Utilizing the case-based reminding theory in analogical reasoning, this research proposes that the

evaluation of a brand extension can be improved by reminding consumers of a similar brand in

the extension category. This effect is derived from a brand-to-brand similarity, in addition to the
product-to-product and brand-to-product similarity identified in prior literature. Experiment 1
explores the idea that the effect of similar case reminders is most pronounced in moderately
similar extensions than in highly similar or highly dissimilar extensions due to schema congruity.
Experiments 2 and 3 distinguish the levels of similarity (i.e., product-to-product, brand-to-
product, and brand-to-brand similarity) as a source of consumers' evaluations on a brand
extension and further eliminate an alternative explanation of instantaneous learning in the
advertisement setup of Experiment 1. © 2010 Wiley Periodicals, Inc.
Consumer evaluations of brand extension have been extensively investigated in consumer
behavior and marketing research. From a psychological perspective, brand extension evaluation
is a categorization process in which the extension evaluation is determined by the perceived
category overlap between the new extension and the core brand or product attributes (Felcher,
Malaviya, & McGill, 2001; Mao & Krishnan, 2006). Among the research related to
categorization, analogical reasoning recently has received attention from the consumer research
community. In a review article, Gregan-Paxton and Roedder (1997) suggested that psychological
research in analogical reasoning may be a fruitful area for consumer research (e.g., Moreau,
Lehman, & Markman, 2001; Moreau, Markman, & Lehman, 2001). Brand extension was one of
the areas on their list that could potentially benefit from psychological research in analogical
reasoning.
The purpose of the present paper is to apply the theory of analogical reasoning to consumers'
brand extension evaluations. Specifically, it is suggested that consumers' evaluations of brand
extension can be manipulated by reminding them of a similar case in the extension category. The
knowledge about the brand in the extension category (i.e., the “case”) can thus be analogically
transferred to the proposed extension, elevating the extension evaluation. The following sections
present a brief literature review, the research hypotheses, and finally the three experiments to test
these hypotheses.

LITERATURE BACKGROUND AND RESEARCH HYPOTHESES

1. Top of page
2. Abstract
3. LITERATURE BACKGROUND AND RESEARCH HYPOTHESES
4. EXPERIMENT 1
5. EXPERIMENT 2
6. EXPERIMENT 3
7. GENERAL DISCUSSION AND CONCLUSIONS
8. REFERENCES
Consumer Evaluations of Brand Extension
Brand extension is a major research interest in the academic community of branding and
consumer studies (Boush, 1993; Hennig-Thurau, Houston, & Heitjans, 2009; Nan, 2002; Oakley
et al., 2008; Shine, Park, & Wyer, 2007; Volckner & Sattler, 2006). Among the topics
concerning consumer evaluations of brand extensions, one central issue is the psychological
mechanisms through which consumers evaluate brand extensions (Kumar, 2005). In prior
literature, there were two major lines of research that explored the determinants of consumers'
brand extension evaluations. The first was product category similarity, which asserted that
consumers' brand extension evaluations are determined by the similarity between the core
product category and the extension product category (Aaker & Keller, 1990; Yeo & Park, 2006).
Because the basis of brand extension evaluation relies on the similarity between two product
categories, it can be called product-to-product similarity. The second was brand concept
consistency, which argued that in addition to the product category similarity, the conceptual
consistency between the brand name and the extension category was also a major factor
determining consumers' brand extension evaluations (Broniarczyk & Alba, 1994; Mao &
Krishnan, 2006; Park, Milberg, & Lawson, 1991). Because this line of research emphasized the
importance of the relationship between specific brand associations and the extension products,
the basis of the similarity for brand extension evaluations can be called the brand-to-product
similarity.
The present study approaches the issue of consumers' evaluations of brand extensions from a
different perspective. As described above, brand extension evaluations are essentially a
psychological process of categorization (Medin & Schaffer, 1978; Rosch & Mervis, 1975;
Rosch, 1978). Overlap in features between the core product/brand and the extension category can
assist the categorization process (Tversky, 1977). Both product category similarity and brand
concept consistency are sources of computing such overlap of features. The source of the
overlapping features in this study is different from those in prior literature. Because the source of
the overlapping features in the present study involves similarities of a core brand to a case brand
in the extension category, it can be called brand-to-brand similarity.
Brand-to-brand similarity has been discussed in the study of the co-branding strategy (Walchli,
2007), where the essential concept was the between-partner congruity. Therefore, the congruity
or fit of the two brands was the major factor in determining consumers' evaluations of a new
product launched by two brands together (Walchli, 2007). However, in the present study the
brand-to-brand similarity is employed from a different perspective, one that goes beyond the
application of co-branding. One brand serves as a cue or a reminding case for the focal brand that
the manufacturer intends to extend to a new product category.
Analogical Reasoning and Case-Based Reminding
Analogical reasoning is a pivotal research area for psychologists interested in the cognitive
process of knowledge transfer. Many theorists (Gentner, 1983; Gilboa & Schmeidler, 1995;
Holyoak, 1985; Holyoak & Thagard, 1989; Kolodner, 1993; Ross, 1987, 1984) have delineated
how internal knowledge was transferred from a base domain to a target domain during the
relational mapping process of analogy (Green et al., 2008; Krawczyk, Holyoak, & Hummel,
2004). Case-based reminding theory is among the various theoretical models of analogy that
emphasized the importance of content in analogical transfer. Case-based reminding theory (Ross,
1987, 1984) argues that when making inductive inferences, people sometimes rely on similarity-
based heuristics rather than normatively valid rules to draw inductive inferences. One such rule
is to use a prior case that is similar to the present case as a basis for judgment, and then draw
inductive inferences by analogically transferring the knowledge of the similar case to the present
case in question. A doctor, for example, may diagnose a patient based on observed symptoms
(Ross, 1984). Instead of using normative rules, such as the Bayesian Rule, the doctor may use a
judgmental heuristic based on the similarities in symptoms between the present patient and a
prior patient (i.e., the “case”). The symptoms of the present patient may remind the doctor of a
similar prior case, and the doctor may analogically transfer the knowledge of the prior patient to
the present patient as the basis of diagnostic judgments.
Previous studies have explored the idea that consumer evaluations of a new product could be
manipulated by reminding consumers of a similar example to enhance the corresponding product
attributes with the analogy process (Goode, Dahl, & Moreau, 2010; Gregan-Paxton & Roedder,
1997; Gregan-Paxton et al., 2002). Therefore, the present research approaches the issue of
consumer evaluations of brand extension from the perspective of case-based reminding and
suggests that the proposed brand extension evaluations can be manipulated by reminding
consumers of a similar brand in the extension category. Consider the following example: Assume
Apple Inc. plans to extend its product lines to the automobile category. According to prior
research in brand extension evaluation, the overlapping features between the personal computer
and the automobile category (i.e., product-to-product similarity), or between the Apple brand
concept and the automobile category (i.e., brand-to-product similarity) are necessary for
consumers to accept this extension. It is not readily apparent that either type of comparative
features between these two products would overlap at all. However, consumers might still find it
appealing to have an Apple car if they were reminded by a case brand, say the Smart car, in the
automobile category. How does this perceived fit come about?
This study suggests that consumers' perceived fit may arise as a result of the aforementioned
case-based reminding process. It is postulated that the existence of a particular car brand like
Smart, which shares a similar brand concept with Apple, may help elevate the evaluation of
Apple cars. Apple produces chic and stylish Mac computers, and other trendy products such as
the iPad and iPhone. Smart is a chic and cute car in the automobile market. That is, the
similarities between Smart and Apple help clarify a defining and desirable characteristic of the
Apple car. An Apple car would look essentially like a Smart car: stylish, fun, colorful, cute,
versatile, and creative. Therefore, reminding consumers of the Smart car can elevate the
evaluation of the Apple car. The brand concept of Smart is analogically transferred in the process
of evaluating the proposed Apple car: An Apple car is just like a Smart car.
The Smart car is a special example because Smart's brand concept is not similar to the
prototypical car (Rosch, 1978). A prototypical car may look like the one in your garage: four-
door, five-seat sedan with a conservative, single color. If Smart did not exist in the market, then
an Apple car might not be seen as a reasonable or desirable product.
Note that in the above hypothesized case, the source of similarity for evaluating the brand
extension occurs between two brands (i.e., Apple and Smart). This brand-to-brand similarity is
different from the product-to-product similarity (Aaker & Keller, 1990) and the brand-to-product
similarity (Broniarczyk & Alba, 1994; Park, Milberg, & Lawson, 1991) discussed in previous
research. The brand-to-brand similarity through case-based reminding can be viewed as a
complementary source to the product-to-product and brand-to-product similarities. However, a
simple demonstration of the effect of case-based reminding would be a result of a mixture of all
three sources of similarities. Thus, it is advantageous to be able to separate and identify the
effects of different sources of similarities.
To summarize, the present study predicts that the brand extension evaluations of consumers who
are reminded of a similar brand in the proposed extension would be more favorable than those
who are not reminded of such a brand. Furthermore, because the basis of the analogical transfer
depends on the perceived similarity (Markman & Gentner, 1993), the present study also attempts
to distinguish the case-based reminding effect due to the brand-to-brand similarity from other
kinds of similarities (e.g., brand-to-product and product-to-product similarities).
The present research consists of three experiments. Experiment 1 first demonstrates the basic
hypothesis of case-based reminding. Experiment 2 and 3 address the issue of a possible
confounding effect in Experiment 1 and further clarify the effects of the different levels of
similarities (i.e., product-to-product, brand-to-product, and brand-to-brand similarities).

