant difference be-
tween accurate and inaccurate acknowledgment (M = 4.28, SD = 1.22)
for the negative emotion condition, p < .001. In contrast, we did not nd
a signicant difference between accurate and inaccurate acknowledg-
ment for the positive emotion condition (M = 4.97, SD = 1.07), p = .78.
These results support H6a.
Expresser displays excitement Expresser displays anxiety
Perceiver acknowledges emotion Perceiver does not acknowledge emotion
Fig. 6. Still shots from video stimuli (Study 6).
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
130
As exploratory analyses, we also compared the inaccurate and no
acknowledgment conditions by valence. In the negative emotion con-
dition, the difference between inaccurate and no acknowledgment was
not signicant, p = .44. However, in the positive emotion condition,
inaccurate acknowledgment led to higher perceptions of trust than no
acknowledgment, p < .001.
In sum, these results suggest that accuracy mattered more in
response to negative than positive emotions. When negative emotions
were expressed, perceivers needed to acknowledge accurately in order
to reap a signicant benet. However, when positive emotions were
expressed, perceivers benetted from acknowledgment regardless of
whether it was accurate or not. This is consistent with a costly signaling
explanation. When the needs of the expresser were higher, as was the
case for negative emotions, participants may have looked for costlier
signals that the perceiver was willing to allocate resources to the
expresser, and thus acknowledgment had to be accurate in order to
increase trust. However, when the needs of the expresser were lower, as
was the case for positive emotions, the less costly signal of inaccurate
acknowledgment was enough to signicantly increase trust.
8.2.2. Mediation and moderated mediation
We conducted mediation analyses to test H2 and H5. Using 5000
bootstrapped samples, we found a signicant mediation pathway from
emotional acknowledgment (accurate vs. no) to interpersonal trust
through perceived costliness (indirect effect = 0.80, SE = 0.10, 95% CI
= [0.62, 1.00]). This replicates the results for H2 from Study 5.
Furthermore, we also found a signicant mediation pathway from
acknowledgment accuracy (accurate vs. inaccurate) to interpersonal
trust through perceived costliness (indirect effect = 0.30, SE = 0.09,
95% CI = [0.13, 0.49]). These results support H2 and H5, respectively.
Next, we tested H3b and H6b. When comparing accurate and no
acknowledgment, moderated mediation was supported (index of
Fig. 7. Interactive effects of acknowledgment accuracy and emotional valence on third-party trust perceptions (Study 6). Note: The depicted bars represent mean
values of interpersonal trust, and error bars represent 95% condence intervals.
Fig. 8. Additive effects of content and accuracy of emotional acknowledgment on third-party trust perceptions (Study 6). Note: The depicted bars represent mean
values of interpersonal trust, and error bars represent 95% condence intervals. The interaction was not signicant.
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
131
moderated mediation = 0.54, SE = 0.18, 95% CI = [0.17, 0.89]).
Replicating previous ndings, the effect of emotional acknowledgment
on interpersonal trust through perceived costliness was stronger for
negative (indirect effect = 1.06, SE = 0.13, 95% CI = [0.82, 1.32]) than
positive emotions (indirect effect = 0.52, SE = 0.13, 95% CI = [0.26,
0.79]), supporting H3b. When comparing accurate and inaccurate
acknowledgment, moderated mediation was also supported (index of
moderated mediation = 0.70, SE = 0.18, 95% CI = [0.35, 1.04]). The
effect of acknowledgment accuracy on interpersonal trust through
perceived costliness was signicant for negative (indirect effect = 0.65,
SE = 0.13, 95% CI = [0.40, 0.89]), but not positive emotions (indirect
effect = −0.05, SE = 0.12, 95% CI = [−0.28, 0.20]), supporting H6b.
8.2.3. Examining empathic accuracy as an alternative theoretical account
Finally, we wanted to examine whether empathic accuracy could
serve as an alternative explanation. According to an empathic accuracy
account, accurate acknowledgment should lead to higher levels of trust
than inaccurate acknowledgment, regardless of emotional valence,
because perceivers are generally evaluated more favorably when they
are able to take the perspective of others (Goldstein et al., 2014; Ickes,
1993; Klein, 2019). However, our results suggested otherwise. Inaccu-
rate acknowledgment incurred a penalty only for negative, but not
positive emotions. To shed insight into why that might be the case, we
recoded our manipulations to focus on the content of the acknowledg-
ment itself (i.e., the words that the acknowledger used), rather than the
valence of the expresser’s emotional displays. According to a costly
signaling account, acknowledgment using negatively-valenced content
(i.e., “You seem anxious”) should lead to higher levels of trust than
acknowledgment of positively-valenced content (i.e., “You seem
excited), even if the acknowledgment is inaccurate—acknowledgment
of negatively-valenced content should still imply an eagerness and
willingness to expend resources on the expresser. Therefore, by exam-
ining the content of the acknowledgment, we can distinguish the effects
of empathic accuracy from costly signaling.
To test this, we excluded the no acknowledgment condition because
it could not be coded for accuracy or content. Then, we examined how
the accuracy and content of the acknowledgment inuenced trust per-
ceptions (Fig. 8). As aforementioned, accurate acknowledgment led to
higher ratings of trust than inaccurate acknowledgment, F(1, 380) =
28.41, p < .001,
η
2
= 0.03. However, acknowledgment using negatively-
valenced content (M = 5.13, SD = 1.05) also led to higher ratings of trust
than using positively-valenced content (M = 4.54, SD = 1.17), F(1, 380)
= 28.4, p < .001,
η
2
= 0.07. The interaction between accuracy and
content was not signicant, F(1, 380) = 0.79, p = .38,
η
2
= 0.002.
