Journal of Business Research 66 (2013) 933–940
Contents lists available at SciVerse ScienceDirect
Journal of Business Research
What happens when you're lost between happiness and sadness?
Effects on consumers' cognitive responses
Baris Ursavas a,⁎, Ozlem Hesapci-Sanaktekin b,⁎⁎
a
b
Istanbul Bilgi University, Department of Advertising, Turkey
Bogazici University, Department of Management, Santral Istanbul 34060 Eyup Istanbul, Turkey
a r t i c l e
i n f o
Article history:
Received 1 June 2011
Received in revised form 1 August 2011
Accepted 1 October 2011
Available online 28 December 2011
Keywords:
Mixed emotions
Emotional ambivalence
Memory
Temporal perception
a b s t r a c t
Two experiments examine the effects of dramatic contrast between the music and the message of an ad on
consumers' temporal perceptions and memory. Results suggest that individuals' level of discomfort with
ambiguity (DWA) plays a significant role in memory and temporal perceptions when being exposed to auditory stimuli that incorporate two oppositely-valenced affective components. Music that creates dramatic
contrast with the ad message leads to weaker recall and recognition for the ad messages, and lower ad
duration estimates for subjects with high discomfort with ambiguity. Further, results reveal an effect of
prior mood on cognitive responses toward stimuli that create dramatic contrast. Participants in a positive
(vs. negative) mood report better recall and recognition, and lower but more accurate duration estimates
in the case of a stimulus creating dramatic contrast.
© 2012 Elsevier Inc. All rights reserved.
1. Introduction
Music is an integral feature of human culture. Whether listening to
a symphony from a majestic orchestra or humming a simple melody,
music penetrates deep into the soul and enables connections with a
higher, emotional realm. Music may be one of the most ancient
forms of human communication (Geissmann, 2000). As a communication tool, music creates a universal language, uniting complete
strangers around common emotions and ideas.
The ability of music to influence human emotions and behavior has
attracted the attention of advertising practitioners since the early 20th
century. Music took great leaps into advertising, first with radio in
the 1930s and then with television in the 1950s (Kellaris, Cox, & Cox,
1993). Today, however, music goes beyond being an executional
advertising element, becoming a ubiquitous aspect in the everyday
lives of modern consumers. Whether in a bank queue, a supermarket
or at home, music surrounds consumers (Kellaris, 2008). Music penetrates into almost all consumption contexts, and shapes consumers'
experiences profoundly.
The pervasiveness of music in consumers' lives makes music studies an extremely relevant area for consumer research. Not surprisingly, consumer research has increasingly directed scholarly attention to
music since the early work of Smith and Curnow (1966). However,
⁎ Correspondence to: B. Ursavas, Istanbul Bilgi University, Department of Advertising,
Santral Istanbul, 34060 Eyup Istanbul, Turkey. Tel.: +90 212 311 7547.
⁎⁎ Correspondence to: O. Hesapci-Sanaktekin, Bogazici University, Department of
Management, Bebek 34342, Istanbul, Turkey. Tel.: +90 212 359 6503; fax: + 90 212
287 7851.
E-mail addresses: [email protected] (B. Ursavas), [email protected]
(O. Hesapci-Sanaktekin).
0148-2963/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.jbusres.2011.12.013
due to music's complex effects on affective and cognitive mechanisms,
existing research findings are very controversial. The objective of this
paper is to contribute to the existing literature by investigating how
individuals form temporal perceptions and memories in response
to ads that contain dramatic contrast, a specific type of emotional
ambivalence which results from conflicting background music and
ad message. Further, the study explores how discomfort with ambiguity (DWA) as a personality trait and prior mood affect the cognitive
responses toward stimuli that contain emotional ambivalence.
2. Theoretical framework
Consumer research typically investigates the effects of stimulus
congruity in a picture-word consistency context. A general finding
from the studies (Childers & Houston, 1984; Heckler & Childers,
1992; Lutz & Lutz, 1977; Unnava & Burnkrant, 1991) displays that
recall of advertising, brand name, and product attributes enhances
when both pictures and words communicate the same information.
Music-related consumer research begins to focus on stimulus congruity with the seminal work of MacInnis and Park (1991). The authors find a significant effect of congruity on message processing,
positive emotions, and attitudes toward the ad, especially for lowinvolvement individuals. Although their study involves music, the
research mainly explores the effects of congruity between two verbal cues (i.e., song lyrics and ad message).
In contrast to MacInnis and Park (1991), Kellaris et al. (1993)
operationalize stimulus congruity in terms of the music's relevance
to the ad message. In this respect, their study is the first to investigate
the role of music in the context of stimulus congruity. Kellaris
et al. (1993) manipulate music's attention-gaining quality and its
934
B. Ursavas, O. Hesapci-Sanaktekin / Journal of Business Research 66 (2013) 933–940
congruity with the ad arguments. The authors suggest that high congruity between music and message moderates the effect of music's
attention-gaining property on ad recall and recognition. Similarly,
North, MacKenzie, Law, and Hargreaves (2004) show that a high
music–message congruity leads subjects to generate better recall of
ad content.
With regard to incongruous emotions, emotion theory suggests
that simultaneous experience of oppositely-charged emotions is a
frequent phenomenon in the lives of individuals. The advertising
industry uses conflicting emotions to increase the persuasiveness of
ad appeals (Williams & Aaker, 2002). Music–message contrast may
activate different cognitive and affective processing mechanisms. As
a result, consumers may develop quite different reactions toward
music–message contrast in comparison to music–message congruity.
Therefore, the music-related stimulus congruity literature may have a
lot to gain from studies that explore consumer responses toward
stimuli containing contrasting emotions.
2.1. Emotional ambivalence
Since the seminal work of Holbrook and Hirschman (1982), and
Hirschman and Holbrook (1982), emotions occupy a pivotal position
in the field of consumer research. Although research effort traditionally focuses on investigating the effects of pure emotions, a nascent
body of research is emerging on the experience of conflicting emotions (Williams & Aaker, 2002). Some evidence suggests that conflicting emotions can be experienced simultaneously (Aaker, Drolet, &
Griffin, 2008; Cacioppo, Gardner, & Bernston, 1997; Larsen, McGraw,
& Cacioppo, 2001; Priester & Petty, 1996; Thompson, Zanna, &
Griffin, 1995; Williams & Aaker, 2002). According to this viewpoint,
emotional valence represents two independent dimensions; hence,
oppositely charged emotions can cooccur. An alternative viewpoint
postulates that opposite emotions lie on a single bipolar dimension
(Brehm & Miron, 2006; Hesapci-Sanaktekin & Ursavas, 2011; Russell,
1980; Russell & Carroll, 1999). Therefore, opposite emotions cannot
occur simultaneously, but can occur in close sequential proximity.
