ArticlePDF Available

Evaluations of Pleasurable Experiences: The Peak-End Rule

February 2008
Psychonomic Bulletin & Review 15(1):96-8

DOI:10.3758/PBR.15.1.96

Source
PubMed

Authors:

Amy M Do

Dartmouth College

George L Wolford

Dartmouth College

Prior research suggests that the addition of mild pain to an aversive event may lead people to prefer and directly choose more pain over less pain (Kahneman, Fredrickson, Schreiber, & Redelmeier, 1993). Kahneman et al. suggest that pain ratings are based on a combination of peak pain and final pain. Similarly, people rate a happy life that ends suddenly as being better than one with additional years of mild happiness (Diener, Wirtz, & Oishi, 2001), even though the former objectively consists of less pleasure than the latter. Applying these concepts to material goods, we investigated the impact of positivity and timing on the retrospective evaluations of material goods. We found strong evidence that the peak-end rule applies to both material goods and pain.

Content uploaded by George L Wolford

Content may be subject to copyright.

Kahneman, Fredrickson, Schreiber, and Redelmeier

(1993) found that participants preferred 60 sec of 14ºC

ice water followed by 30 sec of 15ºC ice water to 60 sec

of 14ºC ice water alone. This result is counterintuitive,

since water at both 14ºC and 15ºC is unpleasant. People

seemed to prefer and directly choose more pain over less

pain in this situation. The authors suggested that pain rat-

ings are influenced by the peak pain experienced during

the episode and the final level of pain—the peak–end rule.

The peak–end rule for pain ratings has been confirmed

in several other experiments. Redelmeier and Kahneman

(1996) found that patients rated colonoscopies as less un-

pleasant if an interval of mild pain was added to the end of

the procedure. Patients’ memories of the pain experienced

during a colonoscopy were correlated with predictions

of the peak–end rule (Redelmeier & Kahneman, 1996).

Schreiber and Kahneman (2000) showed that ratings of

unpleasantly loud sounds showed clear peak–end effects.

Fredrickson (2000) reviewed a large number of experi-

ments supporting the peak–end rule. The peak–end rule

yields the counterintuitive result that conditions can read-

ily be arranged in which people prefer and choose a condi-

tion with objectively more pain as long as the episode ends

on a relatively less painful note.

In the realm of positive experiences, Diener, Wirtz, and

Oishi (2001) conducted studies in which participants rated

a wonderful life that ended suddenly as better than one

with the addition of mildly pleasant years. They termed

this the James Dean effect. In addition, television com-

mercials that induce positive feelings are rated more

highly by consumers if the commercials have high peaks

of intensity and strong positive endings (Baumgartner,

Sujan, & Padgett, 1997). Participants in Fredrickson and

Kahneman’s (1993) study viewed a series of video clips

that varied in pleasantness and duration. The data indicate

that people essentially disregard the duration of an affec-

tive episode when making retrospective assessments and

that they heavily rely on just one or two moments in the

episode (Fredrickson & Kahneman, 1993).

Simonson, Carmon, and O’Curry (1994) showed that

in some cases, added features and offered promotions can

lead participants to rate a brand lower, even when the fea-

tures are seen as neutral or positive and the promotions are

completely optional. Their research is not directly relevant

to the peak–end rule in that the information about prod-

ucts, promotions, and /or features was available simulta-

neously rather than sequentially. Further, their research

differed from both previous research on the peak–end rule

and our research in that the presence of the features and

promotions affords inferences about the primary product

that could affect the ratings (e.g., “the product must be

deficient if they have to offer promotions”). Our research

does not afford such inferences.

By extrapolating the various peak–end findings to the

domain of pleasurable goods, we predicted that people

would retrospectively report lower levels of overall plea-

sure for a desirable gift if a positive but less desirable gift

were added to it, even though the addition of this second

gift objectively increases the total worth. In this way, peo-

ple would irrationally rate experiences with objectively less

overall pleasure higher than experiences with objectively

greater overall pleasure. We conducted two experiments in

order to investigate this counterintuitive hypothesis.

