Content uploaded by Dina Van Dijk
Author content
All content in this area was uploaded by Dina Van Dijk on Dec 04, 2017
Content may be subject to copyright.
Task type as a moderator of positive/negative
feedback effects on motivation and performance:
A regulatory focus perspective
DINA VAN DIJK
1
*AND AVRAHAM N. KLUGER
2
1
Department of Health Systems Management, Ben-Gurion University of the Negev, Beer-Sheva, Israel
2
The Jerusalem School of Business Administration, The Hebrew University— Mt. Scopus, Jerusalem, Israel
Summary Applying Higgins’ regulatory focus theory, we hypothesized that the effect of positive/negative feedback on
motivation and performance is moderated by task type, which is argued to be an antecedent to situational
regulatory focus (promotion or prevention). Thus, first we demonstrated that some tasks (e.g., tasks requiring
creativity) are perceived as promotion tasks, whereas others (e.g., those requiring vigilance and attention to
detail) are perceived as prevention tasks. Second, as expected, our tests in two studies of the moderation
hypothesis showed that positive feedback increased self-reported motivation (meta-analysis across samples:
N¼315, d¼0.43) and actual performance (N¼55, d¼0.67) among people working on promotion tasks,
relative to negative feedback. Positive feedback, however, decreased motivation (N¼318, d¼0.33) and
performance (N¼55, d¼0.37) among individuals working on prevention tasks, relative to negative
feedback. These findings suggest that (a) performance of different tasks can affect regulatory focus and (b)
variability in positive/negative feedback effects can be partially explained by regulatory focus and task type.
Copyright #2010 John Wiley & Sons, Ltd.
Keywords: feedback; feedback sign; motivation; performance; regulatory focus; prevention and promotion
foci; task type
Introduction
Although the effects of various types of feedback (formal performance appraisals, grades, etc.) on motivation and
performance have received considerable theoretical and empirical attention, no general principle predictive of the
effectiveness of feedback interventions has arisen (DeNisi & Kluger, 2000; Kluger & DeNisi, 1996, 1998).
Surprisingly, not even the feedback sign— whether the feedback message conveys success or failure—was found to
moderate the effectiveness of feedback interventions (Kluger & DeNisi, 1996). That is, both negative and positive
feedback can either increase or decrease performance. Trying to explain these puzzling findings, Van-Dijk and
Kluger (2004) suggested that regulatory focus (Higgins, 1997, 1998) moderates the effect of feedback sign (negative/
positive) on motivation. Specifically, using scenario experiments, they showed that positive feedback is a greater
motivator than negative feedback when individuals are promotion-focused, whereas negative feedback motivates
more than positive feedback when people are prevention-focused.
Here, we seek to transfer Van Dijk and Kluger’s (2004) results into the domain of tasks and suggest an argument
with two components: First, different types of tasks induce different regulatory foci. Specifically, tasks that require
creativity and open-mindedness induce promotion focus, whereas tasks requiring vigilance, accuracy and adherence
Journal of Organizational Behavior, J. Organiz. Behav. 32, 1084–1105 (2011)
Published online 7 September 2010 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/job.725
* Correspondence to: Dina Van Dijk, Health Systems Management, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel.
E-mail: dinav@bgu.ac.il
Copyright #2010 John Wiley & Sons, Ltd.
Received 22 January 2008
Revised 10 July 2010, Accepted 25 July 2010
to rules induce prevention focus. Second, the regulatory focus induced by the task moderates the relationship
between the sign of the feedback and motivation, such that when performing a task that requires creativity and open
mindedness (i.e., promotion task), one would benefit more from positive feedback than negative feedback. On
contrary, when performing a task that requires accuracy and adherence to rules (i.e., prevention task), one benefits
more from negative rather than positive feedback.
Task characteristics were already shown to moderate the effects of the feedback sign on performance by
Vancouver and Tischner (2004). Specifically, they showed that when cognitive resources were needed to perform the
task, feedback sign was positively correlated with performance, whereas when cognitive resources were not needed,
positive feedback did not improve performance and negative feedback increased performance, albeit weakly. We
continue to explore the role played by the nature of the task in moderating feedback-sign effects on performance by
considering the motivational, rather than cognitive demands of the task.
In considering the motivational demands of the task, we first describe regulatory focus theory and discuss its
relevance to feedback sign effects. Next, we explain the link between task type and regulatory focus, suggesting that
the type of task could serve as an antecedent or source for regulatory focus. Finally, we present our moderation
hypothesis, according to which the task type as a source for regulatory focus moderates the effect of feedback sign on
motivation and performance.
Regulatory focus theory
Higgins (1997) suggested that people have two basic self-regulation systems: One system regulates the achievement
of rewards and focuses individuals on promotion goals, while the other regulates the avoidance of punishment and
focuses individuals on prevention goals. Individuals’ regulatory foci are evoked by three factors: (a) The needs that
people seek to satisfy—security versus growth, (b) the nature of the goal that people are trying to achieve — ought
versus ideal, and (c) the psychological context in which people act—loss/non-loss versus gain/non-gain situations
(Brockner & Higgins, 2001; Brockner, Paruchuri, Idson, & Higgins, 2002; Higgins, 1997, 1998). These three
regulatory-foci antecedents have been shown to stem from both chronic individual (i.e., trait-like) differences and
situational factors (i.e., primed through situational cues).
Both situational and chronic regulatory foci have been tested in various ways using different operationalizations
(e.g., Freitas, Liberman, & Higgins, 2002; Higgins, 1997; Higgins, Friedman, Harlow, Idson, Ayduk, & Taylor,
2001; Higgins, Shah, & Friedman, 1997; Kark & Van-Dijk, 2007; Liberman, Molden, Idson, & Higgins, 2001;
Lockwood, Jordan & Kunda, 2002; Shah & Higgins, 1997; Shah, Higgins, & Friedman, 1998; Van-Dijk & Kluger,
2004; Wallace & Chen, 2006). For example, situational regulatory focus has been manipulated by framing an
identical set of task payoffs for success or failure as involving ‘‘gain/non-gain’’ or ‘‘loss/non-loss’’ (e.g., Shah &
Higgins, 1997; Shah et al., 1998). Other ways of manipulating situational regulatory focus include priming ideals or
oughts (e.g., Freitas et al., 2002; Liberman et al., 2001), or priming regulatory goals (positive academic gain or
negative academic outcome; Lockwood et al., 2002). When studied as an individual difference variable, regulatory
focus has been assessed using the Self-Guide Strength Measure (e.g., Higgins et al., 1997; Shah & Higgins, 1997),
which measures the chronic accessibility of people’s ideals and oughts and by other types of measures such as the
Regulatory Focus Questionnaire (Higgins et al., 2001), regulatory focus scale (Lockwood et al., 2002), conservation
versus openness values (Van-Dijk & Kluger, 2004), and two measures of work-specific regulatory focus (Neubert,
Kacmar, Carlson, Chonko, & Roberts, 2008; Wallace & Chen, 2006).
These various operationalizations of regulatory foci reflect the conceptualization of prevention and promotion foci
as rich syndromes with multiple antecedents and consequences (Higgins, 1997). Yet the goals that are salient under
each focus are very different. In addition to its punishment avoidance goal, prevention focus ‘‘is likely to have
minimal goals, short-term perspective, sensitivity to social pressures, and concern with goal maintenance,
conservation, keeping the status quo,’’ and it ‘‘is experienced as a necessity, as an obligation, and as something that
people feel that they have to do.’’ In contrast, in addition to its rewards attainment goal, promotion focus ‘‘is also
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1085
likely to have a tendency toward striving for maximal goals, long-term perspective, attunement to intrinsic and
internal needs, and concerns with development, change, and ideals,’’ and it ‘‘is experienced as a wish, as a desire, and
as something that people feel that they are eager to do’’ (Kluger & Ganzach, 2004: p. 78). This suggests that regulatory
foci are broad constructs that could apply to various aspects of life. The relevance of these foci to the effects of positive
versus negative feedbacks on motivation and performance involves the mechanism of ‘‘regulatory fit.’’
Regulatory fit
According to Higgins, people experience a ‘‘regulatory fit’’ when their regulatory focus (promotion or prevention)
fits their outcome (gain/non-gain or loss/non-loss), and this regulatory fit increases the perceived value of what they
are doing (Higgins, 2000; Sassenberg, Jonas, Shah, & Brazy, 2007; Shah et al., 1998; Spiegel, Grant-Pillow, &
Higgins, 2004). For example, Shah et al. (1998) found that in more promotion-focused individuals, motivation and
performance increase under task incentives framed in terms of gain/non-gain. Conversely, as individuals are more
prevention-focused, motivation and performance increase under task incentives framed in terms of loss/non-loss.
