Dimensions of consumer expertise pdf

Analysis, elaboration, and memory are shown to have more specific interrelationships. The empirical findings related to each dimension are reviewed and, on the basis of those findings, specific research hypotheses about the effects of expertise on consumer behavior are suggested.

Most users should sign in with their email address. If you originally registered with a username please use that to sign in. To purchase short term access, please sign in to your Oxford Academic account above.

Don't already have an Oxford Academic account? Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide. Sign In or Create an Account. Sign In. Advanced Search. Search Menu. Article Navigation. Close mobile search navigation Article Navigation.

Volume Dimensions of Consumer Expertise. This paper reviews and integrates these constructs and presents a framework for the analysis of consumer perceptions of quality. The primary purpose of this study is to develop a model of the … Expand. This article presents the results of a meta-analysis of empirical findings associated with the relationship between objective knowledge OK; i.

Evaluative cues and services: the effect of consumer knowledge. Abstract The purpose of this paper is to provide an insight into the use of evaluative cues by consumers in a services context. In particular, the paper investigates the extent to which consumers' … Expand. Highly Influenced. View 7 excerpts, cites background and methods.

The impact of knowledge types on the consumer search process. Consumer knowledge and its consequences: an international comparison.

By far the theories of consumer knowledge are mainly estimated in the western countries. It is important and necessary to assess theory generalizability and model equivalence across different culture … Expand. Few concepts in the marketing literature have proliferated like the concept of attitude.

However, a closer look at studies investigating attitudes as consumers' responses to marketing efforts reveals … Expand. View 1 excerpt, cites background. A conceptual model of consumer sophistication. Consumers with extensive prior knowledge of a category … Expand.

View 2 excerpts, cites background. Product familiarity is conceptualized in terms of the cognitive structures of knowledge concerning the product that are stored in memory. Such knowledge could have been derived from direct or … Expand. Alba ; Johnson and Raye For all infer- tler distinction and prescribe a drug that is unfamiliar ences, source discriminability decreases as time passes to either consumer.

In each case, expertise affects the Hertel ; Spiro and as interference grows solution. The three individuals differ in their ability to Brockway, Chmielewski, and Cofer Paradoxi- differentiate one cause of pain from another i. Thus, by helping to identify the root cause of a prob- lem, expertise often leads to a faster and better solu- tion. In other cases, however, it has an opposite effect.

Problem Solving That is, it reduces the likelihood that a truly difficult problem will be oversimplified Voss et al. Ex- Problem solving presents an almost unlimited po- perts are more likely to appreciate the complexities of tential for elaboration. In our brief discussion we will a problem and are better equipped to deal with them.

The novice may investigate various solution strategy. Experts are more likely to consider other fac- tors such as risk, security, stability, tax implications, 61n an analogous way, the precise source of the externally obtained and so on, because they possess well-developed con- information may also be difficult to retrieve see, e. Thus, for similar reasons, external in- deal with such information.

The likely outcome is consistent with the classic Finally, research on problem solving has shown that sleeper effect, and our reasoning is consistent with the dissociative even when novices achieve a correct solution, the pro- cue hypothesis often used to explain it Cook et al. Instead, they are more likely to reason backwards from the goal or decompose the problem Interpretive inferences into subproblems, some of which may be irrelevant 4. Consider, for example, a budding photography 4.

Instead of are generated to establish coherence. And although the pur- related information is simplified by the consumer. This also suggests that novices are more susceptible to the influence of external 4. Specific Research Hypotheses 4. Combined with information from Problem solving the previous section, it can be asserted that experts are more likely than novices to elaborate on given infor- 4. Moreover, holding ac- need-in terms of its surface structure and less likely to curacy constant, expertise leads to easier and more ef- represent it in terms of its deep structure.

Expertise enhances a consumer's accuracy in generating 4. It increases efficiency by allowing inference to substitute for search and by enabling the consumer to comprehend easily the benefits of product features. Expertise allows the consumer solving a problem to suggest that it is easier to communicate veridical infor- identify its cause accurately and to avoid oversimpli- mation to experts and more difficult to deceive them.