EXPERIMENT 1

Method

Over the course of several months, three charitable fundraisers

were conducted at Dartmouth College during which undergraduate

students made monetary donations to nonprofit organizations. All of

the money collected went to the designated charities. Donors were

Evaluations of pleasurable experiences:

The peak–end rule

Am y m. Do, A l e x A n D e r V. r u p e r t , A n D Ge o r G e Wo l f o r D

Dartmouth College, Hanover, New Hampshire

Prior research suggests that the addition of mild pain to an aversive event may lead people to prefer and di-

rectly choose more pain over less pain (Kahneman, Fredrickson, Schreiber, & Redelmeier, 1993). Kahneman

et al. suggest that pain ratings are based on a combination of peak pain and f inal pain. Similarly, people rate a

happy life that ends suddenly as being better than one with additional years of mild happiness (Diener, Wirtz, &

Oishi, 2001), even though the former objectively consists of less pleasure than the latter. Applying these concepts

to material goods, we investigated the impact of positivity and timing on the retrospective evaluations of material

goods. We found strong evidence that the peak–end rule applies to both material goods and pain.

Psychonomic Bulletin & Review

2008, 15 (1), 96-98

doi: 10.3758/PBR.15.1.96

G. Wolford, wolford@dartmouth.edu

Pl e a s u r a b l e ex P e r i e n c e s a n d t h e Pe a k –en d ru l e 97

informed that they could enter a “raffle” for a chance to win free

DVDs. All of the donors were later contacted via e-mail and told that

they had won one or two DVDs from the raffle and that they would be

able to choose their free DVDs from the list(s) attached to the e-mail.

Participants were contacted through the e-mail account of a nonprofit

organization in order to eliminate suspicion that the DVD raffle was

part of a psychological experiment.

Participants were presented with the pleasure-inducing opportunity

to select free DVDs from one or two predetermined lists. Two dif-

ferent DVD lists were used: List A consisted of 10 movies that were

rated highly on www.rottentomatoes.com (ratings are based on users’

opinions); List B consisted of 10 movies that were rated positively on

www.rottentomatoes.com, but less so than those on List A. The “very

pleasing” experience involved choosing a free DVD from List A. The

“mildly pleasing” experience involved choosing a free DVD from

List B. Participants voluntarily took part in the experiment by respond-

ing to the DVD offer via e-mail. All participants who were offered a

free DVD—even one from List B—voluntarily chose to receive one,

suggesting that the movies from List B were seen as positive for these

participants.

The participants were randomly divided into one of five groups

(see Table 1). Group A received a DVD from List A only. Group A1B

received a DVD from List A and a second DVD from List B. The par-

ticipants in Group A1B viewed List B via a second e-mail only after

they had chosen their first DVD from List A. There were three addi-

tional control groups to aid in interpreting any observed differences

between Groups A and A1B. The control groups were B, B1A, and

A1A. After making all of their DVD selections, participants were

asked to rate how pleased they were with the overall experience of

receiving the DVD offer on a scale that ranged from 1 (least pleased)

to 7 (most pleased). Participants were also asked to provide an expla-

nation for their rating. We had approval from the Institutional Review

Board (IRB) to carry out this experiment without prior informed con-

sent. The DVDs were delivered to the participants on a later date,

along with a letter that served to debrief them.

Results

The participants recruited from all three fundraisers

were combined in the statistical analysis, for a total of

104 participants. On the basis of their written comments to

the free DVD offer, 4 of the participants were eliminated.

They made comments that indicated that our DVD lists

were not viewed as intended in their cases (one in each

of four different groups). For instance, 1 participant in

Group A1B wrote that one of the movies from List B was

his “all time favorite movie.” A 2nd participant thought

the List A movies were all “establishment.” The analyses

are based on the 100 remaining participants.

The results are displayed in Table 1. Because our data

may not fully have met the assumptions for parametric

tests, we carried out a Kruskal–Wallis H test on the f ive

groups, followed by three Mann–Whitney U tests to ex-

amine three specif ic contrasts. The H test was significant

[χ2(4) 5 41.8, p , .001]. Using Mann–Whitney tests, we

found that the mean rank pleasure rating of Group A was

significantly greater than that of Group A1B (z 5 3.31,

p 5 .001). Group A was not signif icantly different from

Group B1A ( z 5 1.34, p 5 .179). Group A1B was signif-

icantly different from Group B1A (z 5 2.08, p 5 .045).