EXPERIMENT 1
1. Top of page
2. Abstract
3. LITERATURE BACKGROUND AND RESEARCH HYPOTHESES
4. EXPERIMENT 1
5. EXPERIMENT 2
6. EXPERIMENT 3
7. GENERAL DISCUSSION AND CONCLUSIONS
8. REFERENCES
According to Mandler (1982), people are likely to resolve or make sense of moderate schema
incongruities by enacting minor changes in their mind-sets. Not only are such moderate
incongruities thought to be interesting and positively valued in their own right, but the process of
resolving such incongruities tends to be rewarding. Congruent schemas do not require people to
resolve incongruities, and hence do not provide such rewarding experiences. Thus, congruent
schemas are slightly preferred. Moreover, it is unlikely that extremely incongruent schemas can
be resolved even with intensive processing, and hence are not preferred as much as moderately
incongruent schemas (Meyers-Levy, Louie, & Curren, 1994; Meyers-Levy & Tybout, 1989).
Schema incongruity, then, is expected to interact with the effect of case-based reminding in the
present research. The effect of case-based reminding should be most pronounced with
moderately similar extensions than with either extremely similar or extremely dissimilar
extensions. As consumers strive to resolve a schema incongruity when there is a moderately
similar brand extension, reminders of a similar brand can best help them resolve the incongruity;
this, in turn, improves brand extension evaluations, particularly with respect to perceptions as to
what is the better fit and in generating positive attitudes toward the extension product. The
marginal contribution is diminished with extremely similar or extremely dissimilar extensions.
Extremely similar extensions do not require such reminding, and extremely dissimilar extensions
would not register much consideration of fit even with reminding. Therefore, the main effect of
case-based reminding on the brand extension and its interaction with the similarity between the
original and extension categories are proposed in the following hypotheses:
H1a: Consumers tend to perceive a better fit between the core brand and the extension product
category when they are reminded of an existing brand with brand concepts similar to the core
brand in the extension category than when they are not reminded.
H1b: Consumers tend to perceive a better attitude toward the extension product when they are
reminded of an existing brand with brand concepts similar to the core brand in the extension
category than when they are not reminded.
H2a: The effect of case-based reminding on the perceived fit of the extension is stronger when
consumers face a moderately similar extension than a highly similar extension or a dissimilar
extension.
H2b: The effect of case-based reminding on the attitude toward the extension is stronger when
consumers face a moderately similar extension than a highly similar extension or a dissimilar
extension.
Pretests
A series of pretests were conducted to select experimental materials, including the core brand,
the three extension categories varying in the level of similarities to the product category of the
core brand, and the reminding brand in the extension categories similar to the core brand.
The first pretest was conducted to select the core brand for the experiment. A set of actual brands
that college students were familiar with was first collected by a panel of consumer researchers.
Then two candidate brands with the criteria of familiarity and wide potential extensions were
chosen by the panel: Rolex watches and the Discovery Channel. Forty undergraduate students
from a major university were asked about the awareness, possible extensions, and brand–product
associations of the two candidate brands. The results revealed that Rolex enjoyed higher brand
awareness (100%) than the Discovery Channel (80%), more potential extensions (9 suitable
extensions vs. 4 extensions for the Discovery Channel), and stronger brand–product associations
(96% of respondents associated Rolex with the watch category vs. 66% respondents associated
the Discovery Channel with the TV channel category). Thus, Rolex was selected as the core
brand for the experiment.
The second pretest was conducted to choose three target extension categories. A panel of three
researchers in consumer behavior first selected five possibly suitable extensions for Rolex:
handbags, cosmetics, sunglasses, fountain pens, and whiskey. A new sample of 40 undergraduate
students rated the relative similarities of each extension category to the core watch category.
Statistical tests comparing the differences in the similarity ratings between each pair of products
and between the product and the core product showed that, except for whiskey and cosmetics,
most extension categories were found to be significantly different in their similarity to each
other, as well as to the core product, the watch. Thus, the other three product categories were
chosen as the experimental materials. The highly similar extension was the sunglasses (Mean
similarity to the watch category = 1.45,SD= 2.35). The moderately similar extension was the
fountain pen (Mean similarity = 0.16,SD= 2.81). Finally, the dissimilar extension was the
handbag (Mean similarity = −1.14,SD= 2.63). Scheffé's tests revealed significant differences in
the similarity between each of the three extension categories (Mean difference = 1.29,SD= 0.52,
p < 0.05 between the sunglasses and the fountain pen; Mean difference = 2.59,SD= 0.52, p <
0.01 between the sunglasses and the handbag; Mean difference = 1.30,SD= 0.52, p < 0.05
between the fountain pen and the handbag).
The third pretest provided for the selection of the case brand in the respective extension category
as the reminding brand. The panel first inspected the core brand Rolex and came up with two
brands in each extension category that were judged to be similar to the core brands in certain
respects. Another 40 respondents were asked to provide an unaided recall of brands and then
indicated the similar features between the core brand and the case brand (Medin, Goldstone, &
Gentner, 1993). Because the case-based reminding effect was demonstrated when the reminding
group had higher brand extension evaluations than the non-reminding group, it was essential to
make sure that the non-reminding group would not recall the reminding brand spontaneously. All
case brands qualified in this regard. Furthermore, the attribute “prestige” was found in the brand
associations for Rolex, Giorgio Armani's sunglasses, Mont Blanc's fountain pens, and Louis
Vuitton's handbags. The concept of prestige was mentioned by 90% of participants for Giorgio
Armani, 90.5% for Mont Blanc, and 91.2% for Louis Vuitton. Thus these three case brands (i.e.,
Armani, Mont Blanc, and Louis Vuitton) were selected as the reminding case brands for the
respective three extensions.
Finally, because the case brand reminding was manipulated with a priming procedure in which
respondents viewed a series of ads before they were given the brand extension evaluation task, a
set of ads including both the target ads and filler ads was developed. The fourth pretest
measuring the attitude toward these ads revealed no differential preferences among these ads
[F(5,222) = 0.40, n.s.]. Thus these ads were employed for Experiment 1.
Main Study
Experiment 1 involved two sections. In the first section respondents were asked to evaluate three
print ads. The target ad of the case-based reminding brand was embedded in the series of ads for
the reminding group but not for the non-reminding group. After respondents filled out some