A body of research in the emotional ambivalence literature centers
on factors that determine when and how people are likely to experience emotional ambivalence. Findings suggest that such factors as
age (Labouvie-Vief, De-Voe, & Bulka, 1998; Williams & Aaker, 2002),
gender (Bagozzi, Gopinath, & Nyer, 1999; Fong & Tiedens, 2002),
culture (Bagozzi & Pieters, 1998; Williams & Aaker, 2002), and discomfort with ambiguity (Janssens, De Pelsmacker, & Weverbergh,
2007) can play a moderating role in the experience of emotional
ambivalence.
Age and gender are important factors that influence the experience of emotional ambivalence. For instance, research suggests that,
in comparison to younger adults, older adults better deal with complex and conflicting emotions (Carstensen, Isaacowitz, & Charles,
1999; Labouvie-Vief et al., 1998), and better recall mixed emotional
stimuli (Aaker et al., 2008). Gender is another variable that affects
the experience of conflicting emotions. In general, women experience
more emotional ambivalence than men (Fong & Tiedens, 2002). In
addition to age and gender, culture has a strong impact on the experience of emotional ambivalence. Williams and Aaker (2002) and
Bagozzi et al. (1999) report that elderly people and individuals from
Eastern cultures possess relatively higher propensity to cope with
the concept of duality. Because such individuals can more easily process duality, they develop less negative responses toward messages
conveying conflicting emotions.
Research also examines the influence of personality traits in
experiencing emotional ambivalence. Janssens et al. (2007) investigate the moderating role of discomfort with ambiguity (DWA) on
the processing of mixed-emotional ad appeals. DWA refers to a personality trait that determines how individuals experience ambiguous
thoughts and feelings. DWA is a subconstruct of the need for closure
scale (NFCL), which reflects an individual's need for clear, structured
knowledge. Individuals with a high need for closure seek to arrive
at quick conclusions about a stimulus event, and are motivated to
quickly end cognitive information processing related to the event
(Webster & Kruglanski, 1994). Similarly, individuals with a high discomfort with ambiguity possess a lower level of propensity to accept
and process duality. Hence, they experience a higher level of displeasure when confronted with ambivalent emotions and meanings.
Conversely, individuals scoring low on discomfort with ambiguity
tend to react more positively toward stimuli containing ambiguous
emotions and meanings.
2.2. Memory
Research investigating the effects of stimulus congruity on memory generally report that subjects' information processing and recall of
a stimulus event are lower when the elements related to the stimulus
convey different information (Childers & Houston, 1984; Heckler &
Childers, 1992; Lutz & Lutz, 1977; Srull, Lichtenstein, & Rothbart,
1985; Unnava & Burnkrant, 1991). This result may emerge from a
number of factors ranging from the difficulty of encoding incongruent
information (Heckler & Childers, 1992; Meyers-Levy & Tybout, 1989)
to the weaker linkages between unrelated nodes in the associative
network (Schmitt, Tavassoli, & Milliard, 1993). Emotional ambivalence research also suggests a negative relationship between conflicting emotions and information processing (Aaker et al., 2008).
Translating these findings, the present research expects to find a
negative relationship between dramatic contrast and ad processing.
Since individuals experience difficulty in processing conflicting
emotions, an ad that contains dramatic contrast (versus dramatic
congruity) should generate a poorer ad memory. Therefore, the first
hypothesis expects a direct negative effect of dramatic contrast on
memory.
H1. Dramatic contrast (versus dramatic congruity) will generate a
significant negative effect on subjects' a) recognition, and b) recall
of central ad arguments.
Research also suggests that the experience of emotional ambivalence can be moderated by discomfort with ambiguity. As previously
explained, discomfort with ambiguity (DWA) refers to a personality
trait which determines how individuals experience ambiguous emotions and meanings. The greater discomfort with ambiguity implies
a reduced capacity for the processing of ad appeals containing
conflicting emotions. Therefore, the effect of dramatic contrast on
memory should be stronger for those individuals with high DWA.
H2. Subjects high (versus low) on DWA will generate significantly
lower message a) recognition, and b) recall in the presence of dramatic contrast.
2.3. Temporal perception
Perhaps, one of the most magnificent qualities of music is its ability to influence listeners' temporal perceptions. Given that consumers
often sacrifice their time in the form of experiencing delays, waiting
for service, spending time to search for product information and to
shop, researchers conceptualize time as a significant component of
the total cost of such transactions (Kellaris & Kent, 1992). Thus, a possible way to minimize the cost of time for consumers is to manipulate
their temporal perceptions by controlling the properties of external
stimuli that surround and influence them.
Prior research investigates human time perception either from
a prospective or a retrospective paradigm (Zakay, 1990). Findings
suggest that different cognitive mechanisms characterize prospective
and retrospective duration estimates (Sucala, Scheckner, & David,
B. Ursavas, O. Hesapci-Sanaktekin / Journal of Business Research 66 (2013) 933–940
935
H3. Subjects will generate significantly shorter duration estimates in
the presence of an ad conveying dramatic contrast (versus dramatic
congruity).
encroaching of negative thoughts during processing (e.g., Ellis &
Ashbrook, 1988). In general, mood states tend to influence behavior,
judgment and recall. However, negative mood states tend to have
diverse effects in comparison to positive moods (Gardner, 1985).
According to Gardner's review, one might expect positive affect to
enhance recall of information, while less positive affect might decrease coding and storage of information. Gardner (1985) suggests
that positive affect leads to an increase in cognitive elaboration by enhancing coding and storage of information. In other words, positively
valenced external stimuli encourage more information processing
from the event.
Despite the generally accepted evidence of mood-congruent
memory, opposing results exist. While positive mood leads to improved recall of positive memories, negative mood does not always
enhance the recall of negative material. A popular explanation for
these results is that of the “mood repair” perspective (Isen, 1985):
people in bad moods actively resist the remembering of negative
items so as to prevent a further decline in their mood states. Findings
report no such evidence for positive mood.
Experiencing mixed emotions results in increased levels of felt
discomfort (Williams & Aaker, 2002). With the mood-repair perspective, increased levels of felt discomfort, further deteriorating the
initial mood state, may lead people in negative moods to a resistance
in processing information in the emotionally ambivalent communication. Thus one may expect to find people in positive moods with
higher levels of processing of mixed emotional content, resulting in
a stronger memory and longer duration estimates (Kellaris & Kent,
1992) in comparison to people in negative mood states.
H4. Subjects with high (versus low) DWA will report significantly
shorter duration estimates in the presence of an ad conveying dramatic contrast.
H5. Positive mood (vs. negative) will lead to (a) longer duration estimates, better (b) recognition, and (c) recall of the ad message estimates in the presence of an ad conveying dramatic contrast.
2.4. Prior mood
3. Study 1
A substantial literature provides insight on the influence of mood
on information processing (e.g., Bless, Mackie, & Schwarz, 1992;
Schwarz & Bless, 1991; Schwarz & Clore, 1996). Although there are
several competing perspectives on the effects of mood on information
processing (see Forgas, 2002), most research suggests that happy
moods result in heuristic processing, whereas sad moods result in
more effortful processing (Bless, Bohner, Schwarz, & Strack, 1990;
Clore, Schwarz, & Conway, 1994). Alternatively, researchers such as
Matmur and Chattopadhayay (1991) report that positive mood actually increases the elaborative processing of happy ads.
There are three main perspectives to account for mood effects
on processing, namely, mood-repair (Isen, 1987), memory effects
(Mackie & Worth, 1989), and problem signaling (Schwarz & Bless,
1991) approaches. The first perspective focuses on the moodmaintenance hypothesis that suggests people are generally motivated
to maintain positive moods and to repair negative moods (Isen,
1987). Isen mainly argues that individuals seek to maintain positive
mood, and avoid exerting cognitive effort that is not directed at either
maintaining or regaining positive mood. Second, Mackie and Worth
(1989) suggest that moods influence cognitive capacity, where
happy moods result in reduced capacity with less systematic processing. Schwarz and Bless's (1991) motivational interpretation, in
the third approach, posits that negative mood signals problematic
situations, which require more cognitive resource that results in
information processing that is relatively analytical, effortful, and cautious. When people are in a positive mood, their environment seems
to pose no threat, and they are more likely to rely on heuristics (Bless
et al., 1990).
On the contrary, other researchers suggest that negative moods
lead to a reduced effort (Hertel & Hardin, 1990; Lassiter, Koenig, &
Apple, 1996), which can reduce processing capacity through the
The first study manipulates dramatic tone, and categorizes
discomfort with ambiguity (DWA) in a 2 (dramatic tone: contrast,
congruity) × 2 (DWA: high, low) between subjects design. The term
dramatic tone refers to a differentiation between the two types of
dramatic qualities. Depending on the dramatic tone of the ad, the
background music either contrasts (dramatic contrast) or supports
(dramatic congruity) the central ad argument. Dramatic contrast differs from emotional ambivalence in that it refers to the specific state
of experiencing conflicting emotions via conflicting auditory stimuli.
2010; Ursavas & Hesapci-Sanaktekin, 2012). In the prospective paradigm, subjects are aware that the experimental task involves
estimating the duration of a stimulus (Zakay & Block, 1997). Cognitive
research on prospective time perception explains the phenomenon
with attentional models, which suggest a negative relationship between the amount of information processing from an event and the
magnitude of duration estimates (Zakay & Block, 1994). In contrast
to prospective time perception, the retrospective paradigm involves
collecting subjects' duration estimates after the completion of a
stimulus task (Zakay & Block, 1997). In the retrospective research
scenario, individuals base their duration estimates about a stimulus
event on the amount of encoding and retrieval of information from
the event (Block, 1990; Ornstein, 1969; Zakay & Block, 1996). Therefore, retrospective models predict a positive relationship between the
amount of memory processed and the magnitude of duration estimates about a stimulus.
This process should also govern the relationship between dramatic contrast and memory. In other words, because individuals cannot
easily process emotional ambivalence, they should process less information from an ad containing dramatic contrast. When these individuals report duration estimates of an ad, they should produce shorter
estimates because their judgments depend on the amount of information their memory stores. This relationship should be even stronger
for individuals with high discomfort with ambiguity.
3.1. Stimuli development
Four experimental ads are designed for a fictitious NGO. A professional actor voiced the ad copy and a professional musician composed
original music scores for the ads to bolster the dramatic quality. Music
and voiced copy combine to create a single audio track. The ads are
identical in terms of duration (45 s) and the information they contain.
3.2. Pretest
32 participants (50% female, μ. age = 23.4) evaluate the music and
the copies. Participants rate the degree to which each copy and music
evoke certain emotions on a seven point scale (1 = not at all, 7 = very
strongly). For the emotion types, the study uses happiness as an example of positive emotion and sadness as an example of negative
emotion. Prior research displays similar experience, recognition, and
expression of these emotions across different settings (Ekman &
Friesen, 1971; Williams & Aaker, 2002). The emotion scale contains
happiness index (happy, delighted, joyful, pleased, cheerful), sadness
index (sad, downhearted, distressed, sorrowful, dejected, regretful),
as well as 15 filler emotions. Results of the pretest show that two
936
B. Ursavas, O. Hesapci-Sanaktekin / Journal of Business Research 66 (2013) 933–940
musical pieces out of six and two ad copies out of four are successful
in evoking either happiness or sadness.
3.3. Participants and the procedure
137 undergraduate students (61.3% female, μ. age = 22.32) participate in the study for course credit. The procedure involves randomly
assigning subjects to each experimental condition in small groups,
and asking them to complete a self administered questionnaire.
3.4. Measures
As key dependent variables, measures include ad duration estimates and ad message recall. In addition, participants provide information on demographics, measures of discomfort with ambiguity,
and manipulation checks for emotions.
Following prior research (Oakes, 1999, 2003) the present study
measures subjects' duration estimates by asking them to mark a
number from a table that includes numbers ranging from 1 to 180 s.
Following prior research, subjects report recall by listing as many
messages as they remember from the ad (Kellaris et al., 1993).
Responses are then compared against the ad copy with exact or
very similar answers rated as accurate. The authors initially code for
the message recall. A researcher then recodes the answers, which
reveal a significantly high association between the two codings
(Cohen's Kappa = .83). Five items measure DWA (α = .89) on a
seven-point scale (1 = not at all; 7 = very strongly), adapted from
Webster and Kruglanski (1994). The score of the DWA transforms
into a categorical variable by means of median split.
3.5. Manipulation checks
Manipulation checks are performed to understand whether the
experimental ads successfully create dramatic contrast and dramatic
congruity. After viewing one of the four ad conditions, participants
rate the degree to which they experience a set of emotions in response to both ad music and ad message on a seven-point scale
(1 = not at all; 7 = very strongly). The current study adopts emotional items following prior research (e.g., Edell & Burke, 1987; Holbrook
& Batra, 1987; Izard, 1977; Richins, 1997). Five items comprise the
happiness index (happy, delighted, joyful, pleased, and cheerful)
and six items comprise the sadness index (sad, downhearted, distressed, sorrowful, dejected, and regretful). Emotion measures also
include ten filler emotion items.
First, participants rate how much the ad music makes them feel
happy (α = .91) and sad (α = .93). Participants in the positively
valenced ad music condition report significantly higher positive
feelings (F1,135 = 157.77, p b .01) in comparison to the participants in
the negatively valenced ad music condition (μ = 4.17, μ = 2.19,
respectively). Participants in the negatively valenced ad music condition report significantly higher negative feelings (F1,135 = 37.49,
p b .01) in comparison to those in the positively valenced ad music
condition (μ = 4.37, μ = 2.67, respectively).
Second, participants rate how much the ad message makes them
feel happy (α = .92) and sad (α = .92). Participants in the positively
framed message condition report significantly higher positive feelings
(F1,135 = 34.01, p b .01) compared to the participants in the negatively
framed ad message condition (μ = 4.14, μ = 2.96, respectively).
Participants in the negatively framed message condition report significantly higher negative feelings (F1,135 = 25.85, p b .01) than the participants in the positively framed ad message condition (μ = 4.34,
μ = 3.08, respectively). These results suggest that the manipulations
are successful.
In the ad conditions where there is dramatic contrast between the
ad music and the ad message, no significant mean differences exist
with respect to the strength of emotions evoked by the music and
the message. More specifically, an ad conveying positively valenced
music and negatively valenced message generates no significant
mean difference in terms of positive and negative feelings (μ = 4.07,
μ = 3.88; t = .74; p = .46). Similarly, an ad conveying negatively
valenced music and positively valenced message generates no significant mean difference with respect to positive and negative feelings
(μ = 3.99, μ = 4.63; t = 1.75; p = .09).
Analyses also check whether the two dramatic congruity ads are
successful in creating congruent emotions. Under the happy dramatic
congruity ad condition, a significant mean difference exists between
felt happiness (μ = 4.33) and sadness (μ = 2.58) (t = .38; p b .01).
More specifically, subjects viewing the happy dramatic congruity ad
report significantly more positive feelings than negative feelings.
Similarly, under the sad dramatic congruity ad condition, felt sadness
(μ = 4.13) is significantly greater than felt happiness (μ = 2.38,
t = 3.5; p b .01). All together, manipulation checks reveal that the
experimental ads successfully inhibit the two dramatic tone conditions (contrast vs. congruity).
4. Results
Significant intercorrelations exist between duration estimates and
message recall (r = .17, p = .04). A multivariate analysis of variance
(MANOVA) is performed on the intercorrelated measures to avoid a
Type I error. Analysis reveals that dramatic tone has significant main
effects on message recall (F1,133 = 25.75; p b .01) (Wilk's Lambda =
.93). Furthermore, dramatic tone and DWA have a significant interaction effect on message recall (F1, 133 = 4.94; p = .028) and duration
estimates (F1, 133 = 6.06; p = .01) (Wilk's Lambda= .93).
Based on the positive MANOVA findings for main and interaction
effects, a series of univariate analysis of variance (ANOVA) are also
performed on each dependent variable.
4.1. Message recall
Results reveal a significant mean difference on message recall
between two dramatic tone conditions (F1,135 = 23.09, p b .01). The
dramatic contrast condition yields lower recall than the congruent
ad condition (μ = 1.20, μ = 2.07, respectively). Thus, hypothesis 1b
(H1) receives support. Further post hoc comparisons indicate significant mean differences between negative–positive dramatic contrast
condition (μ = 1.18) and positive–positive (μ = 1.94, F1,64 = 8.63,
p = .005) and negative–negative (μ = 2.19, F1,68 = 13.31, p b .01)
congruent ad conditions.
Findings reveal significant mean differences between positive–
negative dramatic contrast (μ = 1.23) and positive–positive (F1,65 =
9.41, p = .01) and negative–negative (F1,69 = 14.31, p b .01) congruent
ad conditions. No significant mean difference exists between two dramatic contrast conditions (positive–negative and negative–positive)
(F1,67 = .05, p = .83). Though negative–negative ad condition yields
higher recall than positive–positive congruent condition, there is no
significant mean difference between these two congruent ad conditions (F1,66 = .91, p = .35).
Discomfort with ambiguity seems to produce a significant main
effect (F1,133 = 9.59; p = .01), such that individuals low on DWA report higher levels of ad message recall (μ = 1.90) than do individuals
with high DWA (μ = 1.38). Dramatic tone also has a significant main
effect (F1,133 = 25.75; p ≤ .01). In general, individuals who view the
congruent ads recall many more ad arguments (μ = 2.07) than those
who view the dramatic contrast ads (μ = 1.20), such that, individuals
low on DWA report higher levels of ad message recall (μ = 1.90)
than do individuals with high DWA (μ = 1.38). DWA and dramatic
contrast produce a significant interaction effect on ad message recall
(F1,133 = 4.94; p = .03). Planned comparisons reveal that in the
congruent ad conditions no significant mean difference in ad message
recall is present (F1,66 = .31, p = .58) among participants with high
B. Ursavas, O. Hesapci-Sanaktekin / Journal of Business Research 66 (2013) 933–940
DWA (μ = 2.00) and participants with low DWA (μ = 2.15). On the
other hand, in the dramatic contrast ad conditions, participants with
high DWA recall fewer ad message items (μ = 0.74) than participants
with low DWA (μ = 1.66; F1,67 = 18.02; p b .01). Thus, hypothesis 2b
receives support.
4.2. Duration estimates
While no significant main effects of DWA (F1,133 = .36, p = .55)
and dramatic tone (F1,133 = .29, p = .59) seem to be present, the
interaction effect of DWA and dramatic tone on duration estimate is
significant (F1,133 = 6.06; p = .01). Therefore, hypothesis 3 (H3) does
not receive support. However, planned comparisons reveal that in
the congruent ad conditions, no significant mean difference in duration estimate variations is present (F1,66 = 1.96, p = .17) among participants with high DWA (μ = 32.46) and participants with low
DWA (μ = 28.30). On the other hand, in the ad conditions creating
dramatic contrast, participants with high DWA report shorter estimates of ad duration (μ = 25.76) than participants with low DWA
(μ = 32.57; F1,67 = 4.21; p = .04) (Fig. 1). Hence, hypothesis 4 (H4)
receives support.
5. Study 2
Study 2 aims to extend the findings of study 1 and explores the
impact of prior mood on cognitive responses toward dramatic contrast. The study entails a 2 (mood: positive or negative) × 2 (dramatic
tone: contrast or congruity) between subject design.
5.1. Pretest
In a pretest to choose music for the mood manipulations, participants (42.86% female, μ. age = 22.1) evaluate a total of eight music
pieces that induce either a positive or a negative mood. Judgments
and advice from music professionals guide the selection of musical
pieces. Participants rate the degree to which each musical piece
makes them feel a certain type of emotion on a seven point scale
(1 = not at all, 7 = very strongly). The pretest shows that a piece
from “Adagio”, by Tomaso Albinoni (negative mood condition), and
the song “Let's Twist Again”, by Chubby Checker (positive mood condition) successfully induce the target moods.
5.2. Participants and the procedure
279 (54.8% female, μ. age = 21.43) undergraduate students participate in the study. The ad music and ad message manipulation materials are the same as study 1.
The study follows a procedure similar to study 1.
937
5.3. Measures
The key dependent variables of the present study are ad duration
estimates, ad message recall, and ad message recognition (see Experiment 1). To measure message recognition, a 22-item checklist asks
subjects to identify the advertised message arguments (Singh,
Rothschild, & Churchill, 1988). In order to increase the sensitivity
of the measure, the checklist uses filler items including ‘I can't
remember’ and ‘none of the above’ (Kellaris et al., 1993). The present
study adopts the items from Webster and Kruglanski (1994) to
measure DWA (α = .856).
5.4. Manipulation checks
After listening to one of the two music conditions for five minutes,
participants rate their mood on a four-item semantic scale (α = .90).
Participants in the positive mood condition report significantly higher
positive mood than participants in the negative mood condition
(μ = 5.96, μ = 3.40; respectively, F1,277 = 183.79, p b .01).
Participants also rate how much the ad music makes them feel
happy (α = .91) and sad (α = .93). Participants in the positively
valenced ad music condition report significantly higher positive
feelings in comparison to the participants in the negatively valenced
ad music condition (F1,277 = 78.68, p b . 01, μ = 4.21, μ = 2.16, respectively). Participants in the negatively valenced ad music condition
report significantly higher negative feelings (F1,277 = 52.53, p b .01)
in comparison to the participants in the positively valenced ad
music condition (μ = 4.98, μ = 2.77, respectively).
Participants also rate how much the ad message makes them feel
happy (α = .94) and sad (α = .89). Participants in the positively
framed message conditions report significantly higher positive feelings (F1,277 = 47.61, p = .02) in comparison to the participants in the
negatively framed ad message conditions (μ = 4.11, μ = 2.67, respectively). Participants in the negatively framed message conditions
report significantly higher negative feelings (F1,277 = 12.32, p b .01)
in comparison to the participants in the positively framed ad message
conditions (μ = 4.61, μ = 3.77, respectively).
In the ad conditions where there is dramatic contrast between the
ad music and the ad message, no significant mean difference exists in
emotions evoked by the music and the message. More specifically,
when participants view the ad with a positively valenced music and
a negatively framed message, no significant mean difference occurs
among positive and negative feelings (μ = 4.68, μ = 4.52; t = −.24,
p = .81). When participants view the ad with a negatively valenced
music and a positively framed message, no significant mean difference exists among positive and negative feelings (μ = 4.51, μ = 5.18;
t = −1.76, p = .84). Findings reveal that the two ad conditions are
successful in creating dramatic contrast.
6. Results
Fig. 1. Study 1 — Two-way interaction of DWA and dramatic contrast on duration
estimates.
A significant correlation exists between the two memory measures (message recognition and message recall) (r = .35, p b .01).
There is also a significant correlation between duration estimates
and message recognition (r = −.16, p b .05). Multivariate analysis of
variance (MANOVA) reveals that dramatic tone has significant main
effects on message recognition (F1, 271 = 8.11; p b .01), and duration
estimates (F1, 271 = 4.09; p b .05; Wilk's Lambda = .94). A significant
main effect of mood exists on message recognition (F1, 271 = 25.91;
p b .01), message recall (F1, 271 = 21.49; p b .01), and duration
estimates (F1, 271 = 12.35; p b .01; Wilk's Lambda = .85). MANOVA
reveals no main effect of DWA on any of the dependent variables.
Furthermore, dramatic tone, DWA and mood seem to have an interaction effect on message recall (F1, 271 = 3.99; p b .05) and message
recognition (F1, 271 = 7.33; p b .01) (Wilk's Lambda = .96).
938
B. Ursavas, O. Hesapci-Sanaktekin / Journal of Business Research 66 (2013) 933–940
6.1. Message recall
ANOVA results reveal a significant main effect of prior mood on ad
message recall (F1,271 = 21.24, p b .01), such that in the positive mood
condition, message recall is higher in comparison to the negative
mood condition (μ = 1.52; μ = .93, respectively). Another main effect
appears to be DWA (F1,271 = 4.07, p b .05). Individuals low on DWA
produce more recall for message items than individuals with high
DWA (μ = 1.36; μ = 1.16, respectively).
The interaction effect of DWA, dramatic tone and prior mood
on ad message recall is significant (F1,271 = 9.64; p b .01). Planned
comparisons for the dramatic contrast conditions reveal that when a
negative mood is present prior to the ad, individuals with high
DWA recall fewer message items (μ = .77) than individuals with
low DWA scores (μ = 1.57; F1,61 = 11.04, p b .01). On the other hand,
in the dramatic contrast conditions, mood appears to have a main
effect on message recall (F1,130 = 5,29, p b .05) such that, when a positive mood (vs. negative mood) is present prior to the ad, participants
report better recall (μ = 1.52 vs. μ = 1.13) of the message arguments.
Therefore, hypothesis 5c (H5) receives support.
6.2. Message recognition
Prior mood, dramatic tone and DWA produce significant main
effects (F1,271 = 25.52, p b .01; F1,271 = 5.10, p b .05; F1,271 = 5.56,
p b .05, respectively). Overall, individuals with high DWA recognize
fewer message items than low DWA individuals (μ = 3.36, μ = 3.71;
respectively). In the positive mood condition (vs. negative), individuals display better recognition for message items (μ = 3.91, μ = 3.04;
respectively). Dramatic contrast conditions (vs. congruity) produce
lower recognition (μ = 3.20, μ = 3.76; respectively). Hypothesis 1a
(H1) receives support.
Significant interaction effect of DWA and dramatic congruity on
message recognition supports hypothesis 2a (F1,271 = 10.99, p b .01).
Planned comparisons reveal that high (vs. low) DWA individuals
recognize fewer message arguments in response to a dramatic contrast ad (μ = 2.94, μ = 3.74; respectively, F1,130 = 8.98, p b .01).
Furthermore, similar to the findings on message recall, DWA,
dramatic tone, and prior mood have a significant interaction effect
on ad message recognition (F1,271 = 4.58; p b .05). Planned comparisons reveal that in the dramatic contrast condition, when a negative
mood is present prior to the ad, high (vs. low) DWA individuals
recognize fewer message items (μ = 2.06, μ = 3.57, F1,61 = 19.29,
p b .01). On the other hand, in the dramatic contrast conditions there
is a main effect of mood on message recognition (F1,130 = 13.54,
p b .01) such that, when a positive mood (vs. negative mood) is present prior to the ad, participants report better recognition (μ = 3.64
vs. μ = 2.73). Therefore, hypothesis 5b (H5) receives support.
Fig. 2. Study 2 — Two-way interaction of DWA and dramatic contrast on duration
estimates in positive mood condition.
reveal a main effect of DWA in the dramatic contrast condition where
a negative mood is present prior to the ad (F1,61 = 4.44, p b .05),
such that high DWA individuals report significantly longer duration estimates than low DWA individuals (μ = 69.91, μ = 56.01, respectively).
No such effect of DWA exists in the positive mood dramatic contrast
condition (F1,67 = .51, p = .48). Furthermore, in the dramatic contrast
condition, mood has a main effect on duration estimates (F1,130 = 4.32,
p b .05), such that, when a positive mood (vs. negative mood) is present
prior to the ad, participants report lower but more accurate duration
estimates (μ = 52.93 vs. μ = 63.06) (Fig. 2). Hence, hypothesis 5a (H5)
does not receive support.
7. Discussion
7.1. Message recognition and recall
The results of two experiments indicate that using contrasting
emotions in ad appeals has distorting effects on consumers' memory.
Consumers, especially those who feel a higher level of discomfort
when confronted with ambiguous meanings, tend to demonstrate
lower recall and recognition for the persuasive arguments conveyed
in the ad. Dual meanings in the ad impair the information processing
capacity of high DWA individuals and, hence, result in lower memory
for the ad messages. As previously discussed, DWA is a subconstruct
of the need for closure scale (NFCL), which reflects an individual's
desire for clear and definite knowledge (Janssens et al., 2007). Such
individuals experience greater difficulty while processing ambiguous
information about stimulus events. In line with previous research,
the present study documents that high DWA subjects generate better
6.3. Duration estimates
ANOVA reveals a main effect of mood (F1,271 = 12.35, p b .01),
implying that negative mood leads to longer duration estimates
than positive mood (μ = 59.30, μ = 49.99, respectively). A significant
main effect of dramatic tone is also present (F1,271 = 4.085, p b .05),
such that participants in the dramatic contrast (vs. congruity) conditions report longer duration estimates (μ = 57.77, μ = 51.44, respectively). Planned comparisons display a main effect of mood and a
marginal main effect of DWA (F1,128 = 4.86, p b .05; F1,128 = 3.62,
p = .06; respectively). Participants in the negative mood condition
report longer time estimates than participants in the positive mood
condition (μ = 63.01, μ = 52.93, respectively) (Fig. 3). On the other
hand, participants with high (vs. low) DWA report longer duration
estimates (μ = 60.49, μ = 53.30, respectively).
Analysis also reveals an interaction effect of DWA with dramatic
tone on duration estimates (F1,271 = 6.21, p b .05). Further comparisons
Fig. 3. Study 2 — Two-way interaction of DWA and dramatic contrast on duration
estimates in negative mood condition.
B. Ursavas, O. Hesapci-Sanaktekin / Journal of Business Research 66 (2013) 933–940
recognition for ads containing congruent emotions. On the other
hand, individuals report much lower recognition for ads that contain
contrasting music and message emotions. While viewing a dramatic
contrast ad, high DWA consumers fail to elaborate on central ad arguments due to the relative difficulty of comprehending two opposite
emotions simultaneously. When music and message emotions support the central ad argument, however, these consumers generate
no less ad memory than low DWA consumers.
Similar effects seem to appear also in situations where high DWA
individuals experience negative mood prior to ad exposure. Such individuals demonstrate a lower recall and recognition for ad messages
as compared to those with a lower level of DWA. These results establish that discomfort with ambiguity moderates the processing of ads
that contain dramatic contrast. Within the mixed emotions literature,
Janssens et al. (2007) investigate the moderating role of DWA on the
effects of mixed emotions on consumer evaluations. The authors elicit
mixed emotions by contrasting the mood generated by the medium
context with the emotions induced by an ad appeal embedded in
the medium. Results in two experiments indicate that DWA has a
moderating effect on attitudinal responses toward ad stimuli that
contain mixed emotions. More specifically, individuals with a high
DWA appear to respond less positively to mixed emotions. Findings
of the current study add to the findings of Janssens et al. (2007) and
report the effect of DWA on memory and temporal perception in
the presence of a mixed emotional stimulus.
7.2. Temporal perception
Data provide no support for the hypothesis that dramatic contrast
in an ad shrinks retrospective duration estimates. Findings imply
that the diminishing effect on consumers' time perceptions occurs
regardless of the dramatic tone of the ad. In general, consumers perceive the duration of an ad as being shorter when emotions evoked
by music and message contrast, rather than support each other.
As study 1 suggests, DWA moderates the effect of dramatic contrast, such that higher DWA individuals tend to have a more distorted
perception of time. This finding seems in line with the previous retrospective time perception research. Retrospective time perception
models postulate a negative relationship between the amount of information processed during an event, and duration estimates about
the event (Boltz, 2004; Ornstein, 1969). According to these models,
individuals typically remember event durations as being shorter
when they store and retrieve less information related to an event
(Zakay & Block, 1997).
As for the mood effects, negative mood appears to inhibit the
duration estimation capacity and lead to less recall and recognition.
On the other hand, in dramatic contrast conditions, positive (vs. negative) mood leads to better recognition and recall, but shorter
duration estimates. These findings break the expected underlying
reasoning on retrospective time perception. Effects of mood on coping mechanisms with contrast may play a role in these findings on
temporal perceptions and need further exploration.
Study 2 also reveals an amplifying effect of dramatic contrast on
duration estimates, which can be in harmony with the conventional
wisdom dictating that “a watched pot never boils”. This phrase expresses the naive theory that when an event involves an arduous
task, time seems to slow down. In fact, previous research also suggests that individuals generate longer duration estimates for stressful
stimuli (Vohs & Schmeichel, 2003). This insight may also help explain
why individuals estimate longer durations for ads conveying dramatic contrast.
8. Limitations and future directions
An important limitation is that the study measures message recall
through single exposure. Low level of recall may stem from this very
939
factor. In real life, however, an ordinary consumer views an ad for
more than a single shot. Therefore, a multiple exposure recall measurement in future research may provide a more realistic picture of
the relationship between dramatic contrast and consumer memory.
A second limitation stems from transforming the score of the DWA
variable into a categorical variable by means of median split to meet
the statistical requirements of ANOVA. Previous research suggests
that this technique may reduce the statistical power to detect interaction (Irwin & McClelland, 2001). Therefore, future research may benefit from employing a different data analysis method (e.g., moderated
regression analysis) that accepts DWA as a numerical variable
(Janssens et al., 2007).
The present research operationalizes dramatic contrast by using a
happiness–sadness emotion couple. The use of such a basic emotion
couple has practical benefits. Future research can evoke dramatic
contrast through more complex emotion couples (e.g., pride, shame,
empathy, anger) (Hesapci-Sanaktekin, 2007) to provide a more thorough investigation of the topic. Although eliciting complex emotions
through music may be a challenging task, one can manage to do so
by carefully controlling the structural elements within the music.
Another avenue for future research is to explore the effects of
dramatic contrast in different contexts. The present research design
involves a music–text context. Future research should attempt to
evoke dramatic contrast in music–picture or music-only contexts.
Developing experimental stimuli in a music–picture context should
not be very difficult. However evoking dramatic contrast in a musiconly context may be a rather difficult task. Still, a music-only context
should be possible through a solid understanding of the nature of
structural musical elements. Finally, previous research suggests
that music conveys not only emotions, but also meanings (Zhu &
Meyers-Levy, 2005). Therefore, future research can explore the construct of dramatic contrast in a wider framework, which includes
emotive meanings as well as emotions.
9. Conclusion
Through two experiments, the present study investigates the
influence of dramatic contrast on consumers' memory and time perceptions. The study provides a genuine contribution to consumer
research in various respects. Drawing a demarcation line between
dramatic contrast and other types of emotional ambivalence is important because the former involves evoking emotional conflict through
auditory cues, whereas the latter operates through visual cues.
These two different types of cues may activate different cognitive
and affective mechanisms that result in the experience of emotional
ambivalence. Therefore, the term dramatic contrast makes a clear distinction between different types of emotional ambivalence.
A second contribution of this study is that, for the first time in
the emotional ambivalence literature, music acts as an independent
variable to operationalize emotional ambivalence. In this respect,
this study also sets a unique methodological example for future investigators for using music as an experimental variable in emotional
ambivalence research. Lastly, the experiments extend the current
state of emotional ambivalence research by introducing to the literature a novel dependent variable: time perception. Given the value of
time in exchange relationships, consumer research should investigate
the topic of time perception in a wider domain. In this respect, this
study is also the first to document empirical evidence on the relationship between dramatic contrast and time perception.
References
Aaker J, Drolet A, Griffin D. Recalling mixed emotions. Journal of Consumer Research
2008;35:268–78.
Bagozzi PR, Pieters R. Goal-directed emotions. Cognition and Emotion 1998;12:1-26.
Bagozzi R, Gopinath M, Nyer P. The role of emotions in marketing. Journal of the
Academy of Marketing Science 1999;27:184–206.
940
B. Ursavas, O. Hesapci-Sanaktekin / Journal of Business Research 66 (2013) 933–940
Bless H, Bohner G, Schwarz N, Strack F. Mood and persuasion: a cognitive response
analysis. Personality and Social Psychology Bulletin 1990;16:331–45.
Bless H, Mackie DM, Schwarz N. Mood effects on encoding and judgmental processes in
persuasion. Journal of Personality and Social Psychology 1992;63:585–95.
Block RA. Cognitive models of psychological time. Hillsdale, NJ: Erlbaum; 1990.
Boltz MG. The cognitive processing of film and musical soundtracks. Memory and
Cognition 2004;32:1194–205.
Brehm JW, Miron AM. Can the simultaneous experience of opposing emotions really
occur? Motivation and Emotion 2006;30:13–30.
Cacioppo JT, Gardner WL, Bernston GG. Beyond bipolar conceptualizations and measure:
the case of attitudes and evaluative space. Personality and Social Psychology Review
1997;1:3-25.
Carstensen LL, Isaacowitz DM, Charles ST. Taking time seriously: a theory of socioemotional selectivity. American Psychologist 1999;54:165–81.
Childers TL, Houston MJ. Conditions for a picture-superiority effect on consumer
memory. Journal of Consumer Research 1984;11:643–54.
Clore GL, Schwarz N, Conway M. Affective causes and consequences of social information processing. In: Wyer RS, Srull TK, editors. Handbook of social cognition.
Hillsdale, NJ: Lawrence Erlbaum Associates, Inc; 1994. p. 323–418.
Edell JA, Burke MC. The power of feelings in understanding advertising effects. Journal
of Consumer Research 1987;14:421–33.
Ekman P, Friesen WV. Constants across cultures in the face and emotion. Journal of
Personality and Social Psychology 1971;17:124–9.
Ellis HC, Ashbrook PW. Resource allocation model of the effects of depressed mood
states on memory. In: Fiedler K, Forgas J, editors. Affect, cognition, and social
behavior. Toronto: C. J. Hogrefe; 1988. p. 25–43.
Fong C, Tiedens L. Dueling experiences and dual ambivalences: emotional and motivational ambivalence of women in high status positions. Motivation and Emotion
2002;26:105–21.
Forgas JP. Feeling and doing: affective influences on interpersonal behavior. Psychological
Inquiry 2002;13:1-28.
Gardner MP. Mood states and consumer behavior: a critical review. Journal of Consumer
Research 1985;12:281–300.
Geissmann T. Gibbon songs and human music from an evolutionary perspective. In:
Wallin N, Merker B, Brown S, editors. The origins of music. Cambridge, Massachusetts:
MIT Press; 2000. p. 103–23.
Heckler SE, Childers TL. The role of expectancy and relevancy in memory for verbal and
visual information: what is incongruency? Journal of Consumer Research 1992;18:
475–92.
Hertel PT, Hardin TS. Remembering with and without awareness in a depressed mood:
evidence of deficits in initiative. Journal of Experimental Psychology 1990;119:
45–59.
Hesapci-Sanaktekin, O. The marketing power of mixed emotions across cultures.
Unpublished Ph.D dissertation, Bocconi University: Italy, (2007).
Hesapci-Sanaktekin O, Ursavas B. The role of self-esteem on the persuasive effects of
emotional valence sequence. Proceedings of 10th International Marketing Trends
Congress 20–22 January, France; 2011.
Hirschman EC, Holbrook MB. Hedonic consumption: emerging concepts, methods and
propositions. Journal of Marketing 1982;46:92-101.
Holbrook BM, Batra R. Assessing the role of emotions as mediators of consumer
responses to advertising. Journal of Consumer Research 1987;14:404–20.
Holbrook MB, Hirschman EC. The experiential aspects of consumption: consumer
fantasies, feelings, and fun. Journal of Consumer Research 1982;9:132–40.
Irwin JR, McClelland GH. Faulty heuristics and moderated multiple regression models.
Journal of Marketing Research 2001;38:100–9.
Isen Alice M. Asymmetry of happiness and sadness in effects on memory in normal
college students. Journal of Experimental Psychology: General 1985;114:388–91.
Isen AM. Positive affect, cognitive processes and social behavior. In: Berkowitz L, editor.
Advances in experimental social psychology. San Diego: Academic Press; 1987.
p. 203–53.
Izard CE. Human emotions. New York: Plenum; 1977.
Janssens W, De Pelsmacker P, Weverbergh M. The Effect of mixed emotions in advertising: the moderating role of discomfort with ambiguity. Advances in International
Marketing 2007;18:63–92.
Kellaris J. Music and consumers. In: Haugvedt C, Herr P, Kardes F, editors. Handbook of
consumer psychology. Mahwah, NJ: Erlbaum; 2008. p. 837–56.
Kellaris JJ, Kent RJ. The influence of music on consumers' temporal perceptions: does time
fly when you are having fun? Journal of Consumer Psychology 1992;1:365–76.
Kellaris JJ, Cox AD, Cox D. The effect of background music on ad processing: a contingency explanation. Journal of Marketing 1993;57:114–25.
Labouvie-Vief G, De-Voe M, Bulka D. Speaking about feelings: conceptions of emotion
across the life span. In: Powell LM, Salthouse TA, editors. Essential papers on the
psychology of aging. New York: New York University Press; 1998. p. 481–510.
Larsen JT, McGraw AP, Cacioppo JT. Can people feel happy and sad at the same time?
Journal of Personality and Social psychology 2001;81:684–96.
Lassiter GD, Koenig LJ, Apple KJ. Mood and behavior perception: dysphoria can increase
and decrease effortful processing of information. Personality and Social Psychology
Bulletin 1996;22:794–810.
Lutz KA, Lutz RJ. Effects of interactive imagery on learning: applications to advertising.
Journal of Applied Psychology 1977;62:493–8.
MacInnis DJ, Park CW. The differential role of characteristics of music on high and low
involvement consumers' processing of ads. Journal of Consumer Research 1991;18:
161–73.
Mackie DM, Worth LT. Processing deficits and the mediation of positive affect in persuasion. Journal of Personality and Social Psychology 1989;57:27–40.
Matmur M, Chattopadhayay A. The impact of moods generated by television programs
on responses to advertising. Psychology and Marketing 1991;8:59–77.
Meyers-Levy J, Tybout AM. Schema congruity as a basis for product evaluation. Journal
of Consumer Research 1989;16:39–54.
North AC, MacKenzie LC, Law RM, Hargreaves DJ. The effects of musical and voice fit on
responses to advertisements. Journal of Applied Social Psychology 2004;34:1675–708.
Oakes S. Examining the relationship between background musical tempo and perceived
duration using different versions of a radio ad. European Advances in Consumer
Research 1999;4:40–4.
Oakes S. Musical tempo and waiting perceptions. Psychology and Marketing 2003;20:
685–705.
Ornstein RE. On the experience of time. New York: Penguin Books; 1969.
Priester JR, Petty RE. The gradual threshold model of ambivalence: relating the positive
and negative bases of attitudes to subjective ambivalence. Journal of Personality
and Social Psychology 1996;71:431–49.
Richins ML. Measuring emotions in the consumption experience? Journal of Consumer
Research 1997;24:127–46.
Russell JA. A circumflex model of affect. Journal of Personality and Social Psychology
1980;39:1161–78.
Russell JA, Carroll JM. On the bipolarity of positive and negative affect. Psychological
Bulletin 1999;125:3-30.
Schmitt B, Tavassoli NT, Milliard RT. Memory for print ads: understanding the relations
among brand name, copy, and picture. Journal of Consumer Psychology 1993;2:55–81.
Schwarz N, Bless B. Happy and mindless, but sad and smart? The impact of affective
states on analytic reasoning. In: Forgas JP, editor. Emotion and social judgments.
London: Pergamon; 1991. p. 55–71.
Schwarz N, Clore GL. Feelings and phenomenal experiences. In: Higgins ET, Kruglanski
A, editors. Social psychology: handbook of basic principles. NewYork: Guilford;
1996. p. 433–65.
Singh SN, Rothschild ML, Churchill GA. Recognition versus recall as measures of television commercial forgetting. Journal of Marketing Research 1988;25:72–80.
Smith PC, Curnow R. Arousal hypothesis and the effects of music on purchase behavior.
Journal of Applied Psychology 1966;50:255–6.
Srull TK, Lichtenstein M, Rothbart M. Associative storage and retrieval processes in person memory. Journal of Experimental Psychology: Learning, Memory, and Cognition 1985;11:316–45.
Sucala M, Scheckner B, David D. Psychological time: interval length judgments and
subjective passage of time judgments. Current Psychology Letters 2010;26:2–9.
Thompson MM, Zanna MP, Griffin DW. Let's not be indifferent about attitudinal ambivalence. In: Petty RE, Krosnick JA, editors. Attitude strength: antecedents and consequences. Mahwah, NJ: Erlbaum; 1995. p. 361–86.
Unnava HR, Burnkrant RE. An imagery processing view of the role of pictures in print
advertisements. Journal of Marketing Research 1991;28:226–31.
Ursavas B, Hesapci-Sanaktekin O. Theoretical overview of temporal perception: Prospective versus retrospective estimates. Proceedings of 11th International Marketing
Trends Congress 19–22 January, Italy; 2012.
Vohs KD, Schmeichel BJ. Self-regulation and the extended now: controlling the self alters the subjective experience of time. Journal of Personality and Social Psychology
2003;85:217–30.
Webster DM, Kruglanski AW. Individual differences in need for cognitive closure. Journal of Personality and Social Psychology 1994;67:1049–62.
Williams P, Aaker JL. Can mixed emotions peacefully co-exist? Journal of Consumer Research 2002;28:636–49.
Zakay D. The evasive art of subjective time measurement: some methodological dilemmas. In: Block RA, editor. Cognitive models of psychological time. Hillsdale,
New Jersey: Erlbaum; 1990. p. 59–84.
Zakay D, Block RA. An attentional-gate model of prospective time perception. In:
Richelle M, De Keyser V, D'Ydewalle G, Van Dierendonck A, editors. Time and the
dynamic control of behavior. Liège, Belgium: Universite de Liege; 1994. p. 167–78.
Zakay D, Block RA. The role of attention in time estimation processes. Advances in Psychology 1996;115:143–64.
Zakay D, Block RA. Temporal cognition. Current Directions in Psychological Science
1997;6:12–6.
Zhu R, Meyers-Levy J. Distinguishing between the meanings of music: when background
music affects product perceptions. Journal of Marketing Research 2005;42:333–45.