What If You Surprise Your Customers ... Will They Be More Satisfied? Findings From a Pilot Experiment

Joelle Vanhamme, Erasmus University Rotterdam-ERIM & LABACC
Dirk Snelders, Delft University of Technology
ABSTRACT - The purpose of this pilot experiment was to investigate the causal relationship between surprise (measured via verbal and non verbal measures) and satisfaction. The results clearly show thatBalthough respondents in the surprise condition did not have higher levels of satisfaction than in the control conditionsB1) consumers evaluated more positively the 'surprising element’ in the surprise condition than in the non surprise condition; and 2) within the surprise condition, the level of surprise was found to have a direct effect on satisfaction.
[ to cite ]:
Joelle Vanhamme and Dirk Snelders (2003) ,"What If You Surprise Your Customers ... Will They Be More Satisfied? Findings From a Pilot Experiment", in NA - Advances in Consumer Research Volume 30, eds. Punam Anand Keller and Dennis W. Rook, Valdosta, GA : Association for Consumer Research, Pages: 48-55.

Advances in Consumer Research Volume 30, 2003     Pages 48-55

WHAT IF YOU SURPRISE YOUR CUSTOMERS ... WILL THEY BE MORE SATISFIED? FINDINGS FROM A PILOT EXPERIMENT

Joelle Vanhamme, Erasmus University Rotterdam-ERIM & LABACC

Dirk Snelders, Delft University of Technology

ABSTRACT -

The purpose of this pilot experiment was to investigate the causal relationship between surprise (measured via verbal and non verbal measures) and satisfaction. The results clearly show thatBalthough respondents in the surprise condition did not have higher levels of satisfaction than in the control conditionsB1) consumers evaluated more positively the 'surprising element’ in the surprise condition than in the non surprise condition; and 2) within the surprise condition, the level of surprise was found to have a direct effect on satisfaction.

INTRODUCTION

It has been argued that "businesses need to move beyond mere satisfaction, to customer delight" (Rust et al. 1996: 229). Customer delightBwhich is defined as the highest level of customer satisfactionBis believed to translate into higher customer retention and loyalty (Oliver et al. 1997, Rust et al. 1996). Interestingly, pleasant surprise is believed to be a privileged way to trigger customer delight (Rust and Oliver 2000).

The idea that surprise and delight are related was first suggested by the empirical work of Plutchik (1980). This author found that delight results from a combination of surprise and joy. Westbrook and Oliver (1991; Oliver and Westbrook 1993) also report some indirect evidence for a link between surprise and satisfaction. Using cluster analysis on the emotions consumers experienced during products/services consumption, both of these studies brought to light a cluster of consumers with high scores of surprise and joy; these consumers were found to be more satisfied than the consumers from any other group. The exploratory surveys by Oliver et al. (1997) and Mano and Oliver (1993) also suggest the existence of a relationship between surprise and satisfaction. Their LISREL analysis supported the causal path "arousal positive emotions [positive affective reactions for the Mano and Oliver study] satisfaction." However, a closer look at their measures for arousal reveals that it is rather surprise that was measured: two out of the three items of the DES scale of Izard (1977) for surprise were used. Finally, the existence of both a direct and indirect (via positive emotions) relationship between surprise and satisfaction was also confirmed in a recent exploratory diary study by Vanhamme (2001), which involved a wide range of product/service purchase/consumption experiences.

Thus, all the studies mentioned above suggest a potentially important influence of surprise on consumer satisfaction. However, they do not provide definitive empirical support for the causal relationship between these two concepts. Furthermore, none of these studies investigated the impact of surprise on the evaluation of the attribute used for eliciting surprise. The purpose of the present pilot study is thus to investigate the causal relationship between surprise and satisfaction in an experiment and assess to what extent the evaluation of the surprising attribute is influenced by the surprise reaction.

