The Distance of Power:
The Implications of Formal Organizational Hierarchies on Employees’ Cognitive Information Processing
Master’s Thesis
Lukas Wegenast #10855475
Supervisor: Dr. Claartje ter Hoeven
Master’s Programme Communication Science Graduate School of Communication
University of Amsterdam
Abstract
Previous research has established a connection between the experiencing of power and
cognition. Using Magee and Smith’s (2013) social distance theory of power, the current study tested this connection in the context of intra-organizational dyadic communication of both task-related and non-task-related information. Including an individual, dispositional factor as moderating the proposed relationship, the study investigated the impact of hierarchically-based power asymmetry on cognitive processing of information. Experimental results
partially supported the assumption, indicating that power asymmetry does have an impact on receivers’ cognitive processing of non-task-related information. Identifying a new moment of potential information distortion in organizational communication, the findings stress the importance of including contextual factors of employees’ working environment in future studies.
The Distance of Power:
The Implications of Formal Organizational Hierarchies on Employees’ Cognitive Information Processing
Imagine the following scenario: A globally operating pharmaceutical company is executing a large-scale merger. The company is trying to communicate the change internally by informing employees of the acquired company about the core values of their new parent company. Employees, however, do not identify with these values and signal feelings of detachment to their supervisors. As it travels upwards, this information is distorted and does not reach upper management. Under the false impression that the internal campaign is reaching socially cohesive outcomes, the company cannot understand why its reputation as employer decreases. Ruling out employee identification as a factor, the company allocates resources into external marketing campaigns to strengthen public reputation. The
misalignment between actual and communicated identity (Balmer, Stuart, & Greyser, 2009) causes stakeholder- and investor-skepticism, causing the company to lose market power.
From a resource-based view on the organization, information is a key element in organizational proceedings (Rumelt & Lamb, 1997). By basing market intelligence, the generation of knowledge, and decisions on information entering and passing through the internal domain, organizations are relying on information for their survival and advancement (Choo, 1996). The distortion of information can have detrimental effects on decision making (O’Reilly & Roberts, 1974) and overall organizational performance to an extent that the transfer of false information is associated with lower performance outcomes than an overall breakdown of communication (Carley & Lin, 1997). These findings highlight the managerial urgency of identifying factors influential to the free flow of information within organizations. Organizational hierarchy has been identified as one of these factors (Reitzig & Maciejovski, 2015). Despite their assumed functional nature, hierarchies carry with them dysfunctional outcomes for organizations.
The power that emerges through hierarchical structures has been shown to affect numerous cognitive processes within both high and low power holders (DeWall, Baumeister, Mead, & Vohls, 2011). It affects individuals’ attentiveness to different types of information (Guinote, 2010) and the omission of irrelevant information (Posner & Deheane, 1994). Furthermore, Overbeck and Park (2001) provide initial evidence that the retrieval process of information differs between the powerful and powerless. Power thus appears to affect several stages of information processing: The attentiveness to information, the omission of
information as it is stored, and how information is retrieved. These finding suggest an understanding of information as a non-neutral resource that is adjusted and omitted as it travels through hierarchical structures.
The current study uses the social distance theory of power (Magee & Smith, 2013) to argue that the perceived power asymmetry between the sender and receiver is related to different outcomes of information transfer through different cognitive processing. Entailing Mast’s (2005a) concept of interpersonal hierarchy expectation, it furthermore incorporates a dispositional cognitive factor of the receiver in the analysis. It aims to add to the existing body of literature in four ways.
First, focusing on individuals’ social perception (Overbeck & Park, 2001) or aspirations of moving upward in the hierarchy (O’Reilly & Roberts, 1974), previous
approaches to the distortion of organizational information have not used the communicative event as a frame of reference. The current approach breaks down the communicative process and identifies a moment of potential information distortion within it: that of a receiver’s processing of information conveyed in work-related messages.
Second, the current study investigates the role of hierarchy on cognitive processing of organizational information; contrary to previous approaches (Carley & Lin, 1997), it
related to information distortion, but as triggering cognitive responses in employees and, thus, giving rise to information distortion.
Third, while the social distance theory of power has helped to explain behavioral and cognitive differences between high and low power holders (Scholl & Sassenberg, 2015), it has not been applied to the context of dyadic communication between them. The current study aims to show that the theory can also help explain the stifling or propelling of specific types of information in vertical and lateral communication.
Finally, the current study adds to the existing literature on hierarchy and information distortion by entailing upwards, downwards, and lateral communication in the concept, whereas many researchers of information distortion limited their studies to the upward distortion of information (Kelley, 1951; O’Reilly & Roberts, 1974).
The current study aims to address these shortcomings of previous approaches and contribute to a deeper understanding of the implications of hierarchically based power differences in the workplace by investigating the following research question:
RQ To what extent does hierarchical differentiation between the receiver and sender
of a message affect the receivers’ cognitive processing of the message’s content?
Theoretical Background Hierarchies and Power
Relationships between people are hierarchically structured (Mast, 2010). Even in contexts where no formal hierarchies (such as reporting structures or rank orders) determine stratification of power or dominance, hierarchies emerge along contextually relevant
dimensions; these dimensions are in accordance with the current goal of the group (Tiedens & Fragale, 2003). Magee and Galinsky (2008) thus define hierarchy as “an implicit or explicit rank order of individuals or groups with respect to a valued social dimension” (p. 354). This is better understood when thinking of a football team: The team’s goal being
winning, players’ abilities to contribute to that goal become socially relevant. Because
technically gifted players have higher abilities in this socially relevant dimension, they obtain higher roles in the hierarchy than technically ungifted players.
Formal organizational hierarchies are recurrent across countries, industries, and organizational cultures and practices (Carroll & Hannan, 2000). Usually, formal hierarchies describe a pyramid of social organization with few at the top and many at lower levels. This structure aids organizations to manage and coordinate internal complexity as it grows to interact with the complexity of its environment (Leavitt, 2005; Luhmann, 1984).
Understanding an organization as “a consciously coordinated social unit […] that functions on a relatively continuous basis to achieve a common goal or set of goals” (Robbins & Judge, 2013, p. 5), hierarchies enable the coordination of shared goals and activities that distinguish a random group of individuals from an organization. This idea is reflected in the observation that hierarchies persistently emerge, even against organizational efforts to suppress them (Diefenbach & Sillince, 2011).
Members of a hierarchy differ in their levels of control over a valued resource. By endowing members with different levels of control and authority, hierarchies give rise to a “dyadic asymmetry in power” (Van der Vegt, De Jong, Bunderson, & Molleman, 2010, p. 347) as lower-level members are depending on higher-level members for valued resources.
Through these differences, hierarchies hold functional merits for organizations. On an organizational level, hierarchies structure social interaction (Halevy, Chou, & Galinsky, 2011) by coordinating actions across business units, departments, and functions (Baron & Pfeffer, 1994). On an interpersonal level, they have been found to reduce conflict and promote cooperation amongst employees (Halevy, Chou, Galinsky, & Murnighan, 2012). Finally, on an individual level, hierarchies create reward systems enhancing employee motivation (Hogg, 2001). At the personal and interpersonal level, formal hierarchies thus
leverage the merits of increased power to encourage cooperation and motivation. However, both the desire for and the abuse of power have been identified to stifle organizational advancement, such as strategic ambiguity to maintain powerful positions (Davenport & Leitch, 2005), the impairing of team learning (Bunderson & Reagans, 2011), or workplace bullying (Liefooghe & Davey, 2001).
