|
The Power of Words:
Another Look at the Verbal and Visual Components in Print Ads
By Yulian Li
Doctoral Student
Room 111, Murphy Hall
School of Journalism and Mass Communication
University of Minnesota
Minneapolis, MN 55104
Tel: 651-625-7564
Email: [log in to unmask]
* paper submitted to the Advertising Division of AEJMC Convention in New
Orleans, 1999
The Power of Words:
Another Look at the Verbal and Visual Components in Print Ads
Introduction
Most people believe that a picture is worth a thousand words. Advertisers are no
exception. They often believe that "vivid information is more persuasive than
pallid information" (Kisielius & Sternthal, 1984, p. 54). As a result, prints
ads usually have a pictorial component, creating a pervasive phenomenon of
word-picture combination in the practical world of advertising (Childers &
Houston, 1984; Childers et al., 1986; Smith, 1991). Even further, many print ads
devote so much space to pictures that they provide "little product attribute
information" (Mitchell, 1986, p. 12). Such a phenomenon has been attributed to
advertisers' assumption that pictures are more effective in persuading consumers
to buy a product (Taylor & Thompson, 1982).
The assumption finds its support in advertising research (Starch, 1966; Mitchell
& Olson, 1981; Mitchell, 1986). Childers and Houston (1984) explained that
pictures are "more memorable" than words and that pictures act as "a rich
mnemonic device that enhances learning and retention of material over such
techniques as sentence elaboration or rote rehearsal" (p. 643).
Believing in the effectiveness of "visually oriented advertising" over "verbally
oriented advertising," Percy and Rossiter (1980) suggested that "those creating
print advertising should be aware that visual imagery can play a significant
part in enhancing attitude for the advertised product" (p. 168).
However, learning and memory theorists seemed to favor words over pictures. They
The Power of Words
argued that processing verbal information was more automatic than processing
visual information, because word reading was a learned process and much more
automatic than picture recognition (Schneider & Shiffrin, 1977; Beck et al.,
1982; Logan, 1988). Reviewing studies on the Stroop effect, MacLeod (1991)
argued that words were much more powerful than pictures when the two types of
stimuli were processed simultaneously.
Some advertising studies found that when a picture was added in a print ad, it
did not necessarily have a more positive effect than a verbal ad, because people
had limited cognitive resources. Such a phenomenon was termed the ceiling effect
(Kisielius & Sternthal, 1984). Furthermore, pictures might interfere with the
processing of verbal information, creating a reverse Stroop effect (Dyer &
Severance, 1972; Dyer, 1973).
Reconciling these two opposing views, Krugman (1977) proposed that processing
verbal information was a high involvement activity while processing visual
information was a low involvement activity. Petty and Cacioppo (1981)
maintained, in their elaboration likelihood model, that people of high
involvement tended to devote much cognitive resources to verbal information
processing while people of low involvement relied on peripheral cues, such as
visual images. This raises the question of whether consumers devote more
attention to the verbal component than to the visual component in a print ad in
high involvement situations while they devote more attention to the visual
component than to the verbal component in low involvement situations. The
purpose of this study is to examine and compare the effects of verbal and visual
components of print ads on consumers' attitude toward the ad, attitude toward
the brand, recall of brand attributes and purchase intention under both high and
low involvement situations.
The Power of Words
As Childers and Houston (1984) pointed out, the study of the effects of verbal
and visual components of print ads is particularly important because in the
practical world of advertising, the use of pictures "adds substantially to
advertising production and media placement costs. Such research would offer a
basis for anticipating when these added costs are justified or unnecessary" (p.
644).
Literature Review
The verbal and visual components of print ads are believed to be processed
separately by the human brain. Paivio (1971) raised the dual-code concept, which
stated that verbal and visual stimuli were encoded separately in memory. He
argued that verbal information was stored as sequences of words while visual
information was stored in imageral records. Santa (1977) tested people's
recognition of stimuli in geometric forms and verbal forms, and found that
people encoded words linearly and pictures spatially.
The dual-code concept was also confirmed by Sperry's (1974) theory of the
hemispheric lateralization that the left and right hemispheres of the human
brain had different functions. In Karsen et al.'s (1972) study, radioisotopes
were injected into subjects' cerebral blood supply. When the subjects were
engaged in a verbal task, bloodflow to the left hemisphere increased, while
bloodflow to the right hemisphere increased when subjects were engaged in a
pictorial task.
Experiments on men who had sustained gunshot wounds to their left brain or right
brain and electroencephalograms on brain activities (Weinstein, 1982) also
indicated that the left brain
The Power of Words
controlled verbal information processing while the right brain controlled, among
other things, the processing of visual information and spatial judgement.
Verbal Versus Visual
As verbal and visual stimuli are encoded differently, their consequences are
bound to be different. Some (Paivio & Csapo, 1973; Nelson et al., 1976) argued
that pictorial processing was qualitatively superior to verbal processing.
Shepard (1967) compared people's memory of pictures, words and sentences and
found that people recognized pictures much better than words or sentences.
