The Effects of Missing Information on Decision Strategy Selection

Sandra J. Burke, The University of Michigan
[ to cite ]:
Sandra J. Burke (1990) ,"The Effects of Missing Information on Decision Strategy Selection", in NA - Advances in Consumer Research Volume 17, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT : Association for Consumer Research, Pages: 250-256.

Advances in Consumer Research Volume 17, 1990      Pages 250-256


Sandra J. Burke, The University of Michigan

[The author gratefully acknowledges comments on earlier drafts of this paper by Susan Heckler, Chris Puto, Youjae Yi, and especially Michael D. Johnson.]

Many consumer decisions are characterized by missing information, but an understanding of how this effects decision processing is lacking. This paper proposes a framework to explain that influence. It is argued that when attribute values are missing, a consumer must infer them in order to use some preferred strategies (attribute-based comparisons). Thus, a consumer may attempt to avoid the increased inferencing effort by switching to (1) alternative-based processing, or (2) hierarchical elimination strategies. Consumer experience, task complexity, category attribute redundancy, and decision importance are all argued to influence the magnitude of the proposed strategy shifts.

Much of the available research on consumer choice strategies has implicitly assumed that experimentally available information completely describes each choice alternative. Yet, in many choice situations consumers face incomplete information or partially described alternatives. This may result from advertisements or packaging which only mention some attributes, because of imperfect attribute memory for nonpresent alternatives, or because some attributes are experiential and can only be evaluated after product use. In these situations, "missing" attribute values must be inferred from available product cues.

While the role inferences play in attribute weighting and valuation has been explored (cf Ford and Smith 1987, Meyer 1981), their role in decision strategy has not yet been researched. Missing attribute inferences may lead consumers to utilize different choice strategies than in situations of complete information. The purpose of this paper is to explore how missing information and particularly inferences may affect decision strategies.

It is proposed that as the amount of externally available attribute information for a given alternative decreases, necessary inference-making increases. In order to adapt to this situation, consumers may alter their processing. Specifically, the presence of missing information may introduce an inference stage of processing in order to perform some preferred strategy (e.g., attribute based comparisons). While, for example, attribute-based strategies are relatively easy to use in certain contexts (Tversky 1969), an added inference stage may increase their complexity. Alternative-based strategies, however, may allow more direct processing, and circumvent any added inference stage. This may result in a relative shift away from attribute-based strategies toward alternative-based strategies as "missing" information increases.

Another way to avoid inferencing may be to continue to use an attribute-based strategy, by hierarchically eliminating alternatives based on those attributes with available values. This may be particularly appropriate when information is available on more diagnostic attributes. Summarizing, it is proposed that in situations with missing information, consumers will attempt to avoid the effort of inferences by switching to (1) alternative-based strategies, and/or (2) hierarchical elimination strategies which focus on diagnostic, available attributes. As detailed below, these shifts may be moderated by consumer experience, decision importance, task complexity, and the category attribute redundancy.


Decision strategies are often viewed as varying both in their level of difficulty and their ability to optimize the accuracy of a given decision (Beach and Mitchell 1978). It has been suggested that processing effort may affect the utilization of a given strategy in a given situation and that decision makers will resort to choice strategies which reduce cognitive strain (Bettman 1979; Newell and Simon 1972). The implicit assumption is that consumers trade-off accuracy and effort or consider the costs and benefits of particular strategies (Payne 1982). These costs and benefits may be influenced by aspects of the task or choice context as well as individual differences (Russo and Dosher 1983; Shugan 1980; and Wright 1975). Compensatory strategies in particular require a good deal of cognitive computations and place demands on short term memory. Using a linear compensatory strategy each alternative must be evaluated separately on each of its relevant attributes, and the resulting utility or value of each attribute must be used to arrive at an overall evaluation. In contrast, sequential elimination strategies are generally easier to apply since the alternative set is reduced and cutoffs are used, restricting attention to only part of the available information about each alternative (Payne 1976). Yet, more compensatory strategies may produce better choice alternatives.

Further, Tversky (1969) and Bettman and Jacoby (1976) suggest that attribute-based processing is easier than alternative-based processing because similar units (attributes) are being compared with one another. Slovic (1975) argues this may be due to an emphasis placed on an attribute's discriminating power. Some empirical evidence supports more widespread use of attribute-based processing (Capon and Burke 1977; Ford 1989; and Russo and Rosen 1975). However, other empirical research supports equal amounts of alternative and attribute processing (Bettman and Jacoby- 1976).