Whether the outcome is functional or dysfunctional, hierarchies permeate the
dimension of social perception and interpersonal behavior to an extent that “social relations within organizations are institutionalized and legitimized […] as hierarchical relations” (Diefenbach & Sillince, 2011, p. 1517). Thus, by creating power asymmetry between their members, hierarchies affect the social interactions in organizations.
Power, Cognition, and Construal
With power being the cognitive manifestation through which organizational members experience hierarchies, its implications on behavior and cognition are manifold (for reviews, see Guinote, 2013; Williams, 2014).
Power facilitates goal-pursuit and increases a tendency towards self-serving behavior (Guinote, 2007a; Williams, 2014). High power individuals (HPI) are inclined to act towards the retention of their power even in contexts in which they don’t hold power (Galinsky, Gruenfeld, & Magee, 2003). They are also less susceptible to social influence (Galinsky, Magee, Gruenfeld, Whitston, & Liljenquist, 2008). In contrast, low power individuals (LPI) are more attentive to their environment and perform worse in executive functions (Smith, Jostmann, Galinsky, & van Dijk, 2008).
These observed behaviors are rooted in and affect individuals’ cognition (Bem, 1972). In general terms, power increases individuals’ well-being and triggers more effortless cognition (Guinote, 2007b; Keltner, Gruenfeld, & Anderson, 2003). Cognition is flexible, allowing individuals to attend to changing environments and contexts (Guinote, 2007b).
In the context of dyadic information transfer within organizations, it is helpful to structure the argumentation into two related dimensions of the cognitive process: First, it is important to understand how power affects what kind of information is attended to, and, secondly, if that information is subject to different cognitive processing.
Regarding the type of information that is being attended to, HPI have been found to make a stronger distinction between information that is goal-relevant and –irrelevant information, whereas LPI view information more equivocally (Guinote, 2013). Once identified as relating to their goals, HPI are better able to direct attention to goal-relevant information and can more easily disregard information they consider irrelevant (DeWall et al., 2011). LPI divide their attention between goal-relevant and –irrelevant information (Guinote, 2010). This effect of power depends on the context and content of the information received: While HPI generally directed their attention to work-related information than to social information, this relationship reversed in contexts where social information became relevant to HPI’s personal goals (Guinote, 2008; Overbeck & Park, 2001). High power thus promotes selective cognition and adaptive action in line with individuals’ goals (Guinote, 2007c). Low power, on the other hand, is associated with a processing focus that is “extensive and responsive to multiple influences” (p. 288), leading LPI to be more susceptible to obstacles and to take more time to reach goals (Guinote, 2007c).
Addressing how power affects the way in which information is processed, Magee and Smith (2013) developed the social distance theory of power, which postulates two principles about power, social distance, and construal. The first principle states that asymmetric
interpersonal dependence produces asymmetric social distance, which they define as “a subjective perception or experience of distance from another person or other persons” (Magee & Smith, 2013, p. 159). This concept rests on the idea of psychological distance (Smith & Trope, 2006) and translates it into the social domain. Psychological distance
describes a subjective perception of the distance to a psychological object (e.g. a person, an event, a memory) with the self, here, and now demarking the center of reference (Trope & Liberman, 2010).
Power asymmetry, according to the first principle of the social distance theory of power, creates an asymmetrical perception of social distance with the HPI feeling more socially distant from LPI than the other way around. The asymmetry of perceived social distance arises from a difference in individuals’ motivations to affiliate with the other and their expectations towards the relationship (Magee & Smith, 2013). Even though LPI want a stronger affiliation with HPI than HPI do with LPI, their (LPI) expectations are proposed to be more realistic. In contrast, HPI’s expectations towards the relationship are less precise while their motivation to affiliate is low (Magee & Smith, 2013). Contrary to interactions between HPI and LPI, Magee and Smith (2013) identify symmetrically dependent
relationships of equal-power individuals (EPI) as a “point of departure” (p. 172). Perceiving the lowest amount of social distance, EPI act to minimize the distance between them (Magee & Smith, 2013). Thus, in case of symmetrical power, no (or low amounts of) social distance emerges. In asymmetrical relationships, social distance emerges asymmetrically. Outcomes that can be explained with social distance should therefore not follow a pattern of low-power < equal-power < high-power, but rather a pattern of equal-power < low-power < high-power (Magee & Smith, 2013).
The distinction between symmetric and asymmetric social distance helps explain findings in previous studies. Chen, Tharp, and Kuehn (2017) found that HPI have higher expectations of social acceptance and lower concerns for potential rejection than LPI. And once HPI do get rejected, they are less inclined to make inferences on their self-esteem (Kuehn, Chen, & Gordon, 2015). Moreover, while power gives rise to social distance, it also has been shown to decrease the need to belong, which suggests that, though socially distant,
HPI do not experience loneliness on the basis of power asymmetry (Waytz, Chou, Magee, & Galinsky, 2015). Findings of negotiator dyads achieving higher joint gains in an equal balance of power than when in an unequal balance of power (Mannix & Neale, 1993) further support the theoretical distinction of symmetric and asymmetric power relationships.
The second principle of the social distance theory of power states that, via social distance, power increases the level of construal on which individuals process information (Magee & Smith, 2013). This assumption rests on Trope and Liberman’s (2010) construal level theory of psychological distance, which conceptualizes individuals to construe objects at a higher level as the psychological distance to these objects increases. At higher levels of construal, individuals omit specific information about the object, thus processing it more abstractly (Trope & Liberman, 2010). Magee and Smith (2013) argue that HPI structure mental representations of psychological objects, maintaining central and omitting peripheral features. In contrast, LPI make more unstructured mental representations of their
environment, themselves, and their actions (Magee & Smith, 2013). Interestingly, Magee and Smith (2013) do not formulate construal-related expectations in the interactions of EPI. As they experience less social distance than LPI (Magee & Smith, 2013), EPI are also expected to process information at the lowest level of construal.
As construal level increases, individuals prioritize the desirability of topics and goals over their feasibility as “desirability reflects the superordinate concern of the desired goal, whereas feasibility reflects the subordinate concern of the means to achieve that goal” (Magee & Smith, 2013, p. 16). HPI are therefore proposed to be more inclined than LPI to follow their abstract visions, as well as seeing the bigger picture in tasks. When confronted with competing goals, HPI will therefore prioritize goals that are connected to long-term rather than short-term rewards (DeWall et al., 2011; Magee & Smith, 2013). While no
power, initial evidence supports the theory’s proposed effects regarding high-power construal. HPI have been found to engage in less prefactual thought than LPI (Scholl & Sassenberg, 2015). Furthermore, participants wearing formal clothing, which itself is an expression of formality and, thus, social distance, have been shown to identify actions at higher levels of abstraction than participants wearing casual clothing (Slepian, Ferber, Gold, & Rutchick, 2014). Smith and Trope (2006) showed that these effects are also evident when individuals experience power more indirectly through priming. Participants primed with high levels of power processed information more abstractly than did participants primed with low levels of power. They categorized information in more inclusive categories than LPI and engaged in more abstract thinking in order to identify inter-item coherence (Smith & Trope, 2006).
