ease of retrieving information, then commitment should be high at the consideration price that was generated in response to the lower or higher offer price and low at the consideration price that was generated in response to the intermediate offer price. Consumers should only rely on the ease-of-retrieval heuristic when information retrieved in response to the product description was ambiguous (i.e., the ease of retrieving information was perceived as diagnostic). These relationships are summarized in Figure 1.

Experiment 1

Experiment 1 investigated the influence of offer prices on consumers consideration prices and consumers willingness to complete the transaction at their consideration prices. Participants were presented with a product and asked whether they were willing to purchase the product at one of three offer prices; a standard price, an intermediate price, or a premium price. Anticipating negative responses to this question (based on pretests), participants were next asked to indicate the amount at which they would be willing to consider the purchase of the product (i.e., their consideration price). Then participants were asked about their willingness to complete the transaction if the product were to be offered at their consideration price.

Method

Participants. There were 172 undergraduate students who participated in the main experiment. An additional 242 students participated in a pretest (n 20), a manipulation check study (n 134), and an ancillary study (n 88). All participants were drawn from a subject pool comprised of introductory marketing and

Research Strategy

We sought to demonstrate that a consideration price judgment depended on the content of retrieved information, whereas the willingness to complete the transaction at this price depended on the ease of retrieving information. To provide a convincing demonstration, the information that influenced the consideration price judgment had to be manipulated independently of the ease of retrieving this information. Our strategy depended on a simple observation. Consumers often organize product category information by price tiers and/or performance levels (Alba & Hutchinson, 1987; Sivakumar & Raj, 1997). Product differentiation and product category maturity often guide the number of tiers within any market (i.e., new markets have a single tier whereas mature markets have multiple tiers). Consumer experience with a product category allows the consumers to learn about performance tiers. Thus, to the extent an offer price corresponds to a performance tier, information about products at that tier should be accessible. To the extent an offer price does not correspond to a performance tier (i.e., the offer price suggests a product between performance tiers), information should be less accessible. Our research strategy was to take product categories with a minimum of two performance tiers and to provide offer price information that corresponded to two adjacent tiers or to somewhere in between. We anticipated that the consideration price judgment (base judgment) would increase as the offer price increased (Lichtenstein, Burton, & Karson, 1991; Nunes & Boatright, 2004), provided the product description was sufficiently ambiguous (i.e., information recruited in response to the offer price was perceived as diagnostic). We also anticipated that information would be more accessible when the offer price corresponded to the one of the tiers, as opposed to neither of the tiers. If this is so, and transaction commitment (contingent judgment) is sensitive to the

Experiment 2

In Experiment 1, it was argued that people were able to retrieve differential amounts of information in response to the offer price because the product descriptions were ambiguous. In Experiment 2, we varied the ambiguity of the product description using the bike replicate. In one condition, we used an abbreviated version of the product description used in Experiment 1. In a second condition, we removed the ambiguity from the abbreviated description so that participants understood the product was unambiguously from a standard performance tier. We anticipated that the participants willingness to complete the transaction would depend on the ease of information retrieval when the product description was

Results

One hundred forty-four participants who owned a bike or were interested in purchasing a bike were retained in the analysis. The raw means for the consideration price and the willingness to complete the transaction at the consideration price are shown in Figure 3.

