Investigating Dimensions of Brand Names That Influence the Percieved Familiarity of Brands

Joan Meyers-Levy, Northwestern University
ABSTRACT - Extant research suggests that products perceived as being familiar are preferred to those perceived as less familiar. The current research examines how features of a product's brand name can influence perceived familiarity. Results of a study imply that the number of associations related to a product's brand name and the distinctiveness of processing the brand name induces can jointly influence perceptions of brand familiarity.
[ to cite ]:
Joan Meyers-Levy (1989) ,"Investigating Dimensions of Brand Names That Influence the Percieved Familiarity of Brands", in NA - Advances in Consumer Research Volume 16, eds. Thomas K. Srull, Provo, UT : Association for Consumer Research, Pages: 258-263.

Advances in Consumer Research Volume 16, 1989      Pages 258-263

INVESTIGATING DIMENSIONS OF BRAND NAMES THAT INFLUENCE THE PERCIEVED FAMILIARITY OF BRANDS

Joan Meyers-Levy, Northwestern University

ABSTRACT -

Extant research suggests that products perceived as being familiar are preferred to those perceived as less familiar. The current research examines how features of a product's brand name can influence perceived familiarity. Results of a study imply that the number of associations related to a product's brand name and the distinctiveness of processing the brand name induces can jointly influence perceptions of brand familiarity.

A major purpose of brand names is to provide a distinctive lat>el by which products and services can be identified (Friedman 1985). Indeed, advertisers typically devote considerable time, effort, and financial resources to selecting brand names that are likely to be memorable (Zinkhan and Martin 1987), and they promote these brand names to enhance consumers' familiarity with them. This practice seems to be predicated on the belief that, all other things being equal, consumers are likely to regard favorably products that they perceive as familiar, and in turn this may enhance their likelihood of purchasing these products.

Consistent with this view, Mandler (1982) suggests that familiar objects generally are positively valued because they enable predictability. The thesis that people are attracted to/prefer that which is familiar also underlies research by Zajonc (e.g., 1971). Indeed. objects or experiences made familiar during childhood often have lasting effects on our tastes, preferences, and yearnings (Rheingold 1985). Thus, in light of the seeming practical importance of developing, brands that are perceived as familiar, a critical question that emerges is whether dimensions of a product's brand name itself might influence the memorableness or perceived familiarity of the brand?

Some insight into this question would seem to be offered by the basic learning principle, which suggests that memory for target concepts/information is enhanced if the concepts can be meaningfully related to many as opposed to few other concepts already stored in memory (Nisbett and Ross 1980). The logic underlying this principle is that concepts that are tied to a large repertoire of diverse concepts or associations will be more memorable because the abundant associations provide a multitude of rich memory traces that can enhance the accessibility of the concepts. These meaningfully related associations or linkages to the target concept will be referred to as its association set. Thus, the implication of this view for brand name selection is that, consistent with findings offered by Kanungo and Dutta (1955), brand names will be more accessible in memory and thus more memorable or familiar to the extent that they possess a large rather than a small association set.

However, another substantial body of literature leads to an opposing inference. According to this research, the probability of accessing any particular concept diminishes as the association set size of the concept increases (Alba and Chattopadhyay 1985; Nelson and Friedrich 1980). Two reasons are often cited for this phenomenon. First, as the number of associations linked to a target concept increases, the strength of each linkage or retrieval path is thought to be weakened because total activation is dissipated over the large association set (Anderson 1976). Second, as the association set size increases, the number of competing associations that might be activated and interfere with access to the target concept elevates (Anderson 1976; Alba and Chattopadhyay 1985). Thus, the larger the number of associations that are encoded in relation to the target concept, the lower is the probability of accessing that item. These findings imply that brand names possessing a relatively small rather than a large association set will be more accessible and thus presumably more familiar. Perhaps it is a logic similar to this that often induces practitioners to search for unique brand names.