Cognitive Investments: Recall and Recognition
Scholars are still divided in their understanding of the exact cognitive processes underlying both measures (for a review, see Yonelinas, 2002). Jacoby (1983) understands familiarity to depend on the fluency with which an item is processed; if the item is processed more fluently relative to other items in a recognition test, individuals consider it familiar and thus recognize it (Jacoby, 1983; Yonelinas, 2002). Recollection, on the other hand, “reflects an analytic, consciously controlled process” (Yonelinas, 2002, p. 445; emphasis added). The recall of an item thus requires a higher amount of cognitive resources than its recognition. This understanding is echoed by researchers in the field (Craik, 1983; Hasher & Zacks, 1979). Craik and McDowd (1987) conclude that recall requires more “self-initiated activity” (p. 478), thus supporting Jacoby’s (1983) conceptualization by implying a conscious or subconscious initiation of cognitive processes.
This distinction becomes crucial when being viewed in the foreground of intra-organizational communication. By measuring information recognition and recall, we can
identify whether the type of information is processed differently across different hierarchical relationships.
Disposed to Hierarchy: The Interpersonal Hierarchy Expectation
The concepts of hierarchy, power, and social distance are relational in nature (Smith & Magee, 2015). Emphasizing the concept of perceived power and defining it as “the
impression an observer or interaction partner gains of a target’s power […, which] can stem from the knowledge of each other’s position power” (p. 5), Mast (2010) underlines the idea of power being enacted in social interaction. This understanding stresses that power is dependent not only on hierarchical roles, but also on individual perception of both the hierarchy and the social interaction partner.
To capture “how characteristics of the perceiver influence social perception” (p. 287), Mast (2005a) developed the model of interpersonal hierarchy expectation (IHE), showing that individuals differ in their expectations of social interactions to be hierarchically structured. Dispositional characteristics thus cause individuals to “see social interactions through a hierarchy lens” (Mast, 2005a, p. 291). Individuals who have a higher IHE are therefore expected to perceive hierarchical differences between them and interactional partners as more pronounced. The concept captures individuals’ perceptions of the unique power relationship between them and their interaction partner and therefore anticipates power-related cognition being a function of both interpersonal and individual factors. Hypothesis development
The theoretical elaborations draw a clear picture of the relationship between formal organizational hierarchies and cognitive information processing. First, they allow the assumption of hierarchy and power being so closely interrelated that a manipulation of hierarchy also entails a manipulation in power.
Second, the perceived power asymmetry between sender and receiver affects cognitive information processing by means of social distance. Distinguishing between task-related information and non-task-related information, expectations can be formulated based on different levels of construal. Higher construal is associated with the setting of goals that are relevant to the situational context and connected to HPI’s abstract desires (Scholl &
Sassenberg, 2015); it is furthermore associated with selective processing of information in connection to a goal that is considered relevant (DeWall et al., 2011). Thus, construing information on a higher level than LPI, HPI are expected to prioritize task-related information over non-task-related information when facing two competing types of
information in a work-related context. LPI, on the other hand, are expected to process more non-task-related information and less task-related-information than HPI. Individuals in a symmetrical power relationship have been argued to experience less social distance than either HPI or LPI (Magee & Smith, 2013). EPI are therefore expected to construe information at the lowest level of abstraction and thus to be equivocally attentive to task-related and non-task-related information.
Third, the conceptualization of recall requiring more cognitive resources than
recognition provides a tool to make tangible the prioritizing of different types of information by HPI, LPI, and EPI. More specifically, different scores of information processing allow inferences that hierarchy affects cognitive processing and, thus, communicative outcomes within organizations. The expected relationships are reflected in the following hypotheses:
H1 Perceived power asymmetry has an effect on a receiver’s cognitive processing of
task-related information; a) HPI process a higher amount of task-related information than LPI, and b) LPI process a higher amount of task-related information than EPI.
H2 Perceived power asymmetry has an effect on a receiver’s cognitive processing of
information than LPI, and b) LPI process a lower amount of non-task-related information than EPI.
Additionally, individuals with a higher IHE are expected to view the social relation to their interaction partner as more structured by the relation of their hierarchical roles.
Therefore, the arising power asymmetry between both actors is perceived as higher, which carries with it a higher perceived social distance, making the effects of social distance more pronounced in terms of higher levels of construal. This expectation is reflected in the final two hypotheses. A visual representation of the expected relationships is provided by the conceptual model (see Figure 1).
H3 The effect of perceived power asymmetry on cognitive processing of task-related
information is moderated by the receiver’s interpersonal hierarchy expectation; such that the effect is stronger for high levels of IHE and weaker for low levels of IHE.
H4 The effect of perceived power asymmetry on cognitive processing of
non-task-related information is moderated by the receiver’s interpersonal hierarchy expectation; such that the effect is stronger for high levels of IHE and weaker for low levels of IHE.
Figure 1. Conceptual model depicting the expected relationships.
Perceived Power Asymmetry Task-Related Information Processing Non-Task-Related Information Processing Interpersonal Hierarchy Expectation
Method Design and Participants
A survey-embedded experiment was conducted using a 3x2 factorial design; it used perceived power asymmetry as the between-subjects factor and IHE as the within-subjects factor. A convenience sample was drawn from the population of employees of elevated levels of hierarchy. Participants were recruited via e-mail, face-to-face contact, as well as via the social networking sites LinkedIn, Reddit, and Facebook. Additionally, several organizations agreed to promote the survey amongst their stakeholders. The final period of recruiting spanned from 5 December, 2017, to 8 January, 2018, and thus covered 35 days.
Of the initial 486 participants starting, 218 met the criteria of participation. Of those eligible, 116 completed the questionnaire. The mortality rate of the study can be considered high as a majority of participants (53.21%) dropped out of the study in the course of
answering the questionnaire. Of the eligible and complete responses, 14 (12.07%) were dropped as they failed the manipulation check.
The final sample used for the analysis consisted of 102 participants (20.99% of started responses) and contained employees from companies operating in 32 distinct industries. Organizational size ranged from ‘1 to 50’ to ‘over 5000’, with the majority of the sample being employed in companies of over 5000 employees (Mo = 8 (>5000 employees)). Participants’ mean age was 33 (M = 33.21, SD = 10.81). 56.86% of them were female (42.16% male, 0.98% did not indicate gender). Participants’ working hours were normally distributed around a mean amount of 41.01 hours (SD = 12.76).
Materials
The study’s stimulus consisted of an e-mail written by a fictive co-worker at the
participants’ company. In the e-mail, a total of fourteen distinct, salient pieces of information were present; seven of which were non-task-related information about a company event, and
seven of which were related to a future meeting the co-worker is informing the respondent about (see Appendix A).