questions about these ads, they moved on to the second section to evaluate brand extensions.
Design and Procedure
The experiment was a 2 (case brand reminding: yes or no) × 3 (product similarity: highly similar,
moderately similar, and dissimilar extension) between-subject factorial design. The core brand
Rolex was employed in all conditions. The highly similar extension was the Rolex sunglasses.
The moderately similar extension was the Rolex fountain pen. The dissimilar extension was the
Rolex handbag. For the reminding case brand, Giorgio Armani was used for the sunglasses
extension, Mont Blanc was used for the fountain pen extension, and Louis Vuitton was used for
the handbag extension.
Participants were given an experimental booklet which opened with a general instruction that
this was a study on consumer behavior. The first part of the experimental booklet consisted of
three print ads. For the reminding group, one of the ads was the target ad of the case-based
reminding brand. The other two ads were filler ads. For the non-reminding group, all three ads
were filler ads. The orders of the ads were randomly arranged. Participants first viewed the ads
and then answered questions about both the content of the ad and their attitudes toward the ad
and the product, such as the design, layout, and information amount. These questions were
designed to engage participants to elaborate on the information in ads more thoroughly.
After they finished rating the three ads in the first section, they moved on to the second section,
which was framed as a new study independent of the first one. Participants were first given the
simple definition of brand extension. They then evaluated the proposed extension for Rolex. The
dependent measures were five items of consumers' perceptions of extension fit (viz., “fit,”
“reasonable,” “connected,” “associated,” and “understandable”) and six items of their attitudes
toward the proposed brand extensions (viz., “likeable,” “attractive,” “of good quality,”
“recommended,” “acceptable,” and “purchase intention”) (Aaker & Keller, 1990; Bhat & Reddy,
2001; Bruke & Edell, 1989; Chen & Li, 1999; Klink & Smith, 2001; Sheinin & Schmitt, 1994).
Seven-point semantic differential scales were employed in which the middle point 0 represented
a neutral attitude. Positive numbers from +1 to +3 represented positive evaluations, and negative
numbers from −1 to −3 represented negative evaluations. The last part of the questionnaire was
the manipulation check, in which participants' brand associations for Rolex and other possible
extensions for Rolex were measured. The brand awareness and associations of the case-based
reminding brands as well as the similarities between the core product (i.e., the wrist watch) and
the three extension categories were also collected.
Results
Two hundred forty undergraduate students from a major university were recruited to participate
in the main study for extra credit in a course. The valid samples were those who had heard of the
core brand (i.e., Rolex) and the reminding case brand (i.e., Giorgio Armani, Mont Blanc, and
Louis Vuitton) prior to the experiment, resulting in 160 valid participants.
Manipulation Check and Reliability.
Pairwise comparisons of the similarities between the core product and the three extensions using
Scheff's tests revealed significant differences between the highly similar, moderately similar, and
dissimilar extensions. (Mean difference = 2.25,SD= 0.73, p < 0.05 between the sunglasses and
the fountain pen; Mean difference = 4.20,SD= 0.73, p < 0.01 between the sunglasses and the
handbag; Mean difference = 1.95,SD= 0.73, p < 0.05 between the fountain pen and the handbag).
The similarity between the watch and the sunglasses was 2.35 (SD = 1.84) on a 7-point scale
from +3 to −3; the similarity between the watch and the fountain pen was 0.1 (SD = 2.38); and
the similarity between the watch and the handbag was −1.85 (SD = 2.62). In addition, both the
core brand, Rolex, and the three reminder brands enjoyed high awareness among participants.
They were also associated with the notion of “prestige,” as expected (92% of the participants
associated Giorgio Armani with prestige, 84% associated Mont Blanc with prestige, and 90%
associated Louis Vuitton with prestige). Thus, the manipulations were successful.
Consumers' attitudes toward the ads were not significantly different [F(5,699) = 2, n.s.].
Moreover, the Cronbach's alphas for both the dependent measures of the perceived fit
(Cronbach's alpha = 0.92) and the attitudes toward the proposed extensions (Cronbach's alpha =
0.86) showed that the reliabilities for both measures were high enough for further analysis.
Hypothesis Testing.
A 2 × 3 ANOVA was performed on the experimental data. Two dependent measures, the
perceived fit of the new extension and the attitude toward the proposed extension, were analyzed
separately. The results revealed, for the perceived fit, a significant main effect for case-based
reminding [F(1,154) = 8.68, p < 0.01], a significant main effect for the product similarity
[F(2,154) = 5.10, p < 0.01], and a significant interaction effect between the two factors [F(2,154)
= 4.17, p < 0.05]. Similarly, for the attitude toward the proposed extension, there was a
significant main effect for case-based reminding [F(1,154) = 5.86, p < 0.05], a significant main
effect for product similarity [F(2,154) = 3.36, p < 0.05], and a significant interaction effect
between the two factors [F(2,154) = 4.54, p < 0.05].
Further analysis addressed the specific hypotheses. The main effect for case-based reminding
was significant. The mean of the perceived fit for the reminding group was 0.79 (SD = 1.31),
which was significantly higher than the perceived fit for the non-reminding group, with a mean =
0.26 (SD = 1.50), supporting H1a. The mean of the attitude toward the new extension for the
reminding group was 0.22 (SD = 1.13), which was also significantly higher than the non-
reminding group, with a mean = −0.15 (SD = 1.25), also as expected in H1b.
To examine the moderating role of product similarities, planned contrasts were employed to
compare the differences between reminding and non-reminding in three different product
similarity conditions. The results showed that the case-based reminding effect on the perceived
fit was larger with the moderately similar extension than with the highly similar extension (F =
7.88, p < 0.01). The effect of reminding on the perceived fit of the moderately similar extension
was also higher than that of the dissimilar extension (F = 4.05, p < 0.05). A similar pattern was
also found for attitudes toward new extensions. The effect of case-based reminding on attitudes
toward the extension with the moderately similar extension was more positive than those with
the highly similar extension (F = 8.44, p < 0.01). The effect of reminding on attitudes toward the
new extension with the moderately similar extension was also more positive than those with the
dissimilar extension (F = 4.71, p < 0.05). Thus, the effect of case-based reminding on brand
extension evaluations was more pronounced with the moderately similar extensions than with
either highly similar or dissimilar extensions. H2a and H2b are both supported. Table 1 lists the
mean values and standard deviations of the brand extension evaluations in the respective product
similarity conditions.
Table 1. Mean Values and Standard Deviations of Brand Extension Evaluations in the
Respective Product Similarity Conditions of Experiment 1.