Applying regulatory fit to the sign of feedback, Van Dijk and Kluger (2004) demonstrated that when the regulatory
focus fits the feedback sign, motivation is higher than in cases in which it does not fit. Specifically, they showed that
both negative feedback under prevention focus and positive feedback under promotion focus increase motivation.
This effect was found in scenario experiments where they operationalized regulatory focus either with chronic
individual differences (e.g., high on ‘‘self-direction’’ values versus ‘‘security’’ values) or with situational
manipulation (e.g., framing a job as one ‘‘you wanted to do’’ versus one ‘‘you had to do’’). These findings provide an
explanation for the puzzling variability of feedback-sign effects on performance.
The present studies advance this earlier research by suggesting that the principle of regulatory fit also operates between
the nature of the task and the feedback sign. We suggest that the nature of the task affects the situational regulatory focus,
such that tasks requiring eagerness and creativity induce situational promotion focus, whereas tasks that require vigilance,
conformity, and attention to detail induce situational prevention focus. Therefore, tasks triggering promotion focus will fit
with positive feedback and tasks triggering prevention focus will fit with negative feedback. To substantiate this hypothesis,
we provide a rationale for the capacity of the nature of the task to affect situational regulatory focus.
Task type as an antecedent of situational regulatory focus
The next step then is to identify the nature of the tasks that are conducive to inducing prevention rather than
promotion foci and vice versa without any additional priming. Recent findings indicate that prevention and
promotion foci lead to different behavior tendencies. Specifically, prevention focus leads to conservative behavior,
repetitiveness, error avoidance, accuracy, safety, and vigilance, whereas promotion focus leads to creativity, risk
taking, speed, production, and eagerness (Bass, De Dreu, & Nijstad, 2008; Crowe & Higgins, 1997; Forster, Higgins,
& Bianco, 2003; Friedman & Forster, 2001; Liberman, Idson, Camacho, & Higgins, 1999; Wallace & Chen, 2006).
Casting these different behavioral tendencies into the realm of tasks, we assume that there are certain tasks that
require prevention tendencies, such as vigilance, accuracy, or error avoidance, while others require promotion
tendencies, such as eagerness, speed, or creativity.
Therefore, performing various types of tasks could prime different types of regulatory focus. In other words, not
only do the regulatory foci affect performance as reviewed above, different types of performance may affect
regulatory foci as well. Therefore, when people perform a task requiring caution, the prevention focus is likely to be
activated, whereas when people perform a task requiring creativity, the promotion focus is likely to be activated.
Some examples could clarify why certain tasks activate prevention focus, while others activate promotion focus.
Tasks requiring vigilance, such as identifying objects on a radar screen, detecting errors in company reports, or
cleaning, keep individuals focused on finding what is wrong and what they should avoid (suspicious objects, errors,
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
1086 D. VAN DIJK AND A.N. KLUGER
or dirt, respectively). Therefore, due to their nature, these tasks produce prevention focus. In contrast, tasks requiring
eagerness, such as seeking out new ideas, initiating organizational changes, or developing innovative products, focus
on finding what is right and what could be gained (ideas, changes, or innovations, respectively). Therefore, these
tasks produce promotion focus. Thus, schematically, promotion-focused tasks require eagerness, creativity, and
openness, whereas prevention-focused tasks require vigilance, attention to detail, and adherence to rules.
Another distinction between different task types in terms of regulatory focus was suggested by Bianco, Higgins,
and Klem (2003). The authors showed that some tasks are perceived as more ‘‘fun,’’ while others are considered
more ‘‘important.’’ (Another group of tasks was perceived as equally fun and important.) Specifically, in a pre-test
they found that people perceive academic activities, financial duties, household chores, and hygiene as more
‘‘important’’ than ‘‘fun,’’ whereas dating, games, partying, and sensual activities were considered more ‘‘fun’’ than
‘‘important.’’ Based on this, they demonstrated that performance was higher when there was a fit between the task’s
implicit perception (as fun or important) and the instructional framing for engaging in the task (as fun or important).
Our distinction between promotion-focused and prevention-focused tasks differs from the ‘‘fun/importance’’
distinction in two ways. First, we refer to tasks taken from the organizational context, such as generating new ideas,
initiating organizational changes, bookkeeping, and detecting errors and do not refer to leisure activities such as
dating, playing games, or partying. Second, our distinction emphasizes the behavioral tendency or strategy called for by
the task (e.g., creativity versus conformity), rather than thevalue or importance of the task as perceived by the performer
(suggested by the fun/importance categorization). Thus, prevention tasks are by nature conservative and necessitate
caution and error avoidance (e.g., proofreading), while promotion tasks are by nature creative and require an open mind
and imagination (e.g., developing new products). Nevertheless, both types of tasks can be equally important. For
example, both creativity and attention to detail are highly important to employees (Miron, Erez, & Naveh, 2004), but
nevertheless, require completely different skills and strategies. However, our distinction between different types of tasks
is similar to that made by Bianco et al. (2003), in the sense that they reflect a similar meaning of doing something out of
desire or ambition rather than out of duty or necessity. Therefore, we assume that tasks requiring eagerness and creativity
will be perceived as desirable and something the individual will want to accomplish, while tasks that require vigilance
and conformity will be considered mandatory and something the individual is obligated to fulfill.
If the type of task can induce specific regulatory focus, previous findings regarding the effect of the regulatory
focus could be applied to task type. Specifically, regulatory focus has been found to moderate the effects of sign of
feedback on both motivation (Van Dijk & Kluger, 2004) and performance (Idson & Higgins, 2000). Thus, we may
assume that task type (prevention versus promotion tasks) will moderate the relationship between feedback sign,
motivation, and performance.
Task type moderates the effect of feedback sign on motivation and performance
If there are indeed promotion-focused tasks and prevention-focused tasks, then we should expect to find an
interaction between task-induced regulatory focus and feedback sign. Specifically, we expect that positive feedback,
more than negative feedback, will contribute to the performance of tasks that require eagerness and creativity, with
the opposite being true for tasks requiring vigilance and adherence to rules. Negative feedback is most important for
tasks aimed at identifying errors or avoiding problems as it indicates to people when they are wrong and what they
should avoid. Conversely, positive feedback is significantly important for people searching for new alternatives or
ideas as it indicates that they are making significant progress.
In fact, instructing people to monitor errors versus successes in a vigilance (prevention) task resulted in better
performances (Wade, 1974). Likewise, an examination of goal regulation over time (Ilies & Judge, 2005) using two
creative promotion tasks (finding different uses for an object and finding a fourth word related to three other words)
demonstrated that negative feedback reduced goal levels over time, contrasted to positive feedback which increased
them. Thus, if the nature of a task influences regulatory focus, our hypothesis can account for the findings of both
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1087
Wade (1974) and Ilies and Judge (2005). In summary, we hypothesize that task type will moderate the effect of
feedback sign on motivation and performance.
H1: For tasks that require vigilance and adherence to rules, negative feedback will increase motivation and
performance more than positive feedback. For tasks that require eagerness and creativity, positive feedback will
increase motivation and performance more than negative feedback.
Next, we describe a pre-test conducted to establish the relationship between the task type and the situational-
regulatory focus. Then we present two studies that test the moderating role of task type on the effect of feedback sign
on self-reported motivation using scenarios (Study 1) and both self-reported motivation and actual performance in a
laboratory study (Study 2).
Pre-Testing the Connection between Task Type and Regulatory Focus
Here we sought to test whether there is a general consensus regarding the link between task type and the regulatory
focus that it supposedly induces. Thus, we presented employed individuals with a list of tasks (pre-classified into
prevention and promotion tasks), asking them to state whether each task was a ‘‘promotion task’’ or a ‘‘prevention
task.’’ If non-experts easily make a distinction between the two types of tasks, then we can deduce that there is a
consensus regarding the type of task that creates certain regulatory focus in the ‘‘real world’’ of working people.
Pre-test method
We first generated a list of 10 various tasks that are performed frequently at work. Afterwards, we gave the list to a
group of six managers. The managers, who are acquainted with the first author, were from different types of
organizations (e.g., public companies, high-tech). They were asked to review the list and to add new tasks or change
current tasks. The managers were unaware of the purpose of the task list as they had only been asked to help us
‘‘compose a list of tasks that people usually perform at work.’’ After receiving the lists from all of the managers, we
compiled one list of 30 tasks by arranging the tasks into ‘‘groups’’ of tasks. We then sent the newly completed list, by
email to the managers, requesting that they review it and make specific comments on the clarity of the tasks. After
reviewing all of the suggestions and comments, we modified the list, reworded some of the tasks, and removed a few
that appeared nearly identical so as to avoid overload for those rating the list. The final list was comprised of 23 tasks.
