Advances in Consumer Research Volume 17, 1990 Pages 41-47
AN EXPLORATORY STUDY OF THE EFFECTS OF IMAGERY PROCESSING AND CONSUMER EXPERIENCE ON EXPECTATIONS AND SATISFACTION
Deborah J. MacInnis, University of Arizona
Linda L. Price, University of Colorado
Research was conducted to investigate whether imagery processing and consumer experience interact to affect consumer expectations and satisfaction. and to explore some individual and situational antecedents of naturally occurring imagery processing. Results revealed the use of elaborated imagery is a prevalent part in the anticipation of vacation activities, particularly for consumers who plan to invest more time and money. As predicted, imagery processing and experience affected the positivity bias of expected outcomes. Confidence that spring vacation would turn out as expected was affected by experience, and the interaction of experience and imagery processing, but not imagery processing alone. Imagery processing and the interaction between experience and imagery processing were related to confirmation of expectations. Contrary to predictions, imagery processing had a positive effect on satisfaction and experience had a negative effect. Of all the groups, low experience individuals who engaged in imagery processing were most satisfied with their spring vacation. Finally, consumers who engaged in higher levels of elaborated imagery were more satisfied with their vacation outcomes, whether or not the event unfolded as imagined.
IMAGERY, PLANS, AND SATISFACTION
Recent theory has suggested that imagery processing may affect a wide range of outcome variables, including consumer expectations, affect, intentions and satisfaction (MacInnis and Price 1987). Despite the important implications of this processing on consumer behavior, almost no research has explored imagery effects on consumer intent (for an exception see Mitchell 1989), and no research has explored the relationship between imagery processing, consumer expectations and satisfaction.
The purpose of this research was to investigate naturally occurring imagery processing and explore its relationship to consumer expectations and satisfaction. Two questions were of substantial interest: 1) Do imagery processing and consumer experience interact to affect consumer expectations and satisfaction? 2) What are the individual and situational antecedents of naturally occurring imagery processing?
THE ROLE OF IMAGERY PROCESSING IN CONSUMER EXPECTATIONS
In their 1987 article, MacInnis and Price propose that elaborated imagery processing can have a powerful role in influencing consumer expectations and satisfaction. Specifically, imagery processing is hypothesized to lead to systematic biases in expectations, with attendant implications for satisfaction. MacInnis and Price (1987) argue that imagery processing will affect both the probability and the valence assigned to purchase outcomes. These biased expectations, in turn, may lead to greater discrepancies between actual and imagined outcomes culminating in higher levels of consumer dissatisfaction (Anderson 1973; Oliver and DeSarbo 1988).
Some research supports that imagining future events leads to a conjunctive bias because a probability is assigned to an entire or imagined scene rather than bet assigned to the individual set of elements that comprise the scene. This leads to an overestimation of the likelihood that a sequence of events will unfold as imagined (Einhorn and Hogarth 1984, Kahneman and Tversky 1982). Other research indicates that elaborated imagery not only biases probabilities assigned to contingent events, it also affects more directly consumers' expectations that imagined events will occur. The very act of creating a sensory analogue of the outcome (via imagery), makes the outcome appear more real and hence, more likely (Anderson 1983; Carroll 1978; Sherman, Cialdini, Schwartzman and Reynolds 1984). For example, Sherman et al (1984) found that students who imagined the symptoms of a given disease rated the likelihood that they would contract the disease as higher than students who did not engage in imagery.
As noted in MacInnis and Price (1987), elaborated imagery may also lead to a valence bias in imagining future events (see O'Neal 1974). Specifically, individuals are more likely to focus on positive imagined future outcomes than negative ones. Again, the sensory analogue properties of imagery processing are an important envoy for this bias. The fact that imagery provides a sensory analogue makes it natural for consumers to focus on outcomes that feel good.
The likelihood and valence biases associated with imagery processing have potentially negative implications for resultant consumption satisfaction. If consumers imagining purchase outcomes focus only on positive ones, and assess upwardly biased probabilities that the imagined event will unfold as imagined, there is a good chance that the actual purchase outcomes will fall short of the expectations formed through imagery. Disconfirmation of expectations has been noted in the satisfaction literature as a major factor influencing consumer dissatisfaction (Oliver and DeSarbo 1988; Tse and Wilton 1988). There is some suggestion that even favorable outcomes, if different than expected, may give rise to dissatisfaction (Anderson 1983). Thus, elaborated imagery, by biasing perceived likelihoods, contingent probabilities and the positivity of future events may reduce consumers' satisfaction with the experienced event.
