Sage Journals: Discover world-class research

Abstract

Processing fluency of stimuli has been shown to impact consumers’ decision-making. We investigate whether inhibiting the processing fluency of an anchor results in a more pronounced anchoring effect, as is proposed in the existing literature. We use a point-of-purchase field experiment to test the hypothesis that a disfluent anchor in a product name influences consumers’ willingness to pay for this product more than a fluent anchor. The results provide strong support against the fluency—willingness to pay relationship. Contrary to theoretical predictions, our study cautions marketing practitioners against the use of low-fluency anchors in product names.

Plain language summary

This study focuses on whether a numerical anchor presented in words produces a larger anchoring effect (i.e., overly relying on initial numeric information when making a decision) than a numerical anchor presented in digits. We conducted a field experiment that measured how much money are consumers willing to pay for a product when various names of the product were presented. Different customers bought craft beer that either had a number in the name written in words (99 golden piglets), numbers (99 golden piglets), or did not contain a number in its name (Golden piglet). Contrary to the previous research, we found that the number in the product name written in words does not affect consumers’ willingness to pay more than the one written in digits. Our study cautions marketing practitioners against the use of low-fluency anchors in product names.

Keywords

anchoring effect willingness to pay processing fluency dual process theory field experiment

Introduction

The numerical anchoring is a cognitive bias where individuals overly rely on initial information presented to them, known as an “anchor,” when making decisions (Tversky & Kahneman, 1974). For instance, if a person is first shown a high number and then asked to estimate the price of an unrelated item, their estimate is likely to be higher than if they were first shown a low number (Jacowitz & Kahneman, 1995). Numerical anchoring is a well-established and robust phenomenon that influences people’s numerical decisions, for example, the price they are willing to pay in various situations (e.g., Critcher & Gilovich, 2008; Dogerlioglu-Demir & Koçaş, 2015; Nunes & Boatwright, 2004). However, less is known about the conditions that strengthen or weaken the effect of an anchor on one’s decision. This study uses the design of a field experiment to compare the effect of two types of the same anchor on customers’ willingness to pay when either digits or words are used to write the anchor.

Anchoring effect is a term describing the assimilation of a numeric judgment toward a previously encountered standard (anchor; Mochon & Frederick, 2013). Anchoring is pervasive across domains and remarkably robust (Bahník et al., 2021). The effect has been demonstrated in probability estimates (Plous, 1989; Tversky & Kahneman, 1974), purchase quantity decisions (Wansink et al., 1998), negotiation outcomes (Galinsky & Mussweiler, 2001; Ritov, 1996), or buying and selling prices (Green et al., 1998; Simonson & Drolet, 2004).

Several strategies have been explored to decrease the impact of numerical anchoring on practitioners’ decision-making. AlKhars et al. (2019) conducted an experimental study to investigate the impact of cognitive biases, including the anchoring effect, on decision-making in operations management. They found that training could significantly reduce these biases, suggesting that improving cognitive fluency could help mitigate the anchoring effect. Fihn (2019) discusses the role of cognitive biases, including the anchoring effect, in clinical diagnosis. The author suggests that these biases can be mitigated through collective intelligence, again highlighting the potential role of cognitive fluency in moderating the anchoring effect. Finally, Rastogi et al. (2020) present and test successful strategies to diminish the anchoring effect and improve collaborative performance in AI-assisted decision making.

The anchoring effect occurs even if the person making the judgment does not realize he/she is exposed to the anchor (e.g., Mussweiler & Englich, 2005), and even when the anchor values are clearly random and irrelevant to the subsequent judgment. For example, the anchor can be determined by spinning a “wheel of fortune” (Tversky & Kahneman, 1974) or by throwing a dice (Englich et al., 2006). In the study of Critcher and Gilovich (2008), people’s willingness to pay for a meal at a restaurant was influenced by a number in the name of the restaurant, and in the study of Nunes and Boatwright (2004) people offered to pay a higher price for a CD when a sweatshirt on display at an adjacent stand cost $80 rather than $10.