Next, we gave the list to a group of independent Subject Matter Experts (SME) that included seven regulatory
focus theory experts (four researchers who study regulatory focus and three doctoral students). The SMEs were
asked to classify the items on the list as either prevention or promotion tasks. The seven experts agreed on 21 of the
23 tasks and were unsure or held different views on two tasks. Eleven tasks were coded as promotion by the SMEs
(e.g., generating ideas, creative problem solving, assimilating new technology, challenging decision making,
initiating changes), 10 tasks were coded as prevention (e.g., detecting errors, maintaining safety, bookkeeping, work
scheduling, maintaining quality control), and the two undetermined (neutral) tasks were: Involvement in
organizational politics and training employees. (See Table 1 for the full list of tasks.)
Questionnaire development
Based on the task list, we developed a questionnaire that included the 23 tasks, listed in random order. There was a
short explanation on promotion versus prevention tasks at the beginning of the questionnaire followed by the list of
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
1088 D. VAN DIJK AND A.N. KLUGER
tasks. To avoid experimenter bias (knowing the researchers’ purpose), the explanation that we provided did not
include words that explicitly indicated the nature of the task (e.g., tasks that require creativity and originality or tasks
requiring accuracy and attention to detail). Rather, we explained the difference between promotion and prevention
tasks by using a more general definition of the tasks as ‘‘promote attaining a goal’’ versus ‘‘avoid not attaining a goal’’
(e.g., doing them will develop and promote the organization versus not doing them will adversely affect
the organization). We also used the difference between doing something ‘‘dutifully’’ versus doing something
‘‘willingly.’’ By wording the explanation for the two types of tasks in a general manner, we ensured that the respondents’
categorization would not be directly affected by our conception of task demands (e.g., creative versus accuracy), but
rather more by respondents’ general sense of which tasks evoke ‘‘prevention’’ and which tasks evoke ‘‘promotion.’’
The explanation (translated from Hebrew) was:
During our day-to-day lives we perform different types of tasks. Below is a description of two types of tasks that
people often perform.
Read the description of the two types of tasks carefully. The first type is ‘‘prevention’’ tasks: These are tasks that
people feel obligated to perform because not doing them will adversely affect the organization.
This type of task is usually performed out of duty and necessity. The second type is ‘‘promotion’’ tasks: These are
Table 1. Means and standard deviations for ratings of the degree to which a task induces prevention focus and promotion focus,
Cohen’s dfor the difference between the prevention-promotion ratings, and rfor their correlations
Task
Prevention
ratings
Promotion
ratings
dr(prev/prom)MSDMSD
Promotion tasks
Exerting creative thought 3.10 1.23 4.42 0.69 1.32
0.03
Generating ideas 3.25 1.24 4.38 0.72 1.11
0.03
Challenging decision making 3.40 1.18 4.27 0.84 0.84
0.06
Creative problem solving 3.52 1.13 4.18 0.84 0.65
0.01
Presenting various alternatives 3.48 1.11 4.01 0.88 0.54
0.08
Developing workers’ career 3.65 .98 4.07 0.82 0.46
0.02
Assimilating new technology 3.82 1.11 4.23 0.90 0.40
0.30
Doing research & development 3.84 1.02 4.20 0.91 0.37
0.15
Initiating changes 3.59 1.03 3.87 0.94 0.27
0.06
Planning a task 3.73 1.07 4.00 0.81 0.29
0.14
Setting goals 3.97 1.05 3.85 1.03 0.11 0.05
Prevention tasks
Bookkeeping 4.41 0.83 2.80 1.17 1.58
0.02
Detecting errors 4.20 0.83 3.26 1.16 1.04
0.03
Budget planning 4.21 1.01 3.32 1.07 0.85
0.12
Maintaining safety 4.40 0.91 3.55 1.13 0.86
0.03
Work scheduling 4.15 0.90 3.26 1.32 0.78
0.26
Handling customer complaints 4.22 1.06 3.46 1.34 0.58
0.08
Cleaning 4.16 0.91 3.54 1.12 0.52
0.21
Following schedules 4.35 0.88 3.75 1.12 0.62
0.04
Supervising 4.06 0.97 3.53 1.03 0.53
0.30
Quality control 4.06 1.06 3.66 1.18 0.36
0.02
Neutral tasks
Involvement in org. politics 2.90 1.25 3.13 1.25 0.18 0.06
Training employees 3.86 1.09 4.00 0.97 0.13 0.07
Note:dis based on formulae for repeated measures (Borenstein et al., 2009: p. 29). Yet given the low correlations between the promotion and
prevention ratings, the d values are almost identical to dvalues obtained when the dependence of the measures is ignored.
p<0.05.
p<0.01, one-tailed; N¼121.
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1089
tasks that people are keen to perform because doing so develops and advances the organization.
This type of task is usually performed out of desire and interest.
No additional explanations were given. The list of tasks followed the explanation. Below each task were two
scales: (a) The extent to which the participant perceived the task as a ‘‘prevention’’ task and (b) the extent to which
the participant perceived the task as a ‘‘promotion’’task. The two were on a five-point scale ranging from ‘‘to a lesser
extent’’ through ‘‘to a greater extent.’’
Pre-test sample
A total of 121 participants responded to an internet questionnaire in exchange for approximately $5.00 (Midgam
Project Website: http://www.midgam.com/info.asp). We defined a priori a sample of 120 participants who had been
employed in their jobs for at least six months. Invitations to participate were sent to 600 respondents who met these
employment criteria out of 22 746 users registered to this Midgam Project. When the quota was filled, the survey was
closed to additional respondents. The sample included 45 males and 75 females; the average age was 38; 43 per cent
had academic degrees, 29 per cent had technical certificates, 23 per cent had high school diplomas, and 5 per cent had
not graduated from high school.
Pre-test findings and discussion
We ran a series of t-tests on the dependent measures to compare the prevention versus promotion scales. The results
of the t-tests, including the effect size calculated by Cohen’s dfor dependent measures (Borenstein, Hedges, Higgins,
& Rothstein, 2009: p. 29), and the correlations between the two scales (which were used to calculate d) are
summarized in Table 1.
As predicted, the average ‘‘prevention’’ rating in all ten of the pre-coded prevention tasks was significantly higher
than the average ‘‘promotion’’ rating, and in 10 of the 11 pre-coded promotion tasks, the average ‘‘promotion’’ rating
was significantly higher than the average ‘‘prevention’’ rating. (The ‘‘goal setting task,’’ which was pre-coded as a
‘‘promotion task,’’ did not yield a significant difference between the two scales.) Finally, no significant difference
between the prevention and promotion scales was found for three tasks, two of which–involving oneself in
organizational politics and training employees–had already been pre-coded as ‘‘undetermined.’’
The variance in the ratings of the prevention and promotion of the various tasks suggests that greater agreement is
found when there is a match between the SME’s classification and the scale. For example, the variance on the
promotion ratings is smaller than the variance on the prevention ratings for all tasks classified by SMEs as promotion.
The opposite holds consistently for the prevention tasks. Second, these variances indicate that the absolute inter-rater
agreements range from very poor to moderate agreement. Using r(wg) estimates, assuming uniform distribution, one
would square the standard deviations in Table 1, divide them by 2 (the expected variance) and subtract from 1. Thus,
the worst inter-rater agreement is observed for promotion ratings of ‘‘Handling customers’ complaints’’ with
SD ¼1.34 or r(wg)¼0.10 and the best inter-rater agreement is observed for promotion ratings of ‘‘Exerting creative
thought’’ with SD ¼0.69 or r(wg)¼0.76. Yet the crucial question for our purpose is whether the mean ratings across
all raters yield a reliable differentiation among the tasks (LeBreton & Senter, 2008). The mean ratings across all of
the judges are highly reliable (ICC(2) ¼0.94), which is not surprising given the large number of judges.
In summary, the purpose of the pre-test was to demonstrate that working people intuitively perceive the difference
between prevention-focused and promotion-focused tasks commonly used in organizations. The results show
that tasks requiring vigilance, accuracy, and adherence to rules (e.g., error detection, bookkeeping, safety
maintenance) are indeed perceived as ‘‘prevention tasks,’’ whereas those that require eagerness, creativity, and open-
mindedness (e.g., generating ideas, creative thinking) are perceived as ‘‘promotion tasks.’’ With this support for our
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
1090 D. VAN DIJK AND A.N. KLUGER
pre-classification of tasks, we conducted Study 1 to test whether the type of task moderates the relationship between
feedback sign and motivation (H1).
