Chair: Chadwick J. Miller, Arizona State University, USA Paper #1: The Heart and the Head: On Choosing Experiences
Intuitively and Possessions Deliberatively
Iñigo Gallo, IESE Business School, Barcelona, Spain Sanjay Sood, University of California, USA Thomas Mann, Cornell University, USA Thomas Gilovich, Cornell University, USA
Paper #2: To Do or To Have, Now or Later? The Preferred Consumption Profiles of Material and Experiential Purchases
Amit Kumar, Cornell University, USA Thomas Gilovich, Cornell University, USA
Paper #3: Simply Desirable, Preferably Complex: Feature-Richness in Experiential Purchases
Chadwick J. Miller, Arizona State University, USA Adriana Samper, Arizona State University, USA Naomi Mandel, Arizona State University, USA
Paper #4: The Peculiarly Persistent Pleasantness of Bizarre Experiences
Robert Latimer, University of Toronto, Canada SESSION OVERVIEW
Research on experiential consumption typically focuses on comparing the relative contributions of experiential and material purchases on post-consumption outcomes (e.g. happiness, identity, and regret). However, much of the rest of the journey regarding ex-periential consumption has been ignored. How do we decide what experiences we want, when we want them, and importantly, whether we made a good choice? In this special session, we combine four papers that address these unanswered questions.
While this session focuses on experiential purchases, our re-search differentially contributes to the existing literature by inves-tigating one or more components of the experiential consumption process: (1) decision making, (2) consumption timing, and/or (3) post-consumption evaluations. Accordingly, our session promises to have broad appeal, as it advances connections from the information processing (Gallo, et al.), anticipation/intertemporal choice (Kumar and Gilovich), feature-richness (Miller, et al.), and consumer satis-faction (Latimer) literatures to research on experiential consump-tion. Furthermore, every paper in this session—all of which are at advanced stages in the research process—provides insights for: (a) consumers to improve their experiential decision making, and (b) managers to improve their experiential offerings.
In the first paper of this session, Gallo and colleagues contrast how consumers process information when making material and ex-periential purchase decisions. They show that consumers naturally use an intuitive mindset when comparing experiences and a delibera-tive mindset when comparing products. Additionally, they show that when consumers try to use the alternative mindsets, they make worse choices and anticipate more regret with the outcome.
In the second paper, Kumar and Gilovich examine how con-sumption timing interacts with purchase type, such that consumers prefer immediate consumption of material goods but delayed con-sumption of experiences. Furthermore, the authors demonstrate that consumers prefer an inferior product now rather than a superior prod-uct later but, conversely, prefer a superior experience later rather than an inferior experience now.
Miller, et al. extend the focus of consumers’ inclinations for experiences to investigate pre- and post-consumption preferences for feature-richness. They find that, contrary to feature-fatigue, con-sumers prefer feature-poor experiences before consumption, but fea-ture-rich experiences after consumption. Additionally, they identify several moderators that align consumers’ pre- and post-consumption preferences.
Finally, Latimer examines a type of experience from which consumers derive increased enjoyment. He shows that, after con-sumption, bizarre experiences are more enjoyable than mundane ex-periences—even if the bizarre experience was less enjoyable during consumption.
This special session combines four papers that extend our knowledge of experiential purchases along different points of the consumption process. We believe the insights provided by this re-search will be helpful to consumers and timely for conference attend-ees given the many experiential offerings available in New Orleans.
The Heart and the Head: On Choosing Experiences Intuitively and Possessions Deliberatively
EXTENDED ABSTRACT
We examine the theoretical and empirical connections between recent work on intuitive and reflective processing (Evans and Sta-novich 2013; Kahneman and Frederick 2002) and material and expe-riential consumption (Caprariello and Reis 2013; Carter and Gilov-ich 2010; 2012; Van Boven and GilovGilov-ich 2003). More specifically, we investigate whether people tend to weigh intuition more heavily when making experiential purchases and rational calculation more heavily when making material purchases. There are at least two rea-sons to believe they do.
First, it is often easier to isolate individual features of products and conduct a feature-by-feature comparison that is likely to cue ra-tional processing (Inbar et al. 2010). Second, there are often more di-mensions to consider when it comes to experiential purchases (Gallo and Sood 2014). A two-systems perspective seems to support the pre-diction that people will weigh intuition more heavily when it comes to experiential choices and rational calculation more heavily when it comes to material choices. Thus, examining how people choose to make these purchases provides an opportunity to test important postulates of a two-systems perspective.
This is not to say that people always choose experiences in-tuitively and material goods rationally. Our thesis is not absolute.