Before the e-mail, participants were shown an information disclaimer. The disclaimer was experimentally manipulated in a total of three conditions. In the low-power condition (LPC), participants were informed that they would be reading an e-mail sent to them by their supervisor; in the equal-power condition (EPC), participants were informed they would be reading an e-mail from one of their peers; and in the high-power condition (HPC), they were informed that they would read an e-mail from a subordinate. While the sender was
manipulated and the order of both paragraphs of the e-mail reversed for every other
participant to avoid order effects, all other parts of the stimulus material were held constant. To make the manipulation truer to respondents’ actual working environment, they were asked to give a name to the fictive co-worker. Depending on the experimental condition, this name was asked to be that of their actual supervisor, peer, or subordinate.
Procedure
Upon opening the online questionnaire, participants were informed that participation in the study was voluntary and that they could withdraw their participation from the study by exiting the questionnaire at any time. Additionally, participants were given the opportunity to request a deletion of their response up to 72 hours after they had completed the questionnaire. It was indicated that participants would not be reimbursed for their participation. They were furthermore informed about the general purpose of the study and about the fact that their anonymity was safeguarded.
After giving their informed consent to participating in the study, participants answered questions ensuring their eligibility. Items asked whether, in their current job, they a) had a direct supervisor, b) were working with a peer of the same hierarchical level, and c) had a direct subordinate. When identified as ineligible, participants were directed to the end of the
questionnaire. Eligible participants were randomly assigned to one of three experimental conditions (LPC, EPC, or HPC).
Participants were then asked to answer nine items measuring IHE. Thereafter,
participants were prompted with the stimulus material. To prevent post-stimulus processing (Liu & Fu, 2007), participants were given a distractor task, namely identifying images meeting certain criteria (e.g. ‘identify all images showing a tree’). Following the distractor task, participants were asked to perform a free recall test followed by a recognition test of the information conveyed in the previously displayed e-mail.
After finishing the recognition test, participants were asked to provide demographic information about themselves, the company they work for, and their co-worker; depending on the experimental condition, this was either their supervisor, their peer, or their subordinate. In a final set of two items, respondents were asked to indicate the power relationship between them and their co-worker.
They then reached the debriefing slide, which informed them about the specific purpose of the study, the experimental manipulation, and were given the opportunity to contact the researcher or the ethics committee of the University of Amsterdam in case of ethical or privacy concerns.
Measures
Interpersonal Hierarchy Expectation. Participants’ IHE was measured using the Interpersonal Hierarchy Expectation Scale, which had been developed by Mast (2005a) and successfully tested thereafter (Carter, Hall, Carney, & Rosip, 2006; Mast, 2005b; Sandage, Jankowski, Crabtree, & Schweer-Collins, 2017). The scale is comprised of nine items measuring participants’ dispositional acceptance and expectation of hierarchical structures (e.g. “If people work together on a task, one person is always taking over the lead”). Items
were measured using a five-point answering scale (1 = strongly disagree and 5 = strongly
agree). To avoid order effects, items were displayed in random order.
A principal component analysis was conducted forcing a one-factor solution (factor range: .36 - .71). A reliability analysis using Cronbach’s alpha revealed that, contrary to previous studies, the reliability of the scale was low, but reasonable ( = .64). After dropping one item (Item 8 of original scale, see Schmid Mast, 2005a) from the scale in order to increase
reliability ( = .65), a composite measure was created (M = 3.15, SD = 0.59).
Perceived power asymmetry. Two items measured the power relationship between participants and their individual co-workers. The items were based on those used by Peiró and Meliá (2003) and measured the perceived amount of power participants believed to hold over their co-worker’s outcomes, as well as the perceived amount of power participants believed their co-worker to hold over their outcomes. Both items were measured using five-point sliding scales (1 = No power at all and 5 = A great deal of power).
Information recall. The recall of both task-related and non-task-related information was measured using a free recall test, as recall requires a high amount of cognitive resources (Ratneshwar & Chaiken, 1991). To answer the open-ended item, twenty fields for text-entry were provided.
Participants’ answers were coded for the presence of each of the fourteen distinct pieces of information provided in the e-mail. This way, the highest score to be achieved in the free recall test was fourteen. Two coders rated the presence of each piece of information independently on a sub-sample of 23.53% (nsub-sample = 24) of the overall sample (see
Appendix B for instructions). Inter-coder agreement was measured using Krippendorff’s alpha (Hayes & Krippendorff, 2007). After two rounds of coder-training, the final test of Krippendorff’s alpha indicated sufficient agreement amongst both coders (see Appendix D).
Information recognition. Information recognition was measured using a three-alternative forced-choice recognition test (Warren & Duff, 2014) comprised of fifteen items (see Appendix C). In addition to items measuring each piece of information provided in the e-mail (e.g. “When is the company dinner scheduled?”), a manipulation check item was included asking participants to indicate by whom the e-mail was sent. Answer categories included two false statements and one true statement. Participants were thus able to reach a maximum score of fourteen in the information recognition test. In the case of the
manipulation check, answer categories corresponded with the three conditions of the experiment (superior, subordinate, or peer).
Scoring
Jacoby (1983) suggests that the recognition of information is an automatic cognitive process whereas the recall of information is based on a process of conscious retrieval. A scoring system anticipating this difference was implemented. Different scoring methods have been used in earlier studies on information recognition and recall, yet they did not meet all requirements of Jacoby’s (1983) conceptualization; a ratio of recognition to recall
(Hollingworth, 1913) was not robust to low scores on both measures, whereas a sum (Norris & Colman, 2013) did not weigh recall stronger than recognition. In the current study,
recognizing an item was coded as one point. Recalling an item was coded as two points, regardless of whether that item was also recognized. This method anticipated the higher weight of recall scores while the method was also robust to low scores on recall and
recognition. Thus, participants had the possibility of scoring 3 points per item. Scores from the individual item-measures were summed up; participants could therefore achieve a final score ranging from 0 to 21 on both the task-related (M = 8.78, SD = 4.37) and the non-task-related (M = 10.25, SD = 3.47) scoring variable.
Results Randomization Check
To assess whether the random assignment of participants was successful, several
randomization checks were conducted. Descriptive statistics of the conducted analyses can be found in Table 1.
Table 1
Descriptive Statistics of Tested Demographic Variables, Listed by Experimental Conditions.
Measure Low-Power Equal-Power High-Power
Age M (SD) 33.41 (9.96) 35.27 (12.06) 30.46 (10.24)
Working Hours M (SD) 39.52 (14.07) 39.97 (10.70) 44.56 (12.75)
Agent gender nmale 21 13 9
nfemale 19 20 19
Target Gender nmale 27 18 18
nfemale 12 14 10 Organizational Size 1-50 8 7 4 50-100 4 4 5 100-250 5 3 4 250-500 2 2 4 500-1000 4 0 2 1000-5000 8 6 1 >5000 10 11 8
Note. No significant between-group differences were observed for any of the reported
measures.
The distribution of participants’ gender among the experimental conditions was assessed using Pearson’s chi-square. The analysis returned a non-significant result (2
101) = 3.00, p = .218), indicating that respondents’ gender did not differ significantly across experimental groups.