One reason for these conflicting findings is that a number of variables may mediate differences in the effort required to apply the different strategies. Use of the strategies varies with task complexity (see Payne 1982 for a review), information format (Bettman and Kakkar 1977; Sheluga and Jaccard 1979; and Tversky 1969), the phase of processing (Bettman and Park 1980; Biehal and Chakravarti 1986; Johnson and Meyer 1984; Lussier and Olshavsky 1979; Payne 1976), and individual differences (Bettman and Kakkar 1977; Bettman and Park 1980; Biehal and Chakravarti 1982, 1986; Capon and Burke 1980; Cohen and Basu 1987; Jacoby et al 1976; Kardes 1986; Park and Lessig 1981; Sheluga, Jaccard, and Jacoby 1979; and Sujan 1985).

Finally, paralleling the discussion here, Johnson (1984, 1988) studied the effects of noncomparability (the degree to which alternatives are described by different attributes) among alternatives on choice of decision strategy. These studies show that as comparability between alternatives decreases, subjects increasingly crossover from attribute-based to more alternative-based strategies, possibly due to the increase in effort involved in attribute-based processing as comparability decreases. What these studies suggest is that on the whole, situational factors mediate the effort required to use various decision strategies and lead to shifts in strategy use. We now explore the particular role that inferences regarding missing information play in strategy selection.


While the effect of missing information on decision strategy selection remains unclear, a few studies have looked at its effects on evaluation with mixed results. Slovic and MacPhillamy (1974) suggest that when making pairwise comparative judgments, an attribute is weighted more heavily if attribute values are available for both alternatives than just one alternative. They suggest that comparisons involving dimensions of complete versus missing information may demand less cognitive strain and be more appealing. In an unpublished paper discussed by Bettman (1979), Wright also looked at decisions between two alternatives each with one common and one different attribute, and found opposing results.

One reason for these conflicting results may be that the Slovic and MacPhillamy studies utilized linear models to estimate the weights assigned to the two cues without considering that different models may have been utilized in the two situations. If, as suggested here, subjects switched to more alternative-based processing when information was missing and inferencing was difficult, then simply looking at the regression weights from linear models could be problematic. The effect of missing information on decision strategy selection was not fully explored in these studies.

Following Slovic and MacPhillamy, a number of studies have suggested that missing and subsequently inferred information is discounted or ignored (Meyer 1981; Yates, Jagacinski, and Faber 1978). Yet other studies report opposite or mixed results (Ford and Smith 1987; Johnson and Levin 1985). Overall, these studies suggest that individuals do in fact make inferences when faced with missing information, and that inferential processing requires cognitive effort (Alba and Hutchinson 1985; Meyer 1981).

However, the emphasis throughout this stream of research has been on attribute weighting and valuation and the cues used to infer missing information. Just how missing information may affect decision processing is unclear. The primary methodologies used in these studies were linear model estimates of attribute weights based on likely-to-buy indications from alternative attribute descriptions. Discounted weights, and overall decreases in satisfaction ratings on alternatives with missing information, may reflect some consumer switching from attribute-based to alternative-based processing.

If the effort required to infer the missing information is high, as when cue redundancy is low (Hagerty and Aaker 1984; Johnson and Katrichis 1988), a switch to more holistic, alternative-based processing could have led to an implicitly discounted attribute weight. However, these methods of analysis impose an additive or linear evaluation method which may not reflect how the subjects actually were evaluating the alternatives and thus may not detect a difference in evaluative procedures.