$1,750

Standard
$1,500 Consideration Price $1,250 $1,000 $750 $500 $250 $0 Ambiguous

Intermediate Premium

Unambiguous

Ambiguous

Figure 3. The influence of a standard, intermediate, or premium initial offer price on a participants stated consideration price (price at which the transaction would be considered) and the participants willingness to commit to the transaction should the retailer lower the price to this level (Experiment 2). Ambiguous and unambiguous refer the description of the product. Error bars represent standard errors.
The two dependent measures were transformed into z-scores. The three-way interaction between offer price, the ambiguity of the product description, and the type of dependent measure was significant, F(2, 138) 4.03, MSE .74, p .02, 2p .06. The two-way interaction between offer price and the type of dependent measure was significant in the ambiguous product description condition, F(2, 138) 12.84, MSE .74, p .01, 2p .16, but not in the unambiguous product description condition, F(2, 138) 1.10, MSE .74, p .34. Consideration price. There was an interaction between the offer price and the product ambiguity manipulations, F(2, 138) 5.99, MSE 314,153, p .01, 2p .08. The offer price factor exhibited a significant main effect in the ambiguous condition, F(2, 138) 15.27, MSE 314,153, p .01, 2p .18, but not in the unambiguous condition, F(2, 138) 1.80, MSE 314,153, p .17. Simple main effect tests in the ambiguous condition showed the participants consideration price was higher in the premium (M $1,234, SD $1,230) than in the intermediate (M $536, SD $242, F(1, 141) 15.90, MSE 330,416, p .01, 2p .10, offer price condition, but did not differ between the intermediate and the standard (M $356, SD $190), F(1, 141) .62, MSE 340,416, p .43, offer price conditions. Transaction commitment. There was an offer price by-product ambiguity interaction on the transaction commitment judgments, F(2, 138) 2.80, MSE 373, p .06, 2p .04. The price factor exhibited a significant main effect in the ambiguous condition, F(2, 138) 4.97, MSE 373, p .01, 2p .07, but not in the

unambiguous condition, F(2, 138) .03, MSE 373, p .97. Similar to Experiment 1, the price manipulation created a quadratic influence (i.e., U-shaped curve) on the willingness to complete the transaction in the ambiguous product description condition, F(1, 141) 9.02, MSE 375, p .01, 2p .07. Simple main effect tests showed that participants in the ambiguous condition were less willing to complete the transaction at their reported consideration price in the intermediate price condition (M 63.1, SD 20.9) than in the standard price (M 82.1, SD 14.3), F(1, 141) 11.03, MSE 375, p .01, 2p .07, or the premium price (M 74.6, SD 14.2), F(1, 141) 3.98, MSE 375, p .05, 2p .03, conditions. Product-related thoughts (reliability .87). We anticipated that the quantity of product related thoughts (e.g., thoughts mentioning the product, usage situations, usage partners) could be used as an indicator of the ease of retrieving information and that this ease of retrieval would mediate the participants transaction commitment in the ambiguous description conditions. Unfortunately, the average number of thoughts did not vary by condition, F(2, 138) .27, MSE 2.20, p .77, a result that could be attributed to the low number of product-related thoughts (M 1.31, SD 1.51) relative to the total number of thoughts (M 4.77, SD 1.93). Alternatively, the correlation between the number of productrelated thoughts and the consumers willingness to complete the transaction was computed in each of the six conditions (see row one of Table 1). In the ambiguous condition, the correlation was significant in the intermediate offer price condition, r .59, p
Table 1 Correlations between Transaction Commitment and Potential Explanatory Variables in Experiment 2
Ambiguous product description Explanatory variables Product relevant thoughts ARP believability Vendor counteroffer believability Plausible -0.03 0.06 -0.30 High 0.59 0.09 0.25
Unambiguous product description Plausible -0.01 -0.07 0.23 High -0.17 0.07 0.14 Extreme 0.24 0.01 0.08

Extreme 0.45 0.25 -0.19

Note. Correlation of purchase likelihood with listed variable. a p .05.
.01, and the premium offer price condition, r .45, p .03, but not in the standard offer price condition, r .03, p .88. The correlation was not significant in the unambiguous conditions (rstandard .01, p .96; rintermediate .17, p .47; rpremium .24, p .22). Alternative explanations. Offer price believability and vendor counteroffer believability did not exhibit the same pattern of results as transaction commitment. There was no offer price by ambiguity interaction on the believability of the offer price, F(2, 138) .24, MSE 915, p .79, or the believability of the vendors counteroffer, F(2, 138) .27, MSE 813, p .77. In addition, these measures did not correlate with transaction commitment in any of the six experimental conditions (all p .10, see Table 1). Aggregating selected experimental conditions (i.e., intermediate and premium conditions in the ambiguous condition) or all experimental conditions to increase the power of the test also resulted in no significant correlations (all p .10).