Another chi-square test was conducted to identify whether participants’ co-workers’ gender differed significantly across the experimental conditions. The analysis returned a non-significant result (2
(2, n = 99) = 1.29, p = .534), indicating that there were no differences between groups in participants’ co-workers’ gender.
A one-way analysis of variance (ANOVA) was conducted to identify potential
between-group differences in age. Levene’s test indicated homogeneity of variances (F(2, 99) = 0.82, p = .442). The ANOVA was not significant (F(2, 99) = 1.53, p = .222), indicating that age was evenly distributed amongst the experimental conditions.
Another one-way ANOVA was conducted to test between-group differences in participants’ amounts of working hours. Levene’s test indicated homogeneity of variances (F(2, 98) = 0.37, p = .692). The ANOVA showed no significant differences between experimental conditions (F(2, 98) = 1,44, p = .241).
To identify whether the distribution of organizational sizes was random across the experimental conditions, a chi-square test was conducted. Results indicated that
organizational size was equally distributed amongst experimental conditions (2
(12, n = 102) = 10.58, p = .565).
Manipulation Check
A one-way ANOVA was conducted to confirm that the proposed hierarchical relationships of the experimental conditions translated into perceived power differences between participants and their co-workers. Before running the analysis, a new variable was created by subtracting the amount of power participants perceived to have over their co-worker’s outcomes from the amount of power participants perceived their co-worker to have over their outcomes (M = -0.20, SD = 1.76). The analysis was significant (F (2, 94) = 38.23,
p = .000), indicating that the perceived power asymmetry differed across the experimental
conditions. A post-hoc analysis revealed that, in fact, between-group differences were evident across all three groups in accordance with the experimental manipulation. Participants in the LPC (M = -1.58, SD = 1.33, 95% CI [-2.02, -1,42]) perceived themselves to have less control over their co-worker’s outcomes than participants in the EPC, who perceived that control over outcomes was relatively evenly distributed between them and their co-worker (M =
0.25, SD = 0.76, 95% CI [-0.25, 0.52]; pLPC-EPC = .000). Participants in the HPC (M = 1.22, SD = 1.76, 95% CI [0.53, 1.92]) perceived themselves as having more control over their
co-worker’s outcomes than participants in the LPC (pHPC-LPC = .000) and the EPC (pHPC-EPC =
.018).
Hypothesis Testing
To test H1, a one-way ANOVA was conducted. Since Levene’s test was non-significant (F(2, 99) = 0.63, p = .532), equal variances could be assumed in spite of group sizes not being near-equal. As the assumptions of equal variances and normality of the dependent variable were met, the analysis was carried out. The ANOVA returned a non-significant result (F(2, 99) = 1.42, p = .246), indicating that no between-group differences exist.
Even though randomization tests had indicated no between-group differences in demographic characteristics of the sample, an analysis of covariance (ANCOVA) was conducted. This decision was based on methodological and theoretical motivations. First, the size of the current sample was small (n = 102); with small sample sizes, the probability of making a type II error increases (Freiman, Chalmers, Smith, & Kuebler, 1978), suggesting that, despite successful randomization tests, between-group differences may exist after all. Second, as Magee and Smith (2013) note, structural factors of organizations “might moderate […] the basic relationship between power and social distance” (p. 177); organizational size
and industry were included to account for these structural factors. Moreover, Mast (2010) notes that diffuse status cues such as age and gender influence performance expectations and social perception. Other findings show that the amount of working hours influences
organizational commitment (Scandura and Lankau, 1997), which – in turn – may affect perceived social distance to the organization and its employees.
Including these covariates, no significant between-group differences could be assessed (Fconditions(2, 88) = 2.01, p = .140). With confounding variables held constant and differences
not being significant, participants in the EPC (M = 9.34, SE = 0.78, 95% CI [7.79, 10.88]) scored slightly higher than participants in the LPC (M = 9.23, SE = 0.71, 95% CI [7.81, 10.65]). Participants in the HPC scored lowest (M = 7.21, SE = 0.87, 95% CI [5.49, 8.93]; see Figure 2). On the basis of the analyses, we failed to reject the null hypothesis; H1 was therefore not supported.
Figure 2. Differences in task-related memory scores between experimental conditions. Agent
and target gender, agent working hours, agent age, organizational size, and organizational industry are held constant. Between-group differences are not significant (p > .05).
Note: Agent and target gender, agent working hours, agent age, organizational size, and organizational industry are held constant.
Another one-way ANOVA was run in order to test H2. Levene’s test indicated equality of variance (F(2, 99) = 0.85, p = .431). Results of the analysis were not significant (F(2, 99) = 2.49, p = .088), indicating no significant differences in the processing of non-task-related information between the three experimental conditions. Another ANCOVA was conducted, including the same covariates as in the ANCOVA testing H1. When controlling for
confounding variables, a significant difference between groups was assessed (Fconditions(2, 88)
= 3.15, p = .048). The effect size was small (2
= .07).
Participants in the LPC processed the highest amount (M = 11.13, SE = 0.57, 95% CI [10.00, 12.26]) of non-task-related information on the basis of the scoring system.
Participants in the EPC processed (relative to both other conditions) a medium amount of non-task-related information (M = 10.29, SE = 0.62, 95% CI [9.06, 11.53]). Participants in the HPC processed the lowest amount of non-task-related information (M = 8.85, SE = 0.69, 95% CI [7.47, 10.22]; see Figure 3).
Figure 3. Differences in non-task-related memory scores between experimental conditions.
Agent and target gender, agent working hours, agent age, organizational size, and organizational industry are held constant. Between-group differences are significant at
p < .05.
Note: Agent and target gender, agent working hours, agent age, organizational size, and organizational industry are held constant.
Thus, even though randomization across groups had been assessed, controlling for confounding variables yielded results allowing for a rejection of the null hypothesis. While statistically significant, the differences of the group means did not align with expectations formulated in the hypothesis. EPI were expected to score highest on non-task-related information processing as they were expected to construe on the lowest level of abstraction and attend to the highest amount of information. This expectation was not met as EPI scored lower than LPI. H2 was therefore partially supported.
To test H3, a multiple regression analysis was conducted. The standardized IHE
variable, standardized dummy variables of the experimental conditions, and interaction terms of the respective dummy variables with the standardized IHE variable were entered into the model as predictors. Participants’ task-related score was entered into the model as outcome variable. The regression model was not significant (F(5, 96) = 0.62, p = .682). Additionally, residuals appeared to be non-normally distributed, further underlining that the model was no appropriate predictor for the proposed relationship. On the basis of the model, we failed to reject the null hypothesis. Therefore, H3 was not supported.
Another multiple regression analysis was conducted to test the proposed moderation effect of IHE on the relationship between perceived power asymmetry and the cognitive processing of non-task-related information. Entering the same predictor variables as in the model testing H3, participants’ non-task-related score was entered as outcome variable. The overall model was not significant (F(5, 96) = 1.37, p = .244). Heteroscedasticity was evident, indicating that the model was not a suited predictor for the proposed relationship. H4 was thus not supported.