The basic proposition here is that missing information and the subsequent need to form inferences may result in a shift to processing strategies that avoid the inference process. In general, attribute-based comparisons are relatively easy, making attribute-based strategies (in which individual attributes are chosen and compared across alternatives one at a time) very attractive to consumers (Tversky 1969). If, however, information is missing on one or more of the attributes of comparison, inferences required to "fill-in" the missing values may require significant incremental effort (Alba and Hutchinson 1985). This additional and potentially effortful inference stage may lead consumers to shift to a simpler strategy. As the required effort necessary to utilize an attribute-based strategy increases due to the need to form inferences on individual attributes, consumers may shift to more straightforward alternative-based strategies. In such strategies, all attribute values for a given alternative are combined to provide overall alternative evaluations which can be compared to make a choice. This combining of values across attributes to form overall evaluations may circumvent the need to make attribute-based comparisons (Johnson 1984). Alternatively, consumers may use hierarchical elimination strategies on available information if this is reasonable. Such eliminations are reasonable when based on particularly important or diagnostic attributes (Klein 1983). The effects of missing information and subsequent inferencing should be characterized by some increase in the number of inferences used in processing in missing information situations, and a relative shift from the use of attribute-based strategies to the use of alternative-based choice strategies. This relative shift may allow attribute-based processing to continue, but decrease proportionately while the use of alternative-based strategies increases as missing information is introduced. Based on this, the following propositions are forwarded:

P1a: As missing information increases, the number of inferences regarding attribute values used in processing should increase.

P1b: As missing information is introduced, the use of attribute-based processing to choose among alternatives should decrease relative to alternative-based processing.

P1c: Any attribute-based processing that does occur should involve more hierarchical eliminations based on diagnostic attributes.


The amount of effort that inferencing requires should affect the magnitude of the proposed strategy shifts. This effort should vary from consumer to consumer based on many factors. One such factor may be the consumer's level of expertise within the product category. As inferences about missing attribute values are typically assumed to be based on cues provided about the same-brand, other brand, or category (Ford, et al 1987), consumers with more expertise in the category may be able to more easily use the available cues to form inferences (Alba and Hutchinson 1987). They may already have well-formed ideas regarding the correlations between the presence and values of different attributes (Johnson and Katrichis 1988). Further, they may have previously been faced with missing information in the category and thus be familiar with using cues to form inferences. Therefore, experienced consumers may require less effort to form inferences regarding missing attribute values. As a result, their choice of a strategy may not be influenced by missing information to the same degree as novice consumers.

This idea is supported by research in the area of consumer knowledge and its effects on processing. It has been argued that over time and repeated exposures, consumers may develop a set of expectations about a product category (Sujan 1985). Further, these expectations may provide consumers with a set of hypotheses regarding which attributes go together, typical configurations of attributes, and expected performance levels of products that can be matched to these categories. Extending this view, if a consumer has such expectations, classification of a product with missing information to a category based on the available information permits one to adduce or predict the missing information (Nisbett and Ross 1980). Therefore, experience with the product category can lead to less effortful inferencing.

Alba and Hutchinson (1987) suggest a sin consequence. They suggest that consumer knowledge has two parts; familiarity, or the number of product related experiences accumulated by the consumer; and expertise, or the ability to perform product-related tasks. They hypothesize that pro familiarity increases the likelihood of analytic processing. Consequently, highly familiar consumers will be more likely to perform tasks which require analytic processing. Schema-based inferencing, as discussed above is one such task. very low levels of knowledge, novices lack the ability to make schema-based inferences. Therefore when inferences are required, expertise should make them easier and more likely to be generated.

Alternatively, Alba and Hutchinson (1987 and Sujan (1985) suggest that while some level of knowledge is necessary for all inferencing, expel may be more judicious in their use of nonanalytic inferencing. It has been argued that consumers with at least a rudimentary knowledge structure may prefer to use prior evaluations based on simplistic criteria to make choices (Bettman and Park 1980). They may tend to use schema-based affective processes nonanalytic inferencing more often than consumers with very low or high levels of category knowledge (Sujan 1985).

Overall, these arguments suggest that the likelihood of inferencing increases with experience When faced with missing information, consumers with high levels of experience in the category may have less incentive to switch to alternative-base processing since analytic inferencing may not involve significant effort. This may also occur for consumers with moderate levels of experience will may more readily engage in nonanalytic inferencing. Conversely, those with lower levels of experience will require more effort to infer missing values and should show a greater shift to alternative-based processing. Thus the following propositions are forwarded:

P2a: The number of inferences used in processing when missing information is introduced should increase less for novices than for consumers with moderate and high levels of category experience.

P2b: Novices should exhibit a relatively stronger shift from the use of attribute-based processing when missing information is introduced than consumers with moderate or high levels of category experience.