Efficiency of the process is in- creased by the expert's ability to generate solutions in- MEMORY ternally and to avoid purchase of unneeded products or product features.

The final dimension of consumer expertise that we Experts enjoy these advantages as a result of their will discuss is memory. In particular, we examine the greater factual knowledge, more highly differentiated likely effects of product familiarity on memory for spe- knowledge, and superior analytic skills. From the cific product information. We have chosen to dis- assumption is granted, there are several well-established tinguish between memory for simple information e.

This distinction reflects As might be expected, previous exposures to a word clear differences in experimental paradigms as well as increase the ease with which it is recognized. This phe- likely areas of investigation in consumer research. Thus, nomenon has been labeled perceptual enhancement we will initially focus on the processes of brand name Jacoby a. The issues of principal interest for con- recognition and recall.

We then will discuss memory sumer research are I the number of exposures needed for more complicated product information such as ad- for enhancement, and 2 the types of exposures that vertising messages. Many brand names begin as nonsense e. Baker et al. Each of these will be discussed brand name. Word recognition studies have shown that in turn. When consumers quickly tially more difficult to identify than real words Feustel, scan a typical supermarket shelf display, they are ex- Shiffrin, and Salasoo ; however, after several ex- posed to a large number of brand names that are posures approximately five , the differences between embedded in an even larger number of nonbrand, words and pseudowords disappear Salasoo, Shiffrin, product-related words.

It seems likely that only a small and Feustel a. More importantly, even a year after proportion of these words are even read. As was dis- exposure, pseudowords show no recognition deficits cussed earlier, repeat purchases by brand loyal con- relative to real words. Salasoo et al. On the from the first few exposures. The exact nature of this other hand, a new or infrequent purchase, such as buy- code is still a matter of considerable debate Jacoby ing a cold remedy, may require at least brand name a, b; Jacoby and Brooks ; Johnston, Van recognition for the brand to be considered.

Therefore, Santen, and Hale ; McClelland and Rumelhart it is reasonable to expect that the ease of reading brand , ; Morton , ; Salasoo et al. However, most re- Direct research on in-store processing of verbal in- searchers would agree that this code plays a crucial role formation is rare cf. Hoyer ; Russo et al. However, there is an extensive cognitive literature on Therefore, the establishment of the code is a funda- word recognition in laboratory settings.

Word recog- mental first step in acquiring brand knowledge. However, most of the observed extremely brief exposure, with missing letters, etc. For example, exposure dicate whether it is an actual English word.

Therefore, it is natural to assume that 'We have adopted this approach because 1 visual recognition a new memory code must be established whenever an processes are covered to some degree in our discussions of automa- existing unrelated word is used as a brand name.

It is ticity and concept formation in the Appendix and in our discussion not known whether this is easier or more difficult than in this section of brand name recognition, and 2 empirical inves- establishing the code for a pseudoword. Similarly, little tigations of recall have been dominated by verbal stimuli perhaps because investigations of recall for visual stimuli are greatly compli- cated by the lack of viable nonverbal response measures.

It seems that identifying the exposure context is seldom ognition tasks used in most verbal learning experiments e. This does not mean that con- ; Mandler The latter require subjects to retrieve infor- textual information does not affect word recognition see Jacoby mation that will determine whether the word in question occurred a, b and our subsequent discussion or that retrieving con- in a particular context e.

We suspect that but similar visual appearances Murrell and Morton previous meanings interfere with the establishment of ; Witherspoon and Allan Moreover, en- unique memory codes; however, research is needed to hancement does generalize to words that have the same make clear predictions.

First, exposure to a partic- served but not specifically named Jacoby and Brooks ular word "primes" other words that are semantically Finally, there is no generalization to printed or episodically associated with it, facilitating their rec- words from pictures, spoken words, or self-generated ognition for a very brief period of time i. Such priming from recent in-store advertising exposures and product usage exposures.

Be- exposures seems unlikely. However, the broad gener- cause the brand name is usually presented identically ality of the priming stimuli i. Exposures to product ap- such as the product class name. As we have discussed pearance but not brand name during usage are not earlier, the ways in which consumers categorize brands, likely to enhance brand recognition unless product ap- and therefore the brands that are primed during search, pearance is a salient part of packaging.