Standing (1973) presented 10,000 pictures to his subjects and found that they
could recognize 83 percent of them, a recognition rate that was impossible to
achieve with words or sentences.
Nisbett and Ross (1980) argued that people made judgements on the basis of
information available. As vivid information was more effectively encoded, it was
more retrievable and more readily available; therefore, vivid information had a
greater impact on people's attitudes than non-vivid information.
Another reason given for the superiority of visual information is that it is
encoded more extensively. Kisielius and Sternthal (1984) explained that the
vividness of visual stimuli increased the number of "associative pathways in
memory" and that the more associative pathways there were, the more easily an
individual could gain access to the information (p. 55).
In advertising, some researchers (Childers et al., 1986) agreed that pictorial
ads would be "more elaboratively encoded in memory" (p. 139). They found that
visual information was not only better recalled but also better recognized than
verbal information. Smith (1991) claimed that
The Power of Words
when words and pictures in ads focused on different product attributes, pictures
disproportionately influenced inferences.
However, the notion of visual information being superior to verbal information
apparently ran against some theories of learning and memory. In the theory of
automaticity, Logan (1988) maintained that the more practice a person had in
carrying out a task, the quicker the person would be in doing the same task,
because the person accumulated more strength in his memory with each practice.
Schneider and Shiffrin (1977) proposed the concept of automatic versus
controlled processing of information. They maintained that automatic processing
operated through "a relatively permanent set of associative connections in
long-term store" without "active control or attention by the subject," while the
controlled processing was "a temporary sequence of nodes activated under control
of, and through attention by the subject" (p. 2). The main consequence of an
automatic process is that it dominates in a dual-task (Pashler, 1998), such as
the task of processing reading words and recognizing pictures.
Reading words is considered a learned process and therefore an automatic one
(Stahl, 1991). This automatic process was demonstrated by Healy (1980) in her
experiments on people's detection of misspellings in the word the. Beck et al.
(1982) also found that words that received more practice were responded to more
rapidly on semantic decision tasks than words that received less practice.
Apparently, word reading, a much practiced task, is superior to picture
recognition, a task that needs little practice, according to the theory of
automaticity.
Applying the theory of automaticity, MacLeod (1991) argued that word reading was
"very automatic" because it was a learned process, while processing pictorial
stimuli was
The Power of Words
controlled and "much less automatic" (p. 189). Therefore, most people had "an
irresistible urge to read the word" (p. 186) because automaticity gave "a
greater role for the left hemisphere" (p. 190).
This argument was supported by Edell and Staelin (1983) who found that a visual
ad resulted in consumers' "generating fewer evaluative thoughts" (p. 47). They
concluded that "when subjects saw the unframed pictorial (picture only)
advertisements, they seemed to forget or to be distracted from their task of
evaluating the brand presented in the ad" (p. 59). In Shepard's (1967) study,
people's recognition of pictures was found to decline dramatically after one
week or more time.
Verbal information was even considered as the sole source of attitudes by
Fishbein and Ajzen (1975), who proposed the expectancy-value model and argued
that verbal information was the "basis of attitude" (p. 222). Their formula of
attitude formation is that "a person's attitude toward an object is a function
of his beliefs about the object's attributes and his evaluation of those
attributes" (p. 253).
Verbal Ads Versus Verbal-Visual Ads
Different views also existed on the effect of a combination of verbal and visual
stimuli. Paivio (1971) proposed the "paired-associate learning" paradigm (p.
329) and maintained that a meaningful association between a verbal claim and a
picture would facilitate the recall or retrieval of information from memory.
Bower et al. (1975) compared people's memory of simple pictures with that of the
pictures attached with explanatory labels. They found that people recognized
labeled pictures much better.
The Power of Words
Some advertising researchers produced similar findings. Using ads from the
Yellow Pages, Lutz and Lutz (1977) found that pictures integrated with brands'
names facilitated the recall of the brands. Rossiter and Percy (1978) found that
changing the size of the picture (large and small) in a print ad could create
different attitudes toward the advertised product.
Mitchell and Olson (1981) associated a picture of a fluffy kitten, connoting
"softness" (p. 322), with a fictitious brand of facial tissue in an ad, and
found that the ad was more favorably rated and considered to be very soft,
compared with ads using pictures of a sunset and an abstract painting.
Similarly, Mitchell (1986) found that positively evaluated pictures created more
favorable attitudes toward the ad and the brand, whereas the negatively
evaluated pictures created unfavorable effects on both attitudes. Childers and
Houston (1984) found that ads with word-picture combinations had a stronger
effect in recall than word-only ads at the sensory level. Even ads with
discrepant words and pictures produced more positive effect on recall (Houston
et al., 1987).
However, some (Gottlieb et al., 1977) found no difference between the effect of
the verbal-visual message and that of a verbal message. Edell and Staelin (1983)
found no difference between a verbal ad and a verbal-visual ad on the measures
of the brand attitude, recall, the attitude toward purchasing the brand, and the
purchase intention, Childers and Houston (1984) found no difference in recall
when the ads were processed at a semantic level.