Another factor which may affect the effort involved in inferring missing attribute values and therefore the magnitude of the inference induced strategy shift may be the overall complexity of the decision task. Studies support the concept that task complexity can directly influence the type of decision strategy employed by the decision maker, as discussed earlier in this paper. We extend this framework by suggesting that one way in which task complexity affects strategy use is via its mediating effect on the effort of inferencing in situations with missing information.

Consider that tasks characterized by little missing information may require that only a few attribute values be inferred, and therefore only a marginal increase in effort is needed to process by attribute. This marginal increase may not be sufficient to induce the consumer to switch to alternative-based processing. However, as the amount of missing information increases and more attribute values must be inferred to process by attribute, the effort required increases as well. Eventually, consumers may reach some threshold of missing information where inferencing effort triggers a change in processing.

Similarly, in tasks involving no time pressure, the effort and time required to infer missing attribute values may not deter the consumer from processing by attribute. However, as time pressure is introduced, the consumer may feel unable to perform the decision task while taking extra time and effort to infer missing values. Consequently, the consumer may avoid the effort and time involved in inferencing by switching to alternative-based processing.

Finally, in decision tasks involving few alternatives, the overall decision effort involved is lower than in tasks involving many alternatives. It could be argued therefore, that the extra effort involved in inferring missing values would be more likely to cause a strategy shift in situations with many alternatives particularly when the missing information is spread across alternatives. However, as discussed earlier, many studies have shown the switch to phased strategies as the number of alternatives increased, in which an elimination phase typically involving an attribute-based strategy (such as EBA) is used first to reduce the alternative set. Therefore, increases in switching to alternative-based strategies to avoid inference effort in the many alternative task versus in the few alternative task may be masked or moderated due to the increased use of elimination strategies. At the same time, any switch to elimination strategies may be enhanced.

In summary, there is evidence that task complexity may directly affect the consumers' use of decision strategies. Further, complexity may also affect decision strategy use indirectly via its mediating effect on the effort involved in inferencing. Based on these arguments, the following propositions are forwarded:

P3a: The shift from the use of attribute-based processing to alternative-based processing when missing information is introduced should be relatively stronger in decision tasks involving more missing information than in tasks involving less missing information.

P3b: The shift from the use of attribute-based processing to alternative-based when missing information is introduced should be stronger in decision tasks involving time pressure versus tasks involving no time pressure.


The last factor to be considered for its effect on the effort involved in inferring missing values is the natural attribute redundancy within a product class. To the extent that a consumer perceives one attribute as an accurate predictor of another they may find inferences regarding those attributes to be more straightforward. When faced with missing information on an attribute from a product category with high redundancy, the consumer is more likely to recognize another given attribute as being an accurate predictor of the missing value. While this is related to the effects of experience as discussed earlier, redundancy or covariation within the product category itself may influence whether or not this learning takes place. And there is evidence that consumers may enter a category perceiving redundant information (Johnson and Katrichis, 1988).

This argument rests on the assumption that the level of attribute redundancy varies between product categories. This idea was supported in a study by Johnson and Katrichis (1988) which looked at attribute level redundancy across 65 product categories. They assessed brand by attribute matrices from Consumer Reports and found significant levels and ranges of attribute redundancy across product categories. Redundancy varied widely both across categories, and across attribute pairs within the categories.

This argument also assumes that consumers are able to perceive covariation. Many early studies in this area suggested that prior theories or expectations may be more important to the perceptions of covariation than observed data (Chapman and Chapman 1967; Jennings, Amabile, and Ross 1980; Nisbett and Ross 1980; and Olson 1974). However, recently these assertions have been challenged. Alloy and Tabachnik (1984) argue that people are reasonably accurate estimators of covariation in situations wherein they have no strong beliefs (see also Bettman, Roedder John, and Scott 1986). If beliefs are absent and the data clear and easily processed, covariation assessment will depend more on data. Johnson and Katrichis (1988) demonstrated that consumers appear to learn attribute relationships and do not perceive redundancy in the absence of actual redundancy, suggesting that expectations do not always dominate consumer perceptions.

