Content uploaded by Sophie Lacoste-Badie
Author content
All content in this area was uploaded by Sophie Lacoste-Badie on Apr 13, 2020
Content may be subject to copyright.
DOI: 10.2501/JAR-2018-004 December 2018 JOURNAL OF ADVERTISING RESEARCH 443
INTRODUCTION
For years, advertisers have sought to discover the
advertising executional devices that inuence a
commercial’s effectiveness (Stewart and Furse,
2000). A book on the eectiveness of television
advertising (Stewart and Furse, 1986) identied
several executional factors—in particular, the pres-
ence of a model in an advertisement. Many studies
have been carried out on the characteristics of the
model, such as celebrity (Zwilling and Fruchter,
2013), attractiveness (Caballero, Lumpkin, and
Madden, 1989), ethnic origin (Appiah, 2001), gen-
der (Freiden, 1984), and age (Kozar, 2012). Few,
however, have been conducted on the inuence of
the model’s presence in an advertisement.
Several psychology studies have shown that the
face and the eyes more likely will capture aention,
compared with other stimuli (Awh, Serrences, Lau-
rey, Dhaliwal et al., 2004; Burton and Bindemann,
2009; Dupierrix, Hillairet, Boisferon, Méary et al.,
2014; Frank, Amso, and Johnson, 2014). In a clut-
tered advertising environment, this information is
highly signicant because, as highlighted by ear-
lier researchers, “aention has become one of the
scarcest resources in the contemporary advertising
economy” (Davenport and Beck, 2001, quoted in
Face Presence and Gaze Direction
In Print Advertisements
How They Influence Consumer Responses—
An Eye-Tracking Study
SAFAA ADIL
ESCEM, Tours, France
safaa_adil04@yahoo.fr
SOPHIE LACOSTE-BADIE
Lille Graduate School
of Management –
University of Lille
sophie.lacoste-badie@
univ-lille.fr
OLIVIER DROULERS
Université de Rennes 1
droulerso@gmail.com
For product manufacturers and advertisers, attracting consumer attention in a cluttered
advertising environment is essential. The eye-tracking study reported herein examines the
impact of face presence and model gaze direction in print advertisements. The ndings show
that face presence and gaze directed toward the product—versus no face and gaze toward
the viewer—have a strong inuence on attention to and memorization of advertisements,
on advertisement and brand evaluation, and on purchase intention. This research provides
meaningful and actionable recommendations for managers.
•
Face presence increases attention paid to advertisement elements, including product and brand.
The product receives even more attention when the model’s gaze direction is toward the product,
versus toward the viewer.
•
Face presence in an advertisement with the model’s gaze directed toward the product is the best
combination to improve product and brand memorization, attitude toward the brand, and purchase
intention.
•
Objective attention measures using an eye-tracking device can benefit managers and should not
be neglected. Regarding the effects of attention on memorization, such measures could be of
great help in advertisement tests.
Submitted September 19, 2016;
revised February 15, 2017;
accepted April 4, 2017;
published online February 19, 2018.
444 JOURNAL OF ADVERTISING RESEARCH December 2018
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS
Smit, Boerman, and Van Meurs, 2015, p.
216). The purpose of the current study was
to evaluate the inuence of face presence
and gaze direction on
• viewer aention;
• memorization;
• evaluations, or the viewer’s attitude
toward the advertisement and aitude
toward the brand;
• purchase intention.
The study focused on print advertisements.
lItERAtuRE REVIEW
The face is one of the stimuli that receives
the highest observer aention (Langton,
Law, Burton, and Schweinberger, 2008;
Theeuwes and Van der Stigchel, 2006). In
psychology, many studies have underlined
that humans preferentially process facial
visual stimuli, to which they automatically
give more aention, as opposed to other
visual stimuli: “Faces receive mandatory
processing when competing for aention
with stimuli of less sociobiological sali-
ence” (Weaver and Lauwereyns, 2011, p.
10). Prior research suggests that from a
very early age, infants show a preference
for human faces, as opposed to “non-face
objects” (Zieber, Kangaz, Hock, Hayden
et al., 2013), and that face preference tends
to increase with age (Frank et al., 2014).
Humans are not equally sensitive to
all parts of the face; the eyes are the most
aended facial feature and are used as a
pivotal source of information (Saether, Van
Bille, Laeng, Brennen, and Overvoli, 2009).
Perception of another person’s gaze thus
has an impact on the observer’s aention
(Tipper and Bayliss, 2011). Several stud-
ies in psychology have shown that when
the model’s gaze is averted, the observer’s
attention rapidly and automatically is
oriented toward the direction indicated
by the gaze (Driver, Davis, Ricciardelli,
Kidd et al., 1999). In addition, detection of
peripheral targets becomes slower when
participants xate on a face with direct,
as opposed to averted, gaze (Senju and
Johnson, 2009).
Many neuroscience studies have shown
that the face is processed in specic brain
areas (Gauthier, Tarr, Moylan, Skudlarski
et al., 2000; Kanwisher, McDermo, and
Chun, 1997; Sergent, Ohta, and MacDon-
ald, 1992) and that several brain regions
are activated dierentially when one is
processing direct gaze and averted gaze.
One review identied brain regions—fusi-
form gyrus, amygdala, temporal sulcus—
displaying dierential activity when the
participant observed a direct or averted
gaze (Senju and Johnson, 2009). Another
study used the single-cell recording
method in monkeys to show the existence
of cells sensitive to gaze direction, with
some cells selected for eye contact and
others for averted gaze (Perre, Hietanen,
Oram, Benson, and Rolls, 1992). Previous
research, therefore, suggests that there are
specic neuron circuits for face and gaze
processing. The functioning of these brain
regions explains the aention-grabbing
capacity of this type of stimulus.