The present paper suggests that the outcomes implied by either of the preceding views concerning the relationship between brand name familiarity/ memory and the size of the brand name's association set may occur. It is proposed that what determines which of these outcomes takes place in a given instance depends on the distinctiveness of processing, that the brand name engenders. As used here, the term distinctiveness of processing refers to the extent that during processing, individuals focus on the unique and distinctive meanings or connotations of the target concept/brand name (Eysenek 1979). The main hypothesis is that access to and thus familiarity of brand names that possess large versus small association sets may be inhibited when consumers are prompted to engage in nondistinctive processing. However, an opposite outcome is hypothesized to occur when consumers engage in distinctive processing. In the following sec ion, a means of making the distinctiveness of processing construct operational is offered and the log c for the preceding predictions is developed.

ASSOCIATION SET SIZE AND DISTINCTIVENESS OF PROCESSING

To examine how the relationship between association set size and distinctiveness of processing may affect familiarity, a way is needed to operationalize the distinctiveness construct. A sizeable body of research suggests that this Can be accomplished via word frequency (Eysenek 19,9; Lockhart, Craik and Jacoby 1976). Rare, infrequently encountered words in the language are thought to stimulate distinctive processing as individuals "struggle to find [distinctive] meaning in them" in relation to their context (Jacoby and Craik 1979, p. 11). By contrast, the meaning of common, frequently encountered words is readily apparent and therefore these words are afforded relatively nondistinct processing (Lockhart et al. 1976).

Support for this thesis comes from evidence that implies infrequently encountered or novel words are not as readily perceived or processed as are common high frequency words (Catlin 1969). And like other novel words or concepts, such low frequency words receive greater attention (Warmer and McGray 1969) and take more time to process (Forster and Chambers 1973) as individuals engage in directed efforts to specify the meaning of, make sense of, and encode these words. Given such evidence that word frequency is an appropriate indicator of distinctiveness of processing, this variable will be adopted both in discussing the relationship between distinctive processing and association set size and in the research that will be reported.

Nelson's sensory-semantic model (1979) provides a starting point in investigating the relationship between association set size and brand name familiarity/memory. This model suggests that when individuals are exposed to a target word, they automatically activate in memory concepts that are related to this word. For example, consumers who are exposed to the brand name "Cheer" will automatically activate meaningfully related concepts or associations to the brand name word (e.g., applaud, joyfulness, comfort, etc.). Moreover, these concepts sometimes may be encoded in memory together with the target word (Nelson et al. 1985, p. 95). Whether these concepts that have been automatically activated are in fact encoded in memory with the target (brand name) word appears to vary as a function of the distinctiveness of processing that takes place during encoding. Employing word frequency as an indicator of distinctive processing, let us examine this critical thesis more fully.

Association Set Size Effects When Processing High Frequency Words.

As research reviewed earlier suggests, high frequency words tend to be readily processed with little effort (Catlin 1969). Thus, high frequency words are afforded relatively nondistinct processing during encoding such that little attempt is made "LO integrate, specify, or restrict target [brand] information" with respect to a limited set of related concepts that uniquely and distinctively specify a particular sense of the words (Nelson et al. 1985, p. 95). From this it follows that a broad spectrum of the heterogeneous concepts in the association set that are automatically activated upon target word exposure will be unselectively encoded in memory together with the target (brand) word (c.f. Jacoby and Craik 1979). Note that because these encoded concepts are generally heterogeneous, their meanings should possess little semantic overlap with the connotations of the target word/brand name in the operating context. Hence, these associations might largely interfere with brand name accessibility. This implies that because, relative to brand names possessing small association sets, brands with large association sets should have more of these diverse associations that might interfere with the accessibility of the brand names, interference effects should be more pronounced for brand names possessing large association sets. Thus, the hypothesis emerges that when brand names are embodied by high frequency words, access to and thus familiarity with the brand names may be poorer for brands that possess a large rather than a small association set.