Study 1
Three samples received three sets of different scenarios in which task type and feedback sign were manipulated using
a22 design. In each sample, the type of task was manipulated by a different pair of tasks (one of prevention and
one of promotion) and with the same feedback-sign manipulation.
Method
Participants
The first set of scenarios was administered to 171 Business Administration graduate students (four classes) at Ben-
Gurion University of the Negev during class (Sample 1: 92 males, 78 females, 1 missing datum; average age: 33.3).
The second set of scenarios was administered to 247 undergraduate students from various departments at the
Hebrew University of Jerusalem (Sample 2: No measures for gender or age were taken). The third set of scenarios
was administered to 106 medical interns from eight hospitals in Israel during their shifts at the hospital (Sample 3:
60 males, 46 females; average age: 29.7).
Instruments
Task Type and Feedback Sign: For the first scenario, we chose a pair of tasks: error detection (prevention) and idea
generation (promotion). For the second scenario, we chose another pair of tasks: a safety project (prevention) and a
career-development project (promotion). These four specific tasks were chosen because they represent different
aspects of prevention and promotion. The error-detection/idea-generation tasks represent different strategies
(vigilance versus eagerness), whereas the safety/career development represents different needs (security versus
self-actualization).
For the third scenario, we used two tasks from the medical world. The study was distributed to medical interns
as part of another extensive study (Van Dijk, Shimer, & Bloch, 2007) aimed at exploring young Israeli
physicians’ motivation. The tasks were a drug control task aimed at detecting errors in prescribing drugs for
patients (prevention) and medical research aimed at improving the quality of patients’lives (promotion). The
two types of tasks were chosen based on a discussion with a group of physicians enrolled in a course taught by
one of the authors. The physicians, candidates in the Master of Health Administration (MHA) program, had
been discussing the relevance of prevention and promotion foci to physicians’ work as part of a class on work
motivation.
In all three scenario experiments, participants were asked to visualize themselves being assigned to work on either
a prevention or promotion task (e.g., error detection or idea generation) for an organization. Feedback sign was
manipulated by telling respondents that after one month they learned that ‘‘thus far’’ their project was either failing
or succeeding.
Motivation (intention to exert effort): In all three scenarios, motivation (after the feedback) was assessed with a
one-item question: ‘‘Relative to your effort on this project thus far, how much effort do you intend to exert next?’’
Respondents were provided with an 11-point scale ranging from ‘‘much less’’ (represented by 5) through ‘‘about
the same’’ (0) to ‘‘much more’’ (represented by 5).
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1091
Procedure
For Sample 1, one of the authors went to each class and asked students to fill out a short questionnaire, collecting
them 15–20 minutes later. For Sample 2, a research assistant asked students to answer a short questionnaire during
their break. The questionnaires were distributed in various departments in different locations on campus. For Sample
3, two research assistants distributed questionnaires to medical interns during their work at the hospital. They handed
the questionnaire to each intern privately, waiting for her/him to complete it (approximately half-an-hour) and
answering questions when necessary.
Study design
We employed a between-subject design (2 2) in which each subject received only one type of task (i.e., promotion
or prevention) and one type of feedback (i.e., positive or negative),with the dependent variable (motivation measured
by effort intention).
Results and discussion
The means, standard deviations, and ranges of the motivation measure (effort intention) were M¼1.83, SD ¼2.29,
range 4toþ5 (Sample 1); M¼2.09, SD ¼2.0, range 5toþ5 (Sample 2); M¼2.27, SD ¼1.80, range 3toþ5
(Sample 3).
To test whether task type moderated the effect offeedback sign on motivation, in each sample we calculated a two-
way ANOVA on the motivation measure, with feedback sign and task type as independent variables (see Table 2). No
main effects were found for feedback sign or task type in any of the samples. As predicted, the two-way interactions
between feedback sign and task type were significant in all three samples. The interaction effects are presented in
Figures 1–3. The values of the Y-axis shown are restricted to include only the range in which the means were found.
Excluding the full range of observed Yvalues helps visualize the direction of the interactions.
To reap the benefit of having three independent samples, we ran two meta-analyses (Borenstein et al., 2009),
assuming fixed effects of the simple effects of feedback sign: Once in the promotion condition and once in the
prevention condition. In the promotion-focus task (N¼260), the weighted mean effect of feedback sign on intention
to invest effort was d¼0.40 (Z¼3.21, p<0.001). A heterogeneity test indicated that the differences among the
three studies are likely to be solely due to sampling error (Q(2) ¼2.68, p>0.26). In the prevention focus-task
(N¼263), the weighted mean effect of feedback sign on intention to invest effort was d¼0.32 (Z¼2.59,
p<0.01). A heterogeneity test indicated that the differences among the three studies are likely to be due solely to
sampling error (Q(2) ¼4.39, p>0.11). That is, the meta-analyses suggest that not only feedback sign interacts with
task type in affecting motivation, but also each of the main effects (under promotion and prevention) is significant,
Table 2. Between subjects ANOVA for the effects of task type (promotion vs. prevention) and feedback sign (positive vs.
negative) on motivation (self-reported intention to invest effort) in three samples
Source
Sample 1: Error detection
vs. idea generation
Sample 2: Safety
vs. career-development
Sample 3: Drug control
vs. medical research
df F h
2
df F h
2
df F h
2
Task type (TSK) 1 0.73 0.004 1 0.82 0.003 1 3.73
y
0.036
Feedback sign (FB) 1 0.29 0.002 1 2.86 0.012 1 3.69
y
0.035
TSK FB 1 4.06
0.024 1 9.29
0.037 1 4.58
0.043
S within-group error 167 (5.19) 243 (4.08) 101 (2.94)
Note: The values enclosed in parentheses represent means of square errors. S, subjects.
p<0.01,
p<0.05,
y
p<0.1.
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
1092 D. VAN DIJK AND A.N. KLUGER
yet in the opposite direction, as predicted. Specifically, the meta-analyses indicate that when people imagine working
on promotion tasks, positive feedback increases their intention to invest effort more than negative feedback, but that
when people imagine working on prevention tasks, negative feedback increases their intention to invest effort more
than positive feedback.
Study 2
Study 2 was conducted to extend the test of the moderation hypothesis from self-report measures to actual task
performance.
Feedback sign
PositiveNegative
Intention to exert effort
3.00
2.00
1.00
0.00
Career-
development
Safety projec
t
Task type
Figure 2. Study 1: Intention to exert effort by feedback sign and task type (sample 2)
Feedback sign
PositiveNegative
Intention to exert effort
3.00
2.00
1.00
0.00
Ideas generation
Error detection
Task type
Figure 1. Study 1: Intention to exert effort by feedback sign and task type (sample 1)
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1093
Method
Participants
The participants were 112 undergraduate students from the Hebrew University of Jerusalem (N¼64) and
Ben-Gurion University of the Negev (N¼48).
Instruments
Task type: Consistent with the results of the pre-test, we employed two tasks that represent prevention focus and
promotion focus.
(1) Error detection in arithmetic calculations (prevention). In this task participants were requested to detect errors in
a list of relatively simple, solved arithmetic calculations that required accuracy and attention to detail. An
example item was ‘‘(þ0.6)–(0.8)–(þ0.7) ¼0.9.’’ Approximately 40 per cent of the calculations contained an
error. Participants were requested to mark the calculations in which they detected an error.
(2) Generating different uses for an object (promotion). In this task, participants were asked to generate as many
uses as possible for a particular object. For example, various uses for a building block could include a doorstop, a
stand for a planter, a base for shelves, a crude weapon, or an athletic weight. This task required creativity and
open-mindedness.
The two tasks were programmed and the entire experiment was conducted in a computer lab.
Feedback sign: After working on the task for 10minutes, participants were asked to pause and fill out a
questionnaire (manipulation check— see below). Meanwhile, the computer informed them that the program was
now processing their task performance. After completing the questionnaire, the participants received a false
normative feedback. The feedback appeared on the screen as a short message: ‘‘Up to now your performance on this
task has been above average‘or’ Up to now your performance on this task has been below average.’’
Manipulation check: Two questions were used to check whether the tasks yielded the situational regulatory focus
that they were meant to produce. The questions measured the extent to which the task (1) required creativity and (2)
required accuracy. The items were rated on five-point scales ranging from 1(to a lesser extent) to 5 (to a greater
extent).