hood of an affirming judgment. Applying this logic to our context, retrieving a single piece of information should not be harder or easier relative to expectations, hence retrieving one piece of information should not impact the consumers transaction commitment. In contrast, retrieving three pieces of information does have the potential to vary in difficulty across offer price conditions. It should be harder to retrieve information in the intermediate offer price condition and easier to retrieve information in the standard and premium offer price conditions. The difficult information retrieval in the intermediate price condition should lead to a lower willingness to complete the transaction at a consideration price generated in response to an offer price.
Participants. There were 249 undergraduate students who participated in the main experiment. All participants were drawn from the same subject pool as prior experiments, but did not participate in prior experiments. Students received extra credit for their participation. Procedure. The experiment used a two-factor, betweensubjects design with three levels of offer price (standard, intermediate, and premium) and two levels of generated thoughts (one and three). The procedure was similar to the procedure used in Experiment 2 with the following modifications. First, the ambiguous product description from Experiment 1 was used. Second, after the participant provided a consideration price, the participant was told
We are interested in the thoughts you are having about this bike. Tell us about the features of the bike or places a person could use the bike. Please list one (three) bike feature(s) or place(s) to use the bike.
The results of Experiment 2 provide further evidence that consideration price judgments depend on the content of retrieved information, whereas the willingness to complete a transaction at a consideration price depends on the ease of retrieving the information, provided information content is not diagnostic. Participants in the ambiguous product description condition generated monotonically increasing consideration prices in response to the increasing offer prices, but exhibited a U-shaped pattern of transaction commitment at the three consideration prices. We contend that the ambiguous product description allowed participants to retrieve different types of information at different offer prices and that this information influenced consideration prices. Transaction commitment to these consideration prices was influenced by the ease of retrieving the information. We also argue that the unambiguous product description encouraged participants to retrieve information with more confidence, so their willingness to complete the transaction at the consideration price was not sensitive to the ease of retrieving information. When the product description was unambiguous, the consideration price judgments and willingness to complete the transaction at these consideration prices depended on the content of the retrieved information. Although the data from Experiment 2 are informative, one remaining issue merits further attention. The relationship between the transaction commitment judgment and the ease of retrieving of information has been documented using correlational data. It would be better to manipulate the ease of retrieving information to influence the consumers transaction commitment. Experiment 3 provided this test.