Overall, these results suggest that empathic accuracy is necessary,
but insufcient, in explaining our results. Inaccurately acknowledging
others’ emotions signals inattention, but inaccurately acknowledging
positive emotions (by saying, “You seem anxious”) may not incur a
penalty because the perceiver is still demonstrating a willingness to
provide resources to the expresser, even if the expresser does not have a
need for them. On the other hand, inaccurately acknowledging negative
emotions (by saying, “You seem excited”) incurs a signicant penalty
because the perceiver appears ill-prepared to meet the expressers’
heightened needs. As such, empathic accuracy and costly signaling
appear to have additive effects on perceptions of trust.
By examining the role of accuracy in emotional acknowledgment in
Study 6, we found support for H4–H6, as well as additional support for
H1–H3, that bolsters our costly signaling account. Differences in trust
levels between acknowledging inaccurately (vs. accurately) was stron-
ger for negative emotions than positive emotions. In line with our
theorizing, we also found that acknowledgment containing negatively-
valenced content led to higher ratings of trust than acknowledgment
containing positively-valenced content, even when the acknowledgment
was inaccurate. These results suggest that empathic accuracy alone
could not explain our pattern of results. Finally, our ndings were
consistent with those in Study 5, even though we substituted happiness
and anger for excitement and anxiety, respectively. This hints that the
relationship between emotional acknowledgment and valence may not
be conned to specic emotions, but rather, may generalize across
multiple positive and negative discrete emotions.
9. General discussion
Despite the importance of responsiveness and using verbal language
in emotional communication, the topic of emotional acknowledgment
has received relatively scarce attention. We utilize Costly Signaling
Theory to posit that emotional acknowledgment increases interpersonal
trust by acting as a costly signal that demonstrates the willingness of
perceivers to expend personal resources to meet the needs of expressers.
We found convergent support for our hypotheses across six studies. Our
results suggest that costlier forms of emotional acknowledgment (i.e.,
acknowledging negative rather than positive emotions) increased trust
to a greater extent. When positive emotions were expressed, even
inaccurate emotional acknowledgment was better for fostering trust
than no acknowledgment. But when negative emotions were expressed,
emotional acknowledgment needed to be accurate to increase trust.
9.1. Theoretical contributions
This research makes key contributions to theory and research on
emotions and relationships in organizations. First, existing work on so-
cial emotions predominantly highlights how the emotions of expressers
affect the inferences drawn by perceivers (e.g., EASI; van Kleef, 2009).
Our investigation extends social theories of emotions by focusing on the
inferences that are made about perceivers based on their responses to the
emotion of expressers. We contend that this perspective is critical to
understanding the role of emotions in social interactions and relation-
ships. Existing social theories of emotions conceptualize the inuence of
emotions as unidirectional (i.e., expresser → perceiver; van Kleef, 2009,
2016). Building on this view, we suggest that emotional communication
requires sensemaking by both parties, and may be better conceptualized
as bidirectional (i.e., expresser ⇄ perceiver). We focus on this perspec-
tive by illuminating how responding verbally to emotions shapes
interpersonal relationships.
Second, we integrate Costly Signaling Theory with social theories of
emotions to elucidate the process by which emotional acknowledgment
shapes social relationships. This novel perspective illuminates why
perceivers reap asymmetric rewards for the acknowledgment of nega-
tive and positive emotions. Signaling a willingness to attend to others’
needs, especially if it comes at a greater personal expense, helps foster
the development of trust. Costly Signaling Theory sheds light on how
people infer relationship quality using acknowledgment and offers new
predictions in social emotions that diverge from existing theories, such
as empathic accuracy (Ickes, 1993) and self-verication theory (Swann,
2012). These theories suggest that people seek information that accu-
rately validates their thoughts and feelings, whereas our results suggest
that emotional acknowledgment can increase trust even when it is
inaccurate (as was the case for positive emotions). Thus, Costly
Signaling Theory offers a new and valuable framework for under-
standing how people draw inferences about one another in emotional
communication.
Third, this research extends previous work on emotional labeling
that focuses almost exclusively on using language to express or manage
one’s own emotions at work (e.g., Brooks, 2014; Wolf et al., 2016).
Although interpersonal emotion labelling techniques are used in psy-
chotherapy and clinical settings (Greenberg, 2004), our research docu-
ments the effects of verbally acknowledging others’ emotions on
important relational outcomes in the workplace (i.e., interpersonal
trust). Furthermore, the asymmetric costs of inaccurately acknowl-
edging negative versus positive emotions suggest that when perceivers
are uncertain about what emotions are being expressed, it may be
strategically advantageous for them to use negatively-valenced (“You
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
132
seem upset”) than positively-valenced (“You seem happy”) forms of
acknowledgment. This prescription stands in contrast to research on
labeling one’s own emotions, which suggests that relabeling negative
emotions as positive emotions leads to more favorable outcomes (Wolf
et al., 2016). In this way, our research highlights the importance of
distinguishing between emotional labeling at the intrapersonal and
interpersonal levels.