THE ROLE OF CONSUMER EXPERIENCE
MacInnis and Price (1987) postulate that imagery processing and consumer experience and expertise in the product category may interact to affect purchase satisfaction. In the case of discursive processing, product experience and knowledge are related to important differences in memory and cognitive structure that affect information search and processing (Alba and Hutchinson 1987). However, almost nothing is known about the effect of consumer experience on imagery processing. MacInnis and Price (1987) speculate that consumers who are product novices are likely to have a relatively modest store of information in memory from which to build scenarios of purchase outcomes. Because knowledge stores are impoverished, individuals with little or no product experience may imagine relatively few purchase outcomes. Moreover, given the positivity bias expected with imagery processing, novice consumers are likely to imagine these few scenarios as being positive in nature. Finally, because imagery processing heightens expectations that the positive events constructed from the limited storehouse will actually unfold as planned (a likelihood that is actually quite slim), there is considerable opportunity for dissatisfaction on the part of the novice consumer.
In contrast, MacInnis and Price (1987) argue that a different set of conditions may operate for more experienced consumers. Because of their richer knowledge base, experienced consumers may imagine the anticipated event as unfolding in many ways. Moreover, although experienced consumers too are likely to favor positive outcomes in their imagining, their past experiences may prompt them to consider negative outcomes as well. By imagining many different outcomes, and incorporating both favorable and unfavorable outcomes, their expectations for any one event occurring are likely to be lower than those for novices. Consequently, for experienced as compared to inexperienced consumers, there is a greater probability that at least one of the imagined
Examining the role of imagery processing and experience on consumer expectations and satisfaction suggests a main effect of imagery processing, and an interaction of imagery processing and consumer experience on a positivity bias in expected outcomes, confidence in expectations, expectancy-disconfirmation and satisfaction. Specifically, based on extant theory, imagery processing when combined with low levels of experience should result in more positivity bias, more expectancy-disconfirmation, and less satisfaction. In contrast, imagery processing when combined with high levels of experience should result in less of a positivity bias, less expectancy-disconfirmation and more satisfaction.
ANTECEDENTS OF IMAGERY PROCESSING
The limited consumer research exploring imagery processing has generally manipulated task stimuli to inducemagery processing (see discussion MacInnis and Price 1987). This provides valuable insights into the separate effects of discursive and imagery processing, and the ability of specific variables (such as pictures, instructions to imagine, etc.) to induce imagery processing. However, such research does riot provide exploratory insights into naturally occurring imagery processing and its antecedents.
Some research has supported individual differences in preferences for imagery processing and the use of imagery in everyday life (Childers, Houston and Heckler 1985; Rossiter and Percy 1978; Singer and Antrobus 1972). For example, Singer and Antrobus (1972) suggest that individuals may differ in both the level and content of their imagery. Thus, individuals will differ in their tendency to engage in vivid imagery, use imagery in anticipating the future and solving problems, and/or use imagery to fantasize.
Several characteristics of the consumption experience may also motivate imagery processing. Although the relationship of preconsumption imagery to hedonic consumption activities has not been explored, some research has suggested substantial levels of imagery when engaged in hedonic consumption activities (Hilgard 1978). Playful consumption has been proposed as a domain where imagery appears to play a particularly important value-enhancing role (Holbrook, Chestnut, Oliva, and Greenleaf 1984). In addition, stimuli high on perceived novelty may stimulate imaging of applications and outcomes (Oliver, Robertson and Mitchell 1989). Finally, large expenditures for events of limited duration (such as weddings, proms, and special vacations) may invite preconsumption imagery as a way of enhancing their valuation beyond the typically conceived duration of the event (MacInnis and Price 1987).
Research was conducted to investigate whether imagery processing and consumer experience interact to affect consumer expectations and satisfaction in a hedonic consumption context, and to explore some individual and situational antecedents of naturally occurring imagery processing. One hundred ninety three undergraduate students from a large western university responded to two questionnaires about spring break. The sample was comprised almost equally of men (51%) and women (49%), with the majority (67%) between the ages of 21 and 24. Twenty three percent were between 18 and 20, and about 10% were over 24. Most of the respondents spent less than $300 on spring break (78%), spent spring break with friends (66%), and considered it to last 6 days or more (62%).
CORRELATION AMONG USE OF IMAGERY MEASURES
Prior to spring break students were asked to indicate what activities they expected to engage in over spring break, and what they hoped spring break would offer (i.e., rest and relaxation, adventure, etc.). After spring break, these same students were again asked to recall what activities they planned to engage in, and also asked what activities they did engage in. A content analysis of the planned activities between the first and second questionnaires revealed very high consistency in reports of expected activities before and after spring break.