In the standard anchoring paradigm, a person is asked to compare the anchor value to the target before providing their estimate of the target (Strack & Mussweiler, 1997). However, the basic anchoring paradigm, in which a person is simply exposed to a number, has also been shown to produce anchoring effects (Nunes & Boatwright, 2004; Wilson et al., 1996). It has been suggested that in contrast to the standard anchoring paradigm, the anchoring effect produced by basic anchoring is weaker, and people must pay sufficient attention to the anchor value for the anchoring effect to occur (Wilson et al., 1996). Some evidence even suggests that both numerical priming and semantic context need to be present for the anchoring effect to take place (Onuki et al., 2021).

Previous research on anchoring used numerical anchors presented in numerals. The effect of numeric anchors written in words remains unclear. One can assume that both types of anchors should have the same effect as they have the same semantic meaning. However, numbers presented in words are harder to read and lead to lower processing fluency in comparison to numerals (Steffel, 2009). Low processing fluency of the anchor (i.e., more difficulty in processing the anchor) is a factor that can attract sufficient attention to the anchor and, therefore, affect the magnitude of the anchoring effect. Disfluent fonts are often used in marketing as they are more salient and capture consumer attention (Motyka et al., 2016). Mead and Hardesty (2018) found that presenting a sale price in a difficult-to-read font made participants’ subsequent price estimates assimilate more toward the sale price than when it was written in an easy-to-read font.

Metacognitive experiences of difficulty or disfluency (i.e., a lack of processing fluency) also serve as a signal that more elaborate System 2 processing is necessary (Alter et al., 2007). Schwarz et al. (2021) note, that, unless directed so, people are rarely aware of the source and impact of their metacognitive experience. That is, if present, the effect of fluency on anchoring is likely to be unnoticed by the decision makers. The activation of System 2 processing caused by disfluency leads to deeper and more careful processing (Alter et al., 2007; Song & Schwarz, 2008) and better subsequent recall of the information (Diemand-Yauman et al., 2011). Facilitation of stimulus processing, on the other hand, leads to a greater reliance on fast and less effortful System 1 processing. The extra effort and time required to process the disfluent anchor could help produce a more pronounced anchoring effect in the basic anchoring paradigm, as demonstrated by Mead and Hardesty (2018). Presenting a numerical anchor written out in words could have an effect similar to the effect of an anchor presented in a difficult-to-read font.

In the context of consumer decision making, Schwarz et al. (2021) argue that fluent processing enhances liking and preference, and vice versa, an effect that might be independent of the anchor processing. Shanks et al. (2020) investigate the impact of incidental environmental anchors on consumer price estimations. They found that anchoring effects were limited to situations that required explicit thinking about the anchor, suggesting a relationship between cognitive fluency and the strength of the anchoring effect.

We present an experiment which explores how consumers’ willingness to pay for a product relates to the presence of a numerical anchor in the product name and the impact of changing the processing fluency of the anchor. We manipulate the anchor’s processing fluency by presenting it in either numerals or words. Numbers written in words, as opposed to numerals, are harder to read and to comprehend. The processing fluency is lower when numbers are written out in words than when they are written in digits (Steffel, 2009). We assume that the increased time and effort required to process the number written in words will lead to a bigger anchoring effect, as in the study of Mead and Hardesty (2018).

H: A numerical anchor presented in words produces a larger anchoring effect than a numerical anchor presented in digits.

Methods

Similarly to Andersson et al. (2021), we elicited subject’s willingness to pay for a food article—namely a bottle of beer. Our field experiment was conducted in the Craftbeer Bottle Shop & Bar located in Brno Market Hall. This establishment is situated among several other vendors, but none of them offer bottled beer for sale. One-liter bottles of honey beer from a microbrewery were used in this experiment. At the microbrewery, the beer was on sale for 65 CZK (cca. 2.6 EUR) a bottle. The subjects were not informed of the price during the experiment, and it is improbable that they knew the actual price of this product. The bottles of beer were arranged at the end of the counter with no prices on them. There were more bottles of beer on sale in the shop, some of which had price tags. Although these few price tags might have served as additional anchors, they were kept constant across treatments. Furthermore, we believe such an experimental setting contributes to the external validity of the study.