Fourth, we extend research on active listening and partner respon-
siveness by focusing on acknowledgment of nonverbal displays. This
complements prior research, which relies heavily on analyzing conver-
sations, and thus, focuses predominantly on responses to what was
explicitly and verbally communicated by the speaker (Jones, 2011;
Maisel, Gable, & Strachman, 2008; Weger, Bell, Minei, & Robinson,
2014). Our research highlights the importance of distinguishing be-
tween responses to nonverbal displays and verbal remarks. Because
nonverbal displays tend to be more ambiguous and diffuse than verbal
remarks, acknowledgment of nonverbal displays may be perceived as
riskier and more volitional, and thus a better indicator of the perceiver’s
benevolent intent. Whereas responses to verbal remarks are well-
documented, future research may benet from organizing and classi-
fying responses to nonverbal displays.
Lastly, we contribute to the literature examining how and when
people draw inferences about others’ trustworthiness in the workplace
(Schilke & Huang, 2018; van der Werff & Buckley, 2017). By identifying
emotional acknowledgment as a previously unexplored behavioral in-
dicator of trust, we heed the call by previous researchers to further
explore the role of emotions in trust formation (Schoorman, Mayer, &
Davis, 2007). Whereas scholars have found positive associations be-
tween trust and emotional expressivity (Boone & Buck, 2003), positive
emotion (Dunn & Schweitzer, 2005), and situationally appropriate
expression of emotion (Kang & Schweitzer, 2020), we extend beyond the
role of emotional states and expressions to examine the role of labeling
emotions on the formation of trust.
9.2. Limitations and future directions
Our investigation into the relationship between emotional
acknowledgment and interpersonal trust can be strengthened in several
ways. First, our experimental studies focus on immediate judgments
formed about those who acknowledge others’ emotions. However,
emotional acknowledgment can also foster trust, not only directly
through the act itself, but also indirectly through self-disclosure. Spe-
cically, we theorize (but do not directly test) that when perceivers
acknowledge the emotions of expressers, it is likely to prompt the
expresser to disclose information about their internal affective states. As
past research suggests, engaging in self-disclosure of emotions may be
more inuential for closeness and social connection than self-disclosure
of facts (Laurenceau et al., 1998). Future research could examine the
extent to which emotional acknowledgment licenses expressers to talk
about their emotions, and whether this emotional self-disclosure shapes
downstream perceptions of trust.
In our experimental studies, it is possible that when perceivers fail to
acknowledge others’ emotions, participants infer a lack of ability to
encode others’ emotional expressions. Research suggests that perceived
decits in emotional intelligence and empathic accuracy may lead to
negative social consequences (Elfenbein, Barsade, & Eisenkraft, 2015;
Zickfeld, Schubert, Seibt, & Fiske, 2017). As such, differences in trust
between acknowledgment and no acknowledgment may be attributed to
this lack of ability, rather than the decision to verbally call out the
emotion. Future research can more explicitly delineate the effects of
noticing versus acknowledging others’ emotions by exploring contexts
in which perceivers are able to make accurate judgments about the other
person’s emotions but choose not to acknowledge it verbally.
The current investigation only examines one type of acknowledg-
ment inaccuracy—perceivers mislabeling the valence of the expresser’s
emotions. However, there are many other ways in which emotions can
be inaccurately acknowledged. For example, perceivers may misidentify
a specic discrete emotion (i.e., mistaking anger for disgust) or over- or
under-estimate the intensity of the displayed emotions (Klein, 2019).
Future research can examine whether various forms of inaccuracy
differentially impact trust perceptions.
In Study 4, we explore reputational concern as it pertains to the
perceiver’s relationship with the expresser. However, perceivers may
strategically engage in emotional acknowledgment as a way to bolster
how they are viewed, not only by the expresser, but by third-party ob-
servers. For example, perceivers may choose to acknowledge in the
presence of others rather than in private. Similar to Study 4, we expect
public acknowledgment to be viewed less favorably than private
acknowledgment, because people may have difculty distinguishing
whether the act was motivated by self- or other-interest. A promising
avenue for future research may be to explore how reputational concerns
involving third-parties affect perceptions of acknowledgment.
In our studies, we examine contexts where the expresser and
perceiver already have an established relationship. However, past
research suggests that people are more likely to express and share
negative emotions with friends and family (Fischer & Manstead, 2008).
As such, when a perceiver and expresser already have a standing history,
the normative expectation may be for perceivers to acknowledge the
expresser’s negative emotions. This may help to explain why we found a
greater penalty for not acknowledging negative emotions than a reward
for doing so. However, if the expresser and perceiver are merely ac-
quaintances or strangers, the normative expectations may shift such that
perceivers would not be expected to acknowledge negative emotions.
But if they do, we suspect that they might receive a greater reward for
acknowledging negative emotions than penalty for not acknowledging.
Future research can explore how the nature of the relationship may
change expectations around emotional acknowledgment.
Although our studies portray emotional acknowledgment in a largely
positive light, many contextual and individual factors may increase the
risks associated with emotional acknowledgment. For example, orga-
nizations often impose display rules that encourage positive emotional
expressions and discourage negative ones (Hochschild, 1983; Van
Maanen, 1991). We suspect that in organizations with strong norms to
suppress negative emotions, acknowledgment of negative emotions may
backre because it may be seen as insensitive or inappropriate, pre-
venting the expresser from “saving face”. For example, if a coworker has
an inadvertent outburst of emotion during a meeting, it may actually be
more benecial to feign ignorance than call greater attention to it
(Elfenbein, 2007). Indeed, one study nds that employees with a higher
ability to “eavesdrop” on others’ negative emotions received worse
performance ratings from peers, presumably because people prefer to
keep their negative emotions hidden at work (Elfenbein & Ambady,
2002). As such, unpacking organizational norms on emotional expres-
sivity may be critical for understanding the inuence of emotional
acknowledgment.