Taylor and Thompson (1982) made a comprehensive review of studies on the
"vividness" effect and concluded that "pictorially illustrated information is no
more persuasive than is equivalent information that is not pictorially
illustrated; videotaped information has no consistent
The Power of Words
impact on judgments compared with equivalent oral or written information" (p.
170).
Kisielius and Sternthal (1984) explained the no-difference phenomenon as the
"ceiling effect" (p. 59). They argued that people had limited cognitive
resources, which were "available when the verbal information is presented alone,
but not when verbal information is accompanied by pictorial analogs" (p. 60).
Furthermore, an ad with both verbal and visual information could have a less
positive effect than a verbal ad. Kisielius and Sternthal (1984) found that "in
the absence of instructions to image, the verbal presentation was found to
induce a more favorable evaluation than the verbal presentation with pictorial
analogs" (p. 58). In their study, a shampoo ad with only verbal information was
rated more favorably than a shampoo ad with both verbal and visual components.
They concluded that "in some instances, verbal information alone induced a more
favorable judgment than verbal information accompanied by pictures" (p. 61).
Childers and Houston (1984) noticed that the use of color in print ads "was more
of a detriment to recall than a facilitator, especially under semantic
conditions. When semantic processing occurs, color may interfere with or detract
from elaborative processing. With less processing to interfere with under
sensory processing, this detrimental effect may be reduced" (p. 653). Pictures
in ads might produce unwanted "cognitive elaboration" and reduce the
favorableness of judgment (Kisielius & Sternthal, 1984, p. 59).
The interference between stimuli was also reported in other areas. Bither and
Wright (1973) found that adding incongruent video images to audio information
reduced people's recall of the audio information. McClure and Patterson (1976)
found that television viewers could
The Power of Words
describe what they saw but could remember little about what they heard. Their
conclusion was that "the visual dimension of television overrides, and perhaps
even blots out, the audio dimension" (p. 26).
The interference between verbal and visual stimuli has been extensively studied
within the dual-task paradigm, which predicted that the reverse Stroop effect
was likely to occur when pictorial stimuli were processed together with words
(Dyer & Severance, 1972; Dyer, 1973), as compared with the Stroop effect,
meaning that verbal information processing interfered with visual information
processing (MacLeod, 1991).
Research Question and Hypotheses
Reconciling the opposing views on the effects of verbal and visual stimuli,
Krugman (1977) maintained that the processing of print ads, a left-brain
function, occurred in "high involvement" situations while and the perception of
visual ads, a right-brain function, occurred in "low involvement" situations (p.
8). Hansen (1981) agreed that involvement might be a "critical variable"
determining whether verbal or visual information processing dominated or not. He
argued that under high involvement conditions, verbal information was processed
and deliberate choice was made, while under low involvement conditions,
pictorial material was processed without much verbal coding.
Petty and Cacioppo (1981) posited in their elaboration likelihood model that
highly- involved people tended to take the central route and engaged themselves
in "diligent consideration of issue-relevant arguments," while lowly-involved
people tended to take the
The Power of Words
peripheral route and rely on non-verbal cues, such as visual images (p. 847).
Chaiken (1980) also maintained in her heuristic-systematic model that highly
involved people would take a systematic route by which they devoted considerable
cognitive effort to processing verbal information, while people who did not
consider an issue to be of much personal relevance would "employ a more economic
heuristic strategy" and rely on non-content cues (p. 762).
Apparently, verbal information processing should dominate in high involvement
situations while visual information processing should prevail in low involvement
situations. Therefore, this study hypothesizes:
1. Verbal ads have a more positive effect than visual ads in the high
involvement situation.
2. Visual ads have a more positive effect than verbal ads in the low involvement
situation.
3. Ads with verbal-visual combinations have a more positive effect than visual
ads in the high involvement situation. This is based on the assumption that the
verbal-visual combination ad has a verbal component which is expected to
dominate in high involvement situations.
4. Ads with verbal-visual combinations have a more positive effect than verbal
ads in the low involvement situation. This is based on the assumption that the
verbal-visual combination ad has a visual component which is expected to
dominate in low involvement situations.
Method
Subjects
198 undergraduate students from three classes in a mid-western university
participated in
The Power of Words
the experiment for extra points toward their grades. 150 of the subjects, mostly
freshmen and sophomores, were from an introductory class of mass communication
and the rest, juniors and seniors, were from one class of advertising research
and one class of advertising psychology. The ratio of male and female students
was 77 to 121.
Experimental Design
This study used a two-factor design. The first factor, involvement, was
manipulated into two levels: high and low. The high involvement situation was
created by using a fictitious brand of computer as the product in the ads. The
low involvement situation was created by using a fictitious brand of pencil in
the ads. As Mueller (1987) argued, high-involvement products were those which
were "higher in price, purchased relatively infrequently," and required "some
pertinent information-searching by the consumer," while the low involvement
products tended to be "packaged goods of a relatively low price" and were
"purchased frequently by the consumer" (pp. 53-54). The second factor,
advertisements, was manipulated into three types: verbal, visual, and
verbal-visual combination.