Highly Similar Moderately Similar Dissimilar

Extension Extension Extension

1. Notes: Standard deviations are in parentheses.

2. Upper and lower case superscripts represent a significant difference in the pair
comparison.
Case brand Perceived fit 0.74 (0.96)A 1.57 (0.89)B 0.23 (1.60)C

Extension
reminding 0.18 (1.15)D 0.81 (0.93)E −0.20 (1.09)F
attitude

No case
Perceived fit 0.71 (1.20)A 0.09 (1.72)b −0.16 (1.45)C
brand

Extension
reminding 0.25 (1.58)D −0.40 (1.40)e −0.41 (1.05)F
attitude

Discussion
Experiment 1 found that reminding consumers of a similar case brand in the extension category
can improve the brand extension evaluations. This reminding effect was more pronounced with a
moderately similar extension than with either a highly similar or a dissimilar extension.
Respondents who were reminded of the Mont Blanc fountain pen (i.e., a moderately similar
extension category) would have a higher evaluation of a Rolex fountain pen. This effect of
reminding respondents of a similar case diminished with a highly similar extension (sunglasses),
or with a dissimilar extension (handbags). The data of Experiment 1 provided support for the role
of case-based reminding in elevating consumers' brand extension evaluations.
Nonetheless, two issues are noteworthy. First, as was argued above, the case-based reminding
stimuli should elicit brand knowledge from consumers' stored memory, which would then be
used to evaluate the proposed brand extension. The prior brand knowledge, it was believed, was
ensured by deleting those who did not know the core or case brand before the experiment.
However, one possibility resulting from the participants in Experiment 1 being given print ads as
the reminding stimuli before they were given the brand extension questions is that participants
could have made the “prestige” connection directly from the ad picture and not from prior
knowledge, as delineated in the case-based reminding theory. Thus, an alternative explanation is
to attribute the effect found here to an instantaneous learning of the content in the print ads (i.e.,
a stimulus-based response), rather than to respondents' memories, as is required by case-based
reminding. In such instances, case-based reminding may not be the underlying process in
Experiment 1. Thus, if the integrity of the position is to be maintained, it is essential to eliminate
the possibility of this alternative explanation. One way to eliminate the confounding explanation
of instantaneous learning would be to use a simple description of the reminding case without
pictorial information, which delivers the image of the reminding brand. Without the picture
generating the unwanted stimulus-based learning (i.e., a picture of a prestigious fountain pen by
Mont Blanc), if the participants did not know the case brand prior to the experiment or the case
did not bring out the analogy, no effect of the brand-to-brand similarity would be revealed.
Second, case-based reminding is a form of analogical reasoning that depends on the perceived
similarity between the case brand and the target brand. In the present setting, there are various
bases of similarity (Yoo & MacInnis, 2004) responsible for the observed brand extension
evaluations, such as product-to-product similarity (see Aaker & Keller, 1990), brand-to-product
similarity (Broniarczyk & Alba, 1994; Park, Milberg, & Lawson, 1991), and brand-to-brand
similarity (Bei & Shen, 2007; Walchli, 2007). Brand-to-brand similarity through case-based
reminding can be viewed as a complementary source to the product-to-product and brand-to-
product similarity in prior research. However, the case-based reminding effect observed in
Experiment 1 was a mixture of all three sources of similarity. For instance, the evaluation of
Rolex making fountain pens prompted by Mont Blanc as a similar case was a result of all three
sources of similarity: product-to-product similarity (i.e., the similarity between the watch and the
fountain pen), brand-to-product similarity (i.e., the similarity between Rolex and the fountain
pen), and brand-to-brand similarity (i.e., the similarity between Rolex and Mont Blanc). It would
be desirable to be able to isolate effects out of the three levels of similarities: product-to-product,
brand-to-product, and brand-to-brand. Experiment 2 manipulated the product association to
control for the effect of brand-to-product similarity. With the reminding case for the brand-to-
brand similarity, the effects of three levels of similarities could be identified.