Despite the frequent presence of faces
in advertisements, there has been limited
research on how face presence and gaze
direction aect aention toward adver-
tisements. Researchers have studied the
impact of a model’s gaze—toward the
product versus toward the viewer—on
aention to print advertisements (Hut-
ton and Nolte, 2011). In that study, “the
advertisements were embedded within
an on-screen magazine that participants
[32 students ages 20 to 29; M = 21.4] were
instructed to browse through, as if they
had picked the magazine up in a waiting
room,” (Huon and Nolte, 2011, p. 888)
while their aention was measured with
an eye-tracking system (Eyelink II eye
tracker, SR-Research, Oawa, Canada).
The results revealed that “participants
spent longer looking at all aspects of the
advertisement when the model’s gaze was
directed toward the product” than when it
was directed toward the viewer, and that
“participants looked to both the product
and brand regions of interest for longer
when the model was looking at the prod-
uct compared to when they were looking
out at the viewer” (Huon and Nolte,
2011, p. 890). Those researchers, however,
only studied aention and did not, for
example, address the eect of gaze direc-
tion on memorization, aitude toward the
advertisement, aitude toward the brand,
or purchase intention (Huon and Nolte,
2011).
A more recent study examined the
impact of face versus no face and gaze
cues—gaze toward the viewer (mutual
gaze) versus gaze toward the advertised
text and product (averted gaze)—on aen-
tion to banner advertisements (Sajjachola-
punt and Ball, 2014). Advertisements were
embedded in webpages presented one
after the other on a computer screen, and
participants’ (72 undergraduate and post-
graduate students ages 18 to 32 years; M =
22.9 years) aention was measured with an
eye-tracking system (ASL 5000, Applied
Science Laboratories, Bedford, MA). Find-
ings showed that the mean dwell time (the
sum of all xation durations within a par-
ticular region of interest [ROI]) on banner
advertisements was signicantly higher
in the “gaze toward the text and prod-
uct” condition and in the “gaze toward
the viewer” condition than in the no-face
condition.
These researchers (Sajjacholapunt and
Ball, 2014) found, however, no signicant
difference was identified between the
“gaze toward the text and product” condi-
tion and the “gaze toward the viewer” con-
dition. For aention specically directed to
the product, “the no face condition had a
signicantly lower mean dwell time on
December 2018 JOURNAL OF ADVERTISING RESEARCH 445
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS THEARF.ORG
the product compared to the averted gaze
condition” (Sajjacholapunt and Ball, 2014,
p. 11). Contrary again to previous research
(Huon and Nolte, 2011), however, there
was no signicant dwell-time dierence
between the “gaze toward the text and
product” condition and the “gaze toward
the viewer” condition.
These researchers (Sajjacholapunt and
Ball, 2014) also measured explicit and
implicit advertisement memorization.
They found that “the mean recognition
score for brand names in the no face con-
dition was signicantly lower than for
products in the mutual gaze condition
and the averted gaze condition” (Sajja-
cholapunt and Ball, 2014, p. 11) and that
“banner advertisements containing faces
with averted gaze looking at advertised
texts have a greater ability to improve
participants’ implicit memory perfor-
mance—word fragment completions
score—for advertising contents than do
banner advertisements containing faces
with mutual gaze cues or advertisements
with no faces.”
Academic research on this topic is
scarce. To the authors’ knowledge there is
no research on the inuence of face pres-
ence and gaze direction on aitude toward
the advertisement, attitude toward the
brand, and purchase intention. Memoriza-
tion research is rather fragmented; there is
no research on advertisement, brand, or
product recall, and research on the impact
that gaze direction has on aention shows
inconsistent results. The current study
sought to enhance understanding of the
impact of face presence and gaze direction
on advertisements.
hyPothESIS DEVEloPMENt AND
RESEARCh quEStIoNS
On the basis of previous research in psy-
chology and marketing, the rst group
of hypotheses postulated that aention
toward the advertisement (entire adver-
tisement, product, and brand) would be
higher in the face condition versus the no-
face condition, and that aention toward
the product and the brand would be higher
in the “gaze toward the product” condi-
tion than in the “gaze toward the viewer”
condition.
For the face condition, the authors
hypothesized the following:
H1: Attention (fixation duration,
xation count, and revisit count)
toward the entire advertisement
will be higher in the face condi-
tion versus the no-face condition.
H2: Attention (fixation duration,
xation count, and revisit count)
toward the product will be
higher in the face condition ver-
sus the no-face condition.
H3: Attention (fixation duration,
xation count, and revisit count)
toward the brand will be higher
in the face condition versus the
no-face condition.
For the gaze condition, the researchers
hypothesized the following:
H4: Attention (fixation duration,
fixation count, and revisit
count) toward the product will
be higher in the “gaze toward
the product” condition than in
the “gaze toward the viewer”
condition.
H5: Attention (fixation duration,
xation count, and revisit count)
toward the brand will be higher
in the “gaze toward the prod-
uct” condition than in the “gaze
toward the viewer” condition.