We produced three variants of labels for the same product, which differed only in the product name displayed on the label. The names used in this experiment were “99 zlatých prasátek” (i.e., 99 golden piglets) in the fluent anchor treatment, “Devadesát devět zlatých prasátek” (i.e., Ninety-nine golden piglets) in the disfluent anchor treatment and “Zlaté prasátko” (i.e., Golden piglet) in the no anchor treatment. We chose the anchor value of 99 for two reasons. First, it was higher than the actual price of the product, yet within the price interval of other products in the category, so it served as a high anchor. Second, when written out, the number 99 is quite long and thus harder to process, making it disfluent.

Before the experiment, we verified that the names differ in perceived fluency. Fifteen volunteers rated the readability of the alternative names on a scale from 1 (“very difficult to read”) to 10 (“very easy to read”). Confirming our conjecture, the pre-test participants reported that the name containing a number spelled out in letters (“Devadesát devět zlatých prasátek”) was harder to read (M = 5.6; Wilcoxon test: Z = −2.09464, p = .018) than the name containing a number in a numerical form (“99 zlatých prasátek”; M = 7.6). The mean readability evaluation of the no-anchor version (“Zlaté prasátko”) was M = 7.933.

We use between-subject design in the experiment: of the three labels, each participant encountered only one. The displays were alternated at 3-hour intervals during the testing period. The experiment was conducted on five consecutive weekdays from 11 a.m. until 8 p.m., and the order in which the variants were presented varied each day to minimize any time-of-the-day effect. The same experimenter conducted the experiment each day.

Similarly to Yoon et al. (2019), we elicited willingness to pay with Becker et al.’s (1964) incentive-compatible procedure. Visitors to the Brno Market Hall saw one of three possible displays of beer. When a person approached the display and showed an interest in buying the beer, the experimenter explained that she was a university student collecting data for her bachelor’s thesis and that she would only sell them the product in a somewhat unusual way. The participants were asked to state the maximum price they were willing to pay for the product. The experimenter explained that after they had determined the maximum price they were prepared to pay, a sale price would be drawn at random from a jar on display (the subjects were asked to draw the sale price themselves to increase their confidence in the randomness). The numbers in the jar ranged from 30 to 80 CZK (25 CZK = cca. 1 EUR), in increments of 1 CZK. The subjects were not told what the price distribution was, even when they asked, as they could have anchored on these numbers. If the participant’s offer was greater than or equal to the randomly drawn number, the participant was obliged to buy the beer for the price drawn. If the offer was less than the price that had been drawn, the beer was not sold to the customer. The participants were told that there would be no further negotiations, and nobody was allowed to participate more than once. The experimenter made it clear that, although participation was completely voluntary, there was no other way to obtain the beer than to participate. During the explanation, the experimenter did not provide any numerical examples in order to avoid anchoring on these numbers.

After the deal had been transacted (or not), the experimenter asked the subjects what information they had used to determine their willingness to pay and how easy or difficult it had been to read the name of the product on a scale from 1 to 10 (1 = hard to read; 10 = easy to read). In cases where the participant’s offer had been greater or equal to the randomly drawn sale price and they had purchased the beer, the experimenter asked if they were happy with their purchase (a binary question). Where the participant’s offer had been lower than the drawn sale price and they had not been allowed to buy the beer, the experimenter asked whether they regretted not proposing a higher price and whether they would have been prepared to buy the beer for the price that had been drawn.

The participants in this study were 43 people who visited a Christmas fair in Brno Market Hall in the Czech Republic. Two participants whose offers were lower than the randomly drawn price said that they regretted not stating a willingness to a pay higher price and that they would have bought the beer for the drawn price. This response indicates that they had understated their true willingness to pay and, therefore, we excluded them from the analyses. The final sample consisted of 41 people. Thirteen participants were in the fluent anchor treatment, 12 in disfluent anchor treatment, and 16 in the no anchor treatment. We did not collect socio-demographic information about the participants, as this was outside of the scope of our research question and would not be practical to collect. However, an approximately equal number of men and women participated.

The cooperation with the brewery on the research was limited to one event. Therefore, the sample was limited by the length of the Christmas fair and by the number of visitors who were interested in the honey beer. We collected the data throughout the entire event, and we included all the customers who bought the honey beer. The small sample enabled us to find only strong effects. ANOVA with one factor (three groups) has at least 80% power for effects of f = 0.51 and larger at a 5% significance level and for effects of f = 0.45 and larger at a 10% significance level. Nevertheless, we consider the data obtained in this field experiment, selling a real product to real customers, to be sufficiently interesting for a research note that can stimulate further investigation.