Transaction Commitment 100

One Thought

Three Thoughts
Figure 4. The influence of a standard, intermediate, or premium initial offer price on a participants stated consideration price (price at which the transaction would be considered) and the participants willingness to commit to the transaction should the retailer lower the price to this level (Experiment 3). One thought and three thoughts refer to the number of product thoughts generated after the consideration price measure but before the transaction commitment measure. Error bars represent standard errors.
interaction of the factors, F(2, 168) .36, MSE 243,446, p .70, had no influence on the consideration price. Simple main effect tests showed the consumers consideration price was higher in the intermediate price condition (M $523, SD $244) than in the standard price condition (M $313, SD $105), F(1, 170) 5.17, MSE 243,446, p .02, 2p .03 and was higher in the premium price condition (M $935, SD $777) than in the intermediate price condition (M $523, SD $244), F(1, 170) 19.79, MSE 243,446, p .01, 2p .10. Transaction commitment. There was an offer price by number of thoughts generated interaction on the transaction commitment judgments, F(2, 168) 2.76, MSE 619, p .07, 2p .03. There was a significant quadratic effect of the offer price manipulation in the three thought condition, F(2, 168) 3.44, MSE 619, p .03, 2p .03. Simple main effect tests showed that consumers were less willing to complete the transaction at their reported consideration price after considering the intermediate price (M 55.6, SD 27.2) than after considering the standard price (M 68.6, SD 22.7), F(1, 170) 3.06, MSE 619, p .08, 2p .02, or the premium price (M 73.5, SD 14.5), F(1, 170) 6.74, MSE 619, p .01, 2p .04. There was no effect of the offer price manipulation on the willingness to complete the transaction in the one thought condition (Mstandard 70.0, Mintermediate 64.6, Mpremium 61.2; F(2, 168) .99, MSE 619, p .37. The analyses of the believability measures replicated the null results of Experiment 2. Difficulty of generating product-related thoughts. The procedure included a measure to assess the difficulty of generating product-related thoughts. It was expected that participants would perceive it as relatively difficult to generate three thoughts in the intermediate price condition, as compared to the standard and premium price conditions. The offer price influenced the perceived difficulty of generating thoughts in the three thought condition, F(2, 168) 4.59, MSE 3.52, p .01, 2p .06. People perceived it as more difficult to generate the three thoughts in the intermediate price condition (M 4.50, SD 1.37) relative to the standard price condition (M 3.39, SD 1.92), F(1, 170) 3.92, MSE 3.52, p .05, 2p .02, or the premium price condition (M 2.94, SD 1.85), F(1, 168) 9.03, MSE 3.52, p .01,
2p .05. There was no difference in the perceived difficulty of generating thoughts in the one thought condition (Mstandard 2.79, Mintermediate 2.68, Mpremium 3.29), F(2, 168) .97, MSE 3.52, p .38. Mediation analysis. The Baron and Kenny (1986) three-part procedure was used to assess whether the difficulty of generating product-related thoughts mediated the participants transaction commitment in the three thoughts condition. First, the offer price was a significant predictor of the willingness to complete the transaction ( 7.51, SE 3.58), t(76) - 2.10, p .04. Second, the offer price was a significant predictor of the difficulty of generating thoughts ( .66, SE .28), t(76) 2.40, p .02. When transaction commitment was regressed on the offer price and the difficultly of generating product-related thoughts, the offer price coefficient became nonsignificant, ( 4.71, SE 3.54), t(76) 1.33, p .19, and the difficulty of generating thought coefficient remained significant ( 4.22, SE 1.43), t(76) 2.96, p .01. A Sobel test for the mediating influence of the difficulty of generating product-related thoughts was marginally significant, z 1.88, p .06.

The results of Experiment 3 showed that the ease of retrieving product-related information influenced the consumers willingness to complete the transaction at the consideration price, provided the ease of retrieval is diagnostic. The perceived ease of retrieval varied in the three thought conditions, as did the participants transaction commitment. Participants found it difficult to retrieve three product-related thoughts in the intermediate offer price condition, but found it easy to retrieve three product-related thoughts in the standard and premium offer price conditions. This ease (difficulty) of retrieval was accompanied by an increased (decreased) transaction commitment. The perceived ease of retrieval did not vary in the one thought conditions, nor did the participants transaction commitment. The results are consistent with the claim that consumers can use the ease of retrieving information to inform a subsequent judgment about the willingness to complete the
transaction at that consideration price, provided that more diagnostic information is not available.

Experiment 4

Experiments 1 through 3 provided evidence that base judgments, and subsequent contingent judgments, depend on different types of information. Experiment 4 replicated Experiment 1 in a shopping context. The experimental context was an Internet shopping trip. Participants were asked to shop for five products using the NexTag shopping bot. During the course of this shopping experience, participants had the opportunity to set a consideration price, called a price alert, for any of the five products they shopped. Subsequently, participants were asked if they had set a price alert for any of the products and the likelihood they would buy the product if the price was met (i.e., transaction commitment).
Participants. There were 404 undergraduate students who participated in the main experiment. An additional 76 students participated in pretests. Pretest participants did not participate in the main experiment. All participants were drawn from the same subject pool as prior experiments, but did not participate in prior experiments. Students received extra credit for their participation. Design and stimuli. The experiment used a one-factor, between-subjects design with three levels of an offer price (standard, intermediate, and premium) and five product category rep-