In the 1800s, the U.S. psychologist William
James claimed that “we cannot deny that
an object once aended to will remain in
memory, whilst one inaentively allowed
to pass will leave no traces behind” (James,
1890, p. 427). More recently, the two-route
(evaluative and learning/comprehension)
hierarchic model of advertising effects
highlighted that “in the learning route,
advertisement characteristics drive aen-
tion, which in turn determines compre-
hension and memory” (Thorson, Chi,
and Leavi, 1992, p. 368). Using an eye-
tracking device, researchers showed that
aention to the advertisement had a posi-
tive inuence on brand recognition (Wedel
and Pieters, 2000).
Another group, which measured
attention through self-reported verbal
measures, also found a positive eect of
attention to banner advertisements on
recognition performance (Yoo and Kim,
2005). In a model of aention capture and
transfer by elements of the advertisements,
other researchers showed that increased
aention to the pictorial in advertisements
was associated with increased aention to
other advertisement elements, in particular
to the brand (Pieters and Wedel, 2004).
Because the link between aention and
memory is well established, and given that
the aention-grabbing property of the face
can be benecial for other advertising ele-
ments, the authors proposed that memo-
rization of advertisement content would
There is no research on the influence of face presence
and gaze direction on attitude toward the advertisement,
attitude toward the brand, and purchase intention.
446 JOURNAL OF ADVERTISING RESEARCH December 2018
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS
• In a print advertisement, do face
presence (RQ1) and gaze direction
(RQ2) influence attitude toward the
advertisement?
• In a print advertisement, do face pres-
ence (RQ3) and gaze direction (RQ4)
inuence aitude toward the brand?
• In a print advertisement, do face pres-
ence (RQ5) and gaze direction (RQ6)
inuence purchase intention?
METHOD
Participants
Seventy-nine adults (41 women, 38 men;
18 students, 61 workers) ages 21 to 52
years (M = 32.28, SD = 8.00) took part in
this study. Thirty-one participants were
ages 21 to 29 years, 30 were ages 30 to 39
years, and 18 were ages 40 to 52 years. The
participants, who were recruited from ini-
tial and lifelong learning programs, vol-
unteered to take part in the study without
any incentive.
Stimuli
To avoid forced exposure to the stimuli, the
researchers embedded the advertisements
in ctitious editorial content that had been
created by an advertising agency: a travel
magazine containing 12 pages with ve
advertisements. All advertisements were
inserted in the same position at the boom
of the right-hand pages. The three target
advertisements (pages 3, 7, and 11 of the
magazine) concerned products that might
be purchased and consumed by both men
and women—orange juice, ice cream, and
yogurt.
The researchers took care to include cti-
tious brands (Doo, Crystal, and Baiko) so
as to avoid eects caused by prior brand
exposure in the test country, France. They
used two well-known brands (Club Med
and Lay’s) as distractive advertisements
(pages 5 and 9 of the magazine) to ensure
be higher in the face condition versus the
no-face condition.
H6: Recall of the advertisement ele-
ments will be higher in the face
condition versus the no-face
condition.
H7: Recall (H7a) and recognition
(H7b) of the product category
will be higher in the face condi-
tion versus the no-face condition.
H8: Recall (H8a) and recognition
(H8b) of the brand will be higher
in the face condition versus the
no-face condition.
Likewise, because perceived gaze direc-
tion orients the observer’s visual aention
(Haxby and Gobbini, 2011), the authors
proposed that memorization of product
category and brand would be higher in the
“gaze toward the product” condition than
in the “gaze toward the viewer” condition.
H9: Recall (H9a) and recognition
(H9b) of the product category
will be higher in the “gaze
toward the product” condition
than in the “gaze toward the
viewer” condition.
H10: Recall (H10a) and recogni-
tion (H10b) of the brand will
be higher in the “gaze toward
the product” condition than in
the “gaze toward the viewer”
condition.
Finally, given the absence of knowledge
on the inuence of face presence and gaze
direction on aitude toward the advertise-
ment and toward the brand, and on pur-
chase intention, the authors proposed six
research questions:
that participants would not wonder why
only unknown brands were present. To
avoid any potential eects caused by the
choice of a specic model, the research-
ers used two models (young Caucasian
women) in both conditions (gaze directed
toward the product and gaze directed
toward the viewer). Facial expressions
(neutral) and head orientation were
identical.
In total, 15 advertisements were created:
three with only a product, six with the
model’s gaze directed toward the viewer
(3 products × 2 models), and six with the
model’s gaze directed toward the product
(3 products × 2 models). When consult-
ing the magazine, each participant was
exposed to three target advertisements
(within-subject procedure):
• one with gaze directed toward the prod-
uct (50 percent Model 1 and 50 percent
Model 2);
• one with gaze directed toward the
viewer (50 percent Model 1 and 50 per-
cent Model 2);
• one with only a product.
The appearance order of the models, prod-
ucts, and gaze conditions was randomized
(See Figure 1).
Procedure
On their arrival at the laboratory, partici-
pants were informed that they would be
viewing a new, on-screen magazine and
that their task was to evaluate it. Exposure
time was not limited, and participants
could move on to read the next page by
pressing the right arrow of the keyboard
or go back to the previous page by press-
ing the left arrow. As participants looked
at the magazine, their eye movements
were recorded with an eye-tracking system
(RED 500, SensoMotoric Instruments, Tel-
tow, Germany) located below the screen.
The benet of using this type of material
December 2018 JOURNAL OF ADVERTISING RESEARCH 447
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS THEARF.ORG
is that no device is aached to the partici-
pants’ face, which gives them, to a certain
degree, freedom of movement.