EXPERIMENT 2

Method and Procedure

A smaller-scale experiment was conducted in order to attempt to

replicate the results of Experiment 1 in a different population and

context. Experiment 2 was conducted at a residential house that was

routinely visited by elementary school and middle school children

on Halloween night (see Santino, 1981, for a description of Hallow-

een customs). Participants consisted of “trick-or-treaters” who were

old enough (mean age 5 10.03, SD 5 1.79) to answer the survey

question “How happy are you with the candy that I gave you?” by

pointing to a “smiley face” on a smiley-face happiness scale. Seven

out of 28 (25%) of the participants were female.

On the night of Halloween, trick-or-treaters were given different

combinations of candy and asked to rate their level of happiness

in relation to the candy that they were given. The “very pleasing”

treat was a full-size Hershey’s chocolate bar. The “mildly pleas-

ing” treat was a piece of bubble gum. Permission was granted by

the IRB to conduct the experiment without the prior consent of the

participants.

Trick-or-treaters were assigned to one of four groups (A, A1B,

A1A, and B). Because the comparison of Groups A and A1B was

the critical assessment in the experiment, and because it was not pos-

sible to accurately predict how many participants would be recruited

that night, group assignment was biased toward the first two groups

in order to ensure that there would be enough participants for a statis-

tical comparison of these two groups. Specifically, participants had a

three-eighths probability of being assigned to Group A, a three-eights

probability of being assigned to Group A1B, and a one-eighth prob-

ability of being assigned to each of the other groups.

All participants who were offered the bubble gum voluntarily ac-

cepted it, which suggests that the experience of receiving the gum

was positive. After participants were presented with their candy, they

were shown a picture of seven smiley faces, each expressing a differ-

ent level of happiness. The first smiley face appeared “neutral” (the

mouth was drawn as a horizontal line), and the seventh smiley face

appeared “most pleased” (the mouth was drawn as a wide, open-

mouthed grin). The other smiley faces varied on a continuum be-

tween these two extremes. Participants were asked to indicate which

smiley face reflected how happy they were with the candy that was

given to them. Afterward, trick-or-treaters and their adult compan-

ions were given a letter that served to debrief them.

Results

The results are displayed in Table 2. Every child in

Groups A and A1A responded with the top rating of 7. We

carried out a Kruskal–Wallis H test on the four groups, fol-

lowed by one planned comparison. The H test was significant

[χ2(3) 5 18.04, p , .001]. Using a Mann– Whitney U test,

we found that the mean rank pleasure rating of Group A

was significantly greater than that of Group A1B (z 5

2.96, p 5 .003).1 The results of Experiment 2 replicated

the key finding of Experiment 1, demonstrating that two

Table 1

Mean Ratings As a Function of Group in Experiment 1

Group N M Median SE

A 29 5.21 5.00 0.17

A1B 21 4.14 4.00 0.26

B1A 17 4.82 5.00 0.23

A1A 19 5.50 5.50 0.22

B 14 2.57 2.50 0.29

Table 2

Mean Ratings As a Function of Group in Experiment 2

Group N M Median SE

A 10 7.00 7.00 0.00

A1B 11 5.91 6.00 0.39

A1A 4 7.00 7.00 0.00

B 3 3.00 3.00 0.00

98 do, ru P e r t , a n d Wo l f o r d

positives are given a lower rating than a single positive if

the second item is less positive than the first.

DISCUSSION

The results of both Experiment 1 and Experiment 2 sup-

port the hypothesis that people report lower levels of overall

pleasure with the addition of mild pleasure at the end of an

experience. In this way, people rate less pleasure higher than

more pleasure. All of the participants in Group B rated the

experience as positively pleasurable, with a mean rating of

2.57 in Experiment 1 and a mean rating of 3.00 in Experi-

ment 2. We made it clear in both experiments that a rating of

“1” was considered neutral. In addition, participants volun-

tarily made a selection from List B movies in Experiment 1

and happily accepted the gum in Experiment 2, which im-

plies that these items were seen as positive. There were even

spontaneous expressions of pleasure from a few of the par-

ticipants in Group B in both Experiments 1 and 2. It is, of

course, possible that a particular participant would find the

List B movies or the “mildly pleasing” candy aversive, but

we doubt that was the case for many of the participants.