Discussion
The objective of the current study was to deepen our understanding of the influence of hierarchically based power on individuals’ cognitive processing of information within the
organizational context. It applied the social distance theory of power (Magee & Smith, 2013) to the organizational domain to test whether different perceptions of power in dyadic internal organizational communication translate into distinct communicative outcomes through receivers’ cognitive information processing. Entailing IHE as a dispositional factor, the study anticipated the individual dynamics of social perception in dyadic communication.
The first hypothesis assumed that perceived power asymmetry affected cognitive information processing of task-related information. HPI were assumed to process the highest amount of non-task-related information, followed by LPI; EPI were assumed to process the lowest amount of task-related information. The results of the analysis neither confirmed the assumption of the general effect, nor did in-sample observations indicate a pattern following that of the hypothesis.
Multiple factors can explain this non-significant finding. First, higher levels of construal are associated with selective information processing in accordance with a
contextually relevant goal (DeWall et al., 2011). Providing an organizational goal as part of the experiment, as has been done in previous studies (Overbeck & Park, 2001), may have accentuated differences in goal-related information. Second, compared to task-related items, non-task-related items may be considered more unusual (e.g. “Lebanese food”). As unusual criteria of perceived objects affect recognition memory (Light, Stuart, & Hollander, 1979), they may have increased participants’ overall attentiveness to non-task-related information. A notion also explaining the overall superiority of non-task-related scores to task-related scores. Third, ambitions to move upward in the hierarchy (Cohen, 1958) and job satisfaction
(O’Reilly, 1978) affect task-related communication behavior. Including measures to account for these intrinsic motivations may have improved the model.
The second hypothesis assumed an effect of perceived power asymmetry on the receiver’s cognitive processing of non-task-related information. HPI were expected to
process less task-related information than LPI, who were expected to process less non-task-related information than EPI. The analysis partially supported these assumptions by indicating an effect, but suggesting a different distribution of scores than expected. Given the subtle manipulation and considering that all participants were shown identical information, this finding is compelling.
Only obtaining significant results when including demographic covariates, paired with its small size, suggest that the observed effect is sensitive to contextual factor and was unstable in the current sample. This pushes organizational culture and its implications on power and hierarchy in the foreground. Being based on perceptions of and differing tendencies towards power (Hofstede, 1980), organizational culture is assumed to become behaviorally evident through organizational climate (Schein, 1990), permeating social interactions at all organizational levels. With organizational cultures imposing more
authoritarian belief systems (Magee & Galinsky, 2008) may therefore come a stronger sense of legitimacy of hierarchy – a factor that Magee and Smith (2013) also suggest to play a role in the social distance theory of power.
While small, the observed effect suggests that the flow of non-task-related information within organizations is affected by power relations rooted in the organizational hierarchy. The cognitive response to power asymmetry is in line with previous research on power and cognition outside the organizational field (Waytz et al., 2015). Findings also follow the social distance theory of power (Magee & Smith, 2013), stating that LPI will be more attentive to HPI. Considering that EPI the lowest amounts of social distance, they appear to show a baseline score in non-task-related information processing. The bidirectional difference of groups experiencing power asymmetry makes overt the effect of hierarchical differentiation: While downwards communication of non-task-related information is propelled by
The bidirectional deviation of LPI and HPI from EPI was contrary to expectations, which assumed EPI to construe at the lowest level (Magee & Smith, 2013). The understanding of hierarchy as stratification along a socially valued dimension (Magee & Galinsky, 2008) can help explain this observation: encountering a high- or low-power individual may prime subjects with their own power(lessness), give rise to perceptions of social distance, and activate the cognitive mechanisms associated with it (Smith & Trope, 2006).
In both types of information tested, no moderation of IHE was found. Multiple reasons can help explain these non-significant findings. First, contrary to previous studies (Carter et al., 2006; Sandage et al., 2017), the reliability of the composite IHE measure was low ( =
.65). It therefore may not have captured the full extent of participants’ actual IHE. Another possible explanation is that IHE does affect social judgment (Mast, 2005a), but not social behavior. A third explanation for the non-significant result may be the relational nature of IHE. Mast (2005a) acknowledges that interpersonal dynamics of IHE may differ from her individually-based observations.
The answer to the initially formulated research question is multi-faceted. Depending on the type of information, hierarchical differentiation between the receiver and sender of a message does affect the receivers’ cognitive processing of the message’s content. More specifically, in the context of non-task-related communication, hierarchical differentiation between sender and receiver gives rise to social distance which, in turn, affects receivers’ processing of information. This becomes manifest in LPI allocating more cognitive resources to the processing of information than HPI do. As EPI share a more symmetric power
relationship, their cognitive processing of non-task-related information is higher than that of HPI, but lower than that of LPI.
Theoretical and Practical Contributions
This study makes multiple contributions to theory and practice. First, the study
contributes to research by showing that the distortion of information not only occurs through senders’ motivations, but also through the receivers’ processing; thus identifying another moment of potential information distortion in internal organizational communication. Second, the current study extends our knowledge of the cognitive implications of hierarchy within the organizational context. It showed that – for non-task-related organizational information – interpersonal power relations affect cognitive processes, which lead to distinct communicative outcomes. Third, in the current study, the context of application of the social distance theory of power was extended; it was shown that the theory can be applied to interpersonal dyadic communication in organizations. Finally, the study further underlined the relational nature of the power and social distance by investigating the effect in the population of employees of elevated hierarchy: all participants take on the role of HPI, LPI, and EPI in their everyday activities. Differences in cognitive information processing are therefore not due to a dispositional feeling of power(lessness), internalized by employees. This notion extends the understanding of social distance as its mechanisms appear to apply within members of all hierarchical levels, and in accordance with the power relation they share with their interactional partner (Mast, 2010).
The managerial implications of the current study are manifold as its findings suggest that non-task-related information processing is subject to different cognitive processing. A trade-off in information processing when communicating across rather than within
hierarchical levels was evident. In contexts of sole information distribution, following the hierarchical reporting structure may therefore be effective. In contexts requiring dialogue, such as employee engagement or team bonding, vertical communication structures could be
loosened. Practitioners are therefore urged to think of hierarchy as a strategic tool that can be pronounced more or less strongly depending on context.
Additionally, reducing general psychological distance between HPI and LPI may – through less physical distance, and more frequent HPI-LPI interactions - contribute to reducing social distance and thus lead to a more symmetric information processing of non-task-related information within the organization.
Limitations and future research perspectives
The current study also had limitations which should be considered when interpreting its findings. First, convenience sampling was associated with drawbacks. As not all employees of elevated hierarchy have an account with or actively engage in activities on a social medium, the external validity of the sample is limited. The instability of the observed effect underlines the importance of the organizational context when analyzing intra-organizational communication. Future studies can overcome these limitations by conducting research within organizations. An organizational mandate could furthermore increase participants’
willingness to participate and to finish the study. Another reason for drawing a sample from select organizations is that of organizational culture, a unique factor varying across all organizations. Increasing external validity by tailoring stimulus material to specific
organizational environments may result in higher ecological validity and reinforce the value of future studies.
Another limitation of the current study was that of power-related language use (Clegg, 1987). Individuals’ tone differs in accordance with the power-relation to their interaction partner. Using the same fictive e-mail for all conditions, a potential misalignment of tone of voice and power-relation to the receiver may thus have led to cognitive dissonance and, as a consequence, to a distortion of results.