The experiment started with a calibra-
tion successfully completed by all par-
ticipants. After participants had consulted
the magazine, a surprise memory task
was proposed. Finally, participants were
exposed to the three target advertisements
to complete aitude toward the advertise-
ment, aitude toward the brand, and pur-
chase intention scales. Participants were
asked also about their product-category
involvement, their personal characteris-
tics, whether they already knew of the
brands, and whether they found the two
models aractive. After the experiment,
the researchers debriefed the participants
to see whether they had guessed the real
purpose of the task, and none of them had
done so.
Measures
Eye-tracking measures. To measure how
aention was distributed across the maga-
zine, the researchers created four ROIs:
editorial text (top-right location), entire
advertisement (bottom-right location),
image of the product, and brand in the
advertisement (Pieters and Wedel, 2004).
For each ROI, three eye-tracking meas-
ures were considered: total xation dura-
tion (the sum of durations of all xations
within an ROI), xation count (total num-
ber of xations detected within an ROI),
and revisit count (number of times the ROI
was revisited after the rst visit; Wedel and
Pieters, 2000).
Declarative measures. For memorization,
the following recall and recognition tests
were used:
• recall of the advertisement elements;
• recall of the product category;
• recall of the brand;
• recognition of the product category from
a list of 12 items (one target and three
distractors);
• recognition of the brand from a list of 12
items.
Then, aitude toward the advertisement
(Holbrook and Batra, 1987), attitude
toward the brand (Spears and Singh,
2004), and purchase intention (Juster,
1966) were measured. Aractiveness was
assessed with a ten-point scale (1 = “not
at all aractive” and 10 = “very arac-
tive”). Product category involvement
was measured in line with previous work
(Strazzieri, 1994).
Figure 1 Illustrative Example of Magazine Pages in the No-Face Condition, “Gaze toward the
Product” Condition, and “Gaze toward the Viewer” Condition
448 JOURNAL OF ADVERTISING RESEARCH December 2018
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS
RESULTS
In this experiment, each participant was
exposed to the three conditions: no face,
gaze toward the viewer (direct gaze), and
gaze toward the product (averted gaze;
within-subject design). Because the same
participants took part in all conditions in
this current experiment, the authors car-
ried out analysis of variance (ANOVA) on
repeated-measures data. Most of the eye-
tracking and memory data did not follow
a normal distribution; therefore, nonpara-
metric tests were implemented (Fried-
man’s ANOVA and Wilcoxon signed-rank
test). For normally distributed data, t tests
for paired samples were used.
As was expected, preliminary analy-
ses showed that the three brands used
in the experiment were unknown to the
participants and that there was an identi-
cal level of aractiveness for each model.
The results revealed no eects of gender
and product involvement on eye-tracking
and declarative measures; accordingly,
these elements were given no further
consideration.
Influence of Face Presence
On Attention and Memorization
Aention to the entire advertisement.
First, although it was not the main objec-
tive of this study, it is interesting to note
that the results of several studies converge
to show similar average aention spans
for advertisements inserted in a magazine.
In this respect, for the average amount of
time spent looking at an advertisement,
one study reported 2.12 seconds per adver-
tisement (Pieters, Wedels, and Batra, 2010),
another reported 3.4 seconds per advertise-
ment (one page; Huon and Nolte, 2011),
and the present study found 2.9 seconds
per advertisement (half a page).
As regards experimental manipula-
tion, the xation duration (Mface = 3,507.32,
Mno face = 1,705.67, p = .00), xation count
(Mface = 14.03, Mno face = 7.26, p = .00), and
revisit count (Mface = 2.32, Mno face = 1.30,
p = .00) for the entire advertisement were
higher in the face condition than in the no-
face condition. The advertisement received
more aention in the face condition than
in the no-face condition, supporting Hypo-
thesis 1.
Face presence in an advertisement
enhanced attention toward the edito-
rial text. The fixation duration (Mface =
12,503.29, Mno face = 9,251.01, p = .00), xation
count (Mface = 48.56, Mno face = 35.83, p = .00),
and revisit count (Mface = 3.15, Mno face = 2.45,
p = 000) for the editorial text were higher
in the face condition than in the no-face
condition. The magazine’s editorial text
received more aention in the face condi-
tion than in the no-face condition.
Aention to the product and the brand.
Fixation duration (Mface = 739.25, Mno face =
596.83, p = .02), xation count (Mface = 3.72,
Mno face = 2.17, p = .00), and revisit count
(Mface = 1.10, Mno face = 0.56, p = .00) for
the image of the product were higher
in the face condition than in the no-face
condition. The image of the product
thus received more aention in the face
condition than in the no-face condition,
supporting H2. Fixation duration (Mface =
309.71, Mno face = 194.63, p = .00), xation
count (Mface = 0.74, Mno face = 0.46, p = .03),
and revisit count (Mface = 0.30, Mno face = 0.07,
p = .00) for the brand were higher in the
face condition than in the no-face condi-
tion. The brand received more aention in
the face condition than in the no-face con-
dition supporting H3.
Memorization of advertisement elements,
product category, and brand. Recall of the
advertisement elements was higher in the
face condition than in the no-face condi-
tion (p < .05), supporting H6. Recall and
recognition of the product category were
higher in the face condition than in the
no-face condition (p < .05), supporting H7.
Recall and recognition of the brand were
higher in the face condition than in the
no-face condition (p < .05), supporting H8
(See Table 1).