In Experiment 1, Group A1A had a higher average

rating than did Group A, although this difference is not

statistically significant. In Experiment 2, the two groups

were equal. The lack of significance could be due to a ceil-

ing effect—especially in Experiment 2—but this result

does eliminate the possibility that simply receiving two

items (e.g., two DVDs or two pieces of candy) is worse

than receiving a single item.

In Experiment 1, the mean rating of Group A was not

significantly different from that of Group B1A, and the

mean rating of Group B1A was significantly higher than

that of Group A1B. These results suggest that order is

crucial, and they confirm that the peak–end rule is appli-

cable to pleasurable experiences, particularly to pleasure

derived from material gains.

The peak–end rule presumably applies to episodes that

are viewed as one joint event. In Experiment 1, participants

were specifically asked to rate how pleased they were with

the “overall” DVD offer. In Experiment 2, participants

were asked to rate how pleased they were “with the candy

that I gave you.” This wording was intended to encourage

the participants in Experiment 1 to perceive the overall

experience of receiving two free DVDs as one cohesive

event, and to encourage those in Experiment 2 to perceive

the overall experience of receiving two pieces of candy as

one joint experience. The strongly significant difference

between Groups A1B and B in both experiments rules out

the possibility that the ratings were determined entirely by

the last item received. However, an implementation of the

peak–end rule in which the two items are combined in a

weighted average—with the latter item receiving a higher

weight—is consistent with our results.

The results of this study support the idea that the effects

found in retrospective evaluations of pain are applicable to

evaluations of pleasure. If you were planning to give mul-

tiple gifts for a future holiday, you might consider giving

only the best one—or at least making sure that you give

the best one last.

AUTHOR NOTE

We thank the Lincoln Filene Fund at Dartmouth College for supporting

this research. Address correspondence to G. Wolford, 6207 Moore Hall,

Dartmouth College, Hanover, NH 03755 (e-mail: wolford@dartmouth

.edu).

REFERENCES

Baumgartner, H., Sujan, M., & Padgett, D. (1997). Patterns of af-

fective reactions to advertisements: The integration of moment-to-

moment responses into overall judgments. Journal of Marketing Re-

search, 32, 219-232.

Diener, E., Wirtz, D., & Oishi, S. (2001). End effects of rated life qual-

ity: The James Dean effect. Psychological Science, 12, 124-128.

Fredrickson, B. L. (2000). Extracting meaning from past affective

experiences: The importance of peaks, ends, and specif ic emotions.

Cognition & Emotion, 14, 577-606.

Fredrickson, B. L., & Kahneman, D. (1993). Duration neglect in ret-

rospective evaluations of affective episodes. Journal of Personality &

Social Psychology, 65, 44-55.

Kahneman, D., Fredrickson, B. L., Schreiber, C. A., & Redel-

meier, D. A. (1993). When more pain is preferred to less: Adding a

better end. Psychological Science, 4, 401-405.

Redelmeier, D. A., & Kahneman, D. (1996). Patients’ memories of

painful medical treatments: Real-time and retrospective evaluations

of two minimally invasive procedures. Pain, 66, 3-8.

Santino, J. (1981). Halloween in America: Contemporary customs and

performances. Western Folklore, 41, 1-20.

Schreiber, C. A., & Kahneman, D. (2000). Determinants of the re-

membered utility of aversive sounds. Journal of Experimental Psy-

chology, 129, 27-42.

Simonson, I., Carmon, Z., & O’Curry, S. (1994). Experimental evi-

dence on the negative effect of product features and sales promotions

on brand choice. Marketing Science, 13, 23-40.

NOTE

1. Because there were so many tied scores, we also carried out a χ2 test

of Group A versus Group A1B. We dichotomized the ratings into those

of 7 versus all of the others [χ2(1) 5 9.545, p 5 .002]. A two-sided

Fisher exact test yielded a p value of .004.

(Manuscript received August 22, 2006;

revision accepted for publication May 15, 2007.)