Another objective of this experiment is to go beyond paper-pencil measures of emotions. The marketing literature has traditionally favored verbal reports of emotions (e.g., Edell and Burke 1987, Richins 1997) but these measures have numerous drawbacks [rationalization of emotions, introspection and retrospection problems, etc., see Derbaix and Poncin (1999)]. Some researchers have, therefore, sought to develop non-verbal scales of emotions (e.g. physiological measures, coding of facial expressions). Vanhamme (2000)Busing galvanic skin response (GSR), electromyography (EMG) and coding of facial expressions beside paper-pencil measuresBhas shown that measures of facial expressions provide the best results for the emotion of surprise in a consumption/purchase context. In the present study, both verbal reports of emotions and coding of behavioral aspects, such as facial expressions, will thus be included.

THE EMOTION OF SURPRISE

Most recent studies carried out on surprise (e.g., Meyer et al. 1997, Reisenzein 2000, Schutzwohl 1998) consider it to be a neutral and short-lived emotion. It is described as a syndrome of reactions, i.e. a specific pattern of reactions at the subjective (e.g. subjective feeling, surprise exclamation), physiological (e.g. changes in the respiration rates) and behavioral levels (e.g. specific facial expression [mainly raised eyebrows and opened eyes], interruption of ongoing activities, focusing of attention on the surprising stimulus) (e.g. Meyer et al. 1997; Reisenzein et al. 1996, Vanhamme 2000).

The emotion of surprise is elicited by either unexpected or misexpected elements (e.g., Ekman and Friesen 1975, Scherer1984) or, more precisely, a "schema discrepancy" (e.g. Meyer et al 1994, Schutzwohl 1998). A schema is a type of private, normally informal, inarticulate, unreflective theory about the nature of objects, events or situations (Rumelhart 1984). Individuals continuously check whether their schemas match the inputs coming from the surrounding environment. As soon as inputs diverge from the schema, surprise is elicited.

As a result of the evaluation of the pleasantness / unpleasantness of the experienceBwhich is subsequent to the evaluation of the schema discrepancy (Scherer 1984) -, the emotion of surprise is often followed by another emotion that colors it either positively (e.g. surprise + joy) or negatively (e.g. surprise + anger) (Ekman and Friesen 1975, Meyer et al. 1994). This explains why people talk about positive or pleasant surprise and negative or unpleasant surprise although surprise itself is neutral.

Finally, it has been found thatBthrough its intrinsic arousalBsurprise is an amplifier of subsequent affective reactions (Charlesworth 1969, Desai 1939). Thus, someone who experiences joy after having been surprised will feel more joyful than if he or she had not been surprised previously.

HYPOTHESES

A general theory about the influence of emotions on judgements (e.g. attitude, satisfaction evaluation, etc.) is the affect-as-information view (Schwarz, 1990)-and more specifically the how-do-I-feel-about-it? heuristic (Pham, 1998; Schwartz et Clore, 1996). In cases where people perceive their affective reactions as relevant to a judgement, the how-do-I-feel-about-it? heuristic posits that they use these reactions (emotions, mood, etc.) as a source of information for making their judgement with an object (Pham, 1998; Schwarz et Clore, 1996). Positive affective reactions lead to a favorable evaluation. And, the intensity of these affective responses influences the strength of the judgement (Gorn, Pham et Sin, 2001; Pham, Cohen, Pracejus et Hughes, 2001). Due to the amplification property of surprise, the element that triggers off surprise (e.g. an attribute of the product, an element that is present during the consumption/purchase experience) will be given a "stronger", i.e. a more polarized evaluation than it would have been without surprise. Hence:

H1. In the context of a consumption/purchase experience, a product attribute that is positively evaluated will be evaluated even more positively if it is surprising.

Furthermore, surprise could also have an impact on product satisfaction. Along with recent definitions (e.g., Aurier and Evrard 1998), satisfaction is defined in this paper as a psychological (i.e. cognitive and affective) state, which results from a buying and/or consumption experience (i.e. transaction specific). Among the antecedents of satisfaction, "disconfirmation" is usually the variable that accounts for the largest part of its variance. Some studies, however, have shown that positive and negative emotions considerably add to the explanatory power of the satisfaction model (e.g., Oliver 1993). However, the specific influence of surprise has not been analyzed before.