An expresser’s willingness to accept the perceiver’s interpretation of
their emotion may also heighten the risk of emotional acknowledgment.
Employees may feel ambivalent (Rothman & Melwani, 2017) or wear
expressions that differ from their internal feeling states (Hochschild,
1983). As a starting point, our studies portray a simplistic view of an
expresser’s emotions—the emotions they display are congruous with
how they feel. How might the consequences of acknowledgment change
when expressers’ internal states and external displays are misaligned, or
when they experience ambivalent emotions? We suspect that the
self-presentational goals of the expresser may be important to consider.
For example, if a coworker attempts to conceal their grief by looking
cheerful, perhaps acknowledging their negative emotions may lead to
negative consequences because it conicts with the coworker’s
self-presentational goals. We also suspect that when expressers are un-
aware of their own nonverbal displays or internal feelings, they are more
likely to challenge the perceiver’s acknowledgment, potentially viewing
it as threatening or offensive. As such, future research may benet from
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
133
examining how the preferences, goals, and abilities of the expresser
shape perceptions of emotional acknowledgment.
Finally, although we focus on emotional acknowledgment from the
perspective of expressers and third-party observers, it is also important
to consider the factors contributing to a perceiver’s decision to
acknowledge. Research suggests that people choose not to ask sensitive
questions because they (inaccurately) predict that they will make others
feel uncomfortable and harm their own interpersonal reputations (Hart,
VanEpps, & Schweitzer, 2021). As such, although acknowledging
negative emotions leads to greater increases in trust, perceivers may be
less willing to do so because negative emotions are presumably more
sensitive in nature. Future studies can examine the contexts and pro-
cesses under which perceivers acknowledge others’ emotions.
10. Practical implications and conclusion
Our research illuminates how emotional acknowledgment may serve
as a useful tool for individuals to build and strengthen connections with
others. Given the prevalence of emotions in everyday life, employees
and managers likely have numerous opportunities to acknowledge
others’ emotions in a typical workweek. Our research helps unpack the
potential benets and risks of acting on these opportunities. Although
emotional acknowledgment may be riskiest in response to negative
emotions, people also stand to benet the most from (accurately)
acknowledging others’ negative emotions. Contrary to the view that
emotions should be downplayed, constrained, and concealed (Ashforth
& Humphrey, 1995), our research highlights the value of acknowledging
emotions openly and explicitly at work. Moreover, our work suggests
that expressers and third-party observers are sensitive to the capacity of
perceivers to attend to others’ emotional needs. This hints that
emotional acknowledgment may produce benets beyond the focal
dyad—acknowledging others’ emotions may help build trust with those
witnessing the interaction as well. Broadly, this research sheds new light
on emotional acknowledgment as a powerful informational signal that
shapes the construction and understanding of our social relationships.
Funding
This research did not receive any specic grant from funding
agencies in the public, commercial, or not-for-prot sectors.
CRediT authorship contribution statement
Alisa Yu: Conceptualization, Methodology, Software, Validation,
Investigation, Resources, Writing - original draft, Writing - review &
editing, Visualization. Justin M. Berg: Conceptualization, Writing -
original draft, Writing - review & editing, Supervision. Julian J. Zlatev:
Conceptualization, Methodology, Writing - review & editing.
Appendix A. Validation study of emotional acknowledgment scale
We conducted a pre-registered study (https://aspredicted.org/qw2qc.pdf) to demonstrate convergent and discriminant validity of our emotional
acknowledgment scale, which was originally devised using our conceptual denition of the construct.
As stated in our preregistration, we hypothesized that emotional acknowledgment would show convergent validity with facets of emotional in-
telligence that are relevant to noticing and responding to others’ emotions (i.e., perception, understanding, and management of others’ emotions), as
well as empathy, warmth, interpersonal responsiveness, and agreeableness. We also hypothesized that emotional acknowledgment would show
discriminant validity with less relevant facets of emotional intelligence (i.e., use of emotions, management of one’s own emotions), as well as
emotional expression, conscientiousness, neuroticism, and competence.
Descriptive statistics, reliabilities, and zero-order correlations for the convergent and discriminant constructs are shown below. As expected, the
correlations between emotional acknowledgment and the convergent constructs were relatively stronger (r’s ranged from 0.31 to 0.61), and corre-
lations between emotional acknowledgment and the discriminant constructs were relatively weaker (r’s ranged from −0.07 to 0.44). There were,
however, two minor exceptions. Contrary to what we expected, emotional acknowledgment correlated relatively strongly with positive affect (r =
0.44, p < .001) and use of emotions (r = 0.42, p < .001), but these two correlations were lower than all but one of the correlations for the convergent
constructs (EI – Understanding), which was r = 0.31. Thus, the emotional acknowledgment scale generally performed as expected.
Descriptive statistics, reliabilities, and zero-order correlations with convergent measures.