Toward a Human-Centered Metric for Evaluating Trust in Artificial Intelligence Systems

Conference Paper

Mar 2025

Andrea Esposito

Trust is a critical human factor in the adoption and effectiveness of artificial intelligence (AI) systems, especially in high-risk domains such as medicine and cybersecurity. This paper introduces a novel probabilistic metric to objectively quantify trust in AI systems, using the degree of reliance on AI by human users as a proxy. The proposed metric leverages changes in human decision-making influenced by AI decisions and can be computed using a flexible data-driven approach. To validate this metric, we present user study protocols applicable to diverse domains, providing examples in the medical field (for Alzheimer's disease detection) and cybersecurity (for phishing detection). These protocols employ a frequentist definition of probability to estimate and compute trust, allowing for a comparison with trust questionnaires to establish correlations. This work aims to contribute a standardized, interpretable, and adaptable method for evaluating trust in human-centered AI systems.

Experienced cognitive load in the emergency department. A prospective study

Article

Full-text available

Jan 2025
PLOS ONE

Background and importance The emergency department (ED) is a hectic place, where many critically ill patients are treated. For residents working in the ED, this environment may be demanding. Objectives The aim of this study was to investigate the [1] cognitive load experienced by residents working in the ED, and [2] differences in cognitive load during the day. Methods In this multicentre, prospective study in three EDs in the Netherlands, the experienced cognitive load was graded by residents on three scales, agreed upon during Delphi meetings: the complexity (low-high: 1–3), mental effort (low-high: 1–9) and comfortability scale (low-high: 0–100%). We applied the scores per decision, 1-hour and 2-hour intervals, patient and shift. Main results We observed 14 residents and analysed 74 1-hour and 45 2-hour intervals, 79 patients, 24 shifts and 592 separate decisions. The experienced cognitive load per decision was low. In contrast, the cognitive load was higher per 2-hour interval (mental effort: median 4.0 (IQR 4.0) and comfortability 80% (IQR 20)) and per shift (mental effort: median 5.5 (IQR 4.0) and comfortability 80% (IQR 20). Complexity was low for all measurements. Mental effort rose from 17h onwards higher values, whereas a decrease in comfortability was seen from 21h onwards. Conclusion From 17h onwards, residents working in the ED experienced rather high mental effort and reported feeling not optimally comfortable when making decisions. The mental effort was highest between 21-23h. This was found when cognitive load was measured per 2-hour interval and per shift, but not per decision. This study may provide an insights to optimise cognitive load by reorganisation of the ED.

The Case for Human-like Scalable Intelligence in the Medical Field

Article

Full-text available

Jan 2025

Kyrtin Atreides

This paper discusses the use case of applying the first working cognitive architectures, built on the independent core observer model (ICOM), to the medical field, among others. The advantages of this approach are compared to the status quo and the limitations of narrow AI systems like LLMs and RL. Noteworthy advantages covered include depth, breadth, updatedness, and fidelity of medical knowledge, the “noise” or inconsistency of diagnoses and treatment, as well as preventative care, cost, time, and ethical considerations. Potent advantages for better integrating, coordinating, and otherwise accelerating medical research, particularly in less developed, underserved, and understudied regions, as aligned with the Sustainable Development Goals (SDGs), are also highlighted. Unique opportunities waiting to be explored, including interdisciplinary advantages, as well as challenges related to the disruption of current systems and processes are covered. These conservatively offer cumulative improvements across multiple dimensions measured in orders of magnitude. The novel value of human-like systems is specifically discussed for hyper-complex knowledge domains and problems, where the effect of integrating them is greatest. Received: 13 May 2024 | Revised: 14 October 2024 | Accepted: 2 January 2025 Conflicts of Interest The author declares that he has no conflicts of interest to this work. Data Availability Statement The data that support the findings of this paper are entirely based upon the cited works, with simple calculations applied to illustrate the data visually, in combination, and in context. The supporting calculations, tables, and charts are openly available in Excel at https://norn.ai/wp-content/uploads/2024/09/Medical-Use-Case-Figures-v1.1.xlsx. Author Contribution Statement Kyrtin Atreides: Conceptualization, Methodology, Investigation, Writing – original draft, Writing – review & editing, Visualization.