The influence of surprise on product satisfaction could occur through a response contagion effect (Vanhamme and Snelders 2001) and would lie in the arousal that is part of the emotion of surprise. The intrinsic arousal of surprise would amplify subsequent emotons elicited by the consumption/purchase experience, such as joy, and these, in turn, would enhance the satisfaction level of consumers (=indirect influence of surprise on satisfaction). Surprise might also amplify satisfaction directly since satisfaction is partly affective and surprise is theoretically able to amplify any affective reaction. Moreover, Oliver (1997) suggests that the higher the level of satisfaction, the higher the intrinsic arousal contained in the satisfaction response. Therefore, since surprise has a high potential of arousal, a surprised customer should experience a level of satisfaction with a high degree of arousal, that is a high level of satisfaction. Surprising consumption/purchase experiences should thus give rise to more polarized evaluations of satisfaction than non surprising experiences. [Another possible explanation for the surprise-satisfaction relationship-and thus H3-relies on an accessibility effect (Vanhamme and Snelders, 2001). This effect is not discussed further here since our experiment was designed to elicit a response contagion effect: respondents knew-prior to performing the experimental tasks-that they would have to evaluate their consumption experience and these evaluations were made immediately after consumption.] Hence:

H2. Consumers who are (pleasantly) surprised during their consumption/purchase experience with a product will have higher levels of satisfaction than consumers who have not been surprised during their consumption/experience with the same product.

H3. For surprising successful buying/consumption experiences, surprise has a positive direct influence on satisfaction and a positive indirect influenceBthrough the amplification of subsequent positive emotions. Positive emotions will thus not fully mediate the influence of surprise on satisfaction.

As mentioned above, disconfirmation is often the variable that accounts for the largest part of explained variance in satisfaction (Churchill and Surprenant 1982). Westbrook (1987) showed, however, that the relationship between emotions and satisfaction was not mediated by disconfirmation. Disconfirmation (i.e. a cognition that summarizes the recognition that the performance of a product/service is better or worse than expected) is therefore also unlikely to encompass all aspects of the emotion of surprise that may influence satisfaction. Hence:

H4. The influence of surprise on satisfaction will remain significant once the influence of disconfirmation is taken into account.

METHODOLOGY

Design, procedure and manipulation.

The experimental design is a "posttest only control group design" including one experimental group and two control groups. The products used during the experiment were strawberry yogurts wrapped in an individual package. Participants were recruited among the members of a non-student panel in a European town in order to enhance external validity. Their age and gender distribution was as follows: 24 percent between 20 and 31, 54 percent between 32 and 43, and 22 percent between 44 and 55; 41 percent male and 59 percent female. Respondents were matched according to sex, age and the frequency of consumption of strawberry yogurt, and then randomly assigned to the three conditions. They were paid a small amount for participation. Participants had to perform four tasks during the experiment; the manipulation of surprise occurredBin the experimental groupBduring the third task. During this task participants had to taste three jars of yogurt. Surprise was elicited by the discovery of a foldable plastic spoon hidden under the wrapping of the third yogurt jar the participants had to eat during the third task. This foldable spoon was a novelty, and its presence was supposed to be a pleasant surprise. The participants of the experimental condition (=Surprise/Spoon condition) did not know that they would discover such a foldable plastic spoon in the wrapping of the jar of yogurt (the two previous yogurts did no contain a foldable spoon and the existence of such a spoon was not mentioned at any time). In one of the control conditions, the foldable spoon was presented as part of the product from the start and there was a foldable spoon in all strawberry yogurtsBas announced (=No Surprise/Spoon condition). In the other control condition, there was no mentioning of a foldable spoon and such a spoon never appeared in any of the yogurts (=No Surprise/No Spoon condition). Thus, nothing was surprising for the participants of the control groups. Pre-testsBcarried out on participants from the same population prior to the experiment in order to check the manipulation and the materialBconfirmed that the foldable spoon’s presence or absence (depending on the condition) never elicited surprise in the two control conditions whereas the presence of the foldable spoon elicited surprise in the experimental group.