In summary, there exists evidence that product categories do vary in their inherent attribute redundancy, and that consumers can perceive and may use these covariation assessments to aid in inferring missing attribute values. It follows that, in highly redundant categories, inferring missing attribute values may require less effort. Further, a consumer may then be less likely to shift to an alternative-based decision strategy to avoid inferences in highly redundant categories because the inferencing effort involved is low versus that in less redundant situations. Therefore, the following proposition is forwarded:

P4: Because inferences are easier in product categories with high attribute redundancy, the shift from attribute-based to alternative-based processing when missing information is introduced should be relatively stronger in low redundancy categories than in highly redundant categories .

Similarly, it could be argued that inherent redundancy across attribute pairs of the specific attribute with the missing value will also influence the effort a consumer is required to exert to infer that value and, therefore the likelihood of a strategy switch.


Another factor which may affect the balance of the cost/benefit relationship involved in the proposed inference induced strategy shift is the importance of the decision. While the previously discussed factors were argued to affect the effort involved in inferencing and therefore the likelihood of a strategy shift, decision importance may affect the motivation of the consumer to strive for accuracy. If an accurate decision is more important or salient to the consumer, it may take a greater increase in effort to induce a shift to a simplifying strategy.

Decision importance may be influenced by the importance of the decision consequences to the decision-maker (Chaiken 1980), or whether or not the decision-maker must justify the decision in some way (Hagafors and Brehmer 1983). Thus defined, it has been argued that the higher the importance, the more analytic, systematic, formal, careful, or complete the decision process (Billings and Scherer 1988; Chaiken 1980; Gabrenya and Arkin 1979; and Hagafors and Brehmer 1983).

As previously discussed, in Beach and Mitchell's (1978) cost/benefit contingency model, the selection of a decision strategy was contingent on the relationship between the value of a correct decision and the negative feeling associated with spending the time and effort required to apply the most accurate strategy. The cost/benefit relationship was argued to be influenced by significance and accountability. As these factors are increased, the use of more formal strategies were hypothesized to increase as well, even though the cost incurred (in terms of time and effort) were also positively related to the use of the formal strategies. This relationship has since been demonstrated in studies looking at the effects of importance on time taken to make a decision (Christenson-Szylanski 1978) and on the use of more analytical decision strategies (McAllister, Mitchell, and Beach 1979). This suggests that the more important the decision, the more likely a consumer will stick with a strategy he feels will produce a more accurate decision even in the face of increased effort. Thus, the following proposition is forwarded:

P5: The shift from the use of attribute-based to alternative-based processing when missing information is introduced should be relatively weaker in situations of high decision importance than in situations of low importance.

Similarly, it could be argued that the importance of the specific missing attribute value to the decision task will also influence the effort a decision-maker may be willing to exert to infer that value, and consequently the likelihood of an inference effort induced strategy shift.


Given that many consumer decisions are characterized by missing information, a better understanding of its influence on consumer decision processing may be important. This paper has proposed a cost/benefit based theory which explains the influence of missing attribute value information and particularly the role of inferences on consumer decision processing. This theory argues that when attribute values are missing, in order to use some preferred strategies (e.g. attribute-based comparisons) a decision-maker must infer the missing attribute values. If attribute value inferences require effort, a decision maker may attempt to avoid the increased effort by switching to (1) alternative-based processing which can be applied directly, or (2) hierarchical elimination strategies which use diagnostic, available information.

Further, consumer experience with the product category, task complexity in the forms of amount of missing information and time pressure, product category attribute redundancy, and decision importance were all hypothesized to influence this cost/benefit relationship, and consequently the magnitude of the proposed strategy shift.

Finally, while we have outlined a theoretical framework for the effects of missing information on decision strategies, future research will test the propositions suggested in this paper.


Alba, Joseph and J. Wesley Hutchinson(1987), "Dimensions of Consumer Expertise," Journal of Consumer Research, 14 (Mar) 411-454.

Alloy, Lauren B. and Naomi Tabachnik(1984), "Assessment of Covariation by Humans and Animals: The Joint Influence of Prior Expectations and Current Situational Information," Psychological Review, 91 (January) 112-149.

Beach, Lee Roy and Terence R. Mitchell (1978), "A Contingency Model for the Selection of Decision Strategies," Academy of Management Review, 3 (July), 439 449

Bettman, James R. (1979), An Information Processing Theory of Consumer Choice, Reading, MA: Addison-Wesley

Bettman, James R. (1980), "Implications of a Constructive View of Choice for Analysis of Protocol Data: A Coding Scheme for Elements of Choice Processes," Advances in Consumer Research, 1980.