Impact of memory on opinion dynamics

Preprint

May 2018

We investigate an agent-based model of opinion dynamics with two types of social response: conformity and independence. Conformity is introduced to the model analogously as in the Sznajd model or q-voter model, which means that only unanimous group exerts peer pressure on individuals. The novelty, in relation to previous versions of the q-voter model, is memory possessed by each agent and external noise T, which plays the role of social temperature. Each agent has its own memories of past experiences related to the social costs and benefits of being independent or conformist. If an agent was awarded in past more for being independent, it will have a greater tendency to be independent than conformist and vice versa. We will show that depending on the social temperature T the system spontaneously organizes into one of two regimes. Below a certain critical social temperature

T_c

, all agents in the society acquire personal traits (so called person state). Some of them become permanent conformists and others start to behave forever independently. This means that initially homogeneous population becomes heterogeneous, and agents respond differently to social influence. For

T>T_c

, all agents with equal probabilities behave independently or conform to peer pressure (so called situation state). This regime change between person and situation state, which reminds the idea of an annealed vs. quenched disorder, affects also public opinion. Particularly interesting results are obtained for individualistic societies, in which public opinion is non-monotonic function of T, which means that there is an optimal social temperature for which an agreement in the society is the highest.

First Impressions or Good Endings? Preferences Depend on When You Ask

Article

Full-text available

Sep 2024

Rewards often unfold over time; we must summarize events in memory to guide future choices. Do first impressions matter most, or is it better to end on a good note? Across nine studies (N = 569), we tested these competing intuitions and found that preferences depend on when rewards occur and when we are asked to evaluate an experience. In our “garage sale” task, participants opened boxes containing sequences of objects with values. All boxes were equally valuable, but rewards were either evenly distributed or clustered at the beginning, middle, or end of the sequence. First, we tested preferences and valuation shortly after learning; we consistently found that boxes with rewards at the beginning were strongly preferred and overvalued. Object-value associative memory was impaired in boxes with early rewards, suggesting that value information was linked to the box rather than the objects. However, when tested after an overnight delay, participants equally preferred boxes with any cluster of rewards, whether at the beginning, middle, or end of the experience. Finally, we demonstrated that evaluating shortly after an experience led to lasting preferences for early rewards. Overall, we show that people summarize rewarding experiences in a nonlinear and time-dependent way, unifying prior work on affect, memory, and decision making. We propose that short-term preferences are biased by first impressions. However, when we wait and evaluate an experience after a delay, we summarize rewarding events in memory to inform adaptive longer term preferences. Preferences depend on when rewards occur and when we first evaluate an experience.

Experimental Verification and Further Reflection on the Peak-End Rule

Article

Full-text available

Aug 2024

People make choices all the time, and sometimes the results of choices are unexpected, and the existence of the peak-end rule can explain the counterintuition in our decision evaluation. In fact, changes in "peak" and "end" experiences affect the choices people make all the time and greatly affect the evaluation of decisions. This article summarizes the characteristics of the Peak-End Rule and provides a large number of real-life examples to explain it. Based on this, three experiments were designed to verify the correctness of the Peak-End Rule from different perspectives. Among these experiments, Experiment 2 was the most important, and all participants participated in two stages of the experiment. In the first stage, they ran 100 meters with full effort, and in the second stage, they slowed down and ran 50 meters after running 100 meters with full effort. It was recorded that all participants chose to participate in the second stage of exercise, and more people chose to complete the second stage. This conclusion is contrary to common sense because the second stage covered an additional distance of 50 meters. But this result precisely confirms the conclusion of the Peak-End Theorem, because the end of the second stage run is easier, and more participants choose to run in the second stage. Meanwhile, through in-depth analysis of the data from Experiment 3, the author believes that the impact of the "endpoint" on decision-making is greater than that of the "peak".

Exploring the Impact of Decoys on Decision-Making by Young Children

Article

May 2024
J BEHAV DECIS MAKING

The asymmetric dominance effect (or decoy effect) is a decision‐making phenomenon that occurs when preference for a target alternative shifts with the addition of a similar, yet inferior alternative dubbed the decoy. Despite the considerable number of studies examining the decoy effect with adult humans and animals, there is comparatively less research on context effects within the developmental domain. In this study, we explored the impact of a decoy on choice behavior by young children (3–9 years old) using a preferential choice task as well as a perceptual discrimination task. Introduction of an inferior decoy impacted choice behavior across 2‐alternative (binary) versus 3‐alternative (trinary) sets, such that inclusion of the dominated decoy in expanded sets decreased selection of the superior target alternative. This pattern of results indicates a reversal of the standard attraction effect, also known as the repulsion effect. We discuss these findings in light of the adult and comparative literatures on decoy effects as well as call for additional developmental studies exploring the impact of inferior alternatives in multialternative decision‐making.