Association Set Size Effects When Processing Low Frequency Words.

A different scenario unfolds when brand names are embodied by low frequency words. Exposure to these relatively uncommon low frequency words, which are difficult to encode and assign meaning to, presumably induces distinctive processing. Such distinctive processing is directed at unambiguously specifying the meaning of the word in its context find discriminating it from other concepts (Lockhart et al. 1976). Thus, while general activation of concepts related to the target word or brand name occurs upon exposure to such low frequency words/brand names, the distinctive processing which these words are subjected to limits the breadth of association set concepts that are encoded in memory. Indeed, encoding of concepts in the association set is restricted to only those items that appear to meaningfully relate the brand name with other brand-relevant information presented in the context (c.f. Nelson et al. 1985). This restrictive encoding should result in the formation of strong linkages between t; e target word and some selected set of relevant concepts in the context or message. And this implies that regardless of the size of the association set, the breadth of associations to the brand name that are encoded in memory will be limited. As such, a large association set should not adversely effect the accessibility of the brand name. In fact. it may be that brand name accessibility and thus familiarity might be somewhat enhanced as the size of the brand name's association set increases because more associations will be available to meaningfully relate to the brand name.

OVERVIEW OF THE STUDY AND HYPOTHESES

The preceding theorizing suggests that accessibility to and thus perceived familiarity with a brand name may vary as a function of the brand name's association set size and the distinctiveness of processing induced by the brand name during encoding. Brand names are likely to be more accessible and thus appear to be more familiar when they are relatively common, high frequency words and have few rather than many meaningful associations related to them, but such effects might be eliminated or possibly reversed when brand names are relatively rare, low frequency words.

This hypothesis was examined in a study. Subjects were exposed to a set of ads for fictitious brands. Some of these brands, hereafter referred to as critical brands, had brand names that varied in terms of their association set size (small or large) and word frequency (high or low) The remaining brands are referred to as filler brands because they were used only for filler purposes and their brand names did not vary along these dimensions. Later after being exposed to the entire set of ads, subjects' perceived familiarity with the critical brands/brand names prior to the time of the study and their certainty about this familiarity were assessed. Hence, subjects were queried about their subjective perceptions of prior familiarity with brands that, in reality, did not exist in the market. Moreover, the robustness of treatment effects on familiarity over time was assessed by administering this measure after either a relatively long or short delay.

Because the theory makes no clear predictions about the effect of delay on certainty of perceived familiarity, no hypotheses concerning this variable are offered. However, formal hypotheses are offered that capture the anticipated interaction of brand name association set size and brand name sword frequency on this measure

H1: When critical brand names are high frequency words, greater certainty of the perceived familiarity of the critical brands/brand names should result when the brand names possess small rather than large association sets.

H2: When critical brand names are low frequency words, the certainty of perceived familiarity of the critical brands/brand names should be somewhat positively related to the brand names' association set size.

METHOD

Stimuli

Ads for eight fictitious critical and filler brands of toiletry products were constructed by combining copy claims obtained from a number of existing ads. The product categories featured in these ads included antiperspirant, blemish medicine, disposable razor, pain reliever, vitamins, soap, suntan lotion and bandages. The first three product categories listed above were randomly selected from the set and employed in ads for the critical brands, while the remaining categories were used in ads for the filler brands. Ads for the critical brands were each of comparable length (64-66 words), while the length of filler brand ads varied (36-83 words). Moreover, in each of the critical brand ads, copy was structured in an identical manner such that the brand name was identified in the first, third and fifth sentences of the text. This same structure was maintained for the filler brand ads with the exception that the brand name occurred only twice in the shorter filler ads.