Results

Preliminary Analyses

The analysis of fluency evaluation conducted during the final phase of the experiment showed that participants perceived the fluency of the treatments to vary as expected, F (2, 38) = 125.14, p < .001, ω2 = 0.52. The disfluent anchor was rated as more difficult to read (M = 5.15; 1 = hard to read; 10 = easy to read) than the fluent anchor version (M = 7.86). The no anchor version was considered the easiest to read (M = 9.44).

At the end of the experiment, all participants whose offers had been higher than the random price and who had bought the product stated that they were happy with the transaction (the opposite would indicate that they had overstated their willingness to pay). None of the participants stated that they had used the number in the name of the product to determine their willingness to pay. This suggests that people were not aware of being influenced by the number in the product name.

Hypothesis Testing

The main effect of the experimental condition was strong, but significant only at a level of 10%, F (2, 38) = 3.13, p = .055, ω2 = 0.09. However, the effect was in the opposite direction to that we had hypothesized. The post-hoc test showed that the willingness to pay for the beer was lower with the disfluent anchor (“Devadesát devět zlatých prasátek”; M = 74.333 CZK; SD = 7.5358), when compared to the fluent anchor (“99 zlatých prasátek”; M = 88.692 CZK; SD = 13.774), post hoc test: t(38) = −2.20, p_Tukey = .085, d = −1.28, 95%Ci_d[−2.14, −0.42]. We ran an equivalence test to analyze whether the effect of the anchor written in words on willingness to pay was equivalent to the effect of the anchor written in numerals. We conducted two one-sided t-tests that tested whether the difference between both conditions was smaller than small (i.e., |d| < 0.20). The first one-sided t-test was not significant, which indicates that a numerical anchor presented in words can produce a smaller anchoring effect than a numerical anchor presented in digits, t(18.9) = 2.76, p = .994. However, the second one-sided test was significant, which means that a numerical anchor presented in words does not produce a larger anchoring effect than a numerical anchor presented in digits, t(18.9) = 3.77, p < .001. Therefore, we found support against our hypothesis.

Supplementary Analyses

The post-hoc test indicated that the anchor presented in words did not produce an anchoring effect at all. The willingness to pay in the disfluent anchor treatment (“Devadesát devět zlatých prasátek”; M = 74.333 CZK) was similar to the no-anchor treatment (“Zlaté prasátko”; M = 75.5 CZK; SD = 21.985), t(38) = −0.19, p_Tukey = .981, d = −0.07, 95%Ci_d[−0.82, 0.68]. However, the equivalence test cannot exclude that there are both positive and negative effects of the written-word anchor on willingness to pay, t-test for d < 0.02: t(19.4) = 0.75, p = .230; for d > −0.02: t(19.4) = −0.36, p = .360.

On the other hand, the post hoc test provided some support for the anchoring effect of the classical anchor written in digits (M = 88.692 CZK), which led to a higher willingness to pay than the no-anchor condition (M = 75.5 CZK), t(27) = 2.16, p_Tukey = .091, d = 0.70, 95%Ci_d[−0.05, 1.46]. However, the effect is significant only at a 10% level due to the low test power (N.B., the difference would be significant at a 5% level if we had not used Tukey correction).

Discussion

The goal of this study was to test the hypothesis of whether a numerical anchor presented in words produces a larger anchoring effect (i.e., overly relying on initial numeric information when making a decision) than a numerical anchor presented in digits. We conducted a field experiment that measured consumers’ willingness to pay at the point of purchase using Becker et al.’s (1964) mechanism. Contrary to the hypothesis, the anchor that was written out in words did not affect consumers’ willingness to pay more than the anchor that was written in the form of digits. The one-sided test provided strong support against the hypothesis, and the supplementary analysis indicated that the anchor presented in words might not have produced any anchoring effect at all.