Bettman, James R. and Jacob Jacoby (1976), "Patterns of Processing in Consumer Information Acquisition," Advances in Consumer Research, 3, 233 -240.

Bettman, James R. and Pradeep Kakkar (1977), "Effects of Information Presentation Format of Consumer Information Acquisition Strategies," Journal of Consumer Research, 3 (Mar) 233-240.

Bettman, James R. and C. Whan Park (1980), "Effects of Prior Knowledge and Experience and Phase of the Choice Process on Consumer Decision Processes: A Protocol Analysis," Journal of Consumer Research, 7,(Dec) 234-248.

Bettman, James R., Deborah Roedder John, and Carol A. Scott(1986), "Covariation Assessment by Consumers," Journal of Consumer Research, 13 (Dec) 316-326.

Biehal, Gabriel and Dipankar Chakravarti (1982), "Information Presentation Format and Learning Goals as Determinants of Consumers' Memory Retrieval and Choice Processes," Journal of Consumer Research, 8 (Mar) 431-441.

Biehal, Gabriel and Dipankar Chakravarti (1986), "Consumers' Use of Memory and External Information in Choice: Macro and Micro Processing Perspectives," Journal of Consumer Researcher, 12 (Mar) 382405.

Billings, Robert S. and Stephen A. Marcus(1983), "Measures of Compensatory and Noncompensatory Models of Decision Behavior: Process Training Versus Policy Capturing," Organizational Behavior and Human Performance, 31 (3) 331-352.

Capon, Noel and Marian Burke (1980), "Individual, Product Class, and Task-Related Factors in Consumer Information Processing," Journal of Consumer Research, 7 (Dec) 314-326.

Chapman, Loren J. and Jean P. Chapman (1967), "Genesis of Popular but Erroneous Psychodiagnostic Observations," Journal of Abnormal Psychology, 72 (June) 193-204.

Chaiken, Shelly(1980), "Heuristic Versus Systematic Information Processing and the Use of Source Versus Message Cues in Persuasion," Journal of Personality and Social Psychology, 39 (5) 752-766.

Christensen-Szalanski, Jay J. (1978), "Problem-Solving Strategies: A Selection Mechanism, Some Implications and Some Data," Organizational Behavior and Human Performance, 22 (2) 307-323.

Cohen, Joel B., Paul W. Miniard, and Peter R. Dickson (1979), "Information Integration: An Information Processing Perspective," Advances in Consumer Research, 6.

Cohen, Joel B. and Kunal Basu (1987), "Alternative Models of Categorization: Toward a Contingent Processing Framework," Journal of Consumer Research, 13 (Mar) 455-472.

Ford, Gary T., and Ruth Ann Smith(1987), "Inferential Beliefs in Consumer Evaluations: An Assessment of Alternative Processing Strategies," Journal of Consumer Research, 14 (Dec) 363-371.

Gabrenya, W.K. and R. M. Arkin (1979), 'The Effect of Commitment on Expectancy Value and Expectancy Weight in Social Decision Making," Personality and Social Psychology Bulletin (5) 86-90.

Gerstner,Eitan (1985), "Do Higher Prices Signal Higher Quality?" Journal of Marketing Research, 22 (May) 209-215.

Hagafors, Roger and Berndt Brehmer (1983), "Does Having to Justify One's Judgments Change the Nature of the Judgment Process?" Organizational Behavior and Human Performance, 31 223-232.

Hagerty Michael R. and David A. Aaker (1984), "A Normative Mode of Consumer Information Processing," Marketing Science, 3 227-246.

Jacoby, Jacob, Robert Chestnut, Karl Weigl, and William Fisher (1976), "Pre-purchase Information Acquisition: Description of a Process Methodology, Research Paradigm, and Pilot Investigation," Advances in Consumer Research, Vol. 3 ed. Beverlee B. Anderson, Ann Arbor, MI, Association for Consumer Research, 306-314.

Johnson, Eric and Robert Meyer (1984), "Compensatory Choice Models of Noncompensatory Choice Processes: The Effect of Varying Context," Journal of Consumer Research, 11 (June) 528-541.

Johnson, Michael D. (1984), "Consumer Choice Strategies for Comparing Noncomparable Alternatives," Journal of Consumer Research, 11 (Dec) 741-753.