M SD 1 2 3 4 5 6 7 8
1. Emotional Acknowledgment 4.62 1.35 (0.90)
2. EI - Perception 4.73 0.93 0.48*** (0.68)
3. EI - Understanding 4.59 1.35 0.31*** 0.49*** (0.92)
4. EI – Managing of others’ emotions 4.71 1.21 0.49*** 0.59*** 0.51*** (0.84)
5. Empathy 4.74 0.88 0.60*** 0.63*** 0.48*** 0.70*** (0.90)
6. Warmth 5.48 1.09 0.49*** 0.54*** 0.47*** 0.69*** 0.74*** (0.86)
7. Responsiveness 5.02 1.13 0.53*** 0.58*** 0.52*** 0.70*** 0.69*** 0.78*** (0.95)
8. Agreeableness 4.95 1.23 0.55*** 0.56*** 0.46*** 0.69*** 0.79*** 0.71*** 0.70*** (0.86)
Note: *p < .05, **p < .01, ***p < .001.
Descriptive statistics, reliabilities, and zero-order correlations with discriminant measures.
M SD 1 2 3 4 5 6 7 8
1. Emotional Acknowledgment 4.62 1.35 (0.90)
2. EI – Use of Emotions 4.08 0.84 0.42*** (0.46)
3. EI - Managing one’s own emotions 4.85 1.18 0.14** −0.03 (0.81)
4. Positive Affect 5.22 0.87 0.44*** 0.25*** 0.36*** (0.87)
5. Negative Affect 2.75 1.06 −0.13** 0.03 −0.58*** −0.30*** (0.90)
6. Competence 5.58 0.93 0.21*** 0.00 0.45*** 0.61*** −0.43*** (0.82)
7. Conscientiousness 4.99 1.14 0.14** 0.00 0.45*** 0.49*** −0.43*** 0.56*** (0.76)
8. Neuroticism 3.24 1.10 −0.07 0.01 −0.64*** −0.22*** 0.61*** −0.30*** −0.33*** (0.72)
Note: *p < .05, **p < .01, ***p < .001.
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
134
Appendix B. Supplementary material
Supplementary data to this article can be found online at https://doi.org/10.1016/j.obhdp.2021.02.002.
References
Abele, A. E., & Wojciszke, B. (2014). Communal and Agentic Content in Social Cognition:
A Dual Perspective Model. Advances in Experimental Social Psychology, 50, 195–255.
https://doi.org/10.1016/B978-0-12-800284-1.00004-7.
Antaki, C. (1988). Explanations, communication and social cognition. In C. Antaki (Ed.),
Analysing everyday explanation: A casebook of methods (pp. 1–14). Sage Publications
Inc.
Ashforth, B. E., & Humphrey, R. (1995). Emotion in the workplace: A reappraisal. Human
Relations, 48, 97–125.
Baker, L. R., McNulty, J. K., & Overall, N. C. (2014). When negative emotions benet
close relationships. In W. G. Parrott (Ed.), The positive side of negative emotions (pp.
101–125). Guilford Press.
Barsade, S. G. (2002). The ripple effects: Emotional contagion and its inuence on group
behavior. Administrative Science Quarterly, 47(4), 644–675. https://doi.org/10.2307/
3094912.
Balliet, D., & Van Lange, P. A. M. (2013). Trust, conict, and cooperation: A meta-
analysis. Psychological Bulletin, 139(5), 1090–1112.
Batson, C. D., Fultz, J., & Schoenrade, P. A. (1987). Distress and empathy: Two
qualitatively distinct vicarious emotions with different motivational consequences.
Journal of Personality., 55, 19–39. https://doi.org/10.1111/j.1467-6494.1987.
tb00426.x.
Becker, T. E., Atinc, G., Breaugh, J. A., Carlson, K. D., Edwards, J. R., & Spector, P. E.
(2016). Statistical control in correlational studies: 10 essential recommendations for
organizational researchers. Journal of Organizational Behavior, 37, 157–167.
Bodie, G. D. (2011). The Active-Empathic Listening Scale (AELS): Conceptualization and
evidence of validity within the interpersonal domain. Communication Quarterly, 59
(277–295).
Bono, J. E., & Vey, M. A. (2005). Toward understanding emotional management at work:
A quantitative review of emotional labor research. In C. E. H
¨
artel, W. J. Zerbe, &
N. M. Ashkanasy (Eds.), Emotions in organizational behavior (pp. 213–233). Lawrence
Erlbaum Associates Publishers.
Boone, R. T., & Buck, R. (2003). Emotional expressivity and trustworthiness: The role of
nonverbal behavior in the evolution of cooperation. Journal of Nonverbal Behavior,
27, 163–182.
Bosma, A., & Kunnen, E. S. (2001). Identity and emotion: Development through self-
organization. Paris: Cambridge University Press.
Brooks, A. W. (2014). Get excited: Reappraising pre-performance anxiety as excitement.
Journal of Experimental Psychology: General, 143, 1144–1158.
Cammann, C., Fichman, M., Jenkins, D., & Klesh, J. (1983). Assessing the attitudes and
perceptions of organizational members. New York: Wiley.
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress.
Journal of Health and Social Behavior, 24(4), 385–396.
Cooper, W. H., & Richardson, A. J. (1986). Unfair comparisons. Journal of Applied
Psychology, 71(2), 179–184.
DePaulo, B. M. (1992). Nonverbal behavior and self-presentation. Psychological Bulletin,
111(2), 203–243. https://doi.org/10.1037/0033-2909.111.2.203.
De Jong, B. A., Dirks, K. T., & Gillespie, N. (2016). Trust and team performance: A meta-
analysis of main effects, moderators, and covariates. Journal of Applied Psychology,
101(8), 1134–1150. https://doi.org/10.1037/apl0000110.