Note. Pretest participants viewed products and provided expected prices in the absense of a list price. Experimental participants saw the product with actual list prices. Three products were discontinued while our study was in progress. Replacement products are listed in notes 2 through 4. One person set a price alert on the replacement toaster oven. Nine people set a price alert on the replacement MP3 player. Their mean likelihood of purchasing (M 7.77) did not differ from the mean of the 15 people that set a price alert on the original MP3 player (M 7.80; F 1.0). Two people set a price alert on the replacement microwave oven. 2 Delonghi 17.Airstream Convection Digital Toaster Oven for n 74 (Price $179.99, Pretest M $76.71). 3 iRiver IFp MB MP3 Player for n 94 (Price $169.99, Pretest M $121.89). 4 GE JX81B Microwave Oven for n 56 (Price $29.00, Pretest M $69.84).
Procedure. After being seated at the computer, participants were told they would be learning about shopping bots. First, they read a four page shopping bot usage manual. The manual discussed the NexTag Web site, the types of products that were sold there, and the advantages of shopping with a shopping bot. Next, a search was illustrated. The format of the search results, along with the products price history, a list of available vendors, and the purchase procedure were discussed. Then the price alert feature was discussed. Participants were told that if the offer price was too expensive, they could use the price alert feature to set a price at which they would consider the purchase. If any of the merchants subsequently offered the item at the named price or lower, a Price Alert email would be sent to the e-mail address listed when entering the price alert. If a consumer received a price alert e-mail, the consumer could revisit the NexTag Web site and decide whether or not to purchase the product. Next, participants were asked to search for five products. Participants were given a set of five products that had standard, intermediate, or premium prices (between-subjects). Participants were encouraged to set a price alert if they would be willing to consider the purchase of the product at a lower price, but were also told that they should not set a price alert if they were not interested in the product. Participants then went to the NexTag Web site and did a search for each product. After each search, participants were asked if they would consider buying the product (yes/no), to list the current lowest price for the product, to indicate whether they set a price alert (yes/no), and to list the price alert price if they did use the option. After searching for all five products, participants considered each item a second time. If the participant had set a price alert, the participant indicated the likelihood of purchasing at the consideration price using a 10-point scale labeled not at all likely and very likely. Then general questions about Internet shopping frequency, past shopping bot usage, and intended shopping bot usage were asked. The procedure also included an e-mail solicitation that was sent four days after the experimental procedure. The e-mail alerted the participant to a price drop and contained a link that ostensibly allowed participants to visit the NexTag Web site and make the purchase. Response rates were too low (n 29) for this dependent

variable to be meaningful (i.e., a floor effect). It will not be discussed further.
Participants could request price alerts on as many products as they wished. Two hundred sixteen participants (53%) requested a price alert on at least one product and were included in the analysis. When the participant requested multiple price alerts, the analysis was limited to the responses for the most appealing product. This approach was consistent with the assumption that an individuals responses across multiple products are not independent. The two dependent measures were transformed into z-scores. The interaction between the offer price and the type of dependent measure was significant, F(2, 213) 68.87, MSE .58, p .01, 2p .39, suggesting that the offer price manipulation exerted a different pattern of influence on the consideration price judgments and the purchase likelihood judgments. Consideration price. The data were transformed into z-scores (see Figure 5) owing to the unequal variances across the five replicates (see Table 3 for raw means). There was a significant influence of the offer price on the consideration price, F(2, 213) 390.91, MSE .21, p .01, 2p .79. Simple main effect tests showed the participants consideration price was higher in the intermediate price condition (M .31, SD .31) than in the standard price condition (M .94, SD .18), F(1, 213) 66.57, MSE .21, p .05, 2p .24. The participants consideration price was also higher in the premium price condition (M 1.17, SD .70) than in the intermediate price condition (M .31, SD .31), F(1, 213) 392.07, MSE .21, p .01, 2p .65. Transaction commitment. There was a significant influence of the offer price on a participants purchase likelihood, F(2, 213) 2.99, MSE 3.89, p .05, 2p .03 (see Figure 5). As predicted, the offer price manipulation created a quadratic influence on the participants purchase likelihood, F(1, 213) 5.96, MSE 3.89, p .02, 2p .03. Simple main effect tests showed that participants were less likely to purchase at their consideration price in the intermediate price condition (M 6.77, SD 2.11) than in the standard price (M 7.49, SD 2.15), F(1, 213) 4.77, MSE