When each participant arrived in the laboratory, the experimenter showed him/her a demonstration tape describing and illustrating the experimental tasks to be performed. Beside the description of the experimental task, the demonstration tape’s purpose was also to create the participants’ schema about the product characteristics. This videotape contained a clear visual and verbal description of the product that would be used in the experiment (i.e. a strawberry yogurt with the foldable spoon in the No Surprise/ Spoon condition and without it in the two remaining conditions). Thus, this videotape secured that the proper schema would be activated during the experiment (this was confirmed during pre-tests).

During the first task, participants were asked to test and rank different plastic spoons by order of preference. To do so, they were given a strawberry yogurt identical to the ones they would have to taste during the third task; the jar was still in its packaging and included a foldable spoon in the No Surprise/Spoon condition. The purpose of this task was 1) to make sure that the participants would not be surprisedBduring the third taskBby other elements (e.g. the taste of the yogurt) than the discovery of the foldable spoon and 2) to reinforce the schema of the product. The second task was a questionnaire with items about involvement with the product-category, mood and intro-/extraversion (sex, age and frequency of strawberry yogurt consumption had been recorded prior to the experiment). The last task was to answer the final questionnaire including a manipulation check section and scales about emotions, satisfaction, disconfirmation and appreciation of the foldable spoon. The participants were left alone while performing the tasks. The whole task lasted for a maximum of 45 minutes, including the debriefing.

Measures.

The Differential Emotion Scale of Izard (1977) was used to capture the subjective experience of surprise and of joy (i.e. verbal reports on a 5 point scale of intensity). Facial expressions were video recorded and coded independently by three judges according to the procedure developed by Ekman and Friesen (1975) and adapted by Reisenzein (2000). Other observable aspects pertaining to the emotion of surprise were also coded (i.e. focus of attention, interruption of ongoing activities, surprise exclamation). Beside the verbal and non verbal measures of emotions, a 3 items (10 point scale) measure of the appreciation/liking of the foldable spoon as well as a multi-items scale of disconfirmation (Oliver 1997) and three scales of satisfaction were administered (the Oliver (1997) scale, the mono-item Delighted-Terrible scale (Westbrook 1980) and the SATEXP scale of Aurier and Evrard (1998), see appendix 1). Control measures of mood (Peterson and Sauber (1983) scale), extraversion/introversion (Saucier (1994) scale) and involvement with the product-category (Jain and Srinivasan, 1990) were also collected.

SUMMARY OF MAIN RESULTS

ResultsBManipulation and design checks.

Thirty-one valid questionnaires were collected for the No Surprise/Spoon and Surprise/Spoon conditions and 34 for the No Surprise/No Spoon condition (a dozen questionnaires had to be discarded due to either "technical" problems, obvious demand effects, problems in the manipulation (e.g. absence of surprise in the experimental group or onset of surprise in a control group), or not following the instructions. None of the remaining 65 participants from the two control groups declared having experienced surprise whereas all remaining 31 participants in the experimental group were surprised (this surprise was pleasant: mean 7/10 on the scale anchored by "very negative (1)" to "very positive (10)"). All but 4 participants in the experimental group displayed at least one observable characteristic of surprise and no participant from the control conditions displayed characteristics of surprise. Highly significant differences were found between the experimental group and the control conditions for the measures of surprise (see also table 1, SUR). Corroborating the work by Reisenzein (2000) and Vanhamme (2000), surprise was most often indicatedBin the experimental groupBby raised eyebrows (35% of the respondents), then opened eyes (13%); opened mouth did not happen very often (6%). Interruption of ongoing activities and focus of the attention/exploratory behavior were observed for, respectively, 61% and 68% of the respondents in the experimental group (a few respondents (10%) had a spontaneous surprise exclamation).

Scales were assessed for uni-dimensionality using Principle Components Analysis. Items were then aggregated and the distributions of each variable checked for normality in each group (the normality assumption did not hold for verbal surprise for the No Surprise/No Spoon condition, verbal negative emotions for all three conditions and all non verbal measuresBexcept the overall measure of observed surprise (OBSS)Bfor the Surprise/Spoon condition; non parametric statistics will thus be used for those variables). Reliability analysis was also performed on all verbal scales. All Cronbach alpha were above .7 except for three of the 5 dimensions of involvement and intro-/extraversion. All inter-judge agreements for facial coding and other observed measured were very high (Rust and Cooil (1994) PRL index>.9). Internal validity of the results was ensured through the design and matching of the participants. As expected no difference was found between the experimental group and the two control groups with respect to the control measures (it was also checked that the correlations between these measures and the other variables did not affect the results for the hypotheses).