Advertising, on change as product familiarity increases. For instance, in television commercials tation in word recognition due to previous exposures the product but not the brand name may be visually to the same word. Words that occur frequently in the presented and the name spoken. These sorts of expe- language are more easily identified than infrequent riences are not likely to enhance visual brand name words. Moreover, a single exposure to a word produces recognition. Finally, if one reads about a brand as part a facilitatory effect that lasts for at least a week Jacoby of advertising copy or publicity, enhancement should a; Tulving, Schacter, and Stark , but not as result even though the visual appearance of the brand long as a year Salasoo et al.

Some authors have name differs in the two contexts. If advertising copy is hypothesized that there are enduring memory codes "skimmed," but not read, the effect of exposure to the whose thresholds are gradually lowered as they are more brand name may be reduced. In summary, we have identified three ways in which Other authors have emphasized the role of particular product familiarity may enhance brand name recog- exposure episodes e. They point out that sideration.

First, a small number of exposures to a new a single recent exposure often overrides the differences brand name are required to establish a permanent between words that presumably have accrued over a memory code. Presumably, this code mediates subse- lifetime. Second, structural aspects familiarity, in the sense of total number of product- of consumer knowledge determine which brand names related experiences, may not have much effect on brand are briefly perceptually enhanced via associative name recognition.

Rather, the most recent experiences priming in various situations. As discussed earlier, cog- will be most influential. The same authors have also nitive structure changes systematically with product fa- emphasized the episodic nature of perceptual enhance- miliarity.

Third, frequent or recent exposures to the ment, arguing that the match between the types of in- printed brand name enhance recognition. The effects formation encoded during previous exposures and the of a single exposure may last for several days. Even if this assumption proves false, however, there is reason to believe ognition environments or, conversely, which types of that perceptual enhancement may affect consumer behavior in at product-related experiences enhance in-store brand least two other ways.

First, several recent studies have provided ev- name recognition. In many consumer situations, terms of memory processes Barsalou and Ross In would have information processing characteristics that essence, the consumer's task is to produce instances of are similar to "automobiles. There gories remain unchanged. Specifically, prototypical brands that the size of the memory effects associated with goal- are recalled more frequently and more rapidly than derived categories increases with product familiarity.

Without strong Kelley, Bock, and Keil ; Rosch and Mervis ; memory effects, we would expect the evoked set of the and Smith, Shoben, and Rips The evoked set of the experienced Nedungadi and Hutchinson Little is known backpacker is likely to be more memory-based.

Another aspect of brand name recall that may affect It seems reasonable to expect, however, that the impact consumer decision making involves the extent to which of frequency of instantiation decreases with expertise different brands are recalled together.

Brands that are because, unlike family resemblance and preference, it frequently recalled together are likely to be included in is not directly related to product attributes and benefits, the same evoked sets and, therefore, compete more di- and it may be affected by irrelevant factors such as fre- rectly with each other than brands that are seldom re- quency of advertising exposure.

If expert recall is guided called together. Moreover, the composition of the set by analytic processes, then frequency of instantiation of considered brands can "bias" choice in several ways should be a lesser determinant for experts relative to e.

Interestingly, Hutchinson found that effects-Huber, Payne, and Puto ; and similarity advei-tising expenditures were highly correlated with effects-Tversky ; Tversky and Sattath Brand names tend to be recalled in categorical clusters Also, Barsalou , a found that goal-derived Hutchinson ; also see Bousefield and Sedgewick categories are much less well established in memory ; Friendly ; Gruenewald and Lockhead In particular, This covariation may arise because the category name prototypicality ratings are less predictive of recall fre- functions as a retrieval cue or simply because brands quency Barsalou a , and automatic detection pro- in the same category are highly associated with each cesses are less developed Barsalou and Ross In either case, the consumer's category structure Thus, prototypicality is not expected to predict inclu- should have a strong influence on brand name clustering sion in the evoked set as well for goal-derived categories during recall.