Because the results showed that face
presence in the advertisement had an inu-
ence on both aention and memorization,
the authors investigated the possible medi-
ating role of aention for the impact of face
TABLE 1
Memorization of Advertisement Elements, Product Category,
And Brand in No-Face and Face Conditions
Variable No Face Face
z pM (SD)M (SD)
Recall:
Executional elements (H6) 0.49 (0.83) 2.68 (1.86) −6.95 .00
Product category (H7a) 0.11 (0.32) 0.46 (0.37) −5.45 .00
Brand (H8a) 0.03 (0.15) 0.12 (0.22) −3.27 .00
Recognition:
Product category (H7b) 0.34 (0.47) 0.72 (0.35) −4.85 .00
Brand (H8b) 0.10 (0.30) 0.37 (0.37) −4.52 .00
Note: N = 79. The gures in the table represent the average number of items recalled and recognized among participants.
December 2018 JOURNAL OF ADVERTISING RESEARCH 449
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS THEARF.ORG
presence on advertisement memorization
by using the PROCESS macro (bootstrap-
ping of 5,000 samples; Hayes, 2013). First,
they examined whether aention to the
entire advertisement mediated the impact
of face presence on recall score (sum of
recall of the advertisement elements, recall
of the product category, and recall of the
brand). The results indicated complemen-
tary mediation (Zhao, Lynch, and Chen,
2010; See Figures 2 and 3).
The indirect path of the eects of face
presence on recall score through aen-
tion to the entire advertisement (xation
duration and xation count) was signi-
cant and positive (ab = 0.18 and ab = 0.39,
respectively), with the 95 percent con-
dence interval excluding zero ([.10, .41],
[.18, .67]). The direct eect of face pres-
ence on recall score was also signicant
and positive (c = 2.39, t = 7.60, p < .000;
c = 2.23, t = 7.10, p < .000), as reported ear-
lier. According to the Kappa-squared test
(Preacher and Kelley, 2011), the eect size
of the indirect eect was medium (k2 = .05,
k2 = .08).
The authors conducted a second test with
face presence as the independent variable,
recognition score (sum of recognition of
the product category and recognition of
the brand) as the dependent variable, and
aention to the entire advertisement as the
mediator. Analyses conducted with boot-
strapped samples (5,000) also indicated
complementary mediation (See Figures 4
and 5). The indirect path of the eects of
face presence on recognition score through
aention to the entire advertisement (xa-
tion duration and xation count) was sig-
nicant and positive (ab = 0.05 and ab = 0.07,
respectively), with the 95 percent con-
dence interval excluding zero ([.00, .09], [.02,
.13]). The direct eect of face presence on
recognition score was also signicant and
positive (c = 0.60, t = 5.97, p < .000; c = 0.57,
t = 5.65, p < .000), as reported earlier. The
eect size was medium (k2 = .03, k2 = .04).
Figure 2 Mediation Analysis of Attention to the Advertisement
(Fixation Duration) on Recall Score
Attention to the Ad
(fi xation duration)
a = 1801.64
t = 3.49, p < .000
Direct Effect: c = 2.39, t = 7.60, p < .000
Indirect Effect: ab =0.18, 95% CI [.10, .41]
b = 0.0001
t = 3.45, p < .000
Face Presence Recall Score
Figure 3 Mediation Analysis of Attention to the Advertisement
(Fixation Count) on Recall Score
Attention to the Ad
(fi xation count)
a = 6.76
t = 4.52, p <.000
Direct Effect: c = 2.23, t = 7.1 0, p < .000
Indirect Effect: ab = 0.39, 95% CI [.18, .67]
b = 0.05
t = 4.47, p <.000
Face Presence Recall Score
450 JOURNAL OF ADVERTISING RESEARCH December 2018
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS
This procedure provided evidence for
the underlying process involved. Face
presence exerted a positive direct impact—
and a positive indirect impact mediated
by aention paid to the entire advertise-
ment—on advertisement memorization.
Influence of Gaze Direction on Attention
And Memorization
Aention to the product and the brand.
Fixation count (Mproduct = 4.25, Mviewer = 3.20,
p = .00) and revisit count (Mproduct = 1.24,
Mviewer = 0.94, p = .02) for the image of the
product were higher in the “gaze toward
the product” condition than in the “gaze
toward the viewer” condition. Fixation
duration was marginally higher in the
“gaze toward the product” condition than
in the “gaze toward the viewer” condition
(Mproduct = 857.58, Mviewer = 620.91, p = .06).
Hypothesis 4, therefore, is supported par-
tially. No dierence was observed between
the “gaze toward the product” condition
and the “gaze toward the viewer” condi-
tion for xation duration or xation count
for the brand (p > .05). Only revisit count
(Mproduct = 0.39, Mviewer = 0.21, p = .04) was
higher in the “gaze toward the product”
condition versus the “gaze toward the
viewer” condition, partially supporting
H5.
Memorization of product category and
brand. Recall and recognition of the prod-
uct category were higher in the “gaze
toward the product” condition than in the
“gaze toward the viewer” condition (p <
.05), supporting H9. Recall and recogni-
tion of the brand were higher in the “gaze
toward the product” condition than in the
“gaze toward the viewer” condition (p <
.05), supporting H10; See Table 2).