Stimulus Materials
Ads were designed on computer. The verbal ads contained a description of the
attributes of a fictitious brand of computer for the high involvement situation
and a description of the attributes of a fictitious brand of pencil for the low
involvement situation. The visual ad contained a picture of a computer for the
high involvement situation and a picture of a pencil for the low involvement
situation. The verbal-and-visual ads contained both a picture and a description.
All ads for a computer contained the slogan: Minstar Computer Will Make Your
The Power of Words
University Life a Success. All ads for a pencil contained the slogan: Minstar
Pencil Will Make Your University Life a Success. All ads were black and white
and were of the same size, occupying about 60 percent of a sheet of 8" x
11"-size paper.
In this study, pictures of products were used instead of "affect-laden" pictures
(Mitchell, 1986, p. 13). Those affect-laden pictures in previous studies might
be interpreted differently by different people, raising the question of
validity. For example, pictures of a sunrise, a cat, a kitten, a cropped field
(Mitchell & Olson, 1981; Mitchell, 1986), and the Superman holding a fence
(Childers et al., 1986) could be interpreted very differently by different
consumers.
Procedure
Experiments were conducted at the end of each of the three classes' lectures.
Based on the three types of ads in both high and low involvement conditions,
subjects were randomly assigned into six groups: high verbal (N=32), high visual
(N=33), high verbal-visual (N=36), low verbal (N=32), low visual (N=32), and low
verbal-visual (N=33). Each subject received an ad of one of the six types.
Subjects were instructed to imagine a situation in which they were going to
purchase either a computer or a pencil, to look at the given ad for two minutes,
and then to fill out a questionnaire measuring the ad attitude, the brand
attitude, the recall of brand attributes, and the purchase intention. Subjects
were instructed not to look back at the ad while finishing the questionnaire.
The whole process took ten minutes on average.
Dependent Variables
The dependent variables in this study were the ad attitude, the brand attitude,
the recall of brand attributes, and the intention of purchasing the brand. The
ad attitude was measured by the
The Power of Words
mean of four unipolar seven-point scales: bad-good, dislike-like,
boring-interesting, and uncreative/creative (Biehal et al., 1992). The brand
attitude was measured by the mean of four unipolar seven-point scales: bad-good,
dislike-like, unpleasant-pleasant, and poor quality-good quality (Edell &
Staelin, 1983; Biehal et al., 1992). The recall of brand attributes was measured
by the number of attributes listed by the subjects. The purchase intention was
measured by one unipolar seven-point scale: not at all likely to buy-very likely
to buy (Edell & Staelin, 1983). The reliability coefficient (alpha) for the
scales measuring the ad attitude and the brand attitude were respectively .921
and 0.898.
Results
Main Effects
Based on the two factors: type of ads (verbal, visual and verbal-visual) and
involvement (high and low), subjects were grouped into six cells: high verbal,
high visual, high verbal-visual, low verbal, low visual, and low verbal-visual.
The ANOVA results showed that there was significant main effect among the six
groups on the measures of the brand attitude: F(5, 197)=3.542, p=.004, the
recall of attributes: F(5, 197)=43.083, p<.0001, and the purchase intention:
F(5, 197)=4.083, p=.002. But there was no significant effect on the measure of
the ad attitude: F(5, 197)=2.084, p=.069.
Within the factor of ad type (verbal: N=64, visual: N=65, and verbal-visual:
N=69), there was significant main effect on the measures of the brand attitude:
F(2, 197)=7.094, p=.001; recall: F(2, 197)=93.848, p<.0001; intention: F(2,
197)=5.503, p=.005. But there was no
The Power of Words
significant effect on the ad attitude: F(2, 197)=2.908, p=.057.
Within the factor of involvement (high involvement: N=101, and low involvement:
N=97), there was significant main effect on the measure of recall: F(1,
197)=7.57, p=.006; and intention: F(1, 197)=8.823, p=.003. But there was no
significant effect on the ad attitude: F(1, 197)=.054, p=.817; and the brand
attitude: F(1, 197)=.66, p=.418.
A check on the gender difference showed that male subjects (N=77) and female
subjects (N=121) did not differ on all of the four measures: the ad attitude
(F=.023, p=.88), the brand attitude (F=2.155, p=.144), recall (F=1.079, p=.3),
and intention (F=.177, p=.674).
Test of Hypotheses
Hypothesis One stated that verbal ads would have a more positive effect than
visual ads in the high involvement situation. It was supported on the measures
of the brand attitude, recall and intention, but not on the ad attitude. Verbal
ads (N=32) had significantly higher means compared with visual ads (N=33) on
three measures (the brand attitude: t=2.241, p=.029; recall: t=6.779, p<.0001;
and intention: t=2.411, p=.019). However, on the ad attitude, verbal ads had a
lower mean than visual ads (t=-1.567, p=.122), but the difference was not
significant (see Table 2).