1.50 Standardized Consideration Price
1.00.50.00 -.50 -1.00 -1.50 Standard Intermediate Premium
6.0 Standard Intermediate Premium
Figure 5. The influence of a standard, intermediate, or premium initial offer price on a participants stated consideration price (price at which the transaction would be considered) and the participants willingness to commit to the transaction should the retailer lower the price to this level (Experiment 4). Means are standardized. Error bars represent standard errors.
Table 3 Experiment 4: Mean Consideration Prices
Product list price Product Toaster oven MP3 player Rice cooker Coffee maker Microwave oven Aggregate (z-score) n Plausible $19.18 $24.21 $12.00 $11.00 $30.80 -0.94 High $52.20 $51.87 $36.67 $21.71 $44.28 -0.31 Extreme $ 65.00 $110.14 $120.71 $ 47.70 $125.50 1.17
Note. Consumers used the NexTag shopping agent to shop for one item in each product category (see Table 2). If the consumer was interested in purchasing the item at a lower price, the consumer could set a price alert (a consideration price). If the price dropped to the alert level at some time in the future, the shopping agent sent the consumer an e-mail about the price drop.
3.89, p .03, 2p .02, or premium price (M 7.41, SD 1.61), F(1, 213) 4.00, MSE 3.89, p .05, 2p .02, conditions.
The results of Experiment 4 showed that standard, intermediate, and premium offer prices evoked consideration prices that were monotonically increasing, whereas also leading to a U-shaped pattern of willingness to complete the transaction at the respective consideration prices. Similar to the prior studies, we claim that the consideration price judgments depended on the content of retrieved information, but that the willingness to complete the transaction at this consideration price depended on the ease of retrieving the information used to generate the consideration price. More specifically, the low transaction commitment in the intermediate price condition was associated with difficult information retrieval, whereas the higher transaction commitment in the standard and premium price conditions was associated with easy information retrieval.

General Discussion

Four experiments investigated judgments about judgments. The results suggest that the generation of a consideration price, and the willingness to complete the transaction at that consideration price, can depend on two different processes. The generation of a consideration price depends on the information content primed by an offer price. The consumers willingness to complete the transaction at the consideration price depends on the content of the information retrieved to generate the consideration price (unambiguous context) or on the ease of retrieving the information used to generate the consideration price (ambiguous context). People are most likely to rely on the ease-ofretrieval heuristic when the content of the retrieved information is perceived to be suspect. The content of retrieved information will be perceived as suspect when (a) the product description is ambiguous and (b) the offer price is higher than expected.
Perhaps the most interesting aspect of our investigation is the use of the ease-of-retrieval heuristic for the contingent judgments. Past demonstrations of the use of an ease-of-retrieval heuristic
have relied on a procedure that makes the ease of information retrieval quite salient. In addition, a persons nave theories about the meaning of the ease of information retrieval can encourage or discourage the use of this type of information (Schwarz, 2004). Our procedures differ from these past demonstrations in that the ease of retrieving information is not made salient (i.e., people are not listing reasons to purchase a product). As such, there should be no nave theory that says the ease of retrieving information (as opposed to reasons) should increase commitment to a purchase. It appears that the ease of retrieving information is exerting an influence in a much more subtle way, similar to the fleeting subjective experiences that influence judgments of recognition (e.g., Whittlesea & Leboe, 2000). A second issue concerns the use of the ease-of-retrieval heuristic in Experiment 3. Schwarz et al. (1991) argue that the ease-ofretrieval heuristic will be used when information content is nondiagnostic, when a person lacks the resources to assess the diagnosticity of the content, or when there is little incentive to be accurate. Our experiments provide support for the first boundary condition. Regarding the third boundary condition, our participants had no extrinsic incentive to make a good purchase decision. However, we anticipate extrinsic incentives should encourage people to consider information content, thereby moderating differences in transaction commitment in the intermediate and premium price conditions. The incentive to make a good purchase decision could be initiated by a scarcity of monetary resources, suggesting that wealthy individuals may be most prone to show the differential willingness to complete the transaction at consideration prices generated in response to differing offer prices. Third, the research addresses the important issue of how people decide whether or not to execute a purchase. Purchase deferral has been attributed to trade-off difficulty (e.g., Tversky & Shafir, 1992), decision conflict created by similarity in the attractiveness of the alternatives (e.g., Dhar, 1997), ill-defined preferences (Anderson, 2003), and anticipated regret (e.g., Tykocinski, Pittman, & Tuttle, 1995). We show that transaction commitment depends on the ease-of-retrieving information. The implication is that decision environments can be structured to encourage information retrieval, enhance transaction commitment, and increase the likelihood of completing the transaction. The inclusion of retrieval cues in a decision environment should enhance transaction commitment, especially when there is ambiguity about the value of the good. Finally, it is useful to consider consumption situations in which contingent judgments are likely to occur. Clearly, shopping agents provide an opportunity for a person to engage in a contingent judgment. We expect that barter environments also provide an opportunity for contingent judgments. When a consumer enters an art gallery, a furniture store, or a high end clothing store, the consumer understands that the price should be negotiated. The consumer also realizes that the retailer will not accept any price (i.e., the consumer should not bid 10% of the list price). Thus, to the extent the consumer encounters more than one acceptable product, the consumer must generate a consideration price, anticipate a settlement price, and then assess whether he or she would be willing to execute the purchase at this price. We contend that consumers do engage in these mental simulations and that the contingent judgments that comprise these simulations can encourage or discourage the initiation and/or completion of a negotiation.