Dovidio, J. F., Piliavin, J. A., Shroeder, D. A., & Penner, L. A. (2006). The social psychology
of prosocial behavior. London: Lawrence Erlbaum Ass.
Dunn, J. R., & Schweitzer, M. E. (2005). Feeling and believing: The inuence of emotion
on trust. Journal of Personality and Social Psychology, 88(5), 736–748. https://doi.
org/10.1037/0022-3514.88.5.736.
Ekman, P., & Keltner, D. (1997). Universal facial expressions of emotion: An old
controversy and new ndings. In U. C. Segerstråle, & P. Moln
´
ar (Eds.), Nonverbal
communication: Where nature meets culture (pp. 27–46). Lawrence Erlbaum Associates
Inc.
Elfenbein, H. A. (2007). Emotions in organizations: a review and theoretical integration.
Academy of Management Annals, 1, 315–386.
Elfenbein, H. A., & Ambady, N. (2002). Predicting workplace outcomes from the ability
to eavesdrop on feelings. Journal of Applied Psychology, 87, 963–971.
Elfenbein, H. A., Barsade, S. G., & Eisenkraft, N. (2015). The social perception of
emotional abilities: Expanding what we know about observer ratings of emotional
intelligence. Emotion, 15(1), 17–34. https://doi.org/10.1037/a0038436.
Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The
broaden-and-build theory of positive emotions. American Psychologist, 56, 218–226.
Fredrickson, B. L., & Levenson, R. W. (1998). Positive emotions speed recovery from the
cardiovascular sequelae of negative emotions. Cognition and Emotion, 12(2),
191–220. https://doi.org/10.1080/026999398379718.
Fischer, A. H., & Manstead, A. S. R. (2008). The social functions of emotion. In M. Lewis,
J. Haviland-Jones, & L. F. Barrett (Eds.), Handbook of emotions (3rd ed.). New York:
Guilford Press.
Fridlund, A. J. (1994). Human facial expression: An evolutionary view. Academic Press.
Frijda, N. H., & Mesquita, B. (1994). The social roles and functions of emotions. In S.
Kitayama & H. R. Markus (Eds.), Emotion and culture: Empirical studies of mutual
inuence (pp. 51–87). American Psychological Association. https://doi.org/
10.1037/10152-002.
Gable, S. L., & Reis, H. T. (2010). Good news! Capitalizing on positive events in an
interpersonal context. In M. P. Zanna (Ed.), Advances in experimental social
psychology. Advances in experimental social psychology (Vol. 42, p. 195–257).
Academic Press. https://doi.org/10.1016/S0065-2601(10)42004-3.
Gino, F., & Schweitzer, M. E. (2008). Blinded by anger or feeling the love: How emotions
inuence advice taking. Journal of Applied Psychology, 93(5), 1165–1173. https://
doi.org/10.1037/0021-9010.93.5.1165.
Goldstein, N. J., Vezich, I. S., & Shapiro, J. R. (2014). Perceived perspective taking: When
others walk in our shoes. Journal of Personality and Social Psychology, 106(6),
941–960. https://doi.org/10.1037/a0036395.
Grandey, A. A. (2000). Emotional regulation in the workplace: A new way to
conceptualize emotional labor. Journal of occupational health psychology, 5(1), 95.
Grandey, A. A., & Gabriel, A. S. (2015). Emotional labor at a crossroads: Where do we go
from here? Annual Review of Organizational Psychology and Organizational Behavior, 2,
323–349. https://doi.org/10.1146/annurev-orgpsych-032414-111400.
Greenberg, L. S. (2004). Emotion-focused therapy. Clinical Psychology and Psychotherapy,
11, 3–16.
Guinot, J., Chiva, R., & Roca-Puig, V. (2014). Interpersonal trust, stress and satisfaction
at work: An empirical study. Personnel Review, 43, 96–115. https://doi.org/10.1108/
PR-02-2012-0043.
Hall, D. L., Cohen, A. B., Meyer, K. K., Varley, A. H., & Brewer, G. A. (2015). Costly
signaling increases trust, even across religious afliations. Psychological Science, 26,
1368–1376.
Hardy, C. L., & Van Vugt, M. (2006). Nice guys nish rst: The competitive altruism
hypothesis. Personality and Social Psychology Bulletin, 32, 1402–1413.
Hart, E., VanEpps, E. M., & Schweitzer, M. E. (2021). The (better than expected)
consequences of asking sensitive questions. Organizational Behavior and Human
Decision Processes, 162, 136–154.
Hateld, E., Cacioppo, J. T., & Rapson, R. L. (1993). Emotional contagion. Current
Directions in Psychological Science, 2(3), 96–99. https://doi.org/10.1111/1467-8721.
ep10770953.
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis:
A regression-based approach. New York: Guilford.
Hobfoll, S. E. (2002). Social and psychological resources and adaptation. Review of
General Psychology, 6(4), 307–324. https://doi.org/10.1037/1089-2680.6.4.307.
Hochschild, A. (1979). Emotion work, feeling rules, and social structure. American
Journal of Sociology, 85, 551–575.
Hochschild, A. R. (1983). The managed heart: Commercialization of human feeling.
Berkeley, CA: University of California Press.
Holmes, J. G., & Rempel, J. K. (1989). Trust in close relationships. In C. Hendrick (Ed.),
Review of personality and social psychology. Close relationships (Vol. 10, pp.