Outcome of Research Questions
The influence of face presence (versus
no face) was signicant and positive on
aitude toward the advertisement (RQ1;
Figure 4 Mediation Analysis of Attention to the Advertisement
(Fixation Duration) on Recognition Score
Attention to the Ad
(fi xation duration)
a= 1801.64
t = 3.49, p < .000
Direct Effect: c = 0.60, t = 5.97, p < .000
Indirect Effect: ab = 0.05, 95% Cl [.00, .09]
b = 0.00
t = 2.27, p < .05
Face Presence Recognition Score
Figure 5 Mediation Analysis of Attention to the Advertisement
(Fixation Count) on Recognition Score
Attention to the Ad
(fi xation count)
a = 6.76
t = 4.52, p <.000
Direct Effect: c = 0.57, t = 5.65, p < .000
Indirect Effect: ab = 0.07, 95% CI [.02, .13]
b = 0.01
t = 2.65, p <.000
Face Presence Recognition Score
December 2018 JOURNAL OF ADVERTISING RESEARCH 451
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS THEARF.ORG
• The 79 participants were ages 21 to 52
years, and most were in employment.
• The study used an eye-tracking system
that allowed freedom of movement (e.g.,
Huon and Nolte, 2011, used a head-
mounted, video-based eye tracker).
• It used four dierent types of measures:
attention (fixation duration, fixation
count, and revisit count), memorization
(recall and recognition), evaluations
(aitudes toward the advertisement and
brand), and behavioral intention.
Several key ndings emerged from the cur-
rent study.
Influence of Face Presence and Gaze
Direction on Attention
First, the results reveal that face presence
in an advertisement is a class of stimuli
that captures viewer aention. Compared
with an advertisement with no face pres-
ence, an advertisement containing a face
had a longer xation duration and a higher
xation and revisit count. The results of
the current study support ndings from
previous studies in psychology showing
that, from a very early age, individuals dis-
play a preference for human faces because
p = .00), but no eect of gaze direction was
observed on this variable (RQ2; p = .11).
The inuence of face presence (versus no
face) was signicant and positive on ai-
tude toward the brand (RQ3; p = .00), and
aitude toward the brand was higher in
the “gaze toward the product” condition
than in the “gaze toward the viewer” con-
dition (RQ4; p = .00). The inuence of face
presence (versus no face) was signicant
and positive on purchase intention (RQ5;
p = .00), and purchase intention was higher
in the “gaze toward the product” condition
than in the “gaze toward the viewer” con-
dition (RQ6; p = .03; See Table 3).
DISCUSSION
In the current study, the authors used a
combination of objective (eye tracking)
and declarative measures to determine, in
advertisements, the eect of face presence
and the model’s gaze direction on aen-
tion, memorization, evaluation (aitude
toward the advertisement and brand), and
purchase intention. The current study dif-
fers from previous research conducted in
the eld of psychology (Huon and Nolte,
2011; Sajjacholapunt and Ball, 2014) in the
following ways:
they spend longer looking at faces than at
objects (Frank et al., 2014; Umilta, Simion,
and Valenza, 1996; Zieber et al., 2013).
Concerning the inuence of gaze direc-
tion on aention, the current study pro-
vides a clear-cut answer to the conicting
results of the very few previous studies
conducted on the topic. Contrary to one
study (Sajjacholapunt and Ball, 2014) but
in keeping with another (Hutton and
Nolte, 2011), the results of the current
study show that when the model’s gaze
was directed toward the product, ver-
sus toward the viewer, the image of the
product had a higher xation and revisit
count. Additionally, this research shows
that more aention to the advertisement
did not aect adversely the editorial text,
because the text located above the adver-
tisement received more aention when the
advertisement contained a face, versus no
face. It, therefore, seems that the more an
advertisement captures the viewer’s aen-
tion, the greater is the aention given to
the editorial text. These results echo those
obtained by other researchers (Pieters and
Wedel, 2004), who showed that when an
element receives more viewer aention,
this aention is transferred also to the sur-
rounding elements.
Influence of Face Presence and Gaze
Direction on Memorization
This research makes a signicant contri-
bution to the topic of the inuence of face
presence and gaze direction on memoriza-
tion. One prior study (Huon and Nolte,
2011) did not measure memorization, and
another (Sajjacholapunt and Ball, 2014)
measured only brand recognition. The
inuence on recall, for example, has not
been studied yet.
The ndings of the current study show
that face presence in the advertisement,
versus no presence, improved the recall
of advertisement elements, product cat-
egory, and brand and that the model’s
TABLE 2
Memorization of Product Category and Brand in the “Gaze toward
The Viewer” and “Gaze toward the Product” Conditions
Variable Gaze toward the
Viewer
Gaze toward the
Product
z pM (SD)M (SD)
Recall:
Product category (H9a) 0.30 (0.46) 0.63 (0.51) −4.21 .00
Brand (H10a) 0.05 (0.22) 0.19 (0.39) −2.66 .00
Recognition:
Product category (H9b) 0.65 (0.48) 0.80 (0.40) −2.44 .00
Brand (H10b) 0.27 (0.44) 0.48 (0.50) −3.05 .00
Note: N = 79. The gures in the table represent the average number of items recalled and recognized among participants.
452 JOURNAL OF ADVERTISING RESEARCH December 2018
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS
a positive impact on aitude toward the
advertisement, aitude toward the brand,
and purchase intention. For the inuence
of gaze direction on aitude toward the
advertisement, no dierence was observed
between gaze toward the viewer and gaze
toward the product.
Although this result needs to be con-
rmed by further research, it shows that
using image-processing software to change
gaze direction did not affect advertis-
ing credibility, which was evaluated in
the same way under both experimental
conditions. The model’s gaze direction
toward the product, however, seemed
to be advantageous to attitude toward
the brand and purchase intention. These
results complement some previous psy-
chological research that has shown a posi-
tive link between aention to an object and
evaluation or choice of this object (Shimojo,
Simion, Shimojo, and Scheier, 2003).