Hypothesis Two stated that visual ads would have a more positive effect than
verbal ads in the low-involvement situation. This was not supported by the data,
which showed that the opposite was true. The verbal ads (N=32) had significantly
higher means than the visual ads (N=32) on three measures except on the ad
attitude: the brand attitude (t=3.018, p=.004), recall (t=11.53, p<.0001),
intention (t=2.316, p=.024), and the ad attitude (t=.322, p=.749) (see Table
The Power of Words
3).
Hypothesis Three stated that ads with verbal-visual combinations would have a
more positive effect than visual ads in the high involvement situation. This
hypothesis was an extension of Hypothesis One and was expected to be supported.
However, the combination of words and pictures led to the disappearance of
significance on two measures: the brand attitude (t=.254, p=.8) and the purchase
intention (t=1.327, p=.189). Prior to the combination, the comparison between
verbal and visual ads was significant on the brand attitude (t=2.241, p=.029)
and the purchase intention (t=2.411, p=.019). Only on the measure of recall was
significance retained (4.222>1.727, t=7.299, p<.0001). On the measure of the ad
attitude, there was no significant difference (3.424>2.833, t=1.736, p=.087)
(see Table 4).
The effect of adding a picture to a print ad was also shown by data in Table 5,
which compared the effect of verbal ads (N=32) with that of verbal-visual ads
(N=36) in the high involvement situation. On the ad attitude, there was high
significance (2.328<3.424, t=3.69, p<.0001), indicating a significant positive
effect of adding pictures to words on the ad attitude. However, there was
negative effect coming from adding pictures to words on the brand attitude
(t=1.904, p=.061) and intention (t=.993, p=.325). Such negative changes were not
shown to be significant because the verbal-visual ads also contained a verbal
component. No significant change was found on recall (t=-.305, p=.761).
Hypothesis Four stated that ads with verbal-visual combinations would have a
more positive effect than verbal ads in the low involvement situation. This was
not supported. In the low involvement situation, no significance was found on
all four measures: the ad attitude (t=-
The Power of Words
.145, p=.885), the brand attitude (t=.705, p=.484), recall (t=.07, p=.945), and
intention (t=.973, p=.334). (see Table 6). On the contrary, data in Table 7
showed that adding pictures to words led to the disappearance of significance on
the measure of the purchase intention (t=1.554, p=.125). Prior to the
combination, the comparison between verbal and visual ads in the low involvement
was significant on intention (t=2.316, p=.024).
Further Analysis
The ANOVA results on the hypotheses indicated that verbal, visual, and
verbal-visual ads had different effects on different measures: the ad attitude,
the brand attitude, recall of attributes, and the purchase intention, both in
high involvement and low involvement situation. In the practical world of
advertising, the ultimate concern is how well an ad serves the purpose of
persuading consumers to make a purchase. Therefore, this study conducted further
analysis on the relationship between the measure of the purchase intention and
the other three measures: the ad attitude, the brand attitude, and recall of
attributes.
It was found that the Pearson correlations (bivariate) between the purchase
intention and the other three measures in all cases (N=198) were significant at
.01-level. The highest was the brand attitude (r=.531, p<.01), followed by the
ad attitude (r=.309, p<.01) and recall (r=.202, p<.01). However, when the
purchase intention was correlated with one measure while the other two measures
were controlled for, the only significant partial correlation was the one
between the purchase intention and the brand attitude (r=.4424, p<.01) (see
Table 8).
Dividing all the cases into high and low involvement groups, this study repeated
bivariate and partial correlations between the purchase intention and the other
three measures and found
The Power of Words
the same results. In the high involvement situation, the highest bivariate
correlation was the brand attitude (r=.572, p<.01), followed by the ad attitude
(r=.327, p<.01) and recall (r=.18, n.s.); in the low involvement situation, the
highest bivariate correlation was the brand attitude (r=.495, p<.01), followed
by recall (r=.338, p<.01) and the ad attitude (r=.299, p<.01). However, partial
correlation analysis showed that in both high involvement and low involvement
situations, the only significant correlation was the brand attitude (high
involvement: r=.505, p<.01; low involvement: r=.3405, p<.01) (see Tables 9 and
10).
Conclusion and Discussion
This study found that verbal stimuli are more powerful and effective than visual
stimuli, confirming the theory of automaticity (Logan, 1988; Schneider &
Shiffrin, 1977) that the automatic process, word reading, is superior to the
controlled process, picture recognition.
In both the high involvement and low involvement situations, a verbal ad (words
only) is more effective than a visual ad (a picture with a brand name) in
creating a favorable attitude toward the product, enhancing consumers' recall of
the product's attributes, and creating a purchase intention. However, a verbal
ad might not be as effective as a visual ad in creating a favorable attitude
toward the advertisement in the high involvement situation. In the low
involvement situation, a verbal ad is just effective as a visual ad on the ad
attitude.