187–220). Sage Publications, Inc.
Huang, K., Yeomans, M., Brooks, A. W., Minson, J., & Gino, F. (2017). It doesn’t hurt to
ask: Question-asking increases liking. Journal of Personality and Social Psychology,
113, 430–452. https://doi.org/10.1037/pspi0000097.
Ickes, W. (1993). Empathic accuracy. Journal of Personality, 61(4), 587–610. https://doi.
org/10.1111/j.1467-6494.1993.tb00783.x.
Jones, S. M. (2011). Supportive listening. International Journal of Listening, 25, 85–103.
Kang, P., & Schweitzer, M. E. (2020). Predictability and the logic of perceived
appropriateness: How expressed emotions inuence trust. https://ssrn.com/abstract
=3647279.
Keltner, D., & Haidt, J. (1999). Social functions of emotions at four levels of analysis.
Cognition and Emotion, 13, 505–521. https://doi.org/10.1080/026999399379168.
Keltner, D., Haidt, J., & Shiota, M. N. (2006). Social functionalism and the evolution of
emotions. In M. Schaller, J. A. Simpson, & D. T. Kenrick (Eds.), Evolution and social
psychology (pp. 115–142). Psychosocial Press.
Klein, N. (2019). Better to overestimate than to underestimate others’ feelings:
Asymmetric cost of errors in affective perspective-taking. Organizational Behavior and
Human Decision Processes, 151, 1–15.
Kristjansson, K. (2010). The self and its emotions. Cambridge, UK: Cambridge University
Press.
Laurenceau, J.-P., Barrett, L. F., & Pietromonaco, P. R. (1998). Intimacy as an
interpersonal process: The importance of self-disclosure, partner disclosure, and
perceived partner responsiveness in interpersonal exchanges. Journal of Personality
and Social Psychology, 74(5), 1238–1251. https://doi.org/10.1037/0022-
3514.74.5.1238.
Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature review and
two-component model. Psychological Bulletin, 107(1), 34–47.
Levine, E. E., & Schweitzer, M. E. (2015). Prosocial lies: When deception breeds trust.
Organizational Behavior and Human Decision Processes, 126, 88–106. https://doi.org/
10.1016/j.obhdp.2014.10.007.
Maisel, N. C., Gable, S. L., & Strachman, A. (2008). Responsive behaviors in good times
and in bad. Personal Relationships, 15, 317–338.
Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey, &
D. J. Sluyter (Eds.), Emotional development and emotional intelligence: Educational
implications (pp. 3–34). Basic Books.
Mayer, R. C., & Davis, J. H. (1999). The effect of the performance appraisal system on
trust for management: A eld quasi-experiment. Journal of Applied Psychology, 84(1),
123–136. https://doi.org/10.1037/0021-9010.84.1.123.
A. Yu et al.
Organizational Behavior and Human Decision Processes 164 (2021) 116–135
135
Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of
organizational trust. Academy of Management Review, 20(3), 709–734. https://doi.
org/10.2307/258792.
Mayer, J. D., Salovey, P., Caruso, D. R., & Sitarenios, G. (2001). Emotional intelligence as
a standard intelligence. Emotion, 1, 232–242.
Morelli, S. A., Ong, D. C., Makati, R., Jackson, M. O., & Zaki, J. (2017). Empathy and
well-being correlate with centrality in different social networks. Proceedings of the
National Academy of Sciences, 114(37), 9843–9847. https://doi.org/10.1073/
pnas.1702155114.
Morris, J. A., & Feldman, D. C. (1996). The dimensions, antecedents, and consequences
of emotional labor. Academy of Management Review, 21(4), 1996.
Murray, S. L., & Holmes, J. G. (2009). The architecture of interdependent minds: A
motivation-management theory of mutual responsiveness. Psychological Review, 116,
908–928.
Pauw, L. S., Sauter, D. A., van Kleef, G. A., & Fischer, A. H. (2019). Stop crying! The
impact of situational demands on interpersonal emotion regulation. Cognition and.
Emotion, 33, 1587–1598.
Pennebaker, J. W. (1997). Writing about emotional experiences as a therapeutic process.
Psychological Science, 8(3), 162–166.
Pennebaker, J. W. (2018). Expressive writing in psychological science. Perspectives on
Psychological Science, 13(2), 226–229. https://doi.org/10.1177/
1745691617707315.
Porter, S., & ten Brinke, L. (2008). Reading between the lies: Identifying concealed and
falsied emotions in universal facial expressions. Psychological Science, 19(5),
508–514.
Rim
´
e, B., Finkenauer, C., Luminet, O., Zech, E., & Philippot, P. (1998). Social sharing of
emotions: New evidence and new questions. European Review of Social Psychology, 9,
145–189.
Reis, H. T., & Shaver, P. (1988). Intimacy as an interpersonal process. In S. Duck,
D. F. Hay, S. E. Hobfoll, W. Ickes, & B. M. Montgomery (Eds.), Handbook of personal
relationships: Theory, research and interventions (pp. 367–389). John Wiley & Sons.
Rempel, J. K., Holmes, J. G., & Zanna, M. P. (1985). Trust in close relationships. Journal
of Personality and Social Psychology, 49, 95–112.
Rempel, J. K., Ross, M., & Holmes, J. G. (2001). Trust and communicated attributions in
close relationships. Journal of Personality and Social Psychology, 81(1), 57–64.