As was discussed earlier, the expert's cat- as it would for common product categories. Similarly, egory structure differs significantly from the novice's. This brand recall protocol was produced by goal-derived categories. The apparent reason for this a single subject in response to instructions to list the difference is that goal-derived categories are used infre- brand names of any products that can be used to treat quently relative to taxonomic categories.

There is evi- the symptoms of the common cold from Hutchinson dence that repeated use of these categories can make Brand name clustering is clearly related to benefit them functionally equivalent to common categories in categories based on product class tables given in the Handbook of Nonprescription Drugs For in- stance, Alka-Seltzer Plus, Congespirin, NyQuil, and Comtrex are all multi-ingredient drugs that contain a nition is subjectively detectable and may be attributed to familiarity decongestant and an analgesic, and can be used to treat by the perceiver or experienced as a feeling of familiarity even when fluency has been caused by stimulus properties; Witherspoon and head colds.

This type of suppressant and, in principle, could have been clustered brand familiarity is likely to have a positive effect on brand attitudes with cough remedies such as Robitussin or Vicks For- Obermiller ; Zajonc and Markus Second, consumers with mula It is of interest to note that Vicks Formula high levels of product familiarity are likely to have read all product- D was clustered with cough remedies even though related words more frequently and recently than consumers with low levels of familiarity.

The aggregate effect of perceptual enhancement it contains a decongestant and is functionally similar on the entire product vocabulary should make reading product in- to NyQuil and Comtrex. This is probably the result of formation easier and more rapid, thereby facilitating further learning.

Hutchinson also found that brand name clustering was more related to ingredients for pharmacy students Memory for Complex Information than for marketing students. Thus, a knowledgeable As noted in our discussion of selective encoding, ex- consumer might categorize Formula D as a head pertise is likely to affect the extent to which consumers cold treatment, or perhaps form a third, "cough-cold," search for and process product-related information category that includes Comtrex, NyQuil, and Formula through its effects on comprehension and cognitive ef- D, but not the others.

Furthermore, such a consumer fort. Naturally, differential exposure should affect recall. On the other recall are strongly related independently of exposure hand, a novice consumer might consider only Comtrex effects see Alba and Hasher This relationship is and NyQuil.

This constitutes an agenda effect similar a robust one, having been observed across a variety of to the ones discussed by Tversky and Sattath Alba ; Au- In conjunction with our earlier discussions of cog- subel and Fitzgerald ; Chiesi, Spilich, and Voss nitive structure and analysis, this example suggests at ; Voss et al. First, categorical clustering is likely to be more based, knowledge may afford the expert an opportunity specific for experts than for novices e.

Second, the attributes that are associated with different from the ones the novice may use. This should be particularly true for informed decision. In this case, the advantage to the experts because they are most sensitive to incongruence expert lies in the sheer quantity of information that can cf. However, a contrast effect occurs when consistency However, in addition to this main effect, expertise is assessed relative to a decision rather than a schema: and recall are likely to interact in other ways that are that is, there is a tendency to recall decision-consistent important to decision making.

One general type of in- over decision-inconsistent facts. This effect occurs teraction involves the nature of the stimulus informa- whether the facts reside in semantic memory Koriat, tion and is driven by the expert's ability to notice rel- Lichtenstein, and Fischhoff or are part of a re- ative differences in the importance, relevance, and con- cently encoded message Dellarosa and Bourne Chakravarti ; Johnson and Russo I , though the latter results probably stem from the differential Importance Eflects.

First, consider the importance attention paid to chosen and rejected brands. Regard- and relevance of particular facts. A substantial body of less, as Koriat et al. Facts that are relevant to the theme of a message Johnson ; Kintsch and Memory Measures. In addition to these stimulus van Dijk , the attainment of a goal Brewer and interactions, the effects of knowledge on memory also Dupree ; Lichtenstein and Brewer ; Owens et interact with the type of memory test that is used.

The al. When cued recall Anderson and Pichert ; as relevance and importance may depend on one's point Koriat et al. Yekovich are employed, however, memory dif- In the context of consumer decision making, per- ferences either disappear or are greatly attenuated. Jacoby et storage, nor does processing it with particular biases al. Due to their superior about what is and what is not important. Thus, in sit- ability to distinguish between relevant and unimportant uations in which recall is not important, as when a product information, experts should recall a greater package or point-of-purchase display reminds consum- amount of important and decision-relevant informa- ers of previously advertised information, the memory tion; that is, the expert enjoys an advantage in terms of advantages enjoyed by experts may diminish somewhat the quality of information recalled.