EXPERIMENT 2
1. Top of page
2. Abstract
3. LITERATURE BACKGROUND AND RESEARCH HYPOTHESES
4. EXPERIMENT 1
5. EXPERIMENT 2
6. EXPERIMENT 3
7. GENERAL DISCUSSION AND CONCLUSIONS
8. REFERENCES
Experiment 2 was modified based on the two concerns cited above. In order to distinguish the
three different sources of similarity and purify the net effect of case-based reminding (i.e., brand-
to-brand similarity), the same core brand, Rolex, and the moderately similar extension, the
fountain pen, were employed in Experiment 2. The moderately similar extension was chosen for
two reasons: (1) the obvious difference between the reminding and non-reminding case in
Experiment 1 would allow the effect of brand-to-brand similarity to be easily observed; (2) the
fountain pen revealed a partially functional and partially prestigious product concept in the
pretest, which provided room to manipulate participants' perceptions of brand-to-product
similarity.
Experiment 2 utilized short “product passages” in which either prestigious or neutral product
associations for fountain pens were embedded. A prestigious product association pointing out
that a fountain pen symbolizes a businessman's social status would evoke a higher level of brand-
to-product similarity between Rolex and fountain pens. A neutral product association described
only the history of a fountain pen and its function, without any prestige-or symbol-related
information, in order to evoke a lower level of brand-to-product similarity between Rolex and
fountain pens. For instance, when the brand extension evaluation of a Rolex fountain pen is
given a prestige association with the fountain pen category, it reflects the brand-to-product
similarity between the prestigious status of the Rolex brand and the prestigious fountain pen
product category. In contrast, when the brand extension evaluation of a Rolex fountain pen is not
given a prestigious association with the fountain pen, it reflects that no such brand-to-product
similarity could be established. Thus, comparing the two conditions would reveal the net effect
of the brand-to-product similarity, and comparing the two conditions would reveal the net effect
of the brand-to-product similarity.
The print ad was discarded to eliminate the possible instantaneous learning due to the picture in
the ad conveying an association with prestige. In addition, another set of short “brand passages”
describing the history of the reminding brand, written in a neutral tone and rhetorical style, were
designed to serve the role of the reminding case in Experiment 2. No direct allusions to prestige
were used in the brand passage to avoid the possible confusion with instantaneous learning. The
effect of case-based reminding due to the presence of the Mont Blanc brand name must come
from respondents' brand schemas, where both Rolex and Mont Blanc shared similar
characteristics. The brand passage, similar to the print ad in Experiment 1, was used to elicit the
perception of a similarity in prestige between Rolex and Mont Blanc. The comparison between
this condition and that without such a brand passage would reveal the net effect of the brand-to-
brand similarity on the brand extension evaluation.
The rationale is as follows (see Table 2). First, for participants who are given the neutral product
passage about the fountain pen (in which the fountain pen was mainly described as a functional
product with no symbolic or prestigious meaning) and without the reminding case, their brand
extension evaluation should mainly reflect the effect of product-to-product similarity—in other
words, the similarity between the watch and the fountain pen. Second, for participants who are
given the product passage in which the fountain pen was described as a prestigious product and
without the reminding case, their brand extension evaluations should be based on both product-
to-product similarity (i.e., similarity between the watch and the fountain pen) and brand-to-
product similarity (i.e., similarity between the prestigious brand image of Rolex and the prestige
association of the watch category). Thus, the difference between these two groups would reflect
the net effect of brand-to-product similarities. Third, for participants who were given the product
passage in which the fountain pen was described as a prestigious product and given the brand
passage with the reminder brand in the fountain pen category, their brand extension evaluation is
the sum of all three levels of similarities (i.e., product-to-product, brand-to-product, and brand-
to-brand). As a result, the difference between the third and second group of participants would
reflect the net effect of brand-to-brand similarities (i.e., the similarity between Rolex and Mont
Blanc).
Table 2. The Combination of Different Similarities in the Design of Experiment 2 and
Experiment 3.

Product Passage or Product

Prestige Product Neutral Product
Association:

Brand Passage: Association Association

• Notes: Prestige versus neutral product associations are manipulated (or measured) by
product passage in Experiment 2 (or product association in Experiment 3).
• PtoP stands for Product-to-Product Similarity, BtoP stands for Brand-to-Product
Similarity, and BtoB stands for Brand-to-Brand Similarity.
• *
*
The combination of PtoP + BtoB is not included in Experiments 2 and 3.
Case brand reminding PtoP + BtoP + BtoB PtoP + BtoB*

No case brand reminding PtoP + BtoP PtoP

Finally, the condition of a neutral product association with the reminder brand is more at issue
(as shown in the upper-right quadrant of Table 2). Logically, the brand extension evaluation from
this group would be based on the product-to-product similarity (i.e., similarity between the watch
and the fountain pen) plus the brand-to-brand similarity (i.e., the similarity between Rolex and
Mont Blanc). Thus, the difference between this group and the group with neutral product
associations without the reminding case (that is, product-to-product similarity only) should
reflect the net effect of brand-to-brand similarity. However, the perceived similarity between the
reminding brand and the core brand may render the neutral product association manipulation
ineffective. The reason for this is that the presence of the Mont Blanc reminding brand may be
sufficient to lead participants to perceive the fountain pen as a prestigious product even though
they are given the manipulation of a neutral product association. Thus, although Experiment 2
did include this condition to create a complete 2 × 2 design, it was not expected to be employed
to test the net effect of B2B similarity.
To summarize, in Experiment 2, it is hypothesized that the effect of providing a reminding case
in the extension category when the product category is introduced as a prestigious product is
stronger on the extension evaluation than the effect when the extension product category is
prestigious but without the reminding case. In addition, both effects are stronger than the effect
when the extension product category is described as neutral and without the reminding case. The
last condition represents the product-to-product similarity, and the net differences from the
previous two levels represent the brand-to-product and brand-to-brand similarities, respectively
(as shown in Figure 1).
Figure 1. The effects of case-based reminding and product associations in Experiment 2.