MANAGERIAL IMPLICATIONS
Identifying the positive effects of face
presence and gaze direction has important
implications for product manufacturers
and advertisers. On the basis of a count of
just over 1,000 press advertisements (N =
1,023) conducted by the authors in France,
it was found that a face was present in 85.2
percent of advertisements for perfume,
66.4 percent for cosmetics, and 38.3 percent
for food products but only 28.8 percent for
gaze toward the product, versus toward
the viewer, improved the recall of product
category and brand. Similarly, recognition
of product category and brand were higher
when a face was present in the advertise-
ment, versus no face, and when the mod-
el’s gaze was directed toward the product,
versus toward the viewer.
The results of this study show that the
inuence of face presence on memoriza-
tion is explained partially by the increased
attention given to the advertisement.
Results in previous work (Zhao et al., 2010)
revealed a complementary mediation,
which meant that both face presence and
aention to the entire advertisement posi-
tively inuenced memorization. This is an
interesting result, given that several previ-
ous marketing studies assumed a positive
impact of aention on memorization meas-
ures but had not tested this impact using
the precise measures enabled by the eye-
tracking methodology.
Influence of Face Presence and Gaze
Direction on Evaluation and Purchase
Intention
As far as the authors are aware, no study
yet has addressed the inuence of face
presence in advertisements on attitude
toward the advertisement, aitude toward
the brand, and purchase intention. The
ndings of the current study show that the
presence of a face in an advertisement has
cars. Face presence in a print advertise-
ment is quite common for certain product
categories but much less so for others (e.g.,
food products). This current research, how-
ever, shows that the presence of a face in
three food-product commercials—yogurt,
orange juice, and ice cream—improved
memorization of the advertisement ele-
ments, product category, and brand, partly
as a consequence of the greater aention
paid to the advertisement. Evidently, the
“learning route” (Thorson et al., 1992) ben-
ets greatly from the presence of a face in
an advertisement.
The “evaluative route” (Thorson et al.,
1992) also benets from face presence in
an advertisement: Face presence improved
advertising evaluation, brand evaluation,
and purchase intention. On the basis of
the current ndings, the authors strongly
recommend that advertisers and market-
ers present their product not alone but
with a face inserted into their advertise-
ments. This will aract viewer aention
to the advertisement, which, in turn,
will improve memorization. It also will
improve advertisement and brand evalu-
ation and purchase intention.
The eyes are the most aended facial fea-
ture, and the direction of the gaze orients
the viewer’s aention. By focusing on the
model’s gaze direction in advertisements,
this study has identied that gaze direc-
tion toward the product had a positive
TABLE 3
Influence of Face Presence and Gaze Direction on Attitude toward the Advertisement (Aad),
Attitude toward the Brand (Ab), and Purchase Intention (PI)
Variable No Face Face
t p
Gaze toward
the Viewer
Gaze toward
the Product
t pM (SD)M (SD)M (SD)M (SD)
Aad RQ1 11.35 (3.82) 12.92 (3.08) −3.33 .00 RQ2 12.59 (3.36) 13.25 (3.80) −1.59 .11
Ab RQ3 13.88 (4.40) 16.61 (3.17) −4.60 .00 RQ4 15.79 (3.95) 17.43 (4.10) −2.91 .00
PI RQ5 28.28 (24.13) 42.32 (19.97) −4.39 .00 RQ6 38.76 (24.31) 45.90 (25.41) −2.14 .03
Note: N = 79.
December 2018 JOURNAL OF ADVERTISING RESEARCH 453
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS THEARF.ORG
Although the sample size of this research
is within the standards for studies that use
eye-tracking methodology (25 to 70 partici-
pants tested in most studies), future stud-
ies with larger samples will improve the
generalizability of the results. In line with
Because only female faces were used in the
current study, future research could exam-
Future research also should investigate
toward the logo also should be investi-
gated in advertisements for services.
-
ence of gaze cues in print advertisements.
According to an earlier author (Brasel,
-
ment depends on the nature of the media
to which the consumer is exposed. Extend-
ing the research to another medium, such
as the Internet, could be a promising line
of enquiry from a management perspec-
tive. Finally, although this study showed
-
activations during exposure to advertise-
ABOUT THE AUTHORS
Safaa Adil is an associate professor of marketing at
L’École Supérieure de Commerce et de Management
(ESCEM), Tours, France. Her current research focuses
on the influence of human presence in advertisement.
Her research has been published in the Journal of
Neuroscience, Psychology, and Economics.
toward the brand. When the model’s gaze
was directed toward the product, versus
toward the viewer, this led to higher pur-
chase intent. On the basis of this study’s
be advised to give serious consideration
to the role of the face in advertisements.
be captured by the presence of a face in an
may be directed toward the product.
lIMItAtIoNS AND FutuRE RESEARCh
-
tunities for future research. One limitation
only faces with neutral expressions were
of gaze direction and other facial elements,
such as facial expressions and head orienta-
tion, cannot be examined. The authors chose
not to use celebrity faces to avoid the “bias
It could be worthwhile to examine whether
direction is strengthened or diminished
when the model in the advertisement is a
celebrity or an unknown endorser.
Another limitation is the relative sim-
plicity of the advertisements created for
the experiment; the researchers chose sim-
ple advertisements to control all of the
advertisement elements and thus to isolate
-
tisements and brands may be necessary,
provided that both product and brand
familiarity are measured beforehand. Meas-
uring purchase intention has its limitations
in the sense that intent to purchase does not
translate necessarily into an actual purchase.