In light of these limitations, the current study also opens new avenues for future research. Scholars are encouraged to confirm the findings of the current study and to
investigate the effect of power asymmetry on task-related information processing. Additional research is also needed to investigate the implications of hierarchy on the qualitative
dimension of cognitive processing. Analyzing the level of abstraction evident in respondents’ recall could also extend the application of the social distance theory of power. Future
research should also investigate the intra-personal dynamics of social distance and its effect on cognitive processing. More specifically, scholars are encouraged to analyze whether and how an individuals’ perception of social distance changes in accordance with their
interactional partner. Finally, further research could investigate whether the effect of hierarchy on cognitive information processing is also evident for different types of
communication media. Analyzing whether hierarchical differentiation between organizational members also translates into perceived power asymmetry and social distance on (enterprise) social media could provide valuable insights for research and practice.
Conclusion
The influence of power on cognitive information processing can be explained by the social distance theory of power. Employees process most non-task-related information received from supervisors. Receiving non-task-related information from a subordinate, employees process least non-task-related information. Receiving it from a peer, employees process more non-task-related information than when receiving it from a subordinate, but less than when receiving it from a supervisor. Finding no significant effect in the context of task-related information stressed the importance of conducting future research within rather than across organizations. Observations did not indicate IHE to play an influential role in the effect.
Bringing together individual, interpersonal, and organizational factors, this study identified another source of information distortion in internal organizational communication, suggesting that non-task-related information changes as it travels through the organizational domain.
References
Balmer, J. M., Stuart, H., & Greyser, S. A. (2009). Aligning identity and strategy: Corporate branding at British Airways in the late 20th century. California Management
Review, 51(3), 6-23. doi: 10.2307/41166491
Baron, J. N., & Pfeffer, J. (1994). The social psychology of organizations and inequality. Social Psychology Quarterly, 57(3), 190-209. doi: 10.2307/2786876
Bem, D. J. (1972). Self-perception theory. Advances In Experimental Social Psychology, 6, 1-62. doi: 10.1016/S0065-2601(08)60024-6
Bunderson, J. S., & Reagans, R. E. (2011). Power, status, and learning in
organizations. Organization Science, 22(5), 1182-1194. doi: 10.1287/orsc.1100.0590 Carley, K. M., & Lin, Z. (1997). A theoretical study of organizational performance under
information distortion. Management Science, 43(7), 976-997. doi: 10.1287/mnsc.43.7.976 Carroll, G. R., & Hannan, M. T. (2000). The demography of corporations and industries.
Princeton, NJ: Princeton University Press.
Carter, J. D., Hall, J. A., Carney, D. R., & Rosip, J. C. (2006). Individual differences in the acceptance of stereotyping. Journal of Research in Personality, 40(6), 1103-1118. doi: 10.1016/j.jrp.2005.11.005
Chen, S., Tharp, J. A., & Kuehn, M. M. (2017). They love me, they love me not?: Social power shapes expectations of acceptance and concerns about rejection. Social
Cognition, 35(5), 563-584. doi: 10.1521/soco.2017.35.5.563
Choo, C. W. (1996). The knowing organization: How organizations use information to construct meaning, create knowledge and make decisions. International Journal of
Information Management, 16(5), 329-340. doi: 10.1016/0268-4012(96)00020-5
Clegg, S. R. (1987). The language of power and the power of language. Organization
Cohen, A. R. (1958). Upward communication in experimentally created hierarchies. Human
Relations, 11(1), 41-53. doi: 10.1177/001872675801100103
Craik, F. I. M. (1983). On the transfer of information from temporary to permanent
memory. Philosophical Transactions of the Royal Society of London Series B, 302, 341-358. doi: 10.1098/rstb.1983.0059
Craik, F. I. M., & McDowd, J. M. (1987). Age differences in recall and recognition. Journal
of Experimental Psychology: Learning, Memory, and Cognition, 13(3), 474-479.
doi: 10.1037/0278-7393.13.3.474
Davenport, S., & Leitch, S. (2005). Circuits of power in practice: Strategic ambiguity as delegation of authority. Organization Studies, 26(11), 1603-1623.
doi: 10.1177/0170840605054627
DeWall, C. N., Baumeister, R. F., Mead, N. L., & Vohs, K. D. (2011). How leaders self-regulate their task performance: Evidence that power promotes diligence, depletion, and disdain. Journal of Personality and Social Psychology, 100(1), 47-65.
doi: 10.1037/a0020932
Diefenbach, T., & Sillince, J. A. A. (2011). Formal and informal hierarchy in different types of organization. Organization Studies, 32(11), 1515-1537.
doi: 10.1177/0170840611421254
Freiman, J. A., Chalmers, T. C., Smith Jr, H., & Kuebler, R. R. (1978). The importance of beta, the type II error and sample size in the design and interpretation of the randomized control trial: Survey of 71 negative trials. New England Journal of Medicine, 299(13), 690-694. doi: 10.1056/NEJM197809282991304
Galinsky, A. D., Gruenfeld, D. H., & Magee, J. C. (2003). From power to action. Journal of
Galinsky, A. D., Magee, J. C., Gruenfeld, D. H., Whitson, J. A., & Liljenquist, K. A. (2008). Power reduces the press of the situation: Implications for creativity, conformity, and dissonance. Journal of Personality and Social Psychology, 95(6), 1450-1466. doi: 10.1037/a0012633
Guinote, A. (2007a). Power and goal pursuit. Personality and Social Psychology
Bulletin, 33(8), 1076-1087. doi: 10.1177/0146167207301011
Guinote, A. (2007b). Power affects basic cognition: Increased attentional inhibition and flexibility. Journal of Experimental Social Psychology, 43(5), 685-697.
doi: 10.1016/j.jesp.2006.06.008
Guinote, A. (2007c). Behaviour variability and the situated focus theory of power. European
Review of Social Psychology, 18(1), 256-295. doi: 10.1080/10463280701692813
Guinote, A. (2008). Power and affordances: When the situation has more power over
powerful than powerless individuals. Journal of Personality and Social Psychology, 95(2), 237-252. doi: 10.1037/a0012518
Guinote, A. (2010). The situated focus theory of power. In A. Guinote & T. K. Vescio (Eds.), The Social Psychology of Power (pp. 141-173). New York, NY: Guilford Press. Guinote, A. (2013). Social power and cognition. In D. E. Carlston (Ed.), Oxford library of
psychology. The oxford handbook of social cognition (pp. 575-589). New York, NY:
Oxford University Press. doi: 10.1093/oxfordhb/9780199730018.013.0028
Halevy, N., Chou, E. Y., & Galinsky, A. D. (2011). A functional model of hierarchy: Why, how, and when vertical differentiation enhances group performance. Organizational
Psychology Review, 1(1), 32-52. doi: 10.1177/2041386610380991
Halevy, N., Chou, E. Y., Galinsky, A. D., & Murnighan, J. K. (2012). When hierarchy wins: Evidence from the national basketball association. Social Psychological and Personality
Hasher, L., & Zacks, R. T. (1979). Automatic and effortful processes in memory. Journal of
Experimental Psychology: General, 108(3), 356-388. doi: 10.1037/0096-3445.108.3.356
Hayes, A. F., & Krippendorff, K. (2007). Answering the call for a standard reliability measure for coding data. Communication Methods and Measures, 1(1), 77-89. doi: 10.1080/19312450709336664
Hofstede, G. (1980). Culture and organizations. International Studies of Management &
Organization, 10(4), 15-41. doi: 10.1080/00208825.1980.11656300
Hogg, M. A. (2001). A social identity theory of leadership. Personality and Social
Psychology Review, 5(3), 184-200. doi: 10.1207/S15327957PSPR0503_1
Hollingworth, H. L. (1913). Characteristic differences between recall and recognition. The
American Journal of Psychology, 24(4), 532-544. doi: 10.2307/1413450
Jacoby, L. L. (1983). Remembering the data: Analyzing interactive processes in reading. Journal of Verbal Learning & Verbal Behavior, 22(5), 485-508. doi: 10.1016/S0022-5371(83)90301-8
Kelley, H. H. (1951). Communication in experimentally created hierarchies. Human
Relations, 4(1), 39-56. doi: 10.1177/001872675100400102
Keltner, D., Gruenfeld, D. H., & Anderson, C. (2003). Power, approach, and inhibition.