Combining a picture with words in a print ad was found to have both negative and
positive effects in the high involvement situation. On one hand, the picture
clearly interferes with the effect of words on consumers' brand attitude and
purchase intention. This was proved by the
The Power of Words
finding that statistically significant difference between verbal and visual ads
on the mentioned two measures disappeared after a picture was combined with
words. Prior to the combination of words and a picture, a verbal ad has a
significantly stronger effect than a visual ad on the brand attitude and
purchase intention. On the other hand, adding a picture to words in a print ad
has a positive effect in the high involvement situation: it creates a more
favorable attitude toward the advertisement per se.
In the low involvement situation, adding a picture to words in a print ad does
not appear to have any positive effect on the ad attitude, the brand attitude,
recall, and the purchase intention. On the contrary, adding a picture has a
negative effect on the purchase intention in the low involvement situation.
This study found that it is the brand attitude that is predicative of consumers'
purchase intention. In other words, the more positive effect on the ad attitude
as a result of adding a picture to words in a print ad is, in the practical
world of advertising, meaningless, while words, which exert a more positive
effect on the brand attitude, are much more useful, because a favorable brand
attitude is crucial in persuading a consumer to make a purchase decision. This
finding would provide advertisers with a rationale for focusing on providing
verbal information, as consumers' knowledge of a product depends on it
(Kisielius & Sternthal, 1984).
This study also produced evidence for the reverse Stroop effect (Dyer, 1973),
indicating that pictorial stimuli may interfere with the cognitive processing of
verbal stimuli. Schneider and Shiffrin (1977) termed the phenomenon as
"selective-attention deficit," meaning that "additional inputs reduce
performance" (p. 4). Edell and Staelin (1983) pointed out, adding an unrelated
The Power of Words
picture to words in a print ad may result in consumers' generating "fewer
evaluative thoughts" and recalling "smaller number of brand items" (p. 59). In
this study, a related picture: a picture of a computer or a pencil accompanied
with a description of attributes of the computer or pencil, also acted as
distractors rather than facilitators on consumers' brand attitude, recall of
attributes and the purchase intention.
The findings of this study indicate that the assumption of pictures selling
products better than words is a flawed one and that the pervasive use of
pictures in print ads might be unjustified and unnecessary.
Findings of this study also has theoretical implications. The argument of the
elaboration likelihood model that people of high involvement would devote
cognitive effort to processing verbal information while people of low
involvement rely on peripheral cues such as visual images was not supported in
this study. In both the high and low involvement situations, subjects' attention
was drawn to the words in ads. Furthermore, subjects were more attentive to
visual component of a verbal-visual ad in the high involvement situation than in
the low involvement situation. This indicates that people of high involvement
devote their mental effort to both the verbal and visual components of an ad
while people of low involvement apparently devote much less of their mental
resources to processing incoming stimuli. The effect of pictures in the high
involvement was explained by Petty and Priester (1994) as a result of the "shift
rule" (p. 113), meaning that a peripheral cue transformed itself into a message
argument. However, the shift rule seemed to emphasize the shift process from one
mode to another. This study found that it was an expansion process, ,meaning
that people of high involvement take multiple routes,
The Power of Words
instead of one or another, to processing incoming information.
Therefore, the mindful-mindless model raised by Langer and Imber (1980) seems to
be appropriate in explaining the state of mind of people in high involvement
situation. They posited that mindless people received incoming messages
passively and did not seek distinctions among stimuli, while mindful people
actively manipulated different elements of stimuli, questioned "old categories,"
or constructed new ones (p. 360). Therefore, it seems inevitable that highly
involved consumers attend to both verbal and visual components of print ads.
This study has its limitations. It did not examine all of the functions the
visual component in print ads may have. It is likely that pictures, though not
very effective in changing consumers' brand attitude, may be more effective than
words in catching consumers' attention to print ads, i.e., inducing awareness.
References
Anderson, J. (1995). Learning and memory: An integrated approach. New York:
Wiley.
Baker, S. (1961). Visual persuasion. New York: McGraw-Hill.
Beck, I., Perfetti, C., & McKeown, M. (1982). Effects of long-term vocabulary
instruction on lexical access and reading comprehension. Journal of Educational
Psychology, 74, 506-521.
Bither, S., & Wright, P. (1973). The self-confidence--advertising response
relationship: A function of situational distraction. Journal of Marketing
Research, 10, 146-152.
Borgida, E., & Howard-Pitney, B. (1983). Personal involvement and the robustness
of perceptual salience effects. Journal of Personality and Social Psychology,
45, 560-570.
Bower, G. (1972). Mental imagery and associative learning. In L. Gregg (Ed.),
Cognition in learning and memory. New York: Wiley.
Bower, G., Karlin, M., & Dueck, A. (1975). Comprehension and memory for
pictures. Memory and Cognition, 3, 216-220.
Chaiken, S. (1980). Heuristic versus systematic information processing and the
use of source versus message cues in persuasion. Journal of Personality and
Social Psychology, 39 (5), 752-766.
Childers, T., Heckler, S., & Houston, M. (1986). Memory for the visual and
verbal components of print advertisements. Psychology and Marketing, 3 (3),
137-150.