The results reported in this study focus on an analysis of the second questionnaire. This survey asked about the use of imagery prior to spring break, individual differences in imagery processing, experience with and expectations regarding the planned event, and spring break outcomes (i.e. whether it turned out as expected and satisfaction). Several descriptors of spring break were also included (i.e., amount of money spent, number of days away, activities engaged in). Finally, basic demographic information was collected.
Use of Imagery. The primary measure of imagery processing was a four item scale (alpha=.84, mean= 4.54) to assess the extent to which consumers generated imagery about spring break (unless otherwise indicated, means are reported on a 1 to 7 scale). For example, "When I thought about spring break, my thoughts included the sights, smells, and/or sounds of the activities I would be engaged in." A five item scale (alpha=.85, mean = 4.20 ) was designed to assess the extent to which they imagined many different scenarios. For example, "In my mind, I played out many different scenarios of what I was going to do." A single item measure of time spent imagining (i.e., "I spent considerable time imagining what I would do," mean= 4.2) was also included. Standard deviations on these measures ranged from 1.27 to 1.65, indicating substantial variation between consumers in their use of imagery prior to spring break. Relationships between-these measures of imagery use are provided in Table 1. As indicated, consumers who generate imagery also seem to spend a great deal of time imagining, and seem to imagine many different scenarios. Based on average levels of the primary measure of imagery processing, respondents were divided into a Low Imagery (n=99, mean=3.5) and High Imagery (n=94, mean=5.6 ) group.
Experience. A four item scale (alpha= .78, mean = 4.4) was used to assess whether consumers were familiar with the activities and events they would be engaged in over their spring break. For example, "My spring break plans included activities I've done many times before." The standard deviation was 1.6 suggesting considerable variability. Based on the average experience level, respondents were divided into a Low Experience (n=91, mean=2.9) and High Experience group (n=102, mean= 5.7). The low and high experience groups did not differ in imagery processing (4.6, 4.5), the extent to which they imagined multiple scenarios (3.8. 3.6) or time spent imagining (4.3, 4.1).
Consumer Expectations. The item "My images of what I would do included only positive experiences," (mean= 4.20) was included to assess whether consumers considered only positive events when imagining their spring break. Confidence in expectations was assessed by a single item "How confident were you that it would happen just as expected?" (mean= 5.4). The relationship between these two items was positive (.12, p <.05), but small.
Spring Break Outcomes. Single items measured the extent to which spring break happened as expected (mean= 5.0), and the extent to which it included many new experiences (mean= 4.0). In addition, a seven item scale (alpha= .88, mean= 4.6) asked how satisfied consumers were with the* spring break. For example, "Compared to other spring breaks, my spring break was not that good," (reverse scored). Standard deviations on these measures ranged from 1.39 to 1.78. The correlations between satisfaction with spring break and both the inclusion of many new experiences (.39) and it turning out as expected (.35) were positive and significant. The relationship between spring break turning out as expected and including new experiences was non-significant (.06).
MEANS FOR EXPECTATIONS, EXPECTATION-CONFIRMATION AND SATISFACTION BY EXPERIENCE AND IMAGERY PROCESSING
Individual Differences. The three scales developed to measure individual differences in imagery processing were based on modification of the Imaginal Process Inventory (Singer and Antrobus 1963, 1972; Huba, Singer, Aneshensel and Antrobus 1982). One was a seven item scale (alpha= .83, mean= 4.7) designed to assess consumers' tendency to engage in vivid imagery. For example, "Compared to most people I believe my dreams and fantasies are more vivid." A second individual difference measure was a seven item scale (alpha= .78, mean= 5.4) assessing consumers' use of imagery to plan future events/activities. For example, "I find my daydreams are useful for planning things in the near future." Finally, a five item scale (alpha= .56, mean= 4.2) assessed consumers' tendency to engage in fantasy imagery. For example, "Lots of times, even in the middle of my work, I'll find myself fantasizing about something." Standard deviations for these three individual difference measures ranged from .89 to 1.1, and correlations between the three measures ranged from .31 to .57.
The major purpose of this study was to test whether experience and imagery processing interact to affect a positivity bias in imagined outcomes, confidence in expectations, expectation-confirmation and satisfaction. A series of ANOVAs were performed to test the central hypothesis that experience and imagery processing interact. The results are given in Table 2. As predicted, imagery processing and experience affect the positivity bias of expected outcomes. Specifically, experience leads to less positivity bias (F= 5.3, p < .05) and imagery processing (F=9.0, p < .005) to more. However, contrary to predictions, the interaction effect is not significant.