As a result, when decisions are memory-based, experts maintain their su- perior ability to be analytic. Moreover, this advantage may grow over time. Initially, all consumers may retain ''It should be noted that the precise role that memory plays in decision making has been questioned. Recent research shows that a reasonably large amount of information.

However, recall of specific information may, at times, be unrelated to judgment as memory fades, the effects of knowledge on recall may cf. Anderson and Hubert ; Carlston ; Kardes ; Lich- increase see below. Thus, the carry-over effects of ad- tenstein and Srull ; Lingle and Ostrom ; Sherman et al.

In such cases, it appears that people engage in "judgment-referral" Alba and Ofir or affect- Ward are likely to be moderated by consumer referral Wright a , thereby relying on previously generated knowledge. In other circumstances, however, recall of information Schema Consistency. A second aspect of the stim- or internal responses to that information does play a significant role ulus that may interact with knowledge is its consistency.

This appears to be especially true when no sistent facts, and both are recalled better than schema- judgment was made at the time of encoding, new alternatives become irrelevant facts Hastie and Kumar ; Srull 1; available, or new information is obtained and must be evaluated in Srull, Lichtenstein, and Rothbart Thus, if fea- light of previously acquired knowledge.

Nonetheless, the ability to rec- Finally, it should be noted that the effects of exposure ognize a previously encountered message may be of extend beyond recall in at least two ways.

First, through limited usefulness if the meaning, importance, or im- its effect on simple recognition, prior exposure may fa- plications of its assertions are not fully understood. Sec- ond, through its effect on perceived familiarity, previous exposure to product-related assertions may increase In the preceding section, we described several em- pirical relationships that exist between knowledge and their perceived validity cf.

Bacon ; Begg, Armour, recall of complex information. Left unspecified, how- and Kerr ; Hasher, Goldstein, and Toppino In It may be assumed that all consumers are affected by this section we will attempt to show that the ability to this phenomenon but that novices are affected to a recall product-related information is a natural by-prod- greater degree because they are less aware of the true uct of both familiarity and the other dimensions of ex- validity of a greater number of assertions.

That is, recall covaries with practice, the de- velopment of cognitive structure, and the consumer's Cognitive Structure and Analysis. Recall is also af- analytic and elaborative skills.

It is well known of those facts, and an understanding of how those facts that repetition enhances retrieval Crowder ; are interrelated.

Hintzman ; Sawyer and Ward And because Consider first the advantages of possessing a schema experts receive greater exposure to particular brands to guide encoding and retrieval. For example, assume and attributes, their memory for that information is that an expert possesses and a novice lacks a well-de- better.

However, for some brands and attributes, both veloped schema for a computer. When processing a de- novices and experts may reach an asymptotic level of scription of a new computer, the expert is able to rec- familiarity. At such high levels of familiarity, retrieval ognize the attributes being described and identify them may be relatively effortless, even if one's overall level according to their importance and typicality.

This af- of expertise is quite low. When two ideas are closely fords the expert several advantages. For example, presen- cause them to process important information more se- tation of the name "Rolls Royce" may cause a con- lectively and intensively than other information, re- sumer to generate thoughts about luxury or wealth.

In- sulting in a corresponding pattern of recall. Moreover, deed, a popular goal of advertising is the creation of an an expert also should be more familiar with the attri- association between a brand and its image, position, or butes of the product. Consequently, when processing major attribute Reeves 1.

Due to the effects of recency on recall, tion load Johnson and Kieras Morrison A person who As two of several determinants of recall, repetition possesses a schema can use its features to access cor- and recency are likely to be very influential in the de- responding information contained in the new infor- cision process, particularly for novices. Because novices mation cf. Brewer and Dupree ; Egan and may engage in less search, lack the expertise to use other Schwartz ; Lichtenstein and Brewer The retrieval cues see later , and are less equipped to make schema may be viewed as a multifaceted retrieval cue product comparisons, they are more likely to be influ- Anderson et al.