The relative nature of our thesis is further mandated by the fact that the distinction between experiential and material purchases is not a clear-cut dichotomy. Both bicycles and television sets are undeniably material goods, but both are also vehicles for an experience. Still, the distinction is sufficiently clear and meaningful that, if our thesis is correct, people are more likely to weigh intuition more heavily when purchasing experiences and reason more heavily when purchasing material goods. All the stimuli used in the five studies we present were pretested so that, in each study, the material set of potential purchases was as desirable as the experiential set.
In study 1 we test consumers’ processing style. Holistic pro-cessing involves the analysis of objects and entities taken as a whole (Nisbett et al. 2001), while analytic processing involves the evalua-tion of objects and entities attribute-by-attribute. We presented
par-ticipants (n = 94) with a material or experiential purchase decision (laptop, champagne flutes; weekend trip, dinner out) and asked them to select the format in which they would want to see the pertinent information. In line with our theoretical framework, those evaluat-ing experiences selected a holistic or intuitive process (84%) more frequently than those evaluating material products (64%), χ2(1, 187)
= 6.20, p < .02.
In our second study, we examine choice. Through an imbed-ded figure task (Monga & John 2008), we primed participants (n
= 167) to process information holistically or analytically and then examined their preferences for material or experiential goods (lap-top vs. weekend trip; dinner vs. champagne flutes). If people tend to think about experiential purchases holistically, then being in a holis-tic mindset should make experiential information easier to process and the experience in question more appealing. As predicted, partici-pants induced to adopt a holistic mindset were more likely to choose the experiential option (68%) than those primed to adopt an analytic mindset (53%), χ2(1, 334) = 6.71, p < .01.
In study 3, we asked participants (n = 25) to rate the extent to which different purchase decisions (8 material and 8 experiential), should be made on the basis of intuition versus reason, on a scale from 1 to 9. As predicted, participants thought the material purchases should be based significantly more on reason (M = 6.94, SD = 1.09) than the experiential purchases (M = 4.60, SD = 1.39), paired t(24)
= 6.86, p < .001.
Next (study 4), we tested whether processing style impacted consumer regret. Participants (n = 20) were informed that they would see a sequence (from study 3) of different types of potential pur-chases—one at a time—and they were to imagine that they were torn between two different options. They were also asked to imagine that their intuition led them to prefer one option and reason the other, and that their ultimate decision went quite badly so that they ended up regretting the choice they made. We asked them which process-ing style they would regret usprocess-ing more. As predicted, participants thought they would have significantly more regret going against rea-son for material purchases (M = 4.80, SD = 1.96) than for experien-tial purchases (M = 3.30, SD = 1.87), t(19) = 3.032, p = .007.
Finally, in study 5 we test whether choosing experiences in-tuitively and products deliberately leads to making better decisions.
We checked this by testing for transitivity violations (von Neuman and Morgenstern 1947). We instructed some participants to make a set of pairwise choices intuitively and others analytically, and then we tabulated the number of intransitive preferences (paradigm from Rusou, Zakay, and Usher [2013]). We then presented participants (n
= 140) with a variety of pairs of possible purchases (adapted from studies 3 and 4), and asked them to choose one option in each pair.
As anticipated, participants told to make their choices deliberatively showed more transitivity violations in the experiential condition (M
= 2.85, SD = 4.07) than in the material condition (M = 0.89, SD = 1.33), F(1,136) = 9.30, p = .003. However, for those told to make their choices intuitively, participants in the material condition had more transitivity violations than those in the experiential condition (though this difference was not statistically significant).
These findings contribute to two growing literatures in social and consumer psychology. First, our results attest to the utility of a two-system model for purchase decisions. Second, our results add to the existing work on the psychological differences inherent in expe-riential and material consumption.
To Do or To Have, Now or Later? The Preferred Consumption Profiles of Material and Experiential
Purchases
EXTENDED ABSTRACT
Van Boven and Gilovich (2003) asked a simple question: “To Do or To Have?” With respect to consumer satisfaction, their answer was clear—the hedonic return is greater for experiential than for ma-terial purchases (i.e. for money spent on doing rather than having).
Here, we ask: When?
Nearly all of the research following the aforementioned seminal paper dealt with utility derived after consumption. More recently, however, Kumar, Killingsworth, and Gilovich (2014) found that there is a difference in the value derived from material and expe-riential purchases even before the good is acquired or the experi-ence is consumed. The anticipation of future consumption tends to be more pleasant, more exciting, and less fraught with impatience for experiential than for material purchases. As Loewenstein (1987) has argued, the utility people get from anticipation can lead them to delay consumption. Accordingly, we investigated whether, as a re-sult of this difference in anticipatory pleasure, people might prefer to delay their consumption of experiences, while preferring to consume products immediately.