Future studies accordingly should examine
direction on real purchases.
SoPHie lacoSte-badie is professor of marketing at
Lille University and a member of Lille Economics and
Management – LEM UMR CNRS 9221. Her research
interests center on attention, emotion, and aging. Her
work has been published in PLOS One; Psychology and
Marketing; the Journal of Neuroscience, Psychology,
and Economics; and Recherche et Applications en
Marketing, among other publications.
olivier droUlerS is professor of marketing at
L’Institut de Gestion de Rennes (Graduate School
of Management), Université de Rennes 1. His
research investigates the application of concepts
and techniques deriving from neuroscience to
management, with a special focus on marketing.
Droulers’s work has been published in such journals as
the Journal of Advertising Research; Psychology and
Marketing; the Journal of Neuroscience, Psychology,
and Economics; and PLOS One.
REFERENCES
-
cents’ Evaluations of Advertisements.” Journal
of Advertising Research
“Unimpaired Face Discrimination Dur-
Processing Channels.” Cognitive Psychology 48, 1
“Nonconscious Drivers of Visual
Journal of Brand Management
473–482.
and“The Role of
View in Human Face Detection.” Vision Research
and C. S.
“Using Physical Attractiveness as
454 JOURNAL OF ADVERTISING RESEARCH December 2018
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS
Regression-Based Approach
Press, 2013.
and“Assessing the
Role of Emotions as Mediators of Consumer
Responses to Advertising.” Journal of Consumer
Research
and
Applied Cognitive Psychology
The Principles of Psychology.
Holt, 1890.
“Consumer Buying Intentions and
Design.” Journal of the American Statistical Asso-
ciation
and
Extrastriate Cortex Specialized for Face Per-
ception.” Journal of Neuroscience
4302–4311.
Female Consumers’ Purchase Intentions and
-
ing.” International Journal of Marketing Studies 4,
and
Faces.” Cognition
and “Organization and
Functions of Cells Responsive to Faces in the
Temporal Cortex.” Philosophical Transactions of
the Royal Society of London B: Biological Sciences
and
Journal of Marketing 68,
Journal of Advertising
Research
and The Aention
Economy: Understanding the New Currency of Busi-
ness.
2001.
-
ospatial Orienting.” Visual Cognition
509–540.
“Preference for Human Eyes in
Human Infants.” Journal of Experimental Child
Psychology
and “The
Harm Brand Recall?” International Journal of
Research in Marketing
and“Vis-
Infancy.” Journal of Experimental Child Psychology
“Advertising Spokesperson
Gender on Two Audiences.” Journal of Advertis-
ing Research
-
“The Fusiform ‘Face Area’ Is Part of a
Network That Processes Faces at the Individual
Level.” Journal of Cognitive Neuroscience 12, 3
and “Distributed
Neural Systems for Face Perception.” In The
Oxford Handbook of Face Perception, A. J. Calder,
G. Rhodes, M. H. Johnson, and J.V. Haxby, eds.
Introduction to Mediation, Mod-
eration, and Conditional Process Analysis: A
and “The
Journal of Market-
ing
and
Psychological Methods
-
and“Anchoring Gaze When
Tracking Data.” Vision Research
2870–2880.
and -
Frontiers in
Psychology
and“The Eye Contact
Trends
in Cognitive Sciences
and
“Functional Neuroanatomy of Face and Object
Processing. A Positron Emission Tomography
Study.” Brain
and C.
Preference.” Nature Neuroscience
1317–1322.
and
“The Power of Direct Context as Revealed by
Eye Tracking.” Journal of Advertising Research 55,
and -
tude toward the Brand and Purchase Inten-
tions.” Journal of Current Issues and Research in
Advertising
and “Analysis of
the Impact of Executional Factors on Advertis-
ing Performance.” Journal of Advertising Research
December 2018 JOURNAL OF ADVERTISING RESEARCH 455
FACE PRESENCE AND GAZE DIRECTION IN PRINT ADVERTISEMENTS THEARF.ORG
and“The Impact of
The
Oxford Handbook of Face Perception, A. J. Calder,
G. Rhodes, M. H. Johnson, and J.V. Haxby, eds.
and“New-
born’s Preference for Faces.” European Psycholo-
gist
and
of the Change Detection Advantage for Faces.”
Psychological Research
and“Eye Fixations on
Model and Findings.” Marketing Science 19, 4
and Eective Tel-
evision Advertising: A Study of 1000 Commercials.
“Mesurer l’Implication Durable
vis-à-vis d’un Produit Indépendamment du
Risque Perçu” [“Measuring the Sustainable
Commitment toward a Product Independently
of Perceived Risk”]. Recherche et Applications en
Marketing
and “Faces
-
tion of Return.” Visual Cognition
657–665.
and
Advertising Hierarchy.” Advances in Consumer
Research
and “Processing of Anima-
Cognitive and Emotional Responses.” Journal of
Interactive Marketing
and“Recon-
About Mediation Analysis.” Journal of Consumer
Research
Face Processing in Infancy.” Journal of Experi-
mental Child Psychology
and “Matching
the Brand.” Journal of Advertising Research 53, 4
CopyrightofJournalofAdvertisingResearchisthepropertyofWarcLTDanditscontent
maynotbecopiedoremailedtomultiplesitesorpostedtoalistservwithoutthecopyright
holder'sexpresswrittenpermission.However,usersmayprint,download,oremailarticlesfor
individualuse.