Results for H1.

As expected the results support H1. In the Surprise/Spoon condition, the liking of the element used for eliciting surprise, i.e. the foldable spoon, is significantly higher than in the No Surprise/ Spoon condition, where the same foldable spoon was used (table 1).

Results for H2.

Contrary to what was expected, none of the scores of satisfaction are significantly higher for the experimental condition than for the two control conditions (table 1). However, it should be noticed that only 5 participants in the experimental condition used the foldable spoon. The others used another spoon (reasons: they thought they were not allowed to use the foldable spoon for the taste test and had to use the 'official spoon’ depicted on the demonstration videotapes, there were plenty of other spoons on the table so they did not need this one, they had already opened another spoon prior to discovering the foldable spoon, etc.). The 5 participants who used the foldable spoon did have higher satisfaction scores than the others in the experimental condition, except for DT (SAT: 6.7 vs. 5.5; SATEXP: 6.9 vs. 6.0; SATGLO: 6.7 vs. 5.7; DT: 4.6 vs. 4.6). However, these differences are not statistically significant and the small sample sizes do not allow for valid inferences. These scores are also higher than those of the control groups (table 1).

Results H3 and H4.

The PLS approach for modeling structural equations, mode A of estimation, [As indicators were reflective, mode A of estimation was used. Note also that Tenenhaus (1998) advised to use mode A systematically because of multicolinearity problems occurring with mode B. Finally, Dijkstra (1981) also recommends mode A when there are a lot of indicators compared to the number of observations (which is the case in this study.] (Wold 1980) was used to test H3 and H4. This method is a good alternative to the maximum likelihood method of estimation (e.g., LISREL) when its underlying assumptions are not respected (e.g. too small sample size, non multi-normality) (see Fornell et Bookstein (1982) for a comparison of the two methods). Furthermore, mode A of estimation is free of multi-colinearity problems, which is an advantage as the indicators of a latent variable can be highly correlated.

The model including all possible indicators for each of the latent variables in the experimental group (i.e. joy, surprise, disconfirmation and satisfaction) is represented in figure 1 (upper part). This model has a good predictive validity (i.e. significant Q2 and R2 for satisfaction, Fornell & Bookstein, 1982; Tenenhaus, 1998; Valette-Florence, 1988; see upper table on the right side of figure 1) and a good convergent validity for joy (0.65>0.5) and satisfaction (0.83>0.5) (Valette-Florence, 1988). For surprise, convergent validity is not good enough (0.41<0.5). The model has also a good discriminant validity (0.36<0.41) (Valette-Florence, 1988). Due to the convergent validity problem for surprise, the model has been re-estimated using only the indicators of surprise that were sufficiently correlated with the estimates of the latent variable as advised by Wold (1980). Results are provided in figure 1 (lower part). This model is very close to the first model and respects all criteria of predictive (significant Q2 and R2), convergent (.66<.5, .65<.5 and .83<.5) and discriminant (.32<.65) validity.

TABLE 1

SCORE COMPARISON BETWEEN EXPERIMENTAL AND CONTROL GROUPS

H3 is only partially supported by the data in that there seems to be only a direct influence of surprise on satisfaction (see figure 1, lower part). Surprise does amplify satisfaction and joy since path coefficients [and correlations] between the latent variables of surprise and satisfaction and between the latent variables of surprise and joy are significant. But the latent variable of joy has no impact on satisfaction (path coefficient and correlation between the latent variables of joy and satisfaction are not significant).

Furthermore, as expected (H4), the direct influence of surprise on satisfaction remains significant despite the presence of disconfirmation in the model.