Parentheti- 1. As such, it should have a large facilitating effect cally, this suggests that problem framing techniques may on recall but a relatively small effect on recognition. Some schema theories also tively reminding them of features on which the adver- make provisions for the recall of atypical features. Spe- tised brand fares well. Experts, on the other hand, are cifically, items that do not match the schema are pre- more likely to recall brand information spontaneously sumed to receive special tags, making them particularly and are less likely to choose a brand on the basis of its memorable Graesser et al.

Whereas expertise a function of time. As time passes, episodic facts should leads to relatively effortless organizing of the stimuli, fade, memory should become increasingly schema- consumers who lack expertise may struggle to under- driven Bartlett ; Dooling and Christiaansen , stand how one attribute is related to another.

Moreover, and the probability of recalling atypical features should by "chunking" related pieces of information, the effec- decrease Arkes and Harkness ; Bower et al. Schmidt and Sherman ; Smith and Graesser 1. Finally, because a major processing difference be- Consider next the relations that exist among product tween experts and novices appears to be in their ability attributes. Just as category structure affects the quantity to organize information, one would expect recall but and quality of brand recall, the manner in which com- not recognition differences on a memory test cf.

Del- plex information is organized in memory should affect larosa and Bourne Information that is organized the way it is recalled see Klein and Kihlstrom First, at- name is an effective recall cue Tulving and Pearlstone tributes may be organized in terms of the brands that Srull has illustrated this effect nicely with possess them.

Research suggests that familiarity me- brandlattribute information. When he presented infor- diates a consumer's ability to organize attribute infor- mation in an organized or "blocked" fashion, recall mation in this way Srull ; also see Pryor and Os- differences between high and low knowledge subjects trom When presented in a "random" Second, attributes themselves may be organized cat- fashion, however, large differences emerged between the egorically in terms of their semantic, functional, and groups.

Because organization typically plays a lesser role decision-related implications cf. Dellarosa and Bourne in recognition than in recall, differences in recognition ; Hamilton, Katz, and Leirer ; Means and as a function of knowledge should again be small see Voss And although some attributes may be fa- Srull In the previous section we described mensions may differ.

For example, the attributes leg how different concepts can be organized via association room, bucket seats, and air conditioning all function to a common higher-order concept. The final major to provide the more abstract benefit of comfort. Here, explanation of knowledge-based recall effects involves knowledge allows one to understand how a particular elaboration and the direct associations it creates among attribute is related to its functional category.

When a facts during encoding. These inferred associations, which impose one. First, comprehension differences Carpenter ; Lesgold, Roth, and Curtis ; Man- should arise across segments because experts are able dler and Johnson The effects are likely to be to perceive how different attributes relate to one an- greatest for complex and technical products cf.

These comprehension differences amplified. Coherence aids comprehension by allowing occur regardless of the familiarity of the individual at- the consumer to understand how one assertion is related tributes. A message that describes a product in terms to another. It aids recall because the associations that of unfamiliar attributes is incomprehensible in the basic are formed provide retrieval cues. If two assertions are sense.

When the attributes are understood but the re- linked, one may cue the other. In addition, increased lations are not, the message appears unorganized but coherence often is characterized by a decrease in the the isolated facts may still have some meaning.

For ex- amount of time required to process the information. If the consumer processes such as memory search and inference making knows what "bits" and "memory" are with respect to computers but does not fully understand the implica- "This point is often ignored by advertising practitioners. Rarely do they take into account co- load on the computing power dimension and will not herence relations but see Thorson and Snyder The psycho- be able to use "power" as a retrieval cue.

The first concerns the time Haberlandt and Graesser Although the effect has been shown to be durable novice differences in recall and comprehension are fur- over short periods of time, over very long periods it ther amplified when the time available for these extra may start to resemble the typicality effect.

That is, as search and inference processes is limited, as it is with recall becomes more schema-driven, memory for in- broadcast media cf. Chaiken and Eagly Finally, consistent facts may become very poor cf. Hastie and because the recall differences between experts and nov- Kumar ; Wyer and Srull When inconsistent information is presented and sub- The other relevant class of associations more closely jects are prevented from elaborating, the effect disap- resembles embellishment.