Method
Materials.
The basic experimental concepts were the same as in Experiment 1, except for the manipulations
of the product category association and the reminding brand. The product association was not
controlled in Experiment 1, but was manipulated by constructing short passages describing the
fountain pen either as a prestigious product or as a neutral product in Experiment 2. The two
versions were about the same length (149 words for the prestige version and 151 words for the
neutral version of the passage). A pretest using a 7-point semantic differential scale on 60
participants showed there was a significant difference in terms of the prestige status of the
fountain pen between the two passages [Mean prestige rating = 5.43,SD= 0.85 for the prestige
passage; Mean prestige rating = 2.08,SD= 0.64 for the neutral passage; between group t(58) =
17.34, p < 0.01].
To manipulate the reminding brand, short informational passages (i.e., brand passages) about
Mont Blanc replaced the role of print ads in Experiment 1 in order to reduce the possible
confounding of instantaneous learning opportunities due to the picture of a prestige pen
conveying the prestige message of the brand. The brand passages only described the history of
Mont Blanc, without referring to any concept of prestige. Two other brand history passages
about Kikkoman soy sauce and Charlie Wang soft drinks were also developed for this purpose.
For the non-reminding group, the passage about Kikkoman was used instead of Mont Blanc. The
passage about Charlie Wang was the filler presented after either Mont Blanc or Kikkoman. A
pretest on 30 participants comparing the amount of information, the persuasiveness,
comprehensibility, and clarity of the message revealed that there were no significant differences
among these passages.
Design and Procedure.
Two independent variables (case-based reminding vs. product association) were manipulated in
Experiment 2. The experiment was a 2 (case brand reminding: yes, i.e., Mont Blanc vs. no, i.e.,
Kikkoman) × 2 (product association: prestige vs. neutral association with the fountain pen)
between-subject factorial design. The dependent measures were the same as in Experiment 1.
Participants were told that they would participate in two studies. The first one was masked as a
survey judging the quality of PR statements. Participants received a booklet in which the first
page was the product association passage, where the fountain pen was described as either a
prestige or a functional product (to manipulate the level of brand-to-product similarity). After
finishing the product passage, they evaluated the amount of information, the persuasiveness,
comprehensibility, and clarity of the passage on four 7-point questions. Participants were also
asked to write down their free associations after reading the PR statements. The second page was
the brand passage for the manipulation of the reminding case, the brand history of either Mont
Blanc or Kikkoman. The final page was the filler passage. The same set of four 7-point questions
and one open-ended question were asked for both passages. These data were collected for the
purpose of a manipulation check.
Next, participants were introduced to the second study, the brand extension evaluation, which
was the same as in Experiment 1. After completing the questions on brand extension evaluations,
participants provided their judgments on the perceived prestigious status of the extension
category (i.e., the prestigious–not prestigious level and the symbolic–functional level) on the 7-
point semantic differential scales and other manipulation check measures, just as in Experiment
1.
Results
One hundred twenty-four undergraduate students from a major university were recruited as
participants. Of these, 104 students who knew the core and case brands composed the valid
sample. They were given extra course credit for their participation.
Manipulation Check.
As in the pretest, the two product passages were not different [F(1,307) = 0.01, p = 0.95]; the
two brand passage manipulations were not different [F(1,307) = 0.30, p = 0.59] either, in terms
of the message clarity, the amount of information, or the message's persuasiveness. Furthermore,
90% of participants answered the prestige status of the fountain pen consistently with the
manipulation. The average of two questions (i.e., the prestigious and symbolic level) measuring
the prestige status of the fountain pen also revealed significant differences between the
conditions of the prestige product association (M = 5.91,SD= 0.71) and the neutral product
association [Mean = 2.59,SD= 0.87; t(102) = 20.71, p < 0.01]. The manipulation of the prestige
product association was successful.
As expected, the manipulation of the condition with the fountain pen as a neutral product and
Mont Blanc as the reminding brand failed. In this condition, all participants viewed the fountain
pen as a prestigious product, albeit they were given the neutral manipulation. As reasoned above,
it was possible that the reminding manipulation of a prestige case brand (i.e., Mont Blanc) would
render the product association manipulation ineffective. As a result, the data of this condition
were removed from further analyses. The same product association manipulation was successful
in other conditions. Participants viewed the fountain pen as a prestigious versus neutral product
as expected.
Hypothesis Testing.
The Cronbach's alphas for both the perceived fit (α = 0.93) and attitude toward the brand
extension (α = 0.89) were both higher than the conventional acceptable level. The average scores
of corresponding items were taken for the following tests of the hypotheses. Because the group
of neutral product associations with a reminding case was not included in the design, the design
became an incomplete design. Thus, the rest of the three groups were treated as three levels of
one variable. The result of the ANOVA showed that for the perceived fit, a significant main
effect [F(2,101) = 16.82, p < 0.01] was found. The same was also true for the attitude toward the
brand extension [F(2,101) = 17.04, p < 0.01].
The brand reminding passage containing neutral information was found to be effective in
eliciting stronger brand extension evaluations. The two non-reminding groups given the prestige
and neutral association manipulation were first combined in order to compare them with the
reminding group. The perceived fit of the extension with the reminding passage (M = 1.68,SD=
1.04; see Figure 1) was higher than without the reminding passage (M = 0.35,SD= 1.38; planned
contrast F = 24.68, p < 0.01). Similarly, the attitude toward the extension with the reminding
passage (M = 0.99,SD= 1.03) was more positive than without the reminding passage (M =
−0.24,SD= 1.24; planned contrast F = 25.17, p < 0.01). These results further supported the
mechanism of case-based reminding by eliminating the confounding effect from the potential
instantaneous learning arising from of the print ad presentations in Experiment 1.
Moreover, to isolate the effects from different levels of similarities, the brand extension
evaluations across consumers who were given different product associations were analyzed. The
perceived fit for participants in the non-reminding group given the prestige association of the
fountain pen (M = 0.75,SD= 1.22) was significantly stronger than the same non-reminding
participants given the neutral associations (M = −0.08,SD= 1.44; Scheffé test p < 0.05).
Similarly, the attitude toward the extension for participants in the non-reminding group given the
prestige association of the fountain pen (M = 0.13,SD= 1.24) was significantly stronger than the
same non-reminding participants given the neutral associations (M = −0.64,SD= 1.13; Scheffé
test p < 0.05). Thus, the data provided support the existence of this source for the effect from the
brand-to-product similarity. These results concurred with the notion of brand concept
consistency in prior literature (Broniarczyk & Alba, 1994; Mao & Krishnan, 2006; Park,
Milberg, & Lawson, 1991).
Most importantly, the case-based reminding mechanism relies on brand-to-brand similarity as the
source of the effect. This effect was revealed when there was a significant difference between the
group given the prestige association of the fountain pen and the reminding brand (i.e., brand-to-
brand, brand-to-product, and product-to-product) and the group given the prestige association but
not given the reminding brand (i.e., brand-to-product and product-to-product). The data analysis
showed that the perceived fit for reminding participants (M = 1.68,SD= 1.04) was significantly
stronger than for the non-reminding participants (M = 0.75,SD= 1.22; Scheffé test p < 0.01).
Similarly, the attitude toward the extension for participants who were given the reminding brand
(M = 0.99,SD= 1.03) was also significantly more positive than the non-reminding participants
(M = 0.13,SD= 1.24; Scheffé test p < 0.01). Therefore, the data supported the existence of the
brand-to-brand similarity as the source of effects in the case-based reminding.
Discussion
Experiment 2 successfully established that the reminding effect observed in Experiment 1 came
from consumers' memory, as opposed to the alternative explanation of instantaneous learning
from the print ads. Reminding consumers with case materials that did not convey the notion of
prestige, as in the pictorial stimuli of Experiment 1, reduced the possibility that the observed
effect of reminding came from instantaneous learning and not from consumers' internal schemas.
Obviously, respondents perceived that Rolex extended to a fountain pen more plausibly and
preferably when the reminder of Mont Blanc was present. Without the picture of a prestige
product, respondents could retrieve the prestige image of Mont Blanc from their memory based
on a simple statement of reminder. The result further supported the theory of case-based
reminding as the mechanism responsible for improving the brand extension evaluation.
Furthermore, Experiment 2 also distinguished the brand-to-brand similarity from the other two
brand-to-product and product-to-product similarities as the sources of effects for the case-based
reminding. It was found that all three sources of similarities contributed to the observed effect of
case-based reminding on improving brand extension evaluations.
In the current study, the brand-to-product similarity was manipulated and somehow was different
from that of previous research (Broniarczyk & Alba, 1994; Park, Milberg, & Lawson, 1991),
where the brand-to-product similarity naturally existed in respondents' perceptions. The fountain
pen in this study was selected as the extension product due to its moderate similarity to the watch
and its indistinguishable position on the prestige–function axis. Experiment 3 intended to elicit
respondents' internal perceptual schema to identify the prestige-function position; they could thus
retrieve the brand-to-product similarity naturally. Furthermore, the effect of case-based
reminding could be fully relieved of the demand artifact caused by manipulating (as opposed to
measuring) the brand-to-product similarity.