Our hypothesis was based on previous studies showing that numbers written in words are associated with lower processing fluency in comparison to numerals (Steffel, 2009) and that lower processing fluency can strengthen the anchoring effect (Mead & Hardesty, 2018). However, our results are consistent with the first study only. Our manipulation check showed that the numeric anchor written in words was judged to be harder to read in comparison to the numeric anchor written in digits (i.e., the fluency was lower). Nevertheless, the lower fluency did not affect the willingness to pay as we hypothesized.

It is important to note that the fluency manipulation used in our experiment (i.e., writing a number in words) was different from the fluency manipulation used by Mead and Hardesty (2018; i.e., using a less readable font). And while their disfluency induction did lead to a bigger anchoring effect, ours did not. Therefore, we suggest caution in the generalization of previous findings on all fluency manipulations.

Also, the effect of anchor fluency on the magnitude of the anchoring effect may differ with various types of anchors. While the sale price presented by Mead and Hardesty contained relevant information about the real price of the product, our anchor was an irrelevant number in the product name. The difference in the relevance of these two anchors to the subsequent judgment might explain the different results obtained.

One explanation of the various effects of various anchor types can be proposed using the dual process theory. As mentioned in the introduction, people are more likely to rely on System 1 when processing is fluent (Alter et al., 2007). Decisions based on System 1 processes are characterized as largely unconscious, fast, automatic, and independent from working memory resources and general intelligence (Barrouillet, 2011; Evans & Stanovich, 2013; Stanovich & West, 2002). Therefore, an anchor that has no relation to the real value of the product (as in our study) is more likely to be part of an automatic judgment about the price if it is presented in fluent rather than disfluent form. However, this explanation is not applicable to anchors which are related to the real value of the product (as in the study of Mead & Hardesty, 2018). Therefore, we assume that the relevance of an anchor might moderate the effect of fluency on the strength of the anchoring effect and should be considered in future studies. Moreover, future studies should focus on the effect of various types of anchors in order to reveal the conditions in which it is worth using a disfluent anchor. We recommend conducting studies with various fluency manipulations and with various types of anchor.

Another explanation for why the disfluent anchor in our study did not produce a more pronounced anchoring effect than the fluent anchor can be found in the literature on processing fluency. Previous studies showed that fluency affects judgments of liking (e.g., Zajonc, 1968) and that decision makers are insensitive to recognizing this effect (Schwarz et al., 2021). Fluent stimuli are liked better than disfluent ones. It is thus possible that inhibiting the fluency of the product name made our beer less attractive to consumers, which resulted in a lower willingness to pay. Two antagonistic effects of a disfluent anchor in product name might have occurred: the anchoring effect of a high anchor, which increased the willingness to pay, and the effect of decreased liking, which reduced the willingness to pay.

Additionally, it is unclear what the optimal level of disfluency is, and given the fact that the number 99 is quite long when written out and the fact that it was written using a rather disfluent font, it is possible that the word anchor was overly disfluent for people to perceive it. In contrast, the digit anchor (99) was probably read effortlessly, as Arabic numerals are quite easy to decipher and translate into their linguistic meaning.

Despite our sample size, our tests were strong enough to find support against the original hypothesis that a numerical anchor presented in words produces a larger anchoring effect than a numerical anchor presented in digits. Nevertheless, studies with small samples can be more easily biased by the non-equivalence of experimental groups than studies with large samples that can rely on randomization. Our study cannot be considered as strong evidence against the positive effect of low fluency on anchoring, but rather as a warning that the effect of a disfluent anchor might vary in various conditions and as an impetus for further research in this area.

Our study contributes to the existing literature in several ways. It is the first replication of the effect of cognitive fluency on numerical anchoring. Methodological contribution lies in a novel fluency manipulation via the use of digits versus words. Our field experiment contributes to the body of literature focused on the incidental environmental anchors and their effects—a field of study of great interest to business practitioners. Moreover, the study raises doubts about the generalizability of the results of previous research on fluency and anchoring.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article:

ORCID iD

Jakub Prochazka

Data availability

Dataset available via OSF; DOI 10.17605/OSF.IO/HJN3D

References

AlKhars

Evangelopoulos

Pavur

Kulkarni

(2019). Cognitive biases resulting from the representativeness heuristic in operations management: An experimental investigation. Psychology Research and Behavior Management, 12, 263–276. https://doi.org/10.2147/PRBM.S193092