Confidence that spring vacation would turn out as expected was affected by experience (F=3.8, p < .05), and the interaction of experience and imagery processing (F=6.6, p < .01), but not imagery processing alone. Specifically, consumers with low experience who generated very little imagery about the experience were significantly less confident that things would turn out as expected, and they were right. That is, Table 2 also reports a significant interaction between experience and imagery processing on whether the vacation turned out as expected (F=7.6, p < .01). Surprisingly, experience did not have a significant main effect on confirmed expectations, but imagery processing had a marginally significant main effect (F=3.9, p < .05). (The belief that imagining can make it so may have some truth to it.) Finally, as indicated in Table 2 there is a significant main effect of imagery processing (F=15.2, p < .001) and experience (F=18.4, p < .001) on satisfaction. In addition, the interaction of imagery and experience is marginally significant (F=3.58, p < .06). However, contrary to expectations, imagery processing had a positive effect on satisfaction and experience had a negative effect. Thus, of all the groups, low experience individuals who engaged in imagery processing were most satisfied with their spring vacation.
PREDICTORS OF IMAGERY PROCESSING
One possible explanation for this finding is that, at least in this context, new experiences contribute to satisfaction and low experience individuals engaged in many more new experiences than high experience individuals. Results are consistent with this explanation. New experiences contribute to satisfaction (F=6.6, p<.01), but new experiences do not interact with experience or imagery in affecting satisfaction. However, low experience individuals reported many more new experiences (5.0) than high experience individuals (3.4) (t= 6.3, d.f.= 191).
An interesting implication of the pattern of results on the relationship between imagery processing and satisfaction is that imagery enhances satisfaction whether the break conforms to expectations or not. An analysis of variance using imagery processing and outcomes as expected or not as the independent variables, and satisfaction as the dependent variable, indicated only main effects for imagery processing (F= 2.9, p < .001). The pattern of means revealed that consumers who engaged in imagery were more satisfied with the outcomes of their spring break (mean = 4.99) than consumers who did not engage in imagery (mean = 4.30), whether their break conformed to expectations or not.
A related purpose of this research was to explore the antecedents of imagery processing. Regression analyses were conducted to identify the predictors of imagery processing. The independent variables in the model included individual difference measures (use of anticipatory imagery, fantasy imagery, vividness of imagery, gender and age), and situational variables related to spring break (prior experience with planned trip, money spent, number of days away). The results are reported in Table 3. Surprisingly, individual differences in imagery had virtually no impact on imagery processing. The novelty of the stimuli, as captured by prior experience with the planned break, also had no impact on imagery processing. However, the amount of investment (both in terms of money and time) was a significant predictor of preconsumption imagery.
We deliberately chose a hedonic consumption activity to explore naturally occurring preconsumption imagery. The present research suggests that consumers engage in imagery processing in anticipation of such activities. Moreover, individual difference variables seem to play a relatively small role in predicting imagery. Situational variables such as the duration of the vacation and money spent seem to be better predictors of imagery. The importance of these variables is consistent with the expectation that preconsumption imagery has value-enhancing properties. Such value-enhancement may be more important in the case of significant investments.
Results on the interaction of imagery and experience in affecting expectations, expectation-disconfirmation and satisfaction were interesting, but surprising. Although, as expected, imagery processing leads to a positivity bias, and leads to greater confidence that events will unfold as expected, it does not lead to greater expectancy disconfirmation, and leads to higher (rather than lower) levels of satisfaction. What's especially provocative is that imagery has a direct effect on satisfaction, whether or not the event unfolds as imagined. This finding supports the argument that a satisfaction model based on extent of disconfirmation of a y e-experience comparison standard is too limited to capture a diversity of consumption experiences (Tse and Wilton 1988). Specifically, when learning, adventure, and novelty motives are important, a consumer may be satisfied regardless of the levels of pre-experience comparison standard and disconfirmation (Cohen and Houston 1972, Price and Fisher 1989).
Especially for low experience individuals, imagery leads to higher levels of expectancy confirmation and satisfaction. That is, despite a positivity bias, imagery appears to help low experience individuals predict what to expect. Low experience individuals were able to generate just as many scenarios as high experience individuals, and use of imagery, and time spent imagining did not differ between the two groups. Thus, contrary to predictions, imagery processing seems to be a positive decision tool for low experience consumers, helping them to play out various outcomes.
In general, experience had an unexpected relationship to satisfaction that is perhaps partially a function of the context. As expected, experience lead to less positivity bias. In addition, experience increased confidence that spring vacation would turn out as expected, although it did not lead to more expectancy confirmation. Moreover, experience contributed to lower levels of satisfaction. Perhaps because spring break is viewed as an adventure, and new experiences are highly prized, the expected relationship between experience and satisfaction was not supported.