Johnson, Michael D. (1988), "Comparability and Hierarchical Processing in Multi-alternative Choice," Journal of Consumer Research, 15 (Dec) 303-314.

Johnson, Michael D. and Jerome Katrichis (1988), "Attribute Redundancy in Consumer Environments," Journal of Consumer Policy, 11 131-157.

Johnson, Richard and Irwin Levin (1985), "More Than Meets the Eye: The Effect of Missing Information on Purchase Evaluations," Journal of Consumer Research, 12 (Sept) 169-177.

Kardes, Frank (1986), "Effects of Initial Product Judgments on Subsequent Memory-Based Judgments," Journal of Consumer Research, 13 (June) 1-11.

Klein, Noreen (1983), "Utility and Decision Strategies: A Second Look at the Rational Decision Maker," Organizational Behavior and Human Performance, 31 (1) 1-25.

Lussier, Denis A. and Richard W. Olshavsky (1979), "Task Complexity and Contingent Processing in Brand Choice," Journal of Consumer Research, 6 (Sept) 154-165 .

McAllister, Daniel W., Terence R. Mitchell, and Lee Roy Beach (1979), The Contingency Model for the Selection of Decision Strategies: An Empirical Test of the Effects of Significance, Accountability, and Reversibility," Organizational Behavior and Human Performance, 24 (2) 228-244.

Meyer, Robert J. (1981), "A Model of Multiattribute Judgments Under Attribute Uncertainty and Informational Constraint," Journal of Marketing Research, 18 (Nov) 428-441.

Newell, Alan and Herbert Simon (1972), Human Problem Solving, Englewood Cliffs, NJ: Prentice Hall.

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

Olson, Jerry C.and Jacob Jacoby (1972), "Cue Utilization in the Quality Perception Process," Advances in Consumer Research M. Venkatesan, ed. Iowa City: Association for Consumer Research 167 - 179.

Park, C. Whan and V. Parker Lessig (1981), "Familiarity and its Impact on Consumer Decision Biases and Heuristics," Journal of Consumer Research, 8 (Sept) 223-230

Payne, John (1976), "Task Complexity and Contingent Processing in Decision Making: An Informational Search and Protocol Analysis," Organizational Behavior and Human Performance, 16 366-387.

Payne, John (1982), "Contingent Decision Behavior, " Psychological Bulletin, 92 (2) 382-402.

Russo, J. Edward and Larry Rosen (1975), "An Eye Fixation Analysis of Multi Alternative Choice," Memory and Cognition, 3 267-276.

Russo, J. Edward and Barbara Dosher (1983), "Strategies for Multiattribute Binary Choice," Journal of Experimental Psychology: Learning, Memory, and Cognition, 9 (4) 676-696.

Sheluga, David A., James Jaccard, and Jacob Jacoby (1979), "Preference, Search, and Choice: An Integrative Approach," Journal of Consumer Research, 6 (Sept) 166-176.

Shugan, Steven (1980), "The Cost of Thinking," Journal of Consumer Research, 7 (Sept) 99-111.

Slovic, Paul (1975), "Choice Between Equally Valued Alternatives," Journal of Experimental Psychology: Human Perception and Performance, 1 (3) 280-287.

Slovic, Paul and Douglas McPhillamy (1974), "Dimensional Commensurability and Cue Utilization in Comparative Judgment," Organizational Behavior and Human Performance, 11 (April) 172-194.

Sujan, Mita (1985), "Consumer Knowledge: Effects on Evaluation Strategies Mediating Consumer Judgments," Journal of Consumer Research, 12 (June) 31-46.

Tversky, Amos (1969), "Intransitivity of Preferences," Psychological Review, 76 (1) 3148.

Wright, Peter (1975), "Consumer Choice Strategies: Simplifying vs. Optimizing," Journal of Marketing Research, 12 (Feb) 60-67.

Yates, J. Frank, Carolyn Jagacinski, and Mark D. Faber (1978), "Evaluation of Partially Described Multiattribute Options," Organization Behavior and Human Performance, 21 240-251.

Yamagishi, Toshio and Charles Hill (1981), "Adding Versus Averaging Models Revisited: A Test of Path-Analytic Integration Model" Journal of Personality and Social Psychology, 41 (1) 1325.