Alter

A. L.

Oppenheimer

D. M.

Epley

Eyre

R. N.

(2007). Overcoming intuition: Metacognitive difficulty activates analytic reasoning. Journal of Experimental Psychology General, 136(4), 569–576. https://doi.org/10.1037/0096-3445.136.4.569

Andersson

Bökman

Wallhagen

Holmgren

Sörqvist

Ahonen-Jonnarth

(2021). Anchoring effect in judgments of objective fact and subjective preference. Food Quality and Preference, 88. https://doi.org/10.1016/j.foodqual.2020.104102

Bahník

Mussweiler

Strack

(2021). Anchoring effect. PsyArXiv Preprints. https://doi.org/10.31234/osf.io/h2wfu

Barrouillet

(2011). Dual-process theories and cognitive development: Advances and challenges. Developmental Review, 31(2-3), 79–85. https://doi.org/10.1016/j.dr.2011.07.002

Becker

G. M.

DeGroot

M. H.

Marschak

(1964). Measuring utility by a single-response sequential method. Behavioral Sciences, 9(3), 226–232. https://doi.org/10.1002/bs.3830090304

Critcher

C. R.

Gilovich

(2008). Incidental environmental anchors. Journal of Behavioral Decision Making, 21(3), 241–251. https://doi.org/10.1002/bdm.586

Diemand-Yauman

Oppenheimer

D. M.

Vaughan

E. B.

(2011). Fortune favors the bold (and the italicized): Effects of disfluency on educational outcomes. Cognition, 118(1), 111–115. https://doi.org/10.1016/j.cognition.2010.09.012

Dogerlioglu-Demir

Koçaş

(2015). Seemingly incidental anchoring: The effect of incidental environmental anchors on consumers’ willingness to pay. Marketing Letters, 26(4), 607–618. https://doi.org/10.1007/s11002-014-9295-0

10.

Englich

Mussweiler

Strack

(2006). Playing dice with criminal sentences: The influence of irrelevant anchors on experts’ judicial decision making. Personality and Social Psychology Bulletin, 32(2), 188–200. https://doi.org/10.1177/0146167205282152

11.

Evans

J. S.

Stanovich

K. E.

(2013). Dual-process theories of higher cognition: Advancing the debate. Perspectives on Psychological Science, 8(3), 223–241. https://doi.org/10.1177/1745691612460685

12.

Fihn

S. D.

(2019). Collective intelligence for clinical diagnosis-Are 2 (or 3) heads better than 1? JAMA network open, 2(3), e191071. https://doi.org/10.1001/jamanetworkopen.2019.1071

13.

Galinsky

A. D.

Mussweiler

(2001). First offers as anchors: The role of perspective-taking and negotiator focus. Journal of Personality and Social Psychology, 81(4), 657–669. https://doi.org/10.1037/0022-3514.81.4.657

14.

Green

Jacowitz

K. E.

Kahneman

McFadden

(1998). Referendum contingent valuation, anchoring, and willingness to pay for public goods. Resource and Energy Economics, 20(2), 85–116. https://doi.org/10.1016/s0928-7655(97)00031-6

15.

Jacowitz

K. E.

Kahneman

(1995). Measures of anchoring in estimation tasks. Personality and Social Psychology Bulletin, 21(11), 1161–1166. https://doi.org/10.1177/01461672952111004

16.

Mead

J. A.

Hardesty

D. M.

(2018). Price font disfluency: Anchoring effects on future price expectations. Journal of Retailing, 94(1), 102–112. https://doi.org/10.1016/j.jretai.2017.09.003

17.

Mochon

Frederick

(2013). Anchoring in sequential judgments. Organizational Behavior and Human Decision Processes, 122(1), 69–79. https://doi.org/10.1016/j.obhdp.2013.04.002

18.

Motyka

Suri

Grewal

Kohli

(2016). Disfluent vs. Fluent price offers: Paradoxical role of processing disfluency. Journal of the Academy of Marketing Science, 44(5), 627–638. https://doi.org/10.1007/s11747-015-0459-0

19.