Words used as brand names for critical products were determined on the basis of their association set size and the words' frequency of occurrence in the language. The brand names employed were selected from words with known association set sizes as determined in the research of Nelson and colleagues (Nelson, Bajo and Casanueva 1985; Nelson and Friedrich 1980; Nelson and McEvoy 1979). In this research, word set sizes were obtained by asking large groups of individuals to specify the first meaningfully related association that came to mind upon hearing each target word. Word set sizes were then derived by tallying the total number of different responses offered by at least two individuals. Next, the frequency with which these words occur in the English language was assessed by employing word usage estimates provided in Thorndike and Lorge (1944). Words were selected for use as critical brand names in the study only if 1) the size of their association set was either small or large (as defined by the standards employed in previous research), 2) the frequency of word occurrence was clearly low or high (low=15 or fewer occurrences and high=100 or more occurrences per one million words), and 3) the words did not convey extremely negative or unusual connotations in the context of a consumer product. In addition, pretest data were obtained (from 15 subjects) to ensure overall equivalence among groups in the image-ability of the critical brand name words and the degree to which these words were related to the product categories employed.

On the basis of these criteria, three words were selected for critical brand name use that fulfilled the word frequency (low/high) and association set size (small/large) treatment requirements. Yard, Lake, and Room were high frequency/large association set size brand names; Cloud, Day, and Round were high frequency/small set size names; Crisp, Moose, and Bribe were low frequency/large set size names; and Cork, Shove, Dusk were low frequency/small set size names. Multiple critical brand names were selected for each treatment condition to reduce the likelihood that any treatment effects that might emerge were attributable to unique content embodied by the brand names. Words selected as brand names for the filler ads were Arise, Friday, Swift, Comment, and incline; these words were relatively moderate on the critical treatment dimensions.

Procedure

One hundred females served as subjects and were given $5.00 to compensate them for their time. Participants took part in the study in groups that ranged in size from about 3 to 20 people.

Upon being seated, subjects were informed that they would be listening to a series of test ads being considered for use on radio. Participants were alerted that the ads had not yet been professionally produced and thus they would feature only the advertising text. Subjects were told that their task was to consider how clearly articulated, grammatically correct, and professionally written the ads were.

Subjects then listened to a series of tape recorded radio ads. On the tape, the three critical brand ads always were presented only once and aired in the fourth, seventh, and tenth positions. Filler ads were inserted in each of the other positions, which in total numbered 12. One of the short filler brand ads was repeated three times on the tape, two such ads were repeated twice, and two filler brand ads were presented only once. The inclusion of the tiller brand ads and the multiple presentations of several of these ads was intended to enhance the complexity of the listening context and thus reduce the likelihood that subjects would devote an unusually high level of attention to the critical brand ads, thereby producing ceiling effects and undermining the external validity of the study.

TABLE 1

CERTAINTY OF PERCEIVED FAMILIARITY FOR THE CRITICAL BRANDS

Subjects performed the dependent measures at their own pace. To eliminate short term memory effects for the advertised brands and to maintain the study guise, all subjects first completed an i.-relevant task that entailed rating the ads as a group with regard to their clarity, grammar, professional writing style, etc. Then half of the subjects were assigned to a treatment designated as the long delay condition. These subjects completed the perceived certainty of familiarity measures after a substantial delay (approximately 30 minutes), which was accomplished by having subjects complete some intervening tasks (e.g., a variety of opinion questions concerning topics unrelated to the current study) prior to filling out the certainty of perceived familiarity measures. The remaining subjects completed the familiarity measures after a short delay and thus participated in the short delay condition.

The measures designed to assess perceived certainty of familiarity were administered separately for each of the three critical brands. This entailed first asking subjects whether they had ever heard of each of the critical brands prior to participating in the study (l=no, 2=yes) and then asking subjects to rate how certain they were about their preceding responses (I=extremely uncertain, 7-extremely certain). Using these same scales another pair of somewhat similar questions also was asked. For each of the critical brands subjects were queried whether they had ever before heard of any product bearing the name of the critical brand and asked how certain they were about their ratings.