DISCUSSION AND CONCLUSIONS

This pilot experiment clearly shows that surprise has an effect on the evaluation of the surprising product attribute. Consumers evaluated more positively the same product attribute (i.e. the foldable spoon) in the surprise condition than in the non-surprise condition. The results further show that for successful consumption/buying experiences, pleasantly surprised consumers did not report higher satisfaction scores than non-surprised consumers. Nonetheless, most of them did not use the spoon for reasons related to the context of the experiment, and those who did use the foldable spoon, did have higher, although non-significant, scores. It is therefore possible that H2 would have been supported if more consumers had used the spoon. Replication of this study should therefore check this issue.

The fact that significantly higher scores emerged for the evaluation of the foldable spoon (the surprising product attribute itself) and not for the scores of satisfaction with the whole product could be explained by the relative relevance of the emotion of surprise for the satisfaction judgment. On the on hand, surprise might have been less relevant for the satisfaction evaluation with the whole consumption experience than for the evaluation of the spoon (yogurt evaluation was probably largely done on the basis of taste). As a result, this emotion might have been granted a place large enough to make a significant increase only in the case of the spoon evaluation. This explanation is consistent with the How-do-I-feel-about-it? heuristic. On the other hand, the higher satisfaction scores of the 5 respondents who used the foldable spoon to eat the yogurt might mean that usage (i.e. making a connection between the surprising attribute and the product) is sufficient for making the emotion of surprise relevant to the satisfaction evaluation of the consumption experience with the product (although respondents who used the spoon may have done so because they considered it an important attribute).

Although people in the surprise condition did not have higher levels of satisfaction than people in the two control conditions, within the surprise condition, variation in the level of surprise (elicited by the product attribute) was found to have a direct effect on satisfaction. The direct effect was the sole effect since surprise did not influence satisfaction indirectly (through its amplification of the emotion of joy). Since 1) the study by Vanhamme (2001) Bon a large set of products/servicesBreveals both a direct and indirect relationship (correlation) between surprise and satisfaction and 2) studies on satisfaction have shown that models of satisfaction formation can vary according to the type of product/service (e.g. Churchill and Surprenant 1982), this experiment should be replicated with other products/services. It could then be checked whether the type of relationship (direct and/or indirect) varies according to types of products/services.

FIGURE 1

PLS APPROACH FOR MODELING STRUCTURAL EQUATIONS WITH LATENT VARIABLES (MODE A)

Surprising customers is not free of costs; companies need to invest time in understanding their customers and in creating surprising aspects inBor aroundBthe product/service delivered. It is worth noting that it can be difficultBfor mass product consumptionBto secure that all consumers will be surprised with the same aspect because surprise depends on the consumers’ individual schemas. In one-to-one relationships with the customers, companies have a better knowledge of their customers’ schemas and are therefore more likely to succeed in surprising them.

APPENDIX 1

REFERENCES

Anderson, E. W. and Fornell, C. (1994), "Customer satisfaction prospectus", in Service Quality: New Directions in Theory and Practice, eds. Rust R.T. and Oliver R.L., London: Sage, 241-268.

Aurier, Ph. and Evrard, Y. (1998), "Elaboration et Validation d’une Echelle de Mesure de la Satisfaction des Consommateurs", in Actes de la ConfTrence de l’Association Frantaise du Marketing, 14, 1, Saporta, B. and Trinquecoste, J. F, Bordeaux: IAE, 51-71.

Cohen, J.B and Areni, C.S. (1991), *Affect and Consumer Behavior+, in Handbook of consumer theory and research, eds. Robertson, T.S. and Kassrjian, HH, Englewood Cliffs, NJ: Prentice Hall, 188-240.

Charlesworth, W.R. (1969), "The Role of Surprise in Cognitive Development", in Studies in cognitive development, essays in honor of Jean Piaget, eds. Elkind, D. and Flavel, J.H., London: Oxford University Press, 257-314.

Churchill, G. A. and Surprenant, C. (1982), "An Investigation into the Determinants of Customer Satisfaction", Journal of Marketing Research, 19 (November), 491-504.

Desai, M. M. (1939), "Surprise: A Historical and Experimental Study", British Journal of Psychology, Monogr. Suppl., 22, 1-124.