Because experts possess a pears Srull 1 ; Srull et al. A corrective ad contains information haps resulting in a more polarized attitude; Tesser and that is inconsistent with previously stated information Leone Such elaboration may create retrieval about a product. Thus, if information about a product paths to the new information that facilitate its recall is recalled long after the corrective ad has been run, the Bradshaw and Anderson ; Craik ; Kisielius association between the corrective facts and general and Sternthal Sawyer Elaboration of this sort also helps to explain the re- And if the consumer is not highly involved when lationship between expertise and the recall of inconsis- processing the corrective ad, the associations may not tent information.

In addition to their ability to under- be formed at all. In addition to enhancing memory for perior ability to detect inconsistencies among facts particular facts in the ways described above, expertise Fiske et al. The enhanced recall of inconsistent also may reduce the extent to which recall of those facts information may be explained in terms of the greater is inhibited by competing information.

A robust finding processing devoted to it. If an attribute is incongruent in memory research is that as the number of items an with other facts or with one's expectations, an attempt individual learns increases, the proportion of those may be made to resolve the inconsistency. This results items that can be recalled decreases e. It also has been demonstrated that once a set ; but see O'Sullivan and Durso In a decision of brands or attributes has been learned, presentation context, however, facts that do not support the final of a subset of them sometimes inhibits recall of the re- decision need not be inconsistent with each other or mainder, with the amount of inhibition increasing with with the decision maker's schema.

Thus, they may not the size of the presented subset Alba and Chattopad- receive greater elaboration and, because they will be hyay a, b. Thus, advertisers may inhibit recall less directly linked to the final decision, facts inconsis- of some competing brands or may frame decisions tent with a decision may be recalled less well than con- through the inhibition of particular attributes by men- sistent facts.

Regardless, because recall is explained in tioning other nonthreatening brands or attributes in terms of associative cuing, much smaller effects should their ads. However, discussion of them is beyond the scope ofthis paper. Instead, we would like to note the relationship between interference and knowledge.

That is, interfer- I2In this case experts outperform novices because they are able to ence effects appear to be strongest early in the learning generate more retrieval cues.

It should be noted that in other cases process. As familiarity with the stimuli increases, and an opposite and paradoxical effect may obtain. That is, when a novice as knowledge of them becomes better organized and needs to engage in additional elaboration or reasoning in order to more differentiated, interference decreases Alba and comprehend the same information that an expert comprehends easily, the novice may actually recall it better cf.

O'Brien and Myers Chattopadhyay a; Anderson a; Hayes-Roth This effect is analogous to our previous suggestion that novices may ; Pirolli and Anderson ; Srull and Brand be more accurate than experts in identifying the source of an inference.

Memory for complex information 5. Explanations and implications 5. Specific Research Hypotheses it during visual search. Repeated or recent exposure to a single brand or attribute may lead to easy retrieval of Research hypotheses regarding memory are presented it in situations requiring unaided recall.

Such effects in Exhibit 5. Clearly, experience and expertise can have require little expertise. Wider experience results in the wide-ranging effects on a consumer's ability to remem- accumulation of more information, which enables con- ber product-related information. Different types and sumers to include more brands in their memory-based amounts of memory facilitation obtain from simple ex- evoked sets and to recall and use more attributes during posure, repeated exposure, varied experience, differ- memory-based decision making.

However, wider ex- ences in cognitive structure, and analysis. Therefore, if the ex- meaningful criteria, and attributes take on task-relevant perimental task requires differentiation or analysis, importance values. Consequently, the quality of the in- performance may be inversely related to purchase, formation recalled by experts and novices also differs.

In addition, The complexity of consumer expertise also means we described many differences that exist between ex- that even when knowledge is considered in a more pre- perts and novices on each one. In the process, an at- cise manner, predictions about its effects on behavior tempt was made to demonstrate how differences in ex- are not simple.

Consider, for example, the relationship pertise are reflected in the most fundamental consumer between expertise and evoked set size. For the simple behavior, namely, decision making.