If this is so, our results suggest that consumers may be more likely to initiate a price negotiation on premium-priced, as opposed to intermediate-priced, merchandise.

Future Research

One obvious question raised by our research relates to the contextual moderators of our findings. Specifically, given an ambiguous purchase context, it is provocative that a retailer may be able to use a premium price to positively affect consumer consideration prices and the willingness to complete the transaction at those consideration prices. Clearly, we can hypothesize about contexts where the effect would be more or less pronounced. First, we would expect the effect to be more pronounced when nave consumers engage in decision making that is not supported by external search. There is much evidence that, despite the reported importance of price in purchase decisions, consumers engage in surprisingly low levels of prepurchase search, even for expensive durable goods (Grewal & Marmorstein, 1994; Moorthy, Ratchford, & Talukdar, 1997). Second, even if consumers engage in search, it is possible that they will increase their confidence knowledge, (i.e., learning product attributes), but not their predictive knowledge, (i.e., their ability to use their confidence knowledge to predict product quality) (Cox, 1962). In this case, confidence knowledge will increase transaction commitment, with no increase in the resistance to the premium price.

References

Abelson, R. P. (1988). Conviction. American Psychologist, 43, 267275. Alba, J. W., & Hutchinson, W. J. (1987). Dimensions of consumer expertise. Journal of Consumer Research, 13, 411 454. Alba, J. W., & Hutchinson, W. J. (2000). Knowledge calibration: What consumers know and what they think they know. Journal of Consumer Research, 27, 123156. Anderson, C. J. (2003). The psychology of doing nothing: Forms of decision avoidance result from reason and emotion. Psychological Bulletin, 129, 139 166. Authorware 7.02 [Computer software]. (2003). San Jose, CA: Macromedia Inc. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 11731182. Burson, K. A., Larrick, R. P., & Klayman, J. (2006). Skilled or unskilled, but still unaware of it: How perceptions of difficulty drive miscalibration in relative comparisons. Journal of Personality and Social Psychology, 90, 60 77. Cox, D. F. (1962). The measurement of information value: A study in consumer decision-making. In W. S. Decker (Ed.), Emerging concepts in marketing (pp. 413 421), Chicago: American Marketing Association. Dhar, R. (1997). Consumer preference for a no-choice option. Journal of Consumer Research, 24, 215231. Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95, 542575. Dunlosky, J., & Nelson, T. O. (1994). Does the sensitivity of judgments of learning (JOLs) to the effects of various study activities depend on when the JOLs occur? Journal of Memory and Language, 33, 545565. Einhorn, H. J., & Hogarth, R. M. (1985). Ambiguity and uncertainty in probabilistic inference. Psychological Review, 92, 433 461. Epstein, S., & Pacini, R. (1999). Some basic issues regarding dual-process theories from the perspective of cognitive-experiential self-theory. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 462 482), New York: Guilford Press.