Effort intention: Effort intention (‘‘How much effort do you intend to exert on the following task?’’) was
measured, as in Study 1, on an 11-point scale ranging from 0 (‘‘very little’’) to 10 (‘‘very much’’). It was measured
Feedback sign
PositiveNegative
Intention to exert effort
3.00
2.00
1.00
0.00
Medical research
Drug control
Task type
Figure 3. Study 1: Intention to exert effort by feedback sign and task type (sample 3)
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
1094 D. VAN DIJK AND A.N. KLUGER
twice: Prior to administering the first part of the task, and after receiving feedback but prior to the second part of the
task.
Performance: Performance on the error detection task was measured by the number of errors detected by the
participants. Performance on the generating uses task was measured by the number of uses suggested by the
participants. The scores of both tasks were converted into Z-scores separately for pre-feedback and post-feedback
performance. This standardization controlled for both differences in performance metric between the tasks and
changes in task performance that were not related to the effects of the manipulated variables.
Procedure
Two research assistants in two locations, The Hebrew University of Jerusalem and Ben-Gurion University of the
Negev, conducted the experiment. In each of the universities, the research assistant advertised a 40-minute
experiment carrying a compensation voucher for coffee and a pastry, redeemable at the university cafeterias.
Participants were randomly assigned to work on one of the two tasks and then worked independently, guided by the
computer. The experiment included a general explanation regarding the study’s procedure without mentioning that
feedback would be given, explicit instructions for the specific task, and the effort measure. Participants then began
the first part of the task. After 10 minutes, participants were asked via instructions on the computer to stop working
on the task and to complete a questionnaire (manipulation check, biographical data) while the computer ‘‘checked’’
the performance level. Then the computer randomly delivered bogus positive or negative feedback on the screen, and
the participants were asked to fill in the effort measure again. Finally, they were asked to carry out the second part of
the task, which included a new set of arithmetic calculations for the prevention task, and a different object for the
generating uses (promotion) task. Upon completion of the task, all participants were debriefed. As part of the
debriefing, participants were asked whether they believed the feedback they had been given. The vast majority of
them said that they did and were very much affected by it. Even when they believed that the feedback differed greatly
from their own perception of their performance, they still believed the feedback and were surprised or disappointed,
depending on the feedback message received.
Study design and analysis
We employed a mixed design (22(2)) to test our hypothesis regarding the effects of task type and feedback sign
on two outcome variables: Performance and motivation. Specifically, type of task (i.e., promotion or prevention) and
the feedback sign (i.e., positive or negative) were between-subject variables, and time was a within-subject variable
(before and after feedback). However, to avoid capitalizing on chance with one outcome variable or the other, we first
tested a 2 2(2) (2) MANOVA where the type of outcome measure (motivation or performance) was one more
within-subject factor. If we found support for our hypothesis with this omnibus test, we could safely inspect whether
the hypothesis was supported separately for each outcome variable.
Results and discussion
The two manipulation check items yielded significant results in the predicted direction. Specifically, the promotion
task was perceived as requiring more creativity (t(111) ¼12.01; p<0.001; d¼2.28), whereas the prevention
task was perceived as a task requiring more accuracy (t(111) ¼10.40; p<0.001; d¼1.97).
Table 3 presents the means, standard deviations, and inter-correlations for Study 2 variables. The performance
measures presented in the table are unstandardized, showing the actual performance levels and standard deviations.
To test the hypothesis regardingthe moderating effect of task type on the effect of feedback sign on motivation and
performance, we ran a repeated-measure MANOVA with the type of outcome measure (motivation and performance)
as one within-subject factor, and time (before and after feedback) as another within-subject factor. Task type and
feedback sign were inserted as between-subject variables. Two participants from the error detection task were
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1095
omitted from the analysis because their performance score on the first part of the task (before feedback) was more
than three standard deviations below the average on this task.
The results of the MANOVA (see Table 4) showed that the interaction between task type, feedback sign, and time
was significant (F(1,106) ¼7.73; p<0.01; h
2
¼0.07). The direction of the interaction supports our moderation
hypothesis. Specifically, in the error detection task, motivation and performance increase following negative
feedback and decrease following positive feedback, while in the uses generation task, the pattern of the results
reverses such that motivation and performance increase following positive feedback and decrease following negative
feedback. All other main or interaction effects were non-significant, except for a significant main effect for the
Table 3. Means, standard deviations, and correlations of Study 2 variables
NMSD 1234567
1. Task creativity 110 3.0 1.4 —
2. Task accuracy 110 3.1 1.4 0.09 —
3. Effort intention (pre FB) 110 9.0 1.8 0.31
0.22
—
4. Effort intention (post FB) 110 9.0 1.7 0.30
0.09 0.59
—
5. Performance error detection (pre FB) 55 14.3 6.5 0.32
0.14 0.37
0.29
—
6. Performance error detection (post FB) 55 17.5 4.1 0.05 0.19 0.06 0.05 0.11 —
7. Performance uses generation (pre FB) 55 6.2 2.5 0.05 0.23
y
0.23
y
0.18 —
8. Performance uses generation (post FB) 55 6.5 2.2 0.18 0.17 0.08 0.27
0.83
p<0.01,
p<0.05,
y
p<0.1
Table 4. MANOVA for the effects of time (before vs. after feedback), task type (error detection vs. creativity), and feedback sign
(positive vs. negative) on actual performance and intention to invest effort (measure type)
Source df F h
2
Between Subjects
Task type (TSK) 1 0.72 0.01
Feedback sign (FB) 1 0.20 0.00
TSK FB 1 0.91 0.01
S within-group error 106 (3.31)
Within Subjects
Time (T) 1 0.051 0.00
TTSK 1 1.75 0.02
TFB 1 1.67 0.02
TTSK FB 1 7.73
0.07
TS within-group error 106 (0.76)
Measure Type (MT) 1 2772.995
0.96
MT TSK 1 0.79 0.01
MT FB 1 0.00 0.00
MT TSK FB 1 0.03 0.00
MT S within-group error 106 (0.09)
TMT 1 0.05 0.00
TMT TSK 1 0.72 0.01
TMT FB 1 2.72 0.03
TMT TSK FB 1 0.98 0.01
TMT S within-group error 106 (0.75)
Note: The values enclosed in parentheses represent means of square errors. S, subjects.
p<0.05.
p<0.01; Sources of significant effects are in boldface.
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
1096 D. VAN DIJK AND A.N. KLUGER
measure type (F(1,106) ¼2773; p<0.01; h
2
¼0.96), reflecting the difference in scales between the two outcome
measures of motivation (unstandardized) and performance (standardized).
After finding support for our moderation hypothesis with an omnibus test, we retested the moderation hypothesis
separately for each outcome variable. To test the hypothesis regarding the moderating effect of task type on the effect
of feedback sign on performance outcome, we ran a mixed ANOVAwith standardized performance before and after
receiving feedback as within-subject factor, and feedback sign and task type as between-subject factors (see Table 5).
Tests of within-subject effects on performance did not reveal a main effect for time (before and after the feedback),
which was expected due to the standardization used. Also, no interaction effects were found for time by feedback
sign and time by task type. As expected, the three-way interaction (time feedback sign task type) was significant.
The pattern of this interaction, as hypothesized, indicates that performance increased after negative feedback in the
error-detection task (prevention) and decreased after positive feedback, whereas in the uses-generation task
(promotion) the pattern was the opposite: Performance increased after positive feedback and decreased after
negative feedback. This three-way interaction is depicted in Figure 4.
Another mixed ANOVA was calculated on the motivation measure (i.e., intention to invest effort). Time (before
and after the feedback) served as a within-subject factor, while feedback sign and task type served as between-
subject factors (see Table 5). Tests of within-subjects effects on the effort intention did not reveal a main effect for
time or interaction effects for time by feedback sign and time by task type. As expected, the three-way interaction
(time feedback sign task type) was significant. This three-way interaction is depicted in Figure 5. The values of
the Y-axis shown are restricted as to include only the range in which the means were found. Excluding the full range
of observed Yvalues helps visualize the direction of the interactions.