Psychological Review, 110(2), 265-284. doi: 10.1037/0033-295X.110.2.265
Kuehn, M. M., Chen, S., & Gordon, A. M. (2015). Having a thicker skin: Social power buffers the negative effects of social rejection. Social Psychological and Personality
Science, 6(6), 701-709. doi: 10.1177/1948550615580170
Leavitt, H. J. (2005). Top down: Why hierarchies are here to stay and how to manage them
Liefooghe, A. P. D., & Davey, K. M. (2001). Accounts of workplace bullying: The role of the organization. European Journal of Work and Organizational Psychology, 10(4), 375-392. doi: 10.1080/13594320143000762
Liu, Y., & Fu, X. (2007, July). How does distraction task influence the interaction of working memory and long-term memory?. In International conference on engineering psychology
and cognitive ergonomics (pp. 366-374). Berlin/Heidelberg: Springer. doi:
10.1007/978-3-540-73331-7_40
Luhmann, N. (1984). Soziale systeme (Vol. 478). Frankfurt am Main: Suhrkamp.
Magee, J. C., & Galinsky, A. D. (2008). 8 Social Hierarchy: The self‐reinforcing nature of power and status. Academy of Management Annals, 2(1), 351-398.
doi: 10.1080/19416520802211628
Magee, J. C., & Smith, P. K. (2013). The social distance theory of power. Personality and
Social Psychology Review, 17(2), 158-186. doi: 10.1177/1088868312472732
Mannix, E. A., & Neale, M. A. (1993). Power imbalance and the pattern of exchange in dyadic negotiation. Group Decision and Negotiation, 2(2), 119-133.
doi: 10.1007/BF01884767
Mast, M. S. (2005a). Interpersonal hierarchy expectation: Introduction of a new construct. Journal of Personality Assessment, 84(3), 287-295.
doi: 10.1207/s15327752jpa8403_08
Mast, M. S. (2005b). The world according to men: It is hierarchical and stereotypical. Sex
Roles, 53(11), 919-924. doi: 10.1007/s11199-005-8310-6
Mast, M. S. (2010). Interpersonal behaviour and social perception in a hierarchy: The interpersonal power and behaviour model. European Review of Social Psychology, 21(1), 1-33. doi: 10.1080/10463283.2010.486942
Norris, C. E., & Colman, A. M. (1992). Context effects on recall and recognition of magazine advertisements. Journal of Advertising, 21(3), 37-46.
doi: 10.1080/00913367.1992.10673374
O'Reilly, C. A., & Roberts, K. H. (1974). Information filtration in organizations: Three experiments. Organizational Behavior and Human Performance, 11(2), 253-265. doi: 10.1016/0030-5073(74)90018-X
O’Reilly, C. A. (1978). The intentional distortion of information in organizational
communication: A laboratory and field investigation. Human Relations, 31(2), 173-193. doi: 10.1177/001872677803100205
Overbeck, J. R., & Park, B. (2001). When power does not corrupt: Superior individuation processes among powerful perceivers. Journal of Personality and Social
Psychology, 81(4), 549-565. doi: 10.1037/0022-3514.81.4.549
Posner, M. I., & Dehaene, S. (1994). Attentional networks. Trends in Neurosciences, 17(2), 75-79. doi: 10.1016/0166-2236(94)90078-7
Ratneshwar, S., & Chaiken, S. (1991). Comprehension's role in persuasion: The case of its moderating effect on the persuasive impact of source cues. Journal of Consumer
Research, 18(1), 52-62. doi: 10.1086/209240
Reitzig, M., & Maciejovsky, B. (2015). Corporate hierarchy and vertical information flow inside the firm—a behavioral view. Strategic Management Journal, 36(13), 1979-1999. doi: 10.1002/smj.2334
Robbins, S. P., & Judge, T. A. (2009). Organizational behavior, 13th edition. Upper Saddle
River, NJ: Pearson Education.
Rumelt, R. P., & Lamb, R. (1997). Towards a strategic theory of the firm. Resources, firms,
Sandage, S. J., Jankowski, P. J., Crabtree, S. A., & Schweer-Collins, M. L. (2017). Calvinism, gender ideology, and relational spirituality: An empirical investigation of worldview differences. Journal of Psychology & Theology, 45(1), 17-32.
Scandura, T. A., & Lankau, M. J. (1997). Relationships of gender, family responsibility and flexible work hours to organizational commitment and job satisfaction. Journal of
Organizational Behavior, 18(4), 377-391. doi:
10.1002/(SICI)1099-1379(199707)18:4<377::AID-JOB807>3.0.CO;2-1
Schein, E. H. (1990). Organizational culture (Vol. 45, No. 2, p. 109). American Psychologist,
45(2), 109-119. doi: 10.1037/0003-066X.45.2.109
Scholl, A., & Sassenberg, K. (2015). Better know when (not) to think twice: How social power impacts prefactual thought. Personality and Social Psychology Bulletin, 41(2), 159-170. doi: 10.1177/0146167214559720
Sheposh, J. P., & Gallo Jr, P. S. (1973). Asymmetry of payoff structure and cooperative behavior in the prisoner's dilemma game. Journal of Conflict Resolution, 17(2), 321-333. doi: 10.1177/002200277301700208
Slepian, M. L., Ferber, S. N., Gold, J. M., & Rutchick, A. M. (2015). The cognitive
consequences of formal clothing. Social Psychological and Personality Science, 6(6), 661-668. doi: 10.1177/1948550615579462
Smith, P. K., Jostmann, N. B., Galinsky, A. D., & Van Dijk, W. W. (2008). Lacking power impairs executive functions. Psychological Science, 19(5), 441-447. doi: 10.1111/j.1467-9280.2008.02107.x
Smith, P. K., & Trope, Y. (2006). You focus on the forest when you're in charge of the trees: Power priming and abstract information processing. Journal of Personality and Social