DSRS: DELIGHT sequential recommender system

Article

Feb 2025
ENG APPL ARTIF INTEL

First impressions vs. The Peak-End Rule: Episodic evaluations in a service experience and the moderating effect of retrospective delay

Article

Dec 2024
J BUS RES

Customizing customer journey in hotels: A focus on human touch

Article

Jan 2025
Int J Hospit Manag

There are criticisms that the hospitality industry has lost its human touch as a competitive edge. While previous studies have examined how high-touch customer experiences should be provided, their recommendations are often related to humanizing service employees and technologies. It remains unclear how human touch should be better leveraged and incorporated into service design to enhance customer experience throughout the customer journey. This study explores how hotel service encounters could be tailored to customers with varying needs for human interactions. Using multigroup and importance-performance map analyses, customer experiences reflecting different levels of human touch were compared across critical service encounters throughout the guest cycle. The findings identify the service encounters where human touch is particularly important, recognize the characteristics of customers who desire a higher level of human touch, and provide explanations of the underlying reasons. Follow-up interviews were conducted to complement the quantitative results. These analyses collectively provide a comprehensive explanation of how high-touch service should be provided in hotels.

Experimental Evidence on the Negative Effect of Product Features and Sales Promotions on Brand Choice

Article

Full-text available

May 1994

Sales promotions and product enhancements are commonly expected to increase a brand's sales, when they do not negatively impact its utility and cost. That is, the purchase probability of consumers who find the promotion or additional feature attractive will increase, whereas the purchase likelihood of other consumers will not be affected. In contrast, we propose that consumers, who perceive a new feature or promotion as providing little or no value, will be less likely to purchase the enhanced brand even when the added feature clearly does not diminish the value of the brand. Thus, a new product feature or promotion may decrease a brand's overall choice probability when the segment of consumers who perceive it as providing little or no value is large compared to the segment that finds the feature attractive. This prediction was supported in three studies using actual promotions that have been employed in the marketplace (e.g., a Doughboy Collector's Plate that buyers of Pillsbury cake mix had the option to purchase for $6.19). We examined five alternative explanations for this effect. The results suggest that, when consumers are uncertain about the values of products and about their preferences, such features and premiums provide reasons against buying the promoted brands and are seen as susceptible to criticism. We discuss the theoretical and practical implications of the findings for segmentation, product, promotional, and pricing strategies.

Duration Neglect in Retrospective Evaluation of Affective Episodes

Article

Full-text available

Jul 1993

Two experiments documented a phenomenon of duration neglect in people's global evaluations of past affective experiences. In Study 1, 32 Ss viewed aversive film clips and pleasant film clips that varied in duration and intensity. Ss provided real-time ratings of affect during each clip and global evaluations of each clip when it was over. In Study 2, 96 Ss viewed these same clips and later ranked them by their contribution to an overall experience of pleasantness (or unpleasantness). Experimental Ss ranked the films from memory; control Ss were informed of the ranking task in advance and encouraged to make evaluations on-line. Effects of film duration on retrospective evaluations were small, entirely explained by changes in real-time affect and further reduced when made from memory. Retrospective evaluations appear to be determined by a weighted average of "snapshots" of the actual affective experience, as if duration did not matter.

Extracting meaning from past affective experiences: The importance of peaks, ends, and specific emotions

Article

Jul 2000

B.L. Fredrickson

This article reviews existing empirical research on the peak-and-end rule. This rule states that people's global evaluations of past affective episodes can be well predicted by the affect experienced during just two moments: the moment of peak affect intensity and the ending. One consequence of the peak-and-end rule is that the duration of affective episodes is largely neglected. Evidence supporting the peak-and-end rule is robust, but qualified. New directions for future work in this emerging area of study are outlined. In particular, the personal meanings associated with specific moments and with specific emotions should be assessed. It is hypothesised that moments rich with self-relevant information will dominate people's global evaluations of past affective episodes. The article concludes with a discussion of ways to measure and optimise objective happiness.