The reported study is obviously exploratory and is perhaps most useful when viewed as an invitation to engage in further research. The study explored only one imagery domain-- one that, students would be quick to offer, invites the imagination. Although the design incorporated pre-spring break and post-spring break measures, the results of the reported study relied on retrospective reporting. Finally, measures were developed for this study. Inadequate measurement of individual differences may partially account for their poor predictive ability.
Alba, Joseph W. and J. Wesley Hutchinson (1987), "Dimensions of Consumer Expertise," Journal of Consumer Research, 13, (March), 411-454.
Anderson, Craig A. (1983), "Imagination and Expectation: The Effect of Imagining Behavioral Scripts on Personal Intentions," Journal of Personality and Social Psychology, 45 (2), 293305.
Anderson, Rolph E. (1973), "Consumer Dissatisfaction: The Effect of Disconfirmed Expectancy on Perceived Product Importance." Journal of Marketing Research, 10 (February), 38-44.
Carroll, John (1978), "The Effect of Imagining an Event on Expectations for the Event: An Interpretation in Terms of the Availability Heuristic," Journal of Experimental Social Psychology, 14, 88-96.
Childers, Terry L., Michael J. Houston and Susan E.Heckler (1985), "Measurement of Individual Differences in Visual Versus Verbal Information Processing," Journal of Consumer Research, 12 (September), 125- 134.
Cohen, Joel B. and Michael J. Houston (1972), "Cognitive Consequences of Brand Loyalty," Journal of Marketing Research, 9 (February), 9799.
Einhorn, Hillel J. and Robin M. Hogarth (1984), "Probable Cause: A Decision Making Framework," unpublished paper, University of Chicago, Graduate School of Business, Center for Decision Research.
Hilgard, Josephine R. (1978), Hypnosis: A Study of Imaginative Involvement. Chicago, IL: University of Chicago Press.
Holbrook, Morris B., Robert W. Chestnut, Terence A. Oliva, and Eric A. Greenleaf (1984), "Play as a Consumption Experience: The Roles of Emotions, Performance, and Personality in the Enjoyment-of Games," Journal of Consumer Research, 11 (September), 728-739.
Kahneman, Daniel R. and Amos Tversky (1982), 'The Simulation Heuristic," in Judgements Under Uncertainty: Heuristics and Biases, eds. Daniel R. Kahneman, Paul Slovic, and Amos Tversky, New York: Cambridge University Press, 201208.
MacInnis, Deborah J. and Linda L. Price (1987), 'The Role of Imagery in Information Processing: Review and Extensions," Journal of Consumer Research, 13 (March), 473491.
Mitchell, Deborah J. (1989), "Mental Simulation in Consumer Choice," Working Paper, Department of Marketing, The Wharton School, University of Pennsylvania.
Oliver, Richard L. and Wayne S. DeSarbo (1988), "Response Determinants in Satisfaction Judgments," Journal of Consumer Research, 14 (March), 495-507.
Oliver, Richard L., Thomas S. Robertson and Deborah 1. Mitchell (1989), "Imaging/Analyzing In Response to Innovative Stimuli," Working Paper, Department of Marketing, The Wharton School, University of Pennsylvania.
O'Neal, Edgar (1974), "Influence of Future Choice Importance and Arousal Upon the Halo Effect," in Thought and Feeling, eds. Harvey London and Richard Nisbett, Chicago, Ill: Aldine, 2743.
Price, Linda L. and Robert J. Fisher (1989), "International Tourism as a Service Encounter: The Role of Consumer Expectations in Affecting Service Encounter Outcomes," Working Paper, Department of Marketing, University of Colorado.
Rossiter, John R. (1982) and Larry Percy (1978), "Visual Imagining Ability as a Mediator of Advertising Response," in Advances in Consumer Research, Vol. 5, ed. H. Keith Hunt, Ann Arbor, MI: Association for Consumer Research, 621629.
Sherman, Steven J., R. B. Cialdini, D. F. Schwartzman and K. D. Reynolds (1984), "Imagining Can Heighten Or Lower the Perceived Likelihood of Contracting a Disease," unpublished working paper, Arizona State University.
Singer, Jerome and John Antrobus (1972), "Daydreaming, Imaginal Processes and Personality: A Normative Study," in The Function and Nature of Imagery, ed. Peter Sheehan, New York: Academic Press, 174-232.
Tse, David K. and Peter C. Wilton (1988), "Models of Consumer Satisfaction Formation: An Extension," Journal of Marketing Research, 25 (May), 204-12.