Lastly, we sought to compare the result of Study 2 to the results of Study 1. Therefore, we first convertedthe results
of Study 2 to the same format used in Study 1. Study 1 was based on between-subject design whereas Study 2 had a
mixed design with time (before and after feedback) as a within-subject factor. Therefore, we calculated for each
outcome the residual post-feedback score controlling for the pre-feedback score. With this conversion, we could
inspect the data with the same 2 2 design. Specifically, we calculated the effect of feedback sign on each outcome
(residual performance, residual motivation) separately for promotion and prevention focus tasks. In the promotion
focus task, the effect of feedback sign on residual performance was d¼0.67 ( p<0.01; one tail) and on the residual
motivation d¼0.59 ( p<0.01, one tail). In contrast in the prevention focus task, the effect of feedback sign on
Table 5. Repeated measure ANOVA for the effects of time (before vs. after feedback), task type (error detection vs. creativity)
and feedback sign (positive vs. negative) on actual performance and motivation (self-reported intention to invest effort)
Performance Motivation
Source df F h
2
Fh
2
Between Subjects
Task type (TSK) 1 0.00 0.000 0.96 0.009
Feedback sign (FB) 1 0.24 0.002 0.06 0.001
TSK FB 1 1.20 0.011 0.26 0.002
S within-group error 106 (1.4) (5.05)
Within Subjects
Time (T) 1 0.00 0.000 0.06 0.001
TTSK 1 0.28 0.003 1.50 0.014
TFB 1 0.14 0.001 2.75 0.025
TTSK FB 1 3.85
0.035 4.55
0.041
TS within-group error 106 (0.32) (1.20)
Note: The values enclosed in parentheses represent means of square errors. S, subjects.
p<0.05.
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1097
residual performance was d¼0.37 ( p<0.10 one tail) and on residual motivation d¼0.11 ( p>0.05, ns).
Although, the last two effects are not statistically significant, they are both in the predicted direction.
The effects of Study 2 are similar to the effects in Study 1. To increase precision in our final estimates, we added
the results of the motivation measure of Study 2 (which was identical to that used in Study 1) to the meta-analyses
results reported for Study 1. This addition did not change the conclusion. Across all four samples in the promotion-
focus task (N¼315), the effect of feedback sign on intention to invest effort was d¼0.43 (Z¼3.80, p<0.001). A
heterogeneity test indicated that the differences among the studies are likely to be due solely to sampling error
(Q(3) ¼3.04, p>0.38). In the prevention focus task (N¼318), the sample-weighted mean effect of feedback sign on
intention to invest effort was d¼0.33 (Z¼2.91, p<0.005). A heterogeneity test indicated that the differences
among the studies are likely to be due solely to sampling error (Q(3) ¼4.41, p>0.22). That is, adding the results of
Study 2 increased slightly the Zscores and our confidence in the results, and decreased the significance of the
heterogeneity tests suggesting that minor differences in patterns results are due solely to sampling error. This pattern
of results indicates that feedback sign has modest effects on motivation, yet these effects are in opposite directions in
the context of prevention versus that of promotion task, such that the overall difference in effects is medium-large
(0.43 (0.33) ¼0.76).
Time
After feedbackBefore feedback
Performance (z-score)
0.30
0.20
0.10
0.00
-0.10
Positive
Negative
Feedback sign
Task type: Error detection
Time
After feedbackBefore feedback
performance (z-score)
0.30
0.20
0.10
0.00
-0.10
Positive
Negative
Feedback sign
Task type: Uses generation
Figure 4. Study 2: Performance before/after feedback-by-feedback sign and task type
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
1098 D. VAN DIJK AND A.N. KLUGER
In summary, the results of Study 2 support our moderation hypothesis by demonstrating that individuals improve
their task performance and increase their motivation after receiving positive feedback in a promotion-focused task,
and after negative feedback in a prevention-focused task.
General Discussion
The results of the pre-test and the two studies yielded two key findings: (1) Task type activates regulatory focus, and
(2) task type moderates feedback-sign effects on motivation and performance. Previous research has demonstrated
that regulatory focus differentially affects the nature of the performance or the strategy used in performing a task
(e.g., fast performance or accurate performance, Forster et al., 2003; creative performance or repetitiveness; Crowe
& Higgins, 1997). We show the opposite — that task performance or the strategy required by the task (e.g., error
detection versus creativity) can affect regulatory focus. Furthermore, the regulatory focus elicited by the task
interacts with other variables (in our case feedback sign) in affecting the level of motivation and performance.
Time
After feedbackBefore feedback
Effort intention
9.40
9.20
9.00
8.80
8.60
8.40
Positive
Negative
Feedback sign
Task type: Error detection
Time
After feedbackBefore feedback
Effort intention
9.40
9.20
9.00
8.80
8.60
8.40
Positive
Negative
Feedback sign
Task type: Uses generation
Figure 5. Study 2: Intention to exert effort before/after feedback-by-feedback sign and task type
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1099
Specifically, we showed that non-experts in regulatory focus theory easily differentiated between predominately
prevention tasks and predominantly promotion tasks in their ratings of 20 out of 21 tasks presumed to evoke
regulatory focus by SME. Moreover, when performing prevention tasks, participants were more motivated and
performed better after receiving negative feedback (an indication of fit between regulatory focus and feedback sign)
than after receiving positive feedback (an indication of lack of fit between regulatory focus and feedback sign).
Similarly, when performing promotion tasks, participants were more motivated and performed better after receiving
positive feedback than they did after receiving negative feedback.
All of the tasks used in our studies were judged by actual managers to be routine ones that are part of their jobs.
Furthermore, the performance measures used in Study 2 were conceptually similar to those that employees do at
work. Specifically, the error detection task is similar to tasks usually carried out by bank officers, accountants, or
even teachers reviewing student exams. The uses-generation task is similar to tasks in which professionals are asked
to think of ideas to problem solve or to brainstorm and present alternatives for attaining organizational goals.
Therefore, our findings are likely to be generalized to many ‘‘real life’’ tasks in organizations, but of course, further
empirical replications are desirable.
Additionally, our findings contribute to two different streams of research: Feedback effects and regulatory focus
theory. One contribution to feedback research is a possible explanation for the variability of feedback-sign effects on
performance (reported in a meta-analysis by Kluger & DeNisi, 1996) and points to an interesting phenomenon
related to negative feedback. Our findings suggest that under prevention focus, negative rather than positive feedback
best enhances performance. However, goal-orientation research has documented the debilitating effect of negative
feedback on self-efficacy and, by extension, on performance (e.g., Butler, 2000; Dweck, 1986; Dweck & Leggett,
1988; Grant & Dweck, 2003; Nicholls, 1984). These conflicting effects of negative feedback under prevention focus
may be explained by two motivational processes. Negative feedback under prevention focus may simultaneously
decrease the expectancy of future success but increase the value of future success (Levontin & Kluger, 2004; Van
Dijk & Kluger, 2001). This suggestion needs further examination.
Another contribution to feedback research is the finding that the regulatory focus of the task could explain the
variability in feedback-sign effect. Feedback intervention theory suggests that feedback effects are moderated by the
nature of the task (Kluger & DeNisi, 1996). However, ‘‘the exact task properties are still poorly understood’’ (Kluger
& DeNisi, 1996: p. 275). Most previous research has examined the ‘‘complexity’’ or novelty of the task as a
moderator for feedback or goal effects on performance (Atkins, Wood, & Rutgers, 2002; Kanfer & Ackerman, 1989;
Locke & Latham, 2002; Wood, 1986; Wood, Mento, & Locke, 1987). Specifically, it was found that when a task is
complex, feedback (or a specific goal) does not necessarily increase performance. Our study suggests another
dimension of the task that could moderate the feedback-sign effect and may help in understanding the effect of
feedback on performance.
Our contribution to regulatory focus theory is an extension of the concept of ‘‘situational regulatory focus’’ to
different task types. The antecedents of regulatory focus identified by Higgins (1997, 1998) are needs (security
versus self-actualization), goals (oughts versus ideals), and the framing of the situation (loss/non-loss versus
gain/non-gain). Our findings indicate that the type of task could serve as another possible antecedent to
regulatory focus. In fact, this new antecedent is closely linked to the other antecedents: Prevention tasks relate
to security or safety needs, to something that one ought to do and to a potential loss, while promotion tasks relate
to the need for self-actualization, to something that one ideally wants to do and to a potential gain. Yet our results
indicate that the type of task is a variable that stands on its own as an antecedent to regulatory focus, which further
affects motivation and performance. Thus, if one is working on a creative task, we can assume that one is
functioning under a situational promotion focus, whereas if one is working on a task that requires vigilance and
attention to error, one is functioning under a situational preventionfocus.Inotherwords,thetypeoftaskcreatesa
situational regulatory focus.
Another contribution to self-regulatory research is the use of our findings to broaden the implications of previous
regulatory focus results. If the type of task can be a situational source of regulatory focus, then many of the self-
regulatory effects documented by Higgins and his colleagues can be generalized to task types. For example,
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
1100 D. VAN DIJK AND A.N. KLUGER
expectancy theories posit that motivation increases in part as a function of the product of value and expectancy.
However, self-regulatory theory suggests that this formulation is valid only under promotion focus and that under
prevention focus it is sufficient to have either high valence or high expectancy to induce high levels of motivation
(Shah & Higgins, 1997). Therefore, one might expect, for example, that increasing both valence and expectancy
would be more important for organizations trying to foster risk-taking and creativity than to those attempting to
foster conformity to organizational rules.