Childers, T., & Houston, M. (1984). Conditions for a picture-superiority effect
on consumer memory. Journal of Consumer Research, 11, 643-654.
Dyer, F. (1973). The Stroop phenomenon and its use in the study of perceptual,
cognitive, and response processes. Memory and Cognition, 1, 106-120.
Dyer, F., & Severance, L. (1972). Effects of irrelevant colors on reading of
color names: A controlled replication of the "reversed Stroop" effect.
Psychonomic Science, 28, 336-338.
Edell, J., & Staelin, R. (1983). The information processing of pictures in print
advertisements. Journal of Consumer Research, 10, 45-61.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An
introduction to theory and research. Reading, Mass: Addison-Wesley.
Gottlieb, D., Taylor, S., & Ruderman, A. (1977). Cognitive bases of children's
moral judgments. Developmental Psychology, 13, 547-556.
Hansen, F. (1981). Hemispheral lateralization: Implications for understanding
consumer behavior. Journal of Consumer Research, 8, 23-36.
Healy, A. (1980). Proofreading errors on the word the: New evidence on reading
units. Journal of Experimental Psychology: Human Perception and Performance, 6,
45-57.
Houston, M., Childers, T., & Heckler, S. (1987). Picture-word consistency and
the elaborative processing of advertising. Journal of Marketing Research, 24,
359-369.
Karsen, N., & Ingver, D. (1972). Radioisotopic assessment of regional cerebral
bloodflows. In Progress in Nuclear Medicine, 1. Baltimore: University Park
Press.
Kisielius, J., & Sternthal, B. (1984). Detecting and explaining vividness
effects in attitudinal judgments. Journal of Marketing Research, 21, 54-64.
Krugman, H. (1977). Memory without recall, exposure without perception. Journal
of Advertising Research, 17 (4), 7-12.
Langer, E., & Imber, L. (1980). Role of mindlessness in the perception of
deviance. Journal of Personality and Social Psychology, 39 (3), 360-367.
Logan, G. (1988). Toward an instance theory of automatization. Psychological
Review, 95, 492-527.
Lutz, K., & Lutz, R. (1977). Effects of interactive imagery on learning:
Application to advertising. Journal of Applied Psychology, 62, 493-498.
McClure, R.D., & Patterson, T.E. (1976). Setting the political agenda: Print vs.
network news. Journal of Communication, 26 (2), 23-28.
Middlestadt, S., Fishbein, M., & Chan, D. (1995). The effect of music on brand
attitudes: Affect- or belief-based change? In E. Clark, T. Brock & D. Stewart
(Eds.), Attention, attitude, and affect in response to advertising (pp.
149-167). Hillsdale, NJ: Lawrence Erlbaum.
Mitchell, A. (1986). The effects of verbal and visual components of
advertisements on brand attitudes and attitude toward the advertisement. Journal
of Consumer Research, 13, 12-24.
Mitchell, A., & Olson, J. (1981). Are product attribute beliefs the only
mediator of advertising effects on brand attitudes? Journal of Marketing
Research, 18, 318-332.
Mueller, B. (1987). Reflections of culture: An analysis of Japanese and American
advertising appeals. Journal of Advertising Research, 27, 51-59.
Nelson, D., Reed, V., & Walling, J. (1976). The pictorial superiority effect.
Journal of Experimental Psychology: Human Learning and Memory, 2, 523-528.
Nisbett, R., & Borgida, E. (1975). Attributions and the psychology of
prediction. Journal of Personality and Social Psychology, 32, 932-943.
Nisbett, R., & Ross, L. (1980). Human inference: Strategies and shortcomings of
social judgment. Englewood Cliffs, NJ: Prentice-Hall.
Paivio, A. (1971). Imagery and verbal processes. New York: Holt, Rinehart, &
Winston.
Paivio, A., & Csapo, K. (1973). Pictorial superiority in free recall: Imagery or
dual coding? Cognitive Psychology, 5, 176-206.
Pashler, H. (1998). The psychology of attention. Cambridge, Mass: The MIT Press.
Percy, L., & Rossiter, J. (1980). Advertising strategy: A communication theory
approach. New York: Praeger.
Petty, R., & Cacioppo, J. (1981). Attitudes and persuasion: Classic and
contemporary approaches. Dubuque, Iowa: Wm. C. Brown Company.
Petty, R., Cacioppo, J., & Schumann, D. (1983). Central and peripheral routes to
advertising effectiveness: The moderating role of involvement. Journal of
Consumer Research, 10, 135-146.
Petty, R., & Priester, J. (1994). Mass media attitude change: Implications of
the elaboration likelihood model of persuasion. In J. Bryant & D. Zillmann
(Eds.), Media effects: Advances in theory and research (pp. 91-122). Hillsdale,
NJ: Lawrence Erlbaum.
Rossiter, J., & Percy, L. (1978). Visual imaging ability as a mediator of
advertising response. In H. Hunt (Ed.), Advances in consumer research, 5 (pp. ).
Ann Arbor, Mich: Association for Consumer Research.
Rossiter, J., & Percy, L. (1980). Attitude change through visual imagery in
advertising. Journal of Advertising, 9, 10-16.
Santa, J. (1977). Spatial transformations of words and pictures. Journal of
Experimental Psychology: Human Learning and Memory, 3, 418-427.
Schneider, W., & Shiffrin, R. (1977). Controlled and automatic human information
processing: I. detection, search and attention. Psychological Review, 84, 1-66.
Shepard, R. (1967). Recognition memory for words, sentence, and pictures.
Journal of Verbal Learning and Verbal Behavior, 6, 156-163.
Smith, R. (1991). The effects of visual and verbal advertising information on
consumers' inferences. Journal of Advertising, 20, 13-23.
Stahl, S. (1991). Beyond the instrumentalist hypothesis: Some relationships
between word meanings and comprehension. In P. Schwanenflugel (Ed.), The
psychology of word meanings (pp. 157-186). Hillsdale, NJ: Lawrence Erlbaum.
Starch, D. (1966). How does the shape of ads affect readership? Media/Scope, 10,
83-85.
Taylor, S., & Thompson, S. (1982). Stalking the elusive "vividness" effect.
Psychological Review, 89, 155-181.
Weinstein, S. (1982). A review of brain hemisphere research. Journal of
Advertising Research, 22 (3), 59-63.
Table 1
Gender Difference (Male=77; Female=121)
Male Female t df significance
Ad Attitude 2.926 2.895 .151 196 .880
Brand Attitude 4.399 4.662 -1.468 196 .144
Recall of Attributes 3.208 2.942 1.039 196 .300
Purchase Intention 3.688 3.793 -0.421 196 .674
Table 2
High Verbal Group (N=32) versus High Visual Group (N=33)
High Verbal High Visual t df significance
Ad Attitude 2.328 2.833 -1.567 63 .122
Brand Attitude 4.875 4.273 2.241 63 .029
Recall of Attributes 4.125 1.727 6.779 63 .000
Purchase Intention 3.875 2.909 2.411 63 .019
Table 3
Low Verbal Group (N=32) versus Low Visual Group (N=32)
Low Verbal Low Visual t df significance
Ad Attitude 2.953 2.831 .322 62 .749
Brand Attitude 5.031 4.047 3.018 62 .004
Recall of Attributes 3.625 .846 11.530 62 .000
Purchase Intention 4.563 3.594 2.316 62 .024
Table 4
High Verbal-Visual Ads (N=36) versus High Visual Ads (N=33)
High Verbal/Visual High Visual t df significance
Ad Attitude 3.424 2.833 1.736 67 .087
Brand Attitude 4.347 4.273 .254 67 .800
Recall of Attributes 4.222 1.727 7.299 67 .000
Purchase Intention 3.444 2.909 1.327 67 .189
Table 5
High Verbal Ads (N=32) versus High Verbal-Visual Ads (N=36)
High Verbal High Verbal/Visual t df significance
Ad Attitude 2.328 3.424 -3.69 66 .000
Brand Attitude 4.875 4.347 1.904 66 .061
Recall of Attributes 4.125 4.222 -.305 66 .761
Purchase Intention 3.875 3.444 .993 66 .325
Table 6
Low Verbal Ads (N=32) versus Low Verbal-Visual Ads (33)
Low Verbal Low Verbal/Visual t df significance
Ad Attitude 2.953 3.008 -.145 63 .885
Brand Attitude 5.031 4.813 .705 63 .484
Recall of Attributes 3.625 3.606 .070 63 .945
Purchase Intention 4.563 4.182 .973 63 .334
Table 7
Low Verbal-Visual Ads (N=33) versus Low Visual Ads (32)
Low Verbal/Visual Low Visual t df significance
Ad Attitude 3.008 2.831 .460 63 .647
Brand Attitude 4.813 4.047 2.69 63 .009
Recall of Attributes 3.606 .846 11.215 63 .000
Purchase Intention 4.182 3.594 1.554 63 .125
Table 8
Purchase Intention's Pearson Correlations with Ad Attitude, Brand Attitude,
Recall of Attributes (N=198)
Purchase Intention
----------------------------
Bivariate Partial
Ad Attitude .309** .1215
Brand Attitude .531** .4424**
Recall of Attributes .202** .0879
** p < .01
Table 9
Purchase Intention's Pearson Correlations with Ad Attitude, Brand Attitude,
Recall of Attributes in the High Involvement Situation (N=101)
Purchase Intention
----------------------------
Bivariate Partial
Ad Attitude .327** .1481
Brand Attitude .572** .505**
Recall of Attributes .18 .1491
** p < .01
Table 10
Purchase Intention's Pearson Correlations with Ad Attitude, Brand Attitude,
Recall of Attributes in the Low Involvement Situation (N=97)
Purchase Intention
----------------------------
Bivariate Partial
Ad Attitude .299** .1196
Brand Attitude .495** .3405**
Recall of Attributes .338** .1665
** p < .01
|