Derbaix, C. and Poncin, I. (1999), "Process Tracing of Affective Reactions elicited by Commercials: Assessing the Explanatory Power of the Outputs of the Feelings Monitor and of the Coding of Facial Expressions", in European Advances in Consumer Research, Volume 4, eds. Dubois, B., Lowry, T. and Shrum, L., 166-173.

Dijkstra, T. K. (1981), Latent Variables in Linear Stochastic Models: Reflections on "Maximum Likelihood" and "Partial Least Squares" Methods, Ph.D. Dissertation, Rijksuniversiteit te Groningen.

Edell, J.A. and Burke, M.C. (1987), "The Power of Feelings in Understanding Advertising Effects", Journal of Consumer Research, 14 (3), 421-433.

Ekman, P. and Friesen, W.V. (1975), Unmasking the face, New-Jersey, Prentice Hall.

Fournier, S. et Mick, D.G. (1999), "Rediscovering satisfaction", Journal of Marketing, 63 (October), 5-23.

Fornell, C. and Bookstein, F.L. (1982), "Two Structural Equation Models: LISREL and PLS applied to Consumer Exit-voice Theory", Journal of Marketing Research, 9 (November), 440-452.

Gorn, G., Pham, M. T. and Sin, L. Y. (2001), "When Arousal influences Ad Evaluation and Valence Does Not (and Vice Versa)", Journal of Consumer Psychology, 11(1), 43-55.

Izard, C.E. (1977), Human Emotions, New-York: Plenum Press.

Jain, K. and Srinivasan, N. (1990), "An Empirical Assessment of Multiple Operationalizations of Involvement", Advances in Consumer Research, 17, 594-602.

Mano, H. and Oliver, R.L. (1993), "Assessing the Dimensionality of Consumption Experience: Evaluation, Feeling, and Satisfaction", Journal of Consumer Research, 20 (3), 451-466.

Meyer, W.-U, Niepel, M., Rudolph, U., and Schntzw√∑hl, A. (1994), "Uberraschung und Attribution", in Attributionstheorie, eds. FOrsterling, F., and Stiensmeier-Pelster, J., G√∑ttingen: Hofgrefe, 105-121.

Meyer, W-U., Reisenzein, R., and Schutzwohl, A. (1997), "Toward a Process Analysis of Emotions: The Case of Surprise, Motivation and Emotion, 21(3), 251-274.

Ohania (1990), "Scale of Expertise", in Handbook of Marketing Scales: Multi-item Measures for Marketing and Consumer Behavior Research, eds. Bearden, W. O. and Netemeyer, R. G. (1999), Thousand Oaks: Sage.

Oliver, R. L. (1993), "Cognitive, Affective and Attributes Bases of the Satisfaction Response" Journal of Consumer Research, 20 (3), 418-430.

Oliver, R.L. (1997), Satisfaction: on a Behavioral Perspective on the Consumer, N-Y: McGraw-Hill.

Oliver, R.L., Rust, R. T., and Varki, S. (1997), "Customer Delight: foundations, findings, and managerial insight. Journal of Retailing, 73 (3), 311-336.

Oliver, R. L. and Westbrook, R. A. (1993), "Profiles of Consumer Emotions and Satisfaction in Ownership and Usage", Journal of Consumer Satisfaction, Dissatisfaction, and Complaining Behavior, 6, 12-27.

Peterson, R. A. and Sauber, M. (1983), "Mood Short Form: MSF", in Handbook of Marketing Scales: Multi-item Measures for Marketing and Consumer Behavior Research, eds. Bearden, W.O. and Netemeyer, R. G. (1999), Thousand Oaks: Sage.

Pham, M. T. (1998), "Representativeness, Relevance, and the Use of Feelings in Decision Making", Journal of Consumer Research, 25, 2, 144-159.

Pham, M. T., Cohen, J.B., Pracejus, J.W. and Hughes, G.D. (2001), "Affect Monitoring and the Primacy of Feelings in Judgment", Journal of Consumer Research, 28 (2), 167-188.

Pine, J.B. and Gilmore, J.H. (1999). The Experience Economy: Work is Theatre and Every Business a Stage, Boston: Harvard business school press.