EXPERIMENT 3
 1. Top of page
2. Abstract
3. LITERATURE BACKGROUND AND RESEARCH HYPOTHESES
4. EXPERIMENT 1
5. EXPERIMENT 2
6. EXPERIMENT 3
7. GENERAL DISCUSSION AND CONCLUSIONS
8. REFERENCES
To further consolidate the conclusions of Experiment 2, Experiment 3 replicated the findings of
Experiment 2, with one modification of the procedure. In Experiment 2 the product association
(prestige vs. neutral) was manipulated by giving undergraduate participants short passages.
However, the product association could also be an individual difference that depends on the
person's knowledge. A fountain pen may be viewed as a prestigious product by many white-
collar workers, for instance, but not by most students. Thus, in Experiment 3 the product
association was measured (rather than manipulated as in Experiment 2).
Another reason to use white-collar workers as samples to further enhance the meaning of the
case brands in Experiment 3 is that they are the targets of fountain pens. As reported in the
previous two experiments, many student samples were classified as invalid because they had not
heard of the reminding case brands. One could criticize the experiment on the grounds that
students might not fully understand the meaning of the product's functional or social value or
comprehend the desired projected image of the brand. Nonetheless, it was expected that the
result patterns of Experiment 2 would be similarly obtained in Experiment 3. In sum, the
hypotheses of Experiment 3 were the same as in Experiment 2. Most experimental procedures
were the same as well, except that the product association was measured rather than manipulated.
Method
Participants.
In order to ensure there were enough participants who could make the product association
between fountain pens and the image of prestige, middle-aged, middle-class consumers were
sampled for this experiment, since they are more likely to be users of prestigious fountain pens.
Out of the 100 respondents, 98 were familiar with both Rolex and Mont Blanc, and therefore
constituted the valid samples. These participants were between 30 and 40 years of age, half of
them males and half females.
Materials.
Most materials were the same as in Experiment 2, except that the product association was
measured rather than manipulated. To measure the product association, participants were asked
whether they thought the fountain pen was a prestigious/symbolic product or a not
prestigious/functional product on two 7-point semantic differential scales, ranging between −3
and +3. The prestige and neutral groups were classified using the conventional median split.
Design and Procedure.
As in Experiment 2, Experiment 3 consisted of one variable with three conditions, excluding the
condition of neutral product associations coupled with case-based reminding. The first condition
was the reminding condition, in which participants were given the brand passage describing
Mont Blanc's brand history as a reminder to them of the similar case in the extension category.
The other condition was the non-reminding condition, in which the brand passage introducing
Charlie Wang's brand history was given. Participants in this condition were divided into two
groups based on their perceptions of fountain pens: those who viewed the fountain pen as a
prestigious product and those who viewed it as a neutral or functional product. The dependent
variables were the same as in Experiment 1 and 2.
The experimental procedure was similar to that in Experiment 2. The booklet for the experiment
first presented the reminding brand passages and questions about the brand/product awareness
and associations, as well as questions about the amount of information and the persuasiveness,
comprehensibility, and clarity of the message. Participants were also asked to write down their
free associations after reading the passages. After participants finished the first section of the
experiment, they moved on to the second section on brand extension evaluations, as in the
previous two experiments; they completed questions on brand extension evaluations, the
perceived prestigious status of the extension category, and other manipulation check measures.
Results
Manipulation Check.
The two passages were not different in terms of the amount of information, the persuasiveness,
comprehensibility, or clarity of the message in the passage [F(1,96) = 0.23, p = 0.63].
Hypothesis Testing.
The Cronbach's alphas for both the perceived fit (α = 0.95) and attitude toward the brand
extension (α = 0.91) were both higher than the conventional acceptable level. The result of a one-
way ANOVA showed that there was a significant main effect for the perceived fit [F(2,95) =
23.01, p < 0.01], as well as for the attitude toward the brand extension [F(2,95) = 29.26, p <
0.01]. The perceived fit of the extension with the reminding passage (M = 2.20,SD= 1.05) was
higher than without the reminding passage (M = 0.34,SD= 1.60; planned contrast F = 49.67, p <
0.01). Similarly, the attitude toward the extension with the reminding passage (M = 1.54,SD=
0.72) was more positive than without the reminding passage (M = −0.21,SD= 1.36; planned
contrast F = 56.79, p < 0.01). This result agreed with that of Experiment 2; it further supported
the mechanism of case-based reminding and eliminating the confounding explanation of
instantaneous learning. The means and standard deviations of three conditions are presented in
Figure 2.
Figure 2. The effects of case-based reminding and product associations in Experiment 3.

The brand-to-product similarity as the source of effects was first investigated. The perceived fit
for the two subgroups of participants in the non-reminding condition were significantly different
(Scheffé test p < 0.05). Those who held a prestige association with the fountain pen (M =
0.87,SD= 1.59) perceived a better fit with the brand extension than those without such prestige
associations (M = −0.22,SD= 1.48). Similarly, the attitude toward the extension for participants
in the non-reminding group maintaining a prestige association with the fountain pen (M =
0.18,SD= 1.39) was significantly better than the same non-reminding participants without such
prestige associations (M = −0.67,SD= 1.21; Scheffé test p < 0.05). Thus, the brand-to-product
similarity naturally existing in participants' perceptions as the source of effects was again
supported as in Experiment 2.
For brand-to-brand similarity as the source of effects, participants who were given the reminding
brand (M = 2.20,SD= 1.05) perceived a significantly better fit for the extension than the
participants with the prestige association and without the reminding case (M = 0.87,SD= 1.59;
Scheffé test p < 0.01). Similarly, participants who were given the reminding brand (M =
1.54,SD= 0.72) would hold more positive attitudes toward the extension product than
participants with the prestige association but not given the reminding case (M = 0.18,SD= 1.39;
Scheffé test p < 0.01). These results were the same as in Experiment 2, supporting brand-to-
brand similarity as the source of the effects on consumers' brand extension evaluations.