On the practical level, our findings suggest that no feedback system can fit all situations. That is, organizations
should maintain congruency between feedback system and task type. Specifically, organizations (or certain units
within organizations) that largely require high vigilance — such as certain military jobs (radar operator), security
companies, and banks—may benefit most from training that emphasizes avoiding poor performance and by a
performance appraisal system that emphasizes error-free performance and from blocking the promotion of
employees with histories of frequent errors and disciplinary records. In contrast, other organizations that require a
state of eagerness, including advertising agencies and research and development units, may benefit most from
awards, prizes, and recognition of excellence. However, in many cases, employees need to switch from one task to
another during the same operation. The study raises practical issues regarding how to help organizations make their
feedback systems more flexible so that they can fit changing tasks. For each specific task, this study could provide
guidelines for shaping training, feedback systems, and reward systems. For example, we recommend punishment for
failure to prevent errors but no praise for detecting errors, and praise for succeeding in creative tasks but no
punishment for doing poorly on these tasks.
The principle of fit between task and positive versus negative feedback may be extended to other variables that
carry valence information. Information carrying valence is common in other interventions in organizations. For
example, positive valence may be carried by bonuses. Indeed, in an unpublished study, Matalon (2000) found in
consistency with the fit hypothesis that bonuses paid for attendance (prevention behavior) increased absenteeism.
Specifically, bonuses paid in a certain month correlated with higher absenteeism the following month in four out of
11 time lags measured while none was significant in the other direction. The logic is that as bonuses that provide a
positive feedback signals to employees that the prevention behavior task of good attendance is achieved (lack of fit
between feedback and the regulatory focus), they opt to reduce effort (keeping good attendance records), consistent
with the findings of the present studies do not fit the prevention behavior of attendance.
Using the same logic, we can predict that various interventions offered by positive psychology (e.g., Seligman,
Steen, Park, & Peterson, 2005) and positive organizational scholarship (e.g., Roberts, Dutton, Spreitzer, Heaphy, &
Quinn, 2005) may contribute to the performance of promotion tasks but may be irrelevant or even debilitating to the
performance of prevention tasks (cf. Fineman, 2006).
The application of our recommendation is tricky because both promotion-focused and prevention-focused tasks
seem to play a significant role in organizational life as, ‘‘Today’s employees are required to be creative, yet also
conform to rules and standards and work efficiently to meet time and budget constraints’’ (Miron et al., 2004: p. 175).
Therefore, it appears that in order to reap benefits from our insights, organizations may need to devise different
feedback, reward, and training systems applied to different tasks in different contexts and timing.
Limitations and future research
Several limitations relating to the current research should be noted. First, we did not directly test the mechanism
through which the nature of the task activates or primes the regulatory focus. Instead, we reported correlations
between different types of tasks and the perceived regulatory focus (pre-test stage). Yet, we replicated the interaction
between regulatory focus and feedback sign (Idson & Higgins, 2000; Van-Dijk & Kluger, 2004) using type of task as
a source of regulatory focus. The regularity in which this interaction emerged (in both Study 1 and Study 2) supports,
albeit indirectly, the notion that regulatory focus is activated by the type of the task. Future research may attempt to
prime regulatory focus with various tasks and directly measure perceptions of regulatory focus combined with
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1101
outcome measures to demonstrate that the interactive effect of task type and feedback sign are mediated by
perception of regulatory focus.
Second, in both Study 1 and Study 2, we measured motivation with a one-item question. Moreover, whereas the
conceptual replication with the actual performance measure reduces the risk that these results are merely perceptual,
future research may wish to assess motivational effects with either multi-item scales or with other behavioral
manifestations of motivation (e.g., time spent on a task).
A question still left unanswered is: What happens when the task is prevention-focused and the person is
promotion-focused and more generally, what happens when the situational regulatory focus contradicts the chronic-
regulatory focus? Future studies should further examine the three-way interaction between feedback sign, situational
regulatory focus, and chronic-regulatory focus. Two alternative suggestions could be tested. One is based on the fit
principle suggested by Higgins and his colleagues (e.g., Higgins, 2000; Sassenberg et al., 2007; Shah et al., 1998;
Spiegel et al., 2004), according to which more congruency between the interacting components (chronic regulatory
focus, situational regulatory focus, and feedback sign) leads to the highest motivation. Specifically, motivation is
highest when the employee is chronically promotion-focused, the task requires creativity, and the feedback is
positive or when the employee is chronically prevention-focused, the task requires attention to detail, and the
feedback is negative. Other combinations of the components will produce less motivation. An alternative suggestion
for the three-way interaction is based on the findings of Freitas, Liberman, Salovey, and Higgins (2002) regarding the
priority of the prevention focus over the promotion focus. They found that prevention focus (both chronic and
manipulated) fosters a sense of urgency and a preference to initiate action earlier than promotion focus. Conversely,
in a promotion situation, people feel little pressure to immediately initiate any single action. Thus, prevention focus
may produce a stronger effect than promotion focus (‘‘safety first principle’’). Therefore, a prevention effect will be
obtained even when only one of the components (the situation or the employee) is prevention focused. However, to
obtain a promotion effect, all of the components need to be congruent.
In summary, this research helps predict which feedback sign should be most effective for tasks that prime
prevention focus, as opposed to those that prime promotion focus. This paper clearly suggests that in designing an
effective feedback intervention, the combination of feedback sign and task types must be taken into account.
Moreover, different task types may not interact solely with feedback sign in affecting motivation but with many other
forms of rewards and punishments used in organizations. For example, prizes, bonuses, and recognitions may be
effective largely for motivating and increasing performance of promotion tasks, whereas threats of dismissals,
punishments, demerits, and reprimands may be effective largely for increasing performance and motivation of
prevention tasks. This extension of the concept of fit (Higgins, 2000) may open a new domain of inquiry for
organizational researchers and provide fresh ideas for fine-tuning various organizational practices for managers and
consultants.
Acknowledgements
The research for this paper was supported by a grant from the Recanati Fund at the Hebrew University of
Jerusalem School of Business Administration and an ARI contract no. DASW01-04-K-0001 for the second
author. The views, opinions, and/or findings contained in this paper are those of the authors and should
not be construed as official Department of Defense position, policy or decision. An earlier version of this paper
was presented at the 2009 Academy of Management meeting in Chicago and won the Best Paper Award that
‘‘recognizes the empirical and/or conceptual paper submitted to the Academy of Management meeting that offers the
most significant contribution to the field of OB [Organizational Behavior].’’ The authors wish to thank three
anonymous reviewers for their significant contribution to the final paper, and to Merav Guri-Uziel for running
Study 2.
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
1102 D. VAN DIJK AND A.N. KLUGER
Author biographies
Dina Van Dijk is an OB researcher in the department of Health Systems Management at Ben-Gurion University of
the Negev. She received her Ph.D. from the School of Business Administration of the Hebrew University.
Her primary interest is the implications of regulatory focus theory for feedback, leadership, and health.
Avraham N. (Avi) Kluger studied the detrimental effects of feedback on performance. As an alternative to feedback
interventions, he recently developed with Dina Nir the feedforward interview, which he applied in strength-based
employee performance appraisals and in selection interviews. Currently, he studies the beneficial effects of active
empathetic listening.
References
Atkins, P. W. B., Wood, R. E., & Rutgers, P. J. (2002). The effects of feedback format on dynamic decision making.
Organizational Behavior and Human Decision Processes,88, 587–604.
Bass, M., De Dreu, C. K. W., & Nijstad, B. A. (2008). A meta-analysis of 25 years of mood-creativity research: Hedonic tone,
activation, or regulatory focus? Psychological Bulletin,134, 779–806.
Bianco, A. T., Higgins, E. T., & Klem, A. (2003). How ‘‘Fun/Importance’’ fit affects performance: Relating implicit theories to
instructions. Personality and Social Psychology Bulletin,29, 1091–1103.
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009). Introduction to meta-analysis. Chichester, West Sussex,
U.K; Hoboken: John Wiley & Sons.
Brockner, J., & Higgins, E. T. (2001). Regulatory focus theory: Implications for the study of emotions at work. Organizational
Behavior and Human Decision Processes,86, 35–66.
Brockner, J., Paruchuri, S., Idson, L. C., & Higgins, E. T. (2002). Regulatory focus and the probability estimates of conjunctive
and disjunctive events. Organizational Behavior and Human Decision Processes,87, 5–24.
Butler, R. (2000). Making judgments about ability: The role of implicit theories of ability in moderating inferences from temporal
and social comparison information. Journal of Personality and Social Psychology,78, 965–978.