Rosenquist, J. N., Fowler, J. H., & Christakis, N. A. (2011). Social network determinants
of depression. Molecular Psychiatry, 16(3), 273–281. https://doi.org/10.1038/
mp.2010.13.
Rossignac-Milon, M., & Higgins, E. T. (2018). Epistemic companions: Shared reality
development in close relationships. Current Opinion in Psychology, 23, 66–71. https://
doi.org/10.1016/j.copsyc.2018.01.001.
Rothman, N. B., & Melwani, S. (2017). Feeling mixed, ambivalent, and in ux: The social
functions of emotional complexity for leaders. Academy of Management Review, 42
(2), 259–282. https://doi.org/10.5465/amr.2014.0355.
Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social
Psychology, 39(6), 1161–1178. https://doi.org/10.1037/h0077714.
Schilke, O., & Huang, L. (2018). Worthy of swift trust? How brief interpersonal contact
affects trust accuracy. Journal of Applied Psychology, 103(11), 1181.
Schoorman, F. D., Mayer, R. C., & Davis, J. H. (2007). An integrative model of
organizational trust: Past, present, and future. The Academy of Management Review,
344–354.
Schwarz, N., & Clore, G. L. (2007). Feelings and phenomenal experiences. In
A. W. Kruglanski, & E. T. Higgins (Eds.), Social psychology: Handbook of basic
principles (pp. 385–407). The Guilford Press.
Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2011). False-positive psychology:
Undisclosed exibility in data collection and analysis allows presenting anything as
signicant. Psychological Science, 22(11), 1359–1366. https://doi.org/10.1177/
0956797611417632.
Simpson, J. A. (2007). Foundations of interpersonal trust. In A. W. Kruglanski, &
E. T. Higgins (Eds.), Social psychology: Handbook of basic principles (pp. 587–607).
The Guilford Press.
Smith, E. A., & Bird, R. L. B. (2000). Turtle hunting and tombstone opening: Public
generosity as costly signaling. Evolution and Human Behavior, 21(4), 245–261.
https://doi.org/10.1016/S1090-5138(00)00031-3.
Staw, B. M., & Barsade, S. G. (1993). Affect and managerial performance: A test of the
sadder-but-wiser vs. happier-and-smarter hypotheses. Administrative Science
Quarterly, 38(2), 304–331. https://doi.org/10.2307/2393415.
Swann, W. B., Jr. (2012). Self-verication theory. In P. A. M. Van Lange,
A. W. Kruglanski, & E. T. Higgins (Eds.), Handbook of theories of social psychology (pp.
23–42). Sage Publications Ltd.. https://doi.org/10.4135/9781446249222.n27
Torre, J. B., & Lieberman, M. D. (2018). Putting feelings into words: Affect labeling as
implicit emotion regulation. Emotion Review, 10(2), 116–124.
van der Werff, L., & Buckley, F. (2017). Getting to know you: A longitudinal examination
of trust cues and trust development during socialization. Journal of Management, 43
(3), 742–770.
van Kleef, G. A. (2009). How emotions regulate social life: The emotions as social
information (EASI) model. Current Directions in Psychological Science, 18, 184–188.
van Kleef, G. A. (2016). The interpersonal dynamics of emotion: Toward an integrative theory
of emotions as social information. Cambridge, UK: Cambridge University Press.
Van Maanen, J., & Kunda, G. (1989). “Real feelings”: Emotional expression and
organizational culture. Research in Organizational Behavior, 11, 43–103.
Van Maanen, J. (1991). The smile factory: Work at Disneyland. In P. J. Frost, L. F. Moore,
M. R. Louis, C. C. Lundberg, & J. Martin (Eds.), Reframing organizational culture (pp.
58–76). Newbury Park, CA: Sage.
Weick, K. E. (1995). Sensemaking in organizations. Thousand Oaks, CA: Sage.
Weger, H. J., Bell, G. C., Minei, E. M., & Robinson, M. C. (2014). The relative
effectiveness of active listening in initial interactions. International Journal of
Listening, 28(1), 13–31. https://doi.org/10.1080/10904018.2013.813234.
Wieselquist, J., Rusbult, C. E., Foster, C. A., & Agnew, C. R. (1999). Commitment, pro-
relationship behavior, and trust in close relationships. Journal of Personality and
Social Psychology, 77(5), 942–966. https://doi.org/10.1037/0022-3514.77.5.942.
Winczewski, L. A., Bowen, J. D., & Collins, N. L. (2016). Is empathic accuracy enough to
facilitate responsive behavior in dyadic interaction? Distinguishing ability from
motivation. Psychological Science, 27(3), 394–404. https://doi.org/10.1177/
0956797615624491.
Wolf, E. B., Lee, J. J., Sah, S., & Brooks, A. W. (2016). Managing perceptions of distress at
work: Reframing emotion as passion. Organizational Behavior and Human Decision
Processes, 137, 1–12.
Zickfeld, J. H., Schubert, T. W., Seibt, B., & Fiske, A. P. (2017). Empathic concern is part
of a more general communal emotion. Frontiers in Psychology, 8, 723.
Zaki, J. (2014). Empathy: A motivated account. Psychological Bulletin, 140(6),
1608–1647. https://doi.org/10.1037/a0037679.
Zahavi, A. (1995). Altruism as a handicap: The limitations of kin selection and
reciprocity. Journal of Avian Biology, 26, 1–3.
A. Yu et al.