GENERAL DISCUSSION AND CONCLUSIONS

1. Top of page
2. Abstract
3. LITERATURE BACKGROUND AND RESEARCH HYPOTHESES
4. EXPERIMENT 1
5. EXPERIMENT 2
6. EXPERIMENT 3
7. GENERAL DISCUSSION AND CONCLUSIONS
8. REFERENCES
Summary of Findings
The role of case-based reminding in consumers' brand extension evaluation was investigated in
three experiments. Experiment 1 first revealed the effect of case-based reminding and its
interaction with extension similarities. Experiment 1 found that consumers' brand extension
evaluations could be improved by reminding them of a similar case brand in the extension
category. A brand-to-brand similarity was then defined. This effect was more pronounced with
the moderately similar extension than with either the highly similar extension or the dissimilar
extension. However, the effects from three levels of similarities (i.e., brand-to-brand, brand-to-
product, and product-to-product similarities) were unfortunately mixed in Experiment 1.
Therefore, follow-up experiments were designed to further purify the effect of brand-to-brand
similarity. The additive effect of the brand-to-brand similarity was illustrated on top of the
effects of brand-to-product and product-to-product similarities.
Experiment 2 first eliminated the possibility of the confounding explanation that the effect came
from an instantaneous learning of a print ad's pictorial presentations used in Experiment 1, and
further explored the three levels of similarities as the potential sources of these effects. Using a
moderately similar extension, Experiment 2 concluded that consumers' perceived fit and attitudes
toward the extension were highest when they were provided with a reminding case with a similar
brand image in the extension category (which revealed the brand-to-brand similarity) and when
the extension product category was introduced as being as prestigious as the core brand (which
represented the brand-to-product similarity).
Experiment 3 consolidated the findings of Experiment 2 by treating the product association as an
individual difference to fit the definition of brand-to-product similarity in previous studies, as
opposed to the manipulation approach used in Experiment 2. Similar and even stronger effects of
the brand-to-brand similarity arising from case-based reminding were found again in Experiment
3. Also, Experiment 3 utilized the preexisting product concept and brand image perceived by
middle-aged consumers to increase the external validity of the results.
Implications for Brand Extension Researchers and Practitioners
The contributions of the present research to academia are twofold. First, the present research
provides a bridge between research in analogical reasoning and brand extension evaluation.
Although literature (Gregan-Paxton & Roedder, 1997; Gregan-Paxton et al., 2002) has identified
the possibility of using theories in analogical reasoning to study new product and brand
extension evaluations, the present research is an effort to realize such a possibility. It also offers
a new perspective for brand extension research in that consumers' brand extension evaluations
can be improved by prompting a similar case brand in the extension category through an
analogical reasoning mechanism. Second, the present research identifies different levels of
similarities as the sources of brand extension evaluations. Prior research (Aaker & Keller, 1990;
Broniarczyk & Alba, 1994; Felcher, Malaviya, & McGill, 2001; Park, Milberg, & Lawson, 1991)
has implicitly employed different levels of similarities (product-to-product and brand-to-product
similarities) as the sources of effects on brand extension evaluations. The present research
summarized these differences and further added brand-to-brand similarity as a complementary
source of effects for brand extension evaluations.
The present research also carries implications for marketing practitioners. In order to grow and
increase revenue successfully, brand owners tend to consider extending a successful brand to
similar product categories. These products are usually supplementary or substitutive (Aaker &
Keller, 1990). After analyzing the core image of a brand, marketers may extend the brand into a
product category such that the product concept fits the brand concept; for example, they might
extend a prestige brand into a luxury product (Park, Milberg, & Lawson, 1991). The present
research demonstrates that an existing brand with similar brand concepts in the new product
category can further enhance consumers' acceptance of the brand extension. Thus it is possible to
improve consumers' brand extension evaluations by aligning the new, albeit dissimilar, extension
with a similar case brand in marketing communication campaigns (Markman & Gentner, 1993).
The result of the present research opens a window for a company that needs to extend its brand
to a dissimilar category.
A practical issue is how to utilize the effect of case-based reminding. Different from the product-
to-product and brand-to-product similarities, which can be applied by selecting the proper
extension product categories or building a more desirable brand image, half of the brand-to-
brand similarity is not controlled by the marketer of the core brand. It is suggested that the
company may place its ads close to or after the ad of the competitor, which is also the perfect
case in the new category to elevate consumers' perceived fit. Another tactic is to employ
competitive advertising and directly bring the similar case into its own ad. This tactic can also
directly position the core brand side-by-side with the competitor. Of course, the marketer of the
core brand should avoid the possible deleterious effect on the original brand image due to the
undesirable “me too” suggestion of such a placement.
Further Discussions, Limitations, and Directions for Future Studies
Case-based reminding is the key theory on which the present study is developed. It should be
noted that case-based reminding is just one form of analogical reasoning that can be applied in
studying brand extension evaluations. It is possible that other major theories of analogical
reasoning such as Structure Mapping (Gentner, 1983) or the Pragmatic Theory of Analogy
(Holyoak, 1985) could also be employed as the basis for studying consumers' evaluations of
brand extensions. Future research may address these theoretical approaches.
On the other hand, this study only employs an analogy theory to introduce the new concept of
brand-to-brand similarity. Analogy theories may also be applied on product-to-product and/or
brand-to-product similarities. Previous studies have explained that a brand extension to a
complementary or substitutive product could generate a higher perceived fit (Aaker & Keller,
1990). It is very likely that one successful brand extension from one product to another category
(which is not complementary or substitutive) may serve as an analogical case and increase the
product-to-product similarity and the perceived fit of other similar extensions. More specifically,
this study found that when given the reminding case Mount Blanc, respondents immediately
associated the image of prestige with the fountain pen product category (Experiment 2). This
reveals that consumers can spontaneously make analogies by connecting a brand with a product
even when they are not primed. Expending the application of case-based reminding to product-
to-product and brand-to-product similarities can be another direction for further studies.
It is also recognized that not every product category has a similar case brand. It may not be easy
for a company to find a similar case brand with which to align its own brand in the intended
extension category. The difficulty can be explained by virtue of the fact that pretests were
necessary to find proper experimental materials in the present research. In addition, a similar
brand in the same product category is usually assumed to be a competitor. However, the results
of the present research suggest the positive assistance one can gain from this so-called
competitor brand. It would be interesting and even indispensable to distinguish and compare the
competing forces and the transference forces of a similar preexisting brand in the extension
product category.
Moreover, the notion of prestige was used as the basis of similarities in the present research.
Other dimensions may or may not have the same effects (Medin, Goldstone, & Gentner, 1993).
Future research can address these issues based on the findings of the present research. Finally,
recent research (Mao & Krishnan, 2006) has related consumer evaluations of brand extensions to
the categorization process. Future research may consider other theories of categorization as the
basis for further developments in marketing.