Results

Treatment effects were analyzed for the full 2 (small,<large association set size) by 2 (low/high word frequency) by 2 (short/long delay) factorial design. Treatment means for the certainty of perceived familiarity measure are reported below. This measure was computed by summing the products of: (1) subjects' reported familiarity for each of the three critical brands and their certainty concerning each of these assessments, and (2) subjects' familiarity with any products bearing each of the three critical brand names and their certainty concerning these ratings. [Separate analyses performed on each of these two measures revealed the same outcomes as those reported on the aggregate measure. That is, the interaction of association set size and word frequency was significant for both measures (p<.009 and p<.003, respectively) and no effects involving the delay manipulation emerged.] Higher numbers indica e greater certainty of familiarity.

Subjects' certainty of perceived familiarity with the critical brands/brand names prior to their exposure to them in the study revealed a main effect of word frequency (F(1,86)=5.39, p<.02) that was qualified by an interaction of association set size by word frequency (Ff 1,86)=10.95, p<.001). All other effects were nonsignificant. Consistent with H1 and H2, when the critical brand names were high frequency words, subjects reported that they had greater certainty of familiarity with the critical brands/brand names when the association set sizes of the brands were small rather than large (F(1,86)=4.65, p<.03), but when the critical brand names were low frequency words, subjects reported that their certainty of familiarity with the critical brands/brand names was greater when the brands' association set size were large rather than small (F(1,86)=6.38, p<.01).

Conclusion

To summarize, the data obtained in this study suggest that the brand names assigned to products can influence the extent to which consumers perceive the products as familiar. Hence, if, as conventional wisdom suggests, consumers are attracted to and thus are more likely to purchase familiar brands, the findings offer important insight to marketers. These findings appear to hold regardless of whether a short or long delay separates exposure to the brand names and assessments of familiarity.

Brand names embodied by high frequency words seem to bc perceived as more familiar when they possess relatively few associations tied to the name. This is thought to occur because these high frequency words stimulate nondistinct processing and therefore promote the unselective encoding in memory of heterogeneous concepts/associations related to the brand name. Moreover. because these diverse associations that are encoded in memory may possess little overlap in meaning, with the relevant connotations of the brand name, they tend to inhibit access to the brand name. Thus, perceived familiarity of products bearing high frequency brand name words appears to be greater when the brand names possess a relatively small versus large association set because such names possess fewer associations that can interfere with access to the brand name in memory.

By contrast, brand names embedded by low frequency words appear to be regarded as more familiar when the brand names' association sets are relatively large. Presumably this occurs because the low frequency brand name words engender distinctive processing and, in turn, this induces selective encoding of associations tied to the brand name words; only those associations are encoded in memory that uniquely specify the meaning of he brand names in relation to their contexts. Thus, because low word frequency brand names that possess many associations are likely to have a greater number of associations to help specify the contextually-relevant meaning of the brand names, access to and hence familiarity with such brand names will tend to be enhanced for names with large rather than small association set sizes.

Future research that conceptually replicates and extends these findings would be useful. For example, distinctiveness of processing, might be operationalized in an alternative manner, perhaps by varying the incongruity of (moderate word frequency) brand names in relation to the products they represent. In addition, it would seem worthwhile to investigate qualifications of the current findings that might well occur when repeated exposure to brand names having the characteristics of the ones examined in the current study takes place.

REFERENCES

Alba, Joseph W. and Amitava Chattopadhyay (1985), "The Effects of Context and Part-Category Cues on the Recall of Competing Brands," Journal of Marketing Research. (August), 340-49.

Anderson, John R. (1976), Language, Memory and Thought. Hillsdale, N.J.: Lawrence Erlbaum.

Baddely, Alan D. (1978), "The Trouble with Levels: A Reexamination of Craik and Lockhart's Framework for Memory Research," Psychological Review, 85 (May), 139- 152.

Catlin, Jack (1969) "Theoretical Notes on the Word-Frequency Effect." Psychological Review, 76 (September), 504-506.