Gilovich, T., & Medvec, V. H. (1995). Some counterfactual determinants of satisfaction and regret. In N. J. Roese & J. M. Olson (Eds.), What might have been: The social psychology of counterfactual thinking (pp. 259 282), Hillsdale, NJ: Erlbaum. Grewal, D., & Marmorstein, H. (1994). Market price variation, perceived price variation, and consumers price search decisions for durable goods. Journal of Consumer Research, 21, 453 460. Haddock, G., Rothman, A. J., & Reber, R. (1999). Forming judgments of attitude certainty, intensity, and importance: The role of subjective experiences. Personality and Social Psychology Bulletin, 25, 771782. Holland, R. W., Verplanken, B., & van Knippenberg, A. (2003). From repetition to conviction: Attitude accessibility as a determinant of attitude certainty. Journal of Experimental Social Psychology, 39, 594 601. Kelley, C. M., & Lindsay, D. S. (1993). Remembering mistaken for knowing: Ease of retrieval as a basis for confidence in answers to general knowledge questions. Journal of Memory and Language, 32, 124. Koehler, D. J. (1991). Explanation, imagination, and confidence in judgment. Psychological Bulletin, 110, 499 519. Koriat, A. (1993). How do we know that we know? The accessibility model of the feeling of knowing. Psychological Review, 100, 609 639. Koriat, A. (1995). Dissociating knowing and the feeling of knowing: Further evidence for the accessibility model. Journal of Experimental Psychology: General, 124, 311333. Koriat, A., & Levy-Sadot, R. (1999). Processes underlying metacognitive judgments: Information-based and experience-based monitoring of ones own knowledge. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 483502), New York: Guilford Press. Koriat, A., Maayan, H., & Nussinson, R. (2006). The intricate relationships between monitoring and control in metacognition: Lessons for the cause-and-effect relation between subjective experience and behavior. Journal of Experimental Psychology: General, 135, 36 69. Kruger, J., & Dunning, D. (1999). Unskilled and unaware of it: How difficulties in recognizing ones own incompetence lead to inflated selfassessments. Journal of Personality and Social Psychology, 77, 11211134. Lichtenstein, D. R., Burton, S., & Karson, E. J. (1991). The effect of semantic cues on consumer perceptions of reference price ads. Journal of Consumer Research, 18, 380 391. Lichtenstein, S., Fischhoff, B., & Phillips, L. (1982). Calibration of probabilities: The state of the art to 1980. In D. Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment under uncertainty: Heuristics and biases (pp. 306 334). New York: Cambridge University Press. Mehle, T. (1981). The availability explanation of excessive plausibility assessments. Acta Psychologica, 49, 127140. Monin, B., & Miller, D. T. (2001). Moral credentials and the expression of prejudice. Journal of Personality and Social Psychology, 81, 33 43. Moorthy, S., Ratchford, B. T., & Talukdar, D. (1997). Consumer information search revisited: Theory and empirical analysis. Journal of Consumer Research, 23, 263277. Moritz, S., & Woodward, T. S. (2006). The contribution of metamemory deficits to schizophrenia. Journal of Abnormal Psychology, 115, 1525. Morris, C. C. (1990). Retrieval processes underlying confidence in comprehension judgments. Journal of Experimental Psychology: Learning, Memory, and Cognition, 16, 223232. Mussweiler, T., & Strack, F. (2000). The use of category and exemplar knowledge in the solution of anchoring tasks. Journal of Personality and Social Psychology, 78, 1038 1052. Mussweiler, T., & Strack, F. (2001). The semantics of anchoring. Organizational Behavior and Human Decision Processes, 86, 234 255. Neslin, S. A. (1981). Linking product features to perceptions: Self-stated versus statistically revealed importance weights. Journal of Marketing Research, 18, 80 86. Nunes, J. C., & P. Boatwright. (2004). Incidental prices and their effect on willingness to pay. Journal of Marketing Research, 31, 457 466.