Crowe, E., & Higgins, E.T., (1997). Regulatory focus and strategic inclinations: Promotion and prevention in decision-making.
Organizational Behavior and Human Decision Processes,69, 117–132.
DeNisi, A. S., & Kluger, A. N. (2000). Feedback effectiveness: Can 360-degree appraisals be improved? Academy of Management
Executive,14, 129–139.
Dweck, C. S. (1986). Motivational processes affecting learning. American Psychologist,41, 1040–1048.
Dweck, C. S., & Leggett, E. L. (1988). A social cognitive approach to motivation and personality. Psychological Review,95, 256–
273.
Fineman, S. (2006). On being positive: Concerns and counterpoints. Academy of Management Review,31, 270–291.
Forster, J., Higgins, E. T., & Taylor-Bianco, A. (2003). Speed/accuracy decision in task performance: Built-in trade-off or separate
strategic concern? Organizational Behavior and Human Decision Processes,90, 148–164.
Friedman, R. S., & Forster, J. (2001). The effects of promotion and prevention cues on creativity. Journal of Personality and
Social Psychology,81, 1001–1013.
Freitas, A. L., Liberman, N., & Higgins, E. T. (2002). Regulatory fit and resisting temptation during goal pursuit. Journal of
Experimental Social Psychology,38, 291–298.
Freitas, A. L., Liberman, N., Salovey, P., & Higgins, E. T. (2002). When to begin? Regulatory focus and initiating goal pursuit.
Personality and Social Psychology Bulletin,28, 121–130.
Grant, H., & Dweck, C. S. (2003). Clarifying achievement goals and their impact. Journal of Personality and Social Psychology,
85, 541–553.
Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist,52, 1280–1300.
Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a motivational principle. In M.P. Zanna (Ed.), Advances in
Experimental Social Psychology (Vol. 30, 1–46). New York: Academic Press.
Higgins, E. T. (2000). Making a good decision: Value from fit. American Psychologist,55, 1217–1230.
Higgins, E. T., Friedman, R. S., Harlow, R. E., Idson, L. C., Ayduk, O. N., & Taylor, A. (2001). Achievement orientations from
subjective histories of success: Promotion pride versus prevention pride. European Journal of Social Psychology,31, 3–23.
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1103
Higgins, E. T., Shah, J., & Friedman, R. (1997). Emotional responses to goal attainment: Strength of regulatory focus as
moderator. Journal of Personality and Social Psychology,72, 515–525.
Idson, L. C., & Higgins, E. T. (2000). How current feedback and chronic effectiveness influence motivation: Everything to gain
versus everything to lose. European Journal of Social Psychology,30, 583–592.
Ilies, R., & Judge, T. A. (2005). Goal regulation across time: The effects of feedback and affect. Journal of Applied Psychology,
90, 453–467.
Kanfer, R., & Ackerman, P. L. (1989). Motivation and cognitive abilities: An integrative aptitude treatment interaction approach
to skill acquisition. Journal of Applied Psychology,74, 657–690.
Kark, R., & Van Dijk, D. (2007). Motivation to lead motivation to follow: The role of the self-regulatory focus in the leadership
processes. Academy of Management Review,32, 500–528.
Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on performance: A historical review, a meta-analysis,
and a preliminary feedback intervention theory. Psychological Bulletin,119, 254–284.
Kluger, A. N., & DeNisi, A. (1998). Feedback interventions: Toward the understanding of a double- edged sword. Current
Directions in Psychological Science,7, 67–72.
Kluger, A., & Ganzach, Y. (2004). Two faces of excellence: Perfection versus eminence. In G.B. Graen (Ed.). New frontiers of
leadership, LMX Leadership: The series (Vol. 2). Greenwich, CT: Information Age Publishing Inc.
LeBreton, J. M., & Senter, J. L. (2008). Answers to 20 questions about interrater reliability and interrater agreement.
Organizational Research Methods,11, 815–852.
Levontin, L., & Kluger, A. N. (2004). A comparison between the predictions of goal orientation theory and self regulation theory
regarding the effect of feedback sign on motivation. 19th Annual Convention of the Society for Industrial and Organizational
Psychology, Chicago, IL.
Liberman, N., Idson, L. C., Camacho, C. J., & Higgins, T. E. (1999). Promotion and prevention choices between stability and
change. Journal of Personality and Social Psychology,77, 1135–1145.
Liberman, N., Molden, D. C., Idson, L. C., & Higgins, E. T. (2001). Promotion and prevention focus on alternative hypotheses:
Implications for attributional functions. Journal of Personality and Social Psychology,80, 5–18.
Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation: A 35-year odyssey.
American Psychologist,57, 705–717.
Lockwood, P., Jordan, C. H., & Kunda, Z. (2002). Motivation by positive or negative role models: Regulatory focus determines
who will best inspire us. Journal of Personality and Social Psychology,83, 854–864.
Matalon, I. (2000). The reward paradox: The effects of performance based feedback on the increase of absenteeism from work.
Unpublished MA Thesis, The Hebrew University, Jerusalem.
Miron, E., Erez, M., & Naveh, E. (2004). Do personal characteristics and cultural values that promote innovation, quality and
efficiency compete or complement each other? Journal of Organizational Behavior,25, 175–199.
Neubert, M. J., Kacmar, K. M., Carlson, D. S., Chonko, L. B., & Roberts, J. A. (2008). Regulatory focus as a mediator of the
influence of initiating structure and servant leadership on employee behavior. Journal of Applied Psychology,93, 1220–1233.
Nicholls, J. G. (1984). Achievement-motivation - conceptions of ability, subjective experience, task choice, and performance.
Psychological Review,91, 328–346.
Roberts, L. M., Dutton, J. E., Spreitzer, C. M., Heaphy, E. D., & Quinn, R. E. (2005). Composing the reflected best-self portrait:
Building pathways for becoming extraordinary in work organizations. Academy of Management Review,30, 712–736.
Sassenberg, K., Jonas, K. J., Shah, J. Y., & Brazy, P. C. (2007). Why some groups just feel better: The regulatory fit of group power.
Journal of Personality and Social Psychology. 92, 249–267.
Seligman, M. E. P., Steen, T. A., Park, N., & Peterson, C. (2005). Positive psychology progress - Empirical validation of
interventions. American Psychologist,60, 410–421.
Shah, J., & Higgins, E. T. (1997). Expectancy x value effects: Regulatory focus as determinant of magnitude and direction.
Journal of Personality and Social Psychology,73, 447–458.
Shah, J., Higgins, E. T., & Friedman, R. (1998). Performance incentives and means: How regulatory focus influences goal
attainment. Journal of Personality and Social Psychology,74, 285–293.
Spiegel, S., Grant-Pillow, H., & Higgins, E. T. (2004). How regulatory fit enhances motivational strength during goal pursuit.
European Journal of Social Psychology,34, 39–54.
Van Dijk, D., & Kluger, A. N. (2001). Goal-orientation versus self-regulation: Different labels or different constructs? A paper
presented at the 16th Annual convention of the Society for Industrial and Organizational Psychology, San-Diego.
Van-Dijk, D., & Kluger, A. N. (2004). Feedback sign effect on motivation: Is it moderated by regulatory focus? Applied
Psychology: An International Review,53, 113–135.
Van Dijk, D., Shimer, S., & Bloch, Y. (2007). The connection between values and medical specializations among medical interns.
A paper presented at the Sixth Annual Health Policy Conference, Hilton Hotel, Tel-Aviv, December, 2007. (Hebrew).
Vancouver, J. B., & Tischner, E. C. (2004). The effect of feedback sign on task performance depends on self-concept
discrepancies. Journal of Applied Psychology,89, 1092–1098.
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
1104 D. VAN DIJK AND A.N. KLUGER
Wade, T. C. (1974). Relative effects on performance and motivation of self-monitoring correct and incorrect responses. Journal of
Experimental Psychology-General,103, 245–248.
Wallace, J. C., & Chen, G. (2006). A multilevel integration of personality, climate, regulatory focus, and performance. Personnel
Psychology,59, 529–557.
Wood, R. E. (1986). Task complexity: Definition of the construct. Organizational Behavior and Human Decision Processes,37,
60–82.
Wood, R. E., Mento, A., & Locke, E. (1987). Task complexity as a moderator of goal effects: A meta-analysis. Journal of Applied
Psychology,72, 416–425.
Copyright #2010 John Wiley & Sons, Ltd. J. Organiz. Behav. 32, 1084–1105 (2011)
DOI: 10.1002/job
TASK TYPE AND FEEDBACK SIGN 1105