Eysenek, Michael W. (1979), "Depth, Elaboration, and Distinctiveness," in Revels of Processing in Human Memory, eds. Laird S. Cermak and Fergus I. M. Craik;, Hillsdale, N.J.: Lawrence Erlbaum, 89-118.

Forster, Kenneth I. and .Susan M. Chambers (1973), "Lexical Access and Naming Time," Journal of Verbal Learning and Verbal Behavior, 12 (December), 627-35.

Friedman, Monroe (1985), "The Changing Language of a Consumer Society: Brand Name Usage in Popular American Novels in the Postwar Era " Journal of Consumer Research, 11 (March), 927-938.

Jacoby, Larry L. and Fergus I. M. Craik (1979), "Effects of Elaboration of Processing at Encoding and Retrieval: Trace Distinctiveness and Recovery of Initial Context," in Levels of Processing in Human Memory, eds. Laird S. Cermak and Fergus I. M. Craik, Hillsdale, N.J.: Lawrence Erlbaum, 121.

Kanungo, Rabindra N. and Satrajit Dutta (1966), -"Brand Awareness as a Function of Its Meaningfulness, Sequential Position, and Product Utility," Journal of Applied Psychology, 50 (February), 20-24.

Lockhart, Robert S., Craik, Fergus I. M. and Larry L. Jacoby (1976), "Depth of Processing, Recognition and Recall: Some Aspects of a General Memory System," in Recall and Recognition, ed. J. Brown, London: John Wiley, 75-102.

.Mandler, George (1982), "Tile Structure of Value: Accounting for Taste," in Affect and Cognition: The 17th Annual Carnegie Symposium," eds. Margaret S. Clark and Susan T. Fiske. Hillsdale, N.J.: Lawrence Erlbaum, 3-36.

Nelson, Douglas L. (1979), "Remembering Pictures and Words: Appearance, Significance, and Name," in Levels of Processing in Human Memory, eds. Laird S. Cermak and Fergus I.M. Craik, Hillsdale, N.J.: Lawrence Erlbaum, 45-76.

Nelson, Douglas L., Maria Teresa Bajo and Diane Casanueva (1985), "Prior Knowledge and Memory: The Influence of Natural Category Size as a Function cf Intention and Distraction," Journal of Experimental Psychology: Learning, Memory, and Cognition, 11, 94-105.

Nelson, Douglas L., and Martha A. Friedrich (1980). "Encoding and Cuing Sounds and Senses," Journal of Experimental Psychology. Human Learning and Memory, 6 (November), 717-731.

Nelson, Douglas L., and Cathy L. McEvoy (1979), "Encoding Context and Set Size," Journal of Experimental Psychology: Human Learning and Memory, 5, 292-314.

Nisbett, Richard E. and L. Ross (1980), Human Inference: Strategies and Shortcomings of Social Judgment. Englewood Cliffs, NJ: Prentice-Hall

Rheingold, Harriet L. (1985), "Development as the Acquisition of Familiarity,'' in Annual Review of Psychology, 36, eds. Mark R. Rosenweig and Lyman W. Porter, Palo Alto: Annual Reviews.

Thorndike, Edward L. and Irving Lorge (19*1). The Teacher's Word Book of 30,000 Words, N.Y.: Columbia University Press.

Tversky, Amos and Daniel Kahneman (1973), Cognitive Psychology, 5 (September), 207-32.

Warmer, Joel S. and McCray, Ronald E. (1969), "Influence of Word Frequency and Length on the Apparent Duration of Tachistoscopic Presentations," Journal of Experimental Psychology, 79 (January), 56-58.

Zinkhan, George M. and Claude R. Martin, Jr. (1987), Journal of Business Research, 15, 157-172.

Zajonc, Robert B. (1971), "Attraction, Affiliation, and Attachment, in Man and Beast: Comparative Social Behavior, ed. J.F. Eisenberg, W.S. Dillon Washington DC: Smithsonian Institute, 143-179.

----------------------------------------