In studies 1a-1c, we demonstrate that when asked to choose be-tween an experience and a possession at different times, people show a marked preference for consuming the material purchase now and the experiential purchase later. As an initial assessment of whether people would rather delay their consumption of experiences than of material purchases, participants in study 1a (N=97) were presented with a choice between a possession now and an experience later or an experience now and a possession later. When they were told to imagine that they were given $1500 to spend on two purchases in the scenario we gave them. 68% of participants indicated that they would rather have a $750 gadget now and a $750 beach vacation later, significantly different from indifference, χ2 = 12.63, p < 0.001.
One potential (but banal) explanation of these results is that people are more willing to put off consumption of an experience because most experiences are “time limited” but most material goods are not: that is, experiences are often over quickly and the only way to stretch out their enjoyment is to put off when they begin. Material goods, in contrast, can usually be enjoyed now and down the road.
To deal with this issue, we conducted two follow-up replications us-ing an experiential and material purchase that were matched on the degree to which they were time limited. In study 1b, 89% of partici-pants reported that they would rather have a gadget now and a non-time-limited experience (a lifetime museum membership) later, χ2 (1, N=99) = 59.89, p < 0.0001. 65% of respondents in study 1c indicated that they would rather have a time-limited gadget (a “loaner” pair of Google Glass) now and do a price-matched outdoor activity later, χ2 (1, N=100) = 9.00, p < 0.01.
When forced to choose, consumers prefer to have their mate-rial purchases sooner and their experiential purchases later. Mov-ing beyond forced choice, we extended these findMov-ings in study 2a (N=98) to a variety of purchases in a different paradigm in which participants simply stated their preferred time of consumption. Par-ticipants in this study were presented with 20 experiential and mate-rial purchases in a randomized order and were asked their preferred consumption time-frame for each purchase. Consumers reported that they would want to delay consumption of the 10 experiential pur-chases longer than the 10 material goods, matched pairs t(97) = 5.70, p < 0.0001. In a follow-up study (study 2b; N=102), we replicated this result using a material purchase and an experiential purchase
that could be consumed repeatedly, matched pairs t(101) = 6.39, p
< 0.0001.
Ninety-five participants in study 3a were presented with a choice between a lesser purchase now and a superior purchase later on and we examined whether they were more likely to choose a less-er matless-erial possession now but opt to consume supless-erior expless-eriential purchases later. Respondents were asked to indicate a material/ex-periential purchase (between-subjects) they intended to make in the near future. They were then asked to imagine that 6 months from now, for the same price, they could get an “upgraded” version of that purchase. We then asked them to indicate their relative preference for the purchase that they could make right now versus the superior purchase in the future. Here, consumers exhibited more patience for experiential purchases, t(93) = 3.38, p = 0.001. Study 3b (N=100) used a similar paradigm to investigate whether the same pattern of results would emerge when the purchase was held constant and we varied whether participants were led to think of it in material or ex-periential terms. Here, a trip to New York was desired more imme-diately when the material aspects of that purchase were highlighted than when participants focused on its experiential elements, t(98) = 1.95, p = 0.05.
Study 3c (N=97) directly replicated these findings and concep-tually replicated the findings from Kumar et al. (2014), showing that the greater value added by anticipation to experiential consumption mediates the differences found in intertemporal choices. Participants in this study indicated that they derived more anticipatory utility from a future experiential purchase, t(95) = 2.36, p = 0.02. Addi-tionally, participants exhibited a stronger preference for an improved but delayed version of their purchase in the experiential rather than material condition, t(95) = 2.50, p = 0.01. Regressing intertemporal preference onto purchase condition and reported anticipatory utility showed that purchase condition was no longer a significant predic-tor of intertemporal choice (p > 0.05) whereas the amount of an-ticipatory utility associated with purchases did predict intertemporal preference, ß = 0.45, p < 0.01—indicating that anticipatory utility mediates this relationship. In a bootstrapping mediation analysis (10,000 samples), the 95% CI on the indirect effect did not include zero [0.05, 0.90].
Finally, in study 4 (N=94), we demonstrate that that our find-ings affect people’s real-life purchasing decisions. When participants were asked between-subjects about which purchases from their own lives they either couldn’t wait for or had delayed in order to savor, the delayed purchases were rated as significantly more experiential, t(92) = 2.43, p < 0.02.
Simply Desirable, Preferably Complex: Feature-Richness in Experiential Purchases
EXTENDED ABSTRACT
Previous research has differentiated between products and ex-periences in many ways (Van Boven and Gilovich 2003) and even though experiences generally provide more happiness than their ma-terial counterparts, researchers have primarily focused on studying the heuristