JBR-08407; No of Pages 8

Journal of Business Research xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Journal of Business Research

Does the presence of a mannequin head change shopping behavior?☆

Annika Lindström b,c,⁎, Hanna Berg a, Jens Nordfält c, Anne L. Roggeveen d, Dhruv Grewal d
a
Stockholm School of Economics, Stockholm, Sweden
b
Stockholm University, Department of Psychology, Stockholm, Sweden
c
Stockholm School of Economics, Center for Retailing, Stockholm, Sweden
d
Babson College, Babson Park, MA 02457, USA

a r t i c l e i n f o a b s t r a c t

Article history: Mannequins are ubiquitous; this research investigates a specific element of mannequin style, namely, the pres-
Received 22 December 2014 ence or absence of a humanized head. Study 1 demonstrates that in physical stores, the presence of a humanized
Received in revised form 27 April 2015 head enhances purchase intentions for the merchandise displayed on that mannequin. However, in online stores,
Accepted 28 April 2015
mannequin styles with and without humanized heads are equally effective. Study 2 confirms the physical store
Available online xxxx
results among customers with less fashion knowledge (novices), but among customers with more fashion
Keywords:
knowledge (experts), the results reverse, such that mannequins without humanized heads enhance purchase
Retail atmospherics intentions. Further, accessories are more likely to be viewed by experts when the mannequin is headless.
Mannequin These results are based on experiments whose dependent measures included both survey and eye-tracking data.
Retail displays © 2015 Elsevier Inc. All rights reserved.
In-store displays

1. Introduction
Store mannequins are a very important element of visual merchandis-
ing, especially by clothing retailers that use them to display clothes and
accessories. In fact, of one hundred and fourteen clothing and accessory
retailers in a large suburban mall, eighty five percent of the window dis-
plays examined included mannequins. Mannequins enable consumers to
visualize the merchandise in a three-dimensional space, providing infor-
mation about the fit and look of the merchandise on a physical body
(e.g., Oh & Petrie, 2012; Sen, Block, & Chandran, 2002). Yet despite this
ubiquity, inquiry into the effects of mannequins is sparse. Related re-
search provides mostly qualitative insights which are based on store dis-
plays in their entirety, rather than focusing specifically on the effects of
mannequins (e.g., Fiore, Yah, & Yoh, 2000; Kerfoot, Davies, & Ward,
2003; Law, Wong, & Yip, 2012; Oh & Petrie, 2012). The results of such
broader store display research highlight the importance of mannequins
and the need for further investigations of their role.
Mannequins vary in their shape, color, and features across retailers
(Schneider, 1997). Of the ninety seven stores in the suburban mall
that displayed mannequins, seventy-eight displayed full sized manne-
quins, thirteen displayed half sized mannequins (on the top or bottom),
while eleven (jewelry stores) showed only the neck (the sum is greater
than ninety-seven because some stores displayed multiple types of
☆ This research was funded by the Stockholm School of Economics, Center for Retailing.
⁎ Corresponding author at: Babson College, 215 Malloy Hall, Babson Park, MA 02457.
Tel.: 781 239 4289.
E-mail addresses: Annika.Lindstrom@innventia.com (A. Lindström),
Hanna.Berg@hhs.se (H. Berg), Jens.Nordfalt@hhs.se (J. Nordfält), aroggeveen@babson.edu
(A.L. Roggeveen), dgrewal@babson.edu (D. Grewal).
mannequins). One of the most salient variations between the manne-
quins was whether the mannequins featured a head or were headless.
Thirty-two percent of the retailers in the suburban mall displayed manne-
quins with heads, while sixty-eight percent of the mannequins were
headless. The type of mannequin used (head, headless or both) was not
affected by whether the store was moderate or high end in terms of fash-
ion. Headless mannequins tend to present an anthropomorphic human
form, ending in a flat surface where the head normally would be. People's
attention is naturally attracted to heads, as head contains as faces (eyes,
nose, and mouth) which provide critical nonverbal cues (Palermo &
Rhodes, 2007). Thus, investigating the impact of the lack of a head is
both theoretically and managerially interesting.
This article seeks to determine how the presence or absence of a
head on a mannequin influences consumer purchase intentions for
displayed merchandise, as well as the potential moderating role of the
location of the mannequin, that is, on an online retail website or in a
physical retail store. In addition, the present study considers how con-
sumers' fashion knowledge (expert versus novice) might moderate
the impact of the presence of a mannequin head and explains the ef-
fects. Finally, eye-tracking measures provide detailed insights into the
processing that takes place when consumers observe mannequins.
Two empirical studies serve to test the fundamental propositions.
These are experimentally tested collecting measures from both surveys
and eye-tracking. In physical stores, purchase intentions are higher for
merchandise displayed on mannequins with heads versus those with-
out heads. However, no differences arise across headed or headless
mannequins for merchandise displayed through online stores. People's
ability to envision themselves wearing the clothing is higher among
novice consumers if the mannequin has a head, which results in higher

http://dx.doi.org/10.1016/j.jbusres.2015.04.011
0148-2963/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: Lindström, A., et al., Does the presence of a mannequin head change shopping behavior?, Journal of Business Research
(2015), http://dx.doi.org/10.1016/j.jbusres.2015.04.011
2 A. Lindström et al. / Journal of Business Research xxx (2015) xxx–xxx
purchase intentions. For expert consumers, the reverse holds: The abil-
ity to envision wearing the clothing and their purchase intentions are
higher when the mannequin does not have a head. Further, accessories
are more likely to be viewed by experts when the mannequin is
headless.
2. Theoretical background
Mannequins have evolved, from simple dressmakers' forms to
poseable wax dolls and painted plaster (Schneider, 1997). Modern re-
tailers even feature mannequins that can serve as electronic observers,
using embedded video technology (Grewal & Levy, 2015). However, the
majority of mannequins in major retail chains are human-like forms, ei-
ther with or without a head. If they have heads, the facial features vary
from abstract to realistic, and the shape and color of the mannequin
also vary from abstract (e.g., silver) to realistic (e.g., skin tone). While pre-
vious research has used mannequins in the investigation of window dis-
plays and visual merchandising (see Table 1), none have specifically
focused on aspects of a mannequin that can impact consumer responses.
The central focus for this study to experimentally manipulate is
the presence or absence of a head on a mannequin to determine how
that impacts purchase intentions. Mannequin heads can contain faces
which are generally realistic (i.e., eyes, hair). Faces attract people's at-
tention more easily than virtually any other visual stimuli and provide
a wealth of information (Palermo & Rhodes, 2007). In their eye-
tracking studies, Hendrickson and Ailawadi (2014) report that cereal
boxes with cartoon figures and faces attract more visual attention in a
grocery store environment than boxes without any visible faces. Follow-
ing this logic, a headed mannequin should draw more visual attention
than a headless one, which in turn should lead consumers to pay more at-
tention to the merchandise displayed on that mannequin. Accordingly,
purchase intentions may be higher for merchandise displayed on these
headed mannequins. In contrast, a mannequin without a head is less like-
ly to garner attention, and the merchandise thus displayed may be less
likely to be noticed.
By noticing the merchandise, it is more likely that the consumer may
mentally simulate the experience of wearing the outfit. Work on mental
simulation (Elder & Krishna, 2012) can be used to predict how cus-
tomers will behave when exposed to sensory-rich cues, such as a man-
nequin. Specifically, mental simulation suggests that customers who are
exposed to these sensory-rich mannequins are more likely to mentally
simulate the experience of wearing the merchandise displayed. As a
function they will better able to envision themselves in the outfit and
thus be more likely to purchase the outfit.
Table 1
Past research which has examined mannequins.
Author Type of Variables of interest
research
Fiore et al. Experiment How graphical drawing of a display and fragrance impact attitude,
(2000) purchase intention, and price.
Sen et al. Survey How store and product category information conveyed by store's
(2002) window related to shopping decisions, and how relationships are
impacted by knowledge of retailer's product(s).
Kerfoot et al. Semi-structured Effects of visual merchandising on purchase behavior and brand
(2003) interviews recognition
Khakimdjanova Content Visual merchandising among e-tailers.
and Park analysis
(2005)
Law et al. Focus group Consumer affective response to visual merchandising
(2012)
Oh and Petrie Experiments How store entry decisions are impacted by store window display
(2012) type (artistic/merchandise focused), shopping motive
(purchase/recreational), and cognitive load (low/high)
Please cite this article as: Lindström, A., et al., Does the presence of a mannequin head change shopping behavior?, Journal of Business Research
(2015), http://dx.doi.org/10.1016/j.jbusres.2015.04.011
2.1. Shopping environment: offline vs. online
Store displays, including mannequins, constitute an important part
of offline visual merchandising; visual merchandising in turn plays a
significant role in creating retail atmospherics, which influence con-
sumers' affective, cognitive, and behavioral responses to store environ-
ments (Grewal, Roggeveen, Puccinelli, & Spence, 2014). Building on
research into the role of nonverbal cues (Spence, Puccinelli, Grewal, &
Roggeveen, 2014) and visual merchandising (Nordfält, Grewal,
Roggeveen, & Hill, 2014), visual merchandising that captures attention
should enhance purchase intentions and sales. Therefore, this study focus-
es on how the head of a mannequin affects consumers' attention and be-
havioral intentions. In physical stores, the presence of a head should
engender positive reactions, more attention, and higher purchase inten-
tions. The effects in an online environment are less evident, because the
retailing context (physical versus online) can strongly moderate consum-
er responses (e.g., Harris, Grewal, Mohr, & Bernhardt, 2006).
Only 18% of major U.S. online clothing retailers create website dis-
plays of their clothing using photos of mannequins, 60% display clothing
in photos with human models, and 31% show the clothing lying on a flat
surface (Khakimdjanova & Park, 2005). The limited use of photographs
or videos of mannequins in online settings is likely a function of the ease
of obtaining pictures of live models or the fact that only some online re-
tailers also maintain physical stores, such that they would have ready
access to mannequins. Additionally, the vast number of stock keeping
units carried online, makes it time consuming to display all the mer-
chandise on mannequins.
With what they display and how, online stores create an atmosphere
that likely influences shoppers' cognitive and emotional reactions
(Eroglu, Machleit, & Davis, 2003; Manganari, Siomkos, & Vrechopoulos,
2009). Because online shoppers cannot touch and feel the physical
merchandise, e-tailers often provide more verbal and visual infor-
mation about their products. For example, a consumer shopping on-
line at H&M.com for a blouse can zoom in to see the shirt up close,
read a description of the product, and assess which colors and
sizes are available. All this information should encourage a more
cognitive mindset, in which case any impact of a mannequin's face,
in terms of drawing attention to the product, may diminish in an on-
line setting. As such we hypothesize a mannequin style by shopping
location interaction.
H1. A mannequin style × shopping location interaction effect causes a
headed (versus headless) mannequin style to enhance purchase inten-
tions in a physical (off-line) store but has no effect in an online store.
Findings related to mannequins
Display drawing (which included a mannequin) and fragrance
resulted in higher purchase intention and willingness to pay.
Consumer envisioning facilitated by store displays. Mannequins
were included as part of store displays.
Consumers with medium levels of clothing knowledge are more
influenced by window displays in their shopping decision than
those with low or high knowledge.
Mannequins are part of the window displays
Generally positive responses to mannequins.
Mannequins are very visual and help consumers visualize how the
clothing will look on.
Mannequins are used frequently in online retailing/online visual
merchandising.
Headless mannequins allow consumers to imagine themselves in
displays and garments.
Store displays impact store entry decision, but its influence is
moderated by display type, motivation, and cognitive load.
Mannequins were included as part of the window displays.
 A. Lindström et al. / Journal of Business Research xxx (2015) xxx–xxx
2.2. Consumer decision stage
Customers go through a number of decision stages as they move from
need identification to purchase (Levy, Weitz, & Grewal, 2014; Puccinelli
et al., 2009). Two of the decision stages that retailers devote considerable
attention and resources to understanding are browsing and shopping. Re-
tailers also recognize that customers often use multiple channels
(e.g., online vs. in-store) to assist with these decision stages and there is
need to understand the joint roles of decision stage and channel type
(Neslin et al., 2006). Given that retailers use mannequins to attract cus-
tomer attention, it is important to understand whether the shopping
goal qualifies the prior interaction or alternatively the role of mannequin
style by channel generalizes across the two goals.
2.3. Fashion expertise
In addition to the retailing context, consumer responses to products
displayed on different mannequin styles likely depend on the level of
knowledge that consumers have about clothing and fashion. Some cus-
tomers, such as market mavens (Clark & Goldsmith, 2005; Feick & Price,
1987), tend to be highly knowledgeable about a variety of products.
Prior research indicates the importance of category knowledge, and re-
search on both knowledge and categorization highlights the different
information processing undertaken by expert and novice consumers
(Hutchinson & Eisenstein, 2008; Sujan, 1985). Expert consumers pos-
sess relatively high levels of category knowledge; novice consumers
have relatively low levels of category knowledge (Alba & Hutchinson,
1987).
Expert consumers have more advanced cognitive structures and bet-
ter memory for product information than novice consumers. They also
have a lower threshold for elaboration in their area of expertise, where-
as novices have less advanced cognitive structures and are less prone to
elaborate on information related to the category. Experts rely less on
heuristics in their decision making and tend to be not to be as influenced
by irrelevant contextual cues compared with novices (Hutchinson &
Eisenstein, 2008). For the clothing product category, category knowl-
edge entails fashion expertise, defined in terms of consumers' level of
knowledge about clothing and fashion (Sen et al., 2002).
In turn, expert and novice consumers likely experience different in-
fluences from contextual cues in store environments, such as manne-
quin styles. To expert consumers, who can discern important from
irrelevant information, the mannequin heads are irrelevant and more
likely to detract from their ability to envision themselves wearing that
particular merchandise. This is consistent with the findings of Law
et al. (2012) who interviewed sixty-four women in eight focus groups
regarding about visual merchandising of intimate apparel retailers.
The authors of that research identified women in a given demo-
graphic that “had a relatively high propensity to spend on luxury or
beauty-related products” (Law et al., 2012, p. 117). They then went on
to describe that “participants for the first two interviews were recruited
outside a lingerie select shop. For the other subjects, they were referred
by the Hong Kong office of an international lingerie retailer” (Law et al.,
2012, p. 117).
Consequently, we infer that the women who participated in these
focus groups appear to have greater fashion expertise. For these
women, mannequin heads appear to distract their visual processing
and ability to envision wearing the displayed merchandise. In fact,
Law et al. (2012) concluded that the headless mannequin (versus head-
ed mannequins) allowed customer to visualize themselves in the cloth-
ing. In this context, envisioning means that consumers can imagine
themselves in the clothing as they view the mannequins, such that
they rely on the mannequins to visualize how the displayed clothes
would look on them and infer the degree of product fit (Sen et al.,
2002). The concept of envisioning is akin to mental simulation, de-
scribed by Elder and Krishna (2012), such that the mannequin enhances
consumers' imagination of wearing the outfit.
Please cite this article as: Lindström, A., et al., Does the presence of a mannequin head change shopping behavior?, Journal of Business Research
(2015), http://dx.doi.org/10.1016/j.jbusres.2015.04.011
3
Kerfoot et al. (2003) note that mannequins are very visual and thus
help consumers visualize how clothing would look on them. Some con-
sumers also report negative reactions to very abstract, headless manne-
quins. Because expert consumers possess a higher elaboration likelihood
though, and their cognitive processing relies more on automaticity
while they take in the details of a visual scene (Alba & Hutchinson,
1987), a headless mannequin may help them pay greater attention to
the merchandise (clothing and accessories) on that mannequin. That
is, experts may display a stronger ability to envision themselves wearing
the merchandise and greater purchase intentions for merchandise
displayed on headless mannequins than on mannequins with heads. Nov-
ices instead have less developed cognitive categorization structures, and
mannequin heads provide them with contextual information, increasing
their ability to envision themselves wearing the ensemble. In this case,
the headed mannequin could increase these consumers' envisioning abil-
ity and purchase intentions.
H2. A mannequin style × fashion knowledge interactive effect on
envisioning and purchase intentions causes a headed mannequin to re-
sult in (a) higher abilities to envision wearing the merchandise and
greater purchase intentions among consumers with less fashion knowl-
edge (b) but lower abilities to envision wearing the merchandise and
lower purchase intentions among consumers with more fashion knowl-
edge, compared with a mannequin without a head.
Two studies tested these hypotheses. Study 1 presents the results of
an experiment that examined the role of mannequin style, its appear-
ance in offline versus online contexts (H1), and the shopping goal
of the consumer. Study 2 explores the moderating effects of fashion
knowledge on the relationship between mannequin style and purchase
intentions in an offline context (H2).
3. Study 1: does the mannequin style matter?
This study investigates if and how mannequin style (headed or head-
less) affects consumers' purchase intentions toward items displayed on
the mannequin. It also explores whether the location of the mannequin
(in-store vs. online) moderates the impact of mannequin style on pur-
chase intentions toward the outfit displayed on the mannequin. Finally,
this study provides some indication of whether the results generalize
across shopping goals of consumers (browsing vs. shopping).
3.1. Method
A 2 (mannequin: head/headless) × 2 (location: in-store/online) × 2
(purpose: browse/shop) between-subjects experiment included 252 fe-
male participants, obtained from a sample, who received a nominal fee
for their participation. In the browsing online (in-store) condition, par-
ticipants were asked to “imagine you have decided to browse through
the [named] website ([named] store). You are not interested in buying
anything, rather just in browsing.” In the shopping online (in-store)
condition, they imagined “you want to buy a new dress. As a result
you have gone to a [named] website (store) to buy a dress.” The scenario
then described the shopping experience (Appendix A) and the outfit
displayed on a mannequin. A picture of the mannequin appeared next.
The mannequin was identical across conditions, with the exception of
whether the picture showed its head. After viewing the stimulus, partic-
ipants completed their likelihood to purchase the outfit (1 = “very un-
likely,” 7 = “very likely”) and whether they would buy it (yes/no). They
also completed manipulation check questions.
3.2. Results
3.2.1. Manipulation checks
All manipulations checks worked as expected. Participants accurate-
ly recalled whether the mannequin had a head (χ2(df = 1) = 211.56,
4 A. Lindström et al. / Journal of Business Research xxx (2015) xxx–xxx
p b .001), whether they were browsing or shopping for a dress
(χ2(df = 1) = 175.00, p b .001), and the shopping location (χ2(df = 1) =
71.91, p b .001).
3.2.2. Results
An analysis of variance (ANOVA) on purchase likelihood showed an
interaction between the shopping location and mannequin style
(F(1,243) = 5.53, p b .05; Fig. 1). The other effects were not significant.
The basic results thus generalize across shopping goals (browsing vs.
shopping). Further decomposing the shopping location × mannequin
style interaction reveals that in stores, the headed mannequin resulted
in higher purchase intentions (Mhead = 2.84, Mheadless = 2.23,
F(1,243) = 4.28, p b .05, r = .13), whereas online, no significant difference
emerged in purchase intentions between the two mannequin styles
(Mhead = 2.66, Mheadless = 3.03, p N .2), in support of H1. Similarly, a bi-
nary logistic regression on the “would you buy the outfit?” item re-
vealed a significant mannequin style × location interaction (Wald =
4.01, p b .05). In stores, the headed mannequin prompted more people
to indicate that they would purchase (Mhead = 33%, Mheadless = 19%,
χ2(df = 1) = 3.31, p b .10). Among participants in the online condition,
the percentage who would purchase did not vary significantly as a func-
tion of mannequin style (Mhead = 26%, Mheadless = 33%, p N .3).
3.3. Discussion
Mannequins are used by the vast majority of clothing retailers to dis-
play merchandise. Our exploratory research at a major shopping mall
highlighted that 85% of retailers used mannequins in their displays.
While 32% of those retailers display mannequins that have heads, 68%
displayed mannequins that did not have a head. The use of mannequins
is less prevalent online where pictures of models wearing the merchan-
dise are used to display clothing items.
We hypothesized that headed mannequins allow customers to bet-
ter envision themselves wearing the merchandise, and as a result the
headed mannequins are likely to enhance purchase intentions relative
to the headless mannequins. However, we expected these effect only
in physical stores and not online where customers have considerable
more search and exploration options.
Likelihood to Buy
headless head
3.03 2.84
2.66
2.23
online in-store
Would you buy the outfit
headless head
33% 33%
26%
19%
online in-store
Fig. 1. Study 1 results.
Please cite this article as: Lindström, A., et al., Does the presence of a mannequin head change shopping behavior?, Journal of Business Research
(2015), http://dx.doi.org/10.1016/j.jbusres.2015.04.011
The results of our first study validate these expectations. Mannequin
style does matter, but only in our physical store conditions. In the store,
the headed mannequins increased purchase intentions for the outfit
displayed. The mannequin style had no effect on consumers' likelihood
to buy or purchase intentions in the online condition.
Customers go through multiple stages as they shop for merchandise;
two of which are browsing through the merchandise and actually shop-
ping for items. We explored whether the decision stage further qualified
our two-way interaction or whether the results would generalize across
the two key customer decision stages. The lack of the significance of the
three-way interaction confirms these expectations.
4. Study 2: moderating role of knowledge
Considering the results of Study 1, Study 2 was a laboratory study fo-
cused on in-store, real-world mannequins that combined eye-tracking
methods with a self-reported questionnaire. The purpose was to ex-
plore whether the impact of mannequin style on purchase intentions
depends on the level of consumers' fashion knowledge, that is, novice
versus expert. This study also notes the mediating role of consumer
envisioning and how mannequin style affects consumers' visual pro-
cessing of the mannequins.
Capturing the visual attention of consumers is vital for retailing
and visual merchandising; recent research often uses eye-tracking
methods to test this capture (e.g., Atalay, Bodur, & Rasolofoarison, 2012;
Hendrickson & Ailawadi, 2014; Wästlund, Otterbring, Gustafsson, &
Shams, 2015). Eye-tracking technology enables researchers to quantify
the visual attention that consumers direct at stimuli and provides insights
into their information processing and decision-making processes (Wedel
& Pieters, 2008).
4.1. Method
A 2 (mannequin: head/headless) × 2 (fashion knowledge: high/low)
between-subjects experiment included 79 female participants. To ma-
nipulate the mannequin style, this study included two mannequins:
one with a head and one without. The mannequin with a head had
painted facial features, a wig, and attached eyelashes. Both mannequins
to fit a standard women's dress size XS/S (extra small to small), had pre-
viously been used in store displays, and were typical to the clothing in-
dustry. They were outfitted with a loose-fitting, tunic-style dress in a
beige color, with shoes and accessories added to create a typical store
display (i.e., black opaque tights, a pair of black high-heeled shoes, a sil-
ver colored bracelet, and a black clutch bag). The dress and accessories
selections reflected the advice of a visual merchandiser, working for a
leading fashion manufacturer and clothing retailer. The merchandise
and mannequins were generously donated by this retailer.
4.1.1. Participants
Seventy-nine women, who did not wear eye glasses and were
between 16 and 54 years of age (mean = 24.7 years) participated. No
significant difference in ages distinguished the treatment groups. After
the experiment, participants received a lottery ticket (approximate
value $1.50) in return for participating in the study.
4.1.2. Procedure
Participants were randomly assigned to the two experiment groups.
Each participant viewed one of the mannequins. The data collection fo-
cused on each participant, one at a time. Before entering the experiment
room to view a mannequin, participants put on eye-tracking glasses
(Tobii glasses). These monocular (right-eye) eye trackers record the co-
ordinates of each participant's fovea at a sampling frequency of 30 Hz.
Participants then entered a room with a black line on the floor, which
they were to keep their eyes on as they entered, following the line
and not looking up until they reached the end. When they looked up,
the mannequin was in front of them.
 A. Lindström et al. / Journal of Business Research xxx (2015) xxx–xxx 5

Valid eye-tracking data came from 70 participants; 9 participants had Purchase Intention
to be excluded from the study due to calibration problems. The analysis of No head Head
the eye-tracking recordings (compiled from cameras in the apparatus, si-
multaneously filming the viewed scene and participants' eyes) from each 4.91
4.58
respondent relied on the software provided by Tobii Studio. Infrared light 3.89
3.43
(IR) markers mounted on the wall slightly behind and around the man-
nequin (see Appendix B) enabled the clear definition of areas of interest
(AOI), such that it was possible to aggregate the eye-tracking data from
all the participants. The mannequins were first divided into eight AOI
each: head area (AOI 1), dress area (AOI 2–AOI 6), accessories area (AOI
7), and leg area (AOI 8). For an illustration of the AOI, see Appendix B. Novice Expert
After viewing the mannequin, participants completed a brief survey
indicating their purchase intentions (1 = “definitely will not buy,” Envisioning
10 = “definitely will buy”); two questions adapted from Sen et al.
(2002) to measure their ability to envision wearing the merchandise No head Head
(“I can imagine how I would look in the dress;” “I can envision myself
wearing the dress”); and their fashion knowledge, using three items 5.82 6.21
from Sen et al.(2002): “I consider myself an expert on fashion,” “I 4.74 4.89
know more than most people about trends in fashion,” and “I am very
knowledgeable about the latest styles and fashions.” A median split on
fashion knowledge created the expert and novice groups.
The viewing time for each AOI was measured as the eye-tracking-
based total visit duration, in seconds. The eye-tracking measurement Novice Expert
share of total viewing time was also calculated for all AOI, equal to the
total visit duration for that AOI, divided by the total visit duration for
the entire mannequin (all eight AOI).
Sum of Duration of Each
4.2. Results Visit for the Bag Area
4.2.1. Manipulation check No head Head
The eye-tracking results revealed that those exposed to a headed
4.88
mannequin looked at the head area; those exposed to a headless man-
2.95 3.46 3.07
nequin tended not to look at that area. These trends applied to both nov-
ices and experts (experts: 94% looked at the head area of the headed
mannequin, 33% looked at the head area of the headless mannequin,
χ2(df = 1) = 14.53, p b .001; novices: 88% looked at the head area of
the headed mannequin, 20% looked at the head area of the headless Novice Expert
mannequin, χ2(df = 1) = 14.24, p b .001). That is, participants in the
headed mannequin condition noticed the mannequin's head, while Fig. 2. Study 2 results.
the majority in the headless condition did not look at the head area.
An ANOVA of the eye-tracking results revealed a significant main ef-
fect of mannequin style on the share of total viewing duration (head
AOI/total viewing duration), such that participants examined the head outfit when the mannequin had a head (Mhead = 5.82, Mno head = 4.74;
AOI more when the mannequin had a head (F(1,65) = 93.87, p b .01). F(1, 75) = 2.67, p = .05, one-tailed), whereas experts had an easier time
The main effect of fashion knowledge was not significant (F(1,65) = envisioning it if they saw the headless mannequin (Mhead = 4.89, Mno
.17, p N .05), nor was the interaction between mannequin style and fash- head = 6.21; F(1, 75) = 3.94, p b .05, one-tailed).

ion knowledge (F(1,65) = .02, p N .05). Thus, mannequin faces attracted

visual attention when present, but the head area did not attract atten-
tion when the mannequin lacked a head.
4.2.2. Purchase intentions
The ANOVA for purchase intentions revealed an interaction between
category knowledge and mannequin style (F(1,75) = 4.95, p b .05, see
Fig. 2). For novices, the headed mannequin induced higher purchase in-
tentions (Mhead = 4.58, Mno head = 3.43; F(1, 75) = 2.83, p b .05, one-tailed,
r = .19). Among experts, the headless mannequin led to directionally
higher purchase intentions (Mhead = 3.89, Mno head = 4.91; F(1, 75) =
2.15, p b .10, one-tailed, r = .17). The marginal significance of these results
is a function of the small sample size, as can be seen from the effect size.
4.2.3. Envisioning
A high correlation arose between the two items measuring
envisioning (r = .61, p b .001), so they can be averaged to form an
envisioning scale. The ANOVA of this envisioning scale revealed an inter-
action between category knowledge and mannequin style (F(1,75) = 6.56,
p b .05; Fig. 2). Novices had an easier time envisioning themselves in the
4.2.4. Moderated mediation analysis
Finally, this study analysis tested the prediction that envisioning wear-
ing the product would mediate the effects of the mannequin
style × fashion knowledge interaction. A moderated-mediation analysis
implemented the bootstrap PROCESS macro-method suggested by
Hayes (2012), which has proven superior to traditional multiple regres-
sion techniques (Baron & Kenny, 1986; Zhao, Lynch, & Chen, 2010).
PROCESS Model 8 estimated the conditional (i.e., moderated by fashion
knowledge) indirect effects of the independent variable (i.e., mannequin
style) on the dependent variable (i.e., purchase intention) through the
proposed mediator (i.e., ability to envision), as well as the conditional di-
rect effect of that independent variable on the dependent variable. The re-
sults showed that the overall envisioning outcome model was significant
(F(3,75) = 2.39, p b .10), as was the purchase intention “outcome” model
(F(4,74) = 14.57, p b .01). The moderated mediation results indicated
that the effects of mannequin style were mediated by the ability to envi-
sion the outfit among people in the low fashion knowledge group (90%
confidence interval [−1.62, −.07]) and among those in the high fashion
knowledge group (90% confidence interval [.22, 1.75]).

Please cite this article as: Lindström, A., et al., Does the presence of a mannequin head change shopping behavior?, Journal of Business Research
(2015), http://dx.doi.org/10.1016/j.jbusres.2015.04.011
6 A. Lindström et al. / Journal of Business Research xxx (2015) xxx–xxx
4.2.5. Eye-tracking analysis
The eye-tracking results also revealed that many participants who
were exposed to a headed mannequin spent the largest share of their
total viewing durations on the head area, among both experts (headed:
41% viewed the head area most, 53% dress area most, 6% accessory area
most, and 0% leg area most; headless: 0% viewed the head area most,
71% dress area most, 24% accessory area most, and 5% leg area most;
χ3(df = 3) = 11.88, p b .01) and novices (headed: 38% viewed the head
area most, 50% dress area most, 6% accessory area most, and 6% leg
area most; headless: 0% viewed the head area most, 87% dress area
most, 13% accessory area most, and 0% leg area most; χ2(df = 3) =
8.50, p b .05).
When the mannequin did not have a head, a larger share of partici-
pants' visual attention addressed the focal product. An ANOVA shows a
significant main effect of mannequin style on the share of total viewing
time (dress AOI/total viewing time) of the mannequins (F(1,65) = 11.96,
p b .01). Neither fashion knowledge (F(1,65) = .64, p N .05) nor the inter-
action between mannequin style and fashion knowledge (F(1,65) = .42,
p N .05) had significant effects.
The eye-tracking results also provide further evidence that the role
of mannequin style on attention paid to the merchandise depends on
the customer's level of fashion knowledge. The eye-tracking results re-
vealed a significant interaction effect between the mannequin style
and fashion knowledge on the amount of time people examined the ac-
cessories on a mannequin (e.g., handbag; F(1,64) = 4.03, p b .05). Experts
exposed to the headless mannequin spent more time examining the
handbag area (Mhead = 3.07, Mheadless = 4.88; F(1,64) = 5.35, p b .05).
For novices, no difference appeared in how long they examined the ac-
cessory, regardless of whether the mannequin had a head (Mhead =
3.46, Mheadless = 2.95; p N .5).
4.3. Discussion
The Study 2 results showed the expected interaction between fashion
knowledge and mannequin style. For novice consumers, as demonstrated
in Study 1, the mannequin with a head invoked higher purchase inten-
tions. For expert consumers, purchase intentions were stronger for the
dress displayed on a headless mannequin. This difference likely reflects
the consumers' varying levels of elaboration, such that they envisioned
themselves in the clothing. The results regarding the interaction between
fashion knowledge and mannequin style were similar for both
envisioning and purchase intentions: Expert consumers envisioned
themselves in the clothing more when the mannequin did not have a
head, but novice consumers did so when the mannequin had a head. Fur-
thermore, envisioning contributed to the effects of mannequin style and
fashion knowledge on purchase intentions. A moderated mediation anal-
ysis confirmed that the effects of mannequin style on purchase intentions
were moderated by fashion knowledge and mediated by consumer
envisioning.
Eye-tracking results also revealed that consumers exposed to a
headed mannequin looked at the head area, whereas consumers
exposed to a headless mannequin tended not to look at that area.
What differed between the mannequins was the amount of attention
that experts spent looking at the accessory items (handbag and brace-
let). The interaction between mannequin style and fashion knowledge
influenced the amount of visual attention paid to the accessory items.
Novices showed no difference between mannequin types, whereas
experts looked at the accessory items longer when the mannequin did
not have a head.
5. General discussion
By addressing the largely unexplored area of how mannequins influ-
ence consumer purchase intentions, this research supports the prediction
that heads increase attention to the mannequin and increase people's
purchase intentions for merchandise displayed on the mannequin in a
Please cite this article as: Lindström, A., et al., Does the presence of a mannequin head change shopping behavior?, Journal of Business Research
(2015), http://dx.doi.org/10.1016/j.jbusres.2015.04.011
physical store. However, the Study 1 results indicate that mannequin
style (head versus headless) does not influence purchase intentions for
online shoppers.
The second study therefore centers only on the physical store and
simulates the experience by having participants view real mannequins
(cf. images of the mannequins in Study 1). The results highlight that
the Study 1 findings apply to novice customers, who have less fashion
knowledge than experts in this study context.
However, as predicted, experts expressed greater purchase inten-
tions and envisioning ability when they viewed a headless mannequin,
rather than a mannequin with a head. The moderated mediation results
support a mediation process of mannequin style on purchase intentions,
through the participants' ability to envision wearing the merchandise.
Finally, this study provides support in the form of eye-tracking
results, demonstrating that accessories are more likely to be viewed
by experts when the mannequin is headless. The use of eye-tracking
to corroborate the survey results provides more support to the underly-
ing process and enhances the contribution as the results are synergistic
across multiple methods and multiple methodological contexts.
The article enhances understanding of the role of mannequins in en-
hancing attention and visual attention and contributes to research into
both store displays stores and visual merchandising. In particular, it pro-
vides theoretical insights into how variations in mannequin style affect
purchase intentions and visual processing of mannequins.
Furthermore, this research contributes to the literature stream
devoted to visual attention in retailing by highlighting the use of eye-
tracking methodology to explore how consumers visually process prod-
ucts displayed on mannequins as opposed to in ads, on packages, or on
displays, as has been studied in grocery contexts (e.g., Atalay et al.,
2012; Hendrickson & Ailawadi, 2014; Wästlund et al., 2015).
In addition to theoretical contributions, this article provides im-
portant insights for retailers regarding which mannequin style they
should use and how the effects of mannequin variations are likely
to vary across consumer segments and retail contexts. In designing
store displays, retailers can benefit from increased knowledge
about how consumers visually inspect mannequins, and how this
process affects their purchase intentions for the displayed products.
Table 2 highlights some of the salient implications of our research for
retailers. First, we highlight the importance of why headed mannequins
enhance purchase intentions in physical stores. They allow the customers
to better envision the merchandise. We also qualify these findings. More
specifically, they help customer with less fashion knowledge. Second, we
demonstrate that the role of mannequin style (headed vs. headless) does
not have an impact when customers are shopping online. Third, through
our eye-tracking results we highlight the importance of displaying acces-
sories on mannequins in addition to the outfit. These results highlight that
experts spent more time on the accessories when they were displayed on
the headless mannequin.
Our results also highlight the importance of multi-method studies.
In Study 2, in addition to conducting an experiment, subjects' responses
were assessed using both surveys and eye-tracking. As mentioned
earlier, the eye-tracking data confirms that heads on mannequins do at-
tract viewing attention. However, they can detract from viewing the
merchandise.
Finally, this article suggests several additional research avenues. The
presence or absence of a head is only one mannequin style variable.
Other research should investigate the role of idealized or aspirational
mannequins versus realistic body types. Some retailers have begun
displaying merchandise on mannequins that are shaped more consis-
tently with the body sizes of average women. This point also brings eth-
ical and public policy concerns to the forefront, in terms of how
mannequins and models might create unrealistic body type and image
expectations. Most fashion-related retailers target younger, image-
conscious consumers; displaying extremely thin mannequins might be
contributing to teen anxiety and eating disorders. Further research
could explore these issues.
 A. Lindström et al. / Journal of Business Research xxx (2015) xxx–xxx 7

Table 2 the store looking at all the available dresses, in order to find one that you
Summary of implications. will buy.
Study details Results Implications for retailers As you are walking through the store you notice an outfit displayed
Study 1: The study In-stores, headed
on a mannequin. Please look at the mannequin and outfit. We will ask
examined the role of mannequins enhance you some questions about it.
mannequin style purchase intentions
(head/headless), relative to the headless
location mannequins. It allows
(in-store/online) and retailers to draw
purpose attention to certain
(browse/shop). merchandise or displays.
However, these effects
do not work for online
retailers where
customers have
considerable more
search and exploration
options.

Study 2: The study Retailers need to design

examined the role of their visual
mannequin style merchandising strategy
(head/headless) and keeping in mind their
fashion knowledge customer expertise
(high/low). level. Our results
highlight that headed
mannequins allow
novice customers to
better envision
themselves wearing the
outfits and accessories.
Thus, headed
mannequins could be
very effective to depict
new styles and
accessories. Whereas as
headless mannequins
are likely to be more
effective for
experts/fashionistas.
Thus, retailers need to
carefully select when to [One group saw the headed mannequin. The other group saw the
use which type of headless mannequin image.]
mannequin.

Experts focused more

attention on accessories
Appendix B. Mannequin, Visual Data Points, and Areas of Interest
displayed on headless
mannequins. Thus,
retailers whose primary
customers have a
reasonable level of
fashion knowledge
would benefit from
displaying accessories
on headless
mannequins.

Baker, Parasuraman, Grewal, and Voss (2002) highlight the impor-

tance of social elements in retail environments. Generally, prior re-
search has focused on how the presence of others in a store affects
outcomes. Customers tend to shop differently when they are alone ver-
sus with others (Grewal, Roggeveen, & Lindsey-Mullikin, 2014); per-
haps the presence of realistic mannequins represents a sort of social
presence that influences how people shop. Because some stores use
very few mannequins, mostly in window displays, whereas others use
many mannequins throughout their stores, it is important to under-
stand the differential effectiveness of each type of strategy, depending
on whether they simulate greater social presence.

Appendix A. Shopping, In-Store, Headed Mannequin Scenario

Imagine you feel that you need a new dress. As a result you have
gone to a [named] store to buy a dress. On this day you are out shopping
on your own. There are not a lot of people in the store. You walk through

Please cite this article as: Lindström, A., et al., Does the presence of a mannequin head change shopping behavior?, Journal of Business Research
(2015), http://dx.doi.org/10.1016/j.jbusres.2015.04.011
8 A. Lindström et al. / Journal of Business Research xxx (2015) xxx–xxx
References
Alba, J.W., & Hutchinson, J.W. (1987). Dimensions of consumer expertise. Journal of
Consumer Research, 13(4), 411–454.
Atalay, A.S., Bodur, H.O., & Rasolofoarison, D. (2012). Shining in the center: Central gaze
cascade effect on product choice. Journal of Consumer Research, 39(4), 848–866.
Baker, J., Parasuraman, A., Grewal, D., & Voss, G. (2002). The influence of multiple store
environment cues on perceived merchandise value and patronage intentions.
Journal of Marketing, 66, 120–141.
Baron, R.M., & Kenny, D.A. (1986). The moderator–mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51(6), 1173–1182.
Clark, R.A., & Goldsmith, R.E. (2005). Market mavens: Psychological influences. Psychology
and Marketing, 22(4), 289–312.
Elder, R.S., & Krishna, A. (2012). The “visual depiction effect” in advertising: Facilitating
embodied mental simulation through product orientation. Journal of Consumer
Research, 38(6), 988–1003.
Eroglu, S.A., Machleit, K.A., & Davis, L.M. (2003). Empirical testing of a model of online
store atmospherics and shopper responses. Psychology and Marketing, 20(2),
139–150.
Feick, L.F., & Price, L.L. (1987). The market maven: A diffuser of marketplace information.
Journal of Marketing, 51(1), 83–97.
Fiore, A.M., Yah, X., & Yoh, E. (2000). Effects of a product display and environmental
fragrancing on approach responses and pleasurable experiences. Psychology and
Marketing, 17(1), 27–54.
Grewal, D., & Levy, M. (2015). Marketing (5th ed.). Burr Ridge, IL: McGraw-Hill Publishing.
Grewal, D., Roggeveen, A.L., & Lindsey-Mullikin, J. (2014a). Contingent effects of semantic
price cues. Journal of Retailing, 90, 198–205.
Grewal, D., Roggeveen, A.L., Puccinelli, N.M., & Spence, C. (2014b). Retail atmospherics
and in-store nonverbal cues: An introduction. Psychology and Marketing, 31(7),
469–471.
Harris, K.E., Grewal, D., Mohr, L.A., & Bernhardt, K.L. (2006). Consumer responses to ser-
vice recovery strategies: The moderating role of online versus offline environment.
Journal of Business Research, 59(4), 425–431.
Hayes, A.F. (2012). PROCESS: A versatile computational tool for observed variable medi-
ation, moderation, and conditional process modeling [white paper]. Retrieved from
http://www.afhayes.com/public/process2012.pdf
Hendrickson, K., & Ailawadi, K.L. (2014). Six lessons for in-store marketing from six years
of mobile eye-tracking research. Shopper Marketing and the Role of In-Store Marketing
(Review of Marketing Research), 11, 57–74.
Hutchinson, W.J., & Eisenstein, E.M. (2008). Consumer learning and expertise. In C.P.
Haugtvedt, P.M. Herr, & F.R. Kardes (Eds.), Handbook of consumer psychology. New
York City, NY: Psychology Press.
Please cite this article as: Lindström, A., et al., Does the presence of a mannequin head change shopping behavior?, Journal of Business Research
(2015), http://dx.doi.org/10.1016/j.jbusres.2015.04.011
Kerfoot, S., Davies, B., & Ward, P. (2003). Visual merchandising and the creation of
discernible retail brands. International Journal of Retail & Distribution Management,
31(3), 143–152.
Khakimdjanova, L., & Park, J. (2005). Online visual merchandising practice of apparel
e-merchants. Journal of Retailing and Consumer Services, 12(5), 307–318.
Law, D., Wong, C., & Yip, J. (2012). How does visual merchandising affect consumer
affective response?: An intimate apparel experience. European Journal of Marketing,
46(1/2), 112–133.
Levy, M., Weitz, B.A., & Grewal, D. (2014). Retailing management. New York: Irwin/
McGraw-Hill.
Manganari, E.E., Siomkos, G.J., & Vrechopoulos, A.P. (2009). Store atmosphere in web
retailing. European Journal of Marketing, 43(9/10), 1140–1153.
Neslin, S.A., Grewal, D., Leghorn, R., Shankar, V., Teerling, M.L., Thomas, J.S., et al. (2006).
Challenges and opportunities in multichannel customer management. Journal of
Service Research, 9(2), 95–112.
Nordfält, J., Grewal, D., Roggeveen, A., & Hill, K. (2014). Insights from in-store marketing
experiments. Shopper Marketing and the Role of In-Store Marketing (Review of Marketing
Research), 11, 127–146.
Oh, H., & Petrie, J. (2012). How do storefront window displays influence entering deci-
sions of clothing stores? Journal of Retailing and Consumer Services, 19(1), 27–35.
Palermo, R., & Rhodes, G. (2007). Are you always on my mind? A review of how face per-
ception and attention interact. Neuropsychology, 45, 75–92.
Puccinelli, N.M., Goodstein, R.C., Grewal, D., Price, R., Raghubir, P., & Stewart, D. (2009).
Customer experience management in retailing: Understanding the buying process.
Journal of Retailing, 85(1), 15–30.
Schneider, S.K. (1997). Body design, variable realisms: The case of female fashion manne-
quins. Design Issues, 13(3), 5–18.
Sen, S., Block, L.G., & Chandran, S. (2002). Window displays and consumer shopping de-
cisions. Journal of Retailing and Consumer Services, 9(5), 277–290.
Spence, C., Puccinelli, N.M., Grewal, D., & Roggeveen, A.L. (2014). Store atmospherics: A
multisensory perspective. Psychology and Marketing, 31(7), 472–488.
Sujan, M. (1985). Consumer knowledge: Effects on evaluation strategies mediating con-
sumer judgments. Journal of Consumer Research, 12(12), 31–46.
Wästlund, E., Otterbring, T., Gustafsson, A., & Shams, P. (2015). Heuristics and resource
depletion: Eye-tracking customers' in situ gaze behavior in the field. Journal of
Business Research, 68(1), 95–101.
Wedel, M., & Pieters, R. (2008). Eye tracking for visual marketing. Foundations & Trends in
Marketing, 1(4), 231–320.
Zhao, X., Lynch, J.G., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truths
about mediation analysis. Journal of Consumer Research, 37(2), 197–206.