Patterns of Affective Reactions to Advertisements: The Integration of Moment-to-Moment Responses into Overall Judgments

Article

May 1997

The authors develop several hypotheses regarding the integration of moment-to-moment emotional responses into overall ad judgments, using the psychological literature dealing with people's preferences for sequences of hedonic outcomes, and they conduct three studies to test these predictions. The results of Study 1 indicate that consumers' global assessments of extended affective episodes elicited by advertisements are dominated by the peak emotional experience and the final moment of the series and also are correlated with the pace at which momentary affective reactions improve over time. Ad duration is related only weakly to overall ad judgments, though longer advertisements have an advantage as long as they build toward a peak emotional experience. In Study 2, the authors replicate these findings under more realistic viewing conditions and demonstrate that the results cannot be attributed solely to memory artifacts that are based on recency. Study 3 implicates adaptation as a possible explanation for the preference for delayed peaks and high ends and further explains the weak effects of ad duration by showing experimentally that longer advertisements can both enhance and depress ad judgments depending on how duration affects the peak emotional experience and the final moment.

Halloween in America: Contemporary Customs and Performances

Article

Jan 1983
West Folklore

Jack Santino

Extracting Meaning From Past Affective Experiences: The Importance of Peaks, Ends, and Specific Emotions

Article

Aug 2010

Barbara L. Fredrickson

When More Pain is Preferred to Less—Adding a Better End

Article

Nov 1993

Subjects were exposed to two aversive experiences: in the short trial, they immersed one hand in water at 14 °C for 60 s; in the long trial, they immersed the other hand at 14 °C for 60 s, then kept the hand in the water 30 s longer as the temperature of the water was gradually raised to 15 °C, still painful but distinctly less so for most subjects. Subjects were later given a choice of which trial to repeat. A significant majority chose to repeat the long trial, apparently preferring more pain over less. The results add to other evidence suggesting that duration plays a small role in retrospective evaluations of aversive experiences; such evaluations are often dominated by the discomfort at the worst and at the final moments of episodes.

Duration neglect in retrospective evaluations of affective episodes

Article

Jul 1993
J PERS SOC PSYCHOL

Documented with 2 experiments a phenomenon of duration neglect in people's global evaluations of past affective experiences. In Study 1, 32 Ss viewed aversive film clips and pleasant film clips that varied in duration and intensity. Ss provided real-time ratings of affect during each clip and global evaluations of each clip when it was over. In Study 2, 96 Ss viewed these same clips and later ranked them by their contribution to an overall experience of pleasantness (or unpleasantness). Experimental Ss ranked the films from memory; control Ss were informed of the ranking task in advance and encouraged to make evaluations on-line. Effects of film duration on retropsective evaluations were small, entirely explained by changes in real-time affects and further reduced when made from memory. Retrospective evaluations appear to be determined by a weighing average of "snapshots" of the actual affective experiences, as if duration did not matter.

Patients' memories of painful medical treatments: Real-time and retrospective evaluations of two minimally invasive procedures

Article

Aug 1996

Patients' memories of painful medical procedures may influence their decisions about future treatments, yet memories are imperfect and susceptible to bias. We recorded in real-time the intensity of pain experienced by patients undergoing colonoscopy (n = 154) and lithotripsy (n = 133). We subsequently examined patients' retrospective evaluations of the total pain of the procedure, and related these evaluations to the real-time recording obtained during the experience. We found that individuals varied substantially in the total amount of pain they remembered. Patients' judgments of total pain were strongly correlated with the peak intensity of pain (P < 0.005) and with the intensity of pain recorded during the last 3 min of the procedure (P < 0.005). Despite substantial variation in the duration of the experience, lengthy procedures were not remembered as particularly aversive. We suggest that patients' memories of painful medical procedures largely reflect the intensity of pain at the worst part and at the final part of the experience.

Determinants of the remembered utility of aversive sounds

Article

Apr 2000

Retrospective evaluations of aversive episodes were studied in the context of a general model of "judgment by prototype" that has been applied in other situations. Unpleasant sounds of variable loudness and duration were the stimuli. In Experiment 1, continuous reports of annoyance closely tracked variations of noise intensity. Hypotheses about the determinants of retrospective evaluation were examined in Experiment 2. Experiment 3 confirmed a prediction of judgment by prototype: The effects of sound duration and intensity are additive in multitrial experiments. Experiment 4 confirmed a robust preference for aversive episodes that are "improved" by adding a period of reduced aversiveness.