Plutchik, R. (190), Emotion: a Psychoevolutionary Synthesis, New-York: Harper and Row.

Reisenzein, R. (2000), "Exploring the Strength of Association between the Components of Emotion Syndrome: the Case of Surprise", Cognition and Emotion, 14(1), 1-38.

Reisenzein, R., Meyer, W-U., and Schutzwohl, A. (1996), "Reactions to Surprising Events: a Paradigm for Emotion Research", in Proceedings of the 9th Conference of the International Society for Research on Emotions, Frijda, N. H., Toronto: ISRE, 292-296.

Richins, M.L. (1997), "Measuring Emotions in the Consumption Experience", Journal of Consumer Research, 24, 127-146.

Rumelhart, D.E. (1984). "Schemata and the Cognitive System", in Handbook of Social Cognition 1, eds. Wyer, R.S. Jr. and Srull, T.K., Hillsdale, NJ: Lawrence Erlbaum Associates Inc, 161-188.

Rust, R.T. and Cooil, B. (1994), "Reliability Measures for Qualitative Data: Theory and Implications", Journal of Marketing Research, 31 (February), 1-14.

Rust, R.T. and Oliver, R.L. (2000), "Should we delight the customer ?", Journal of The Academy of Marketing Science, 28 (1), 86-94.

Rust, R. T., Zahorik, A. J. and Keiningham, T. L. (1996), Service marketing, New York, Harper Collins.

Saucier, G. (1994), "Mini-markers: a Brief Version of Goldberg’s Unipolar Big Five Markers", Journal of Personality Assessment, 63, 506-516.

Scherer, K.R. (1984), "On the Nature and Function of Emotion: a Component Process Approach", in Approaches to emotion, eds. Scherer, K.R. and Ekman, P., Hillsdale, NJ: Lawrance Erlbaum Associates Inc., 293-317.

Schutzwohl, A. (1998), "Surprise and Schema Strength", Journal of Experimental Psychology, Learning, Memory and Cognition, 24(5), 1182-1199.

Schwarz, N. (1990), "Feelings as Information: Informational and Motivational Functions of Affective States", Handbook of Motivation and Cognition: Foundations of Social Behaviour, 2, eds. Higgins, E. T. and Sorentino, R., New-York: Guilford, 527-561.

Schwarz, N. et Clore, G. L. (1996), "Feelings and Phenomenal Experiences", in Social Psychology: Handbook of Basic Principles, eds. Higgins, E. T. and Kruglanski, A. W., New York: Guilford press, 433-465.

Tenenhaus, M. (1998), La Regression PLS: ThTorie et Pratique, Paris: Editions Technip.

Valette-Florence, P. (1988), Analyse Structurelle Comparative des Composantes des Systemes de Valeurs selon Kahle et Rokeach, Recherche et Applications en Marketing, 3 (1), 15-34.

Vanhamme, J. (2000), "The Link Between Surprise and Satisfaction: an Exploratory Research on How Best to Measure Surprise", Journal of Marketing Management, 16, 565-582.

Vanhamme, J. (2001), L’Influence de la Surprise sur la Satisfaction des Consommateurs: Etude Exploratoire par Journal de Bord, Recherche et Applications en Marketing, 16(2),1-32.

Vanhamme, J. and Snelders, D. (2001). The role of surprise in satisfaction judgements? Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 14, 27-45.

Westbrook, R.A. (1980), "A Rating Scale for Measuring Product/Service Satisfaction, Journal of Marketing, 44, 68-72.

Westbrook, R.A. (1987), "Product/consumption-based Affective Responses and Postpurchase Process", Journal of Marketing Research, 24, 258-270.

Westbrook, R. A. and Oliver, R. L. (1991), "The Dimensionality of Consumption Emotion Patterns and Consumer Satisfaction, Journal of Consumer Research, 18, 1, 84-91.

Wold, H. (1980), "Model Construction and Evaluation When Theoretical Knowledge is Scarce: Theory and Applications of Partial Least Squares", in Evaluation of econometric models, eds. Kmenta, J. and Ramsey, J.B, New York: Academic Press, 47-74.

----------------------------------------