Mussweiler

Englich

(2005). Subliminal anchoring: Judgmental consequences and underlying mechanisms. Organizational Behavior and Human Decision Processes, 98(2), 133–143. https://doi.org/10.1016/j.obhdp.2004.12.002

20.

Nunes

J. C.

Boatwright

(2004). Incidental prices and their effect on willingness to Pay. JMR, Journal of Marketing Research, 41(4), 457–466.

21.

Onuki

Honda

Ueda

(2021). What stimuli are necessary for anchoring effects to occur? Frontiers in Psychology, 12(602372), https://doi.org/10.3389/fpsyg.2021.602372

22.

Plous

(1989). Thinking the unthinkable: The effects of anchoring on likelihood estimates of nuclear war¹. Journal of Applied Social Psychology, 19(1), 67–91. https://doi.org/10.1111/j.1559-1816.1989.tb01221.x

23.

Rastogi

Zhang

Wei

Varshney

Dhurandhar

Tomsett

R. J.

(2020). Deciding fast and slow: The role of cognitive biases in AI-assisted decision-making. In: Proceedings of the 2020 conference on fairness, accountability, and transparency, New Orleans, LA, 700-710. https://doi.org/10.1145/3512930

24.

Ritov

(1996). Anchoring in simulated competitive market negotiation. Organizational Behavior and Human Decision Processes, 67(1), 16–25. https://doi.org/10.1006/obhd.1996.0062

25.

Schwarz

Jalbert

Noah

Zhang

(2021). Metacognitive experiences as information: Processing fluency in consumer judgment and decision making. Consumer Psychology Review, 4(1), 4–25. https://doi.org/10.1002/arcp.1067

26.

Shanks

D. R.

Barbieri-Hermitte

Vadillo

M. A.

(2020). Do incidental environmental anchors bias consumers’ price estimations? Collabra, 6(1), https://doi.org/10.1525/collabra.310

27.

Simonson

Drolet

(2004). Anchoring effects on consumers' willingness-to-pay and willingness-to-accept. Journal of Consumer Research, 31(3), 681–690. https://doi.org/10.1086/425103

28.

Song

Schwarz

(2008). Fluency and the detection of misleading questions: Low processing fluency attenuates the Moses illusion. Social Cognition, 26(6), 791–799. https://doi.org/10.1521/soco.2008.26.6.791

29.

Stanovich

K. E.

West

R. F.

(2002). Individual differences in reasoning: Implications for the rationality debate? In T. Gilovich, D. Griffin, & D. Kahneman (Eds.), Heuristics and biases: The psychology of intuitive judgment (pp. 421–440). Cambridge University Press.

30.

Steffel

M. L.

(2009). The impact of choice difficulty on self and social inferences. Princeton University.

31.

Strack

Mussweiler

(1997). Explaining the enigmatic anchoring effect: Mechanisms of selective accessibility. Journal of Personality and Social Psychology, 73(3), 437–446. https://doi.org/10.1037//0022-3514.73.3.437

32.

Tversky

Kahneman

(1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124–1131. https://doi.org/10.21236/ad0767426

33.

Wansink

Kent

R. J.

Hoch

S. J.

(1998). An anchoring and adjustment model of purchase quantity decisions. JMR, Journal of Marketing Research, 35(1), 71–81. https://doi.org/10.2307/3151931

34.

Wilson

T. D.

Houston

C. E.

Etling

K. M.

Brekke

(1996). A new look at anchoring effects: Basic anchoring and its antecedents. Journal of Experimental Psychology General, 125(4), 387–402. https://doi.org/10.1037//0096-3445.125.4.387

35.

Yoon

Fong

N. M.

Dimoka

(2019). The robustness of anchoring effects on preferential judgments. Judgment and Decision Making, 14(4), 470–487.

36.

Zajonc

R. B.

(1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology, 9(2, Pt.2), 1–27. https://doi.org/10.1037/h0025848

Are Fluent Anchors More Effective? Field Experiment on Anchoring,Anchor Fluency,and Willingness to Pay

Abstract

Plain language summary

Keywords

Introduction

Methods

Results

Preliminary Analyses

Hypothesis Testing

Supplementary Analyses

Discussion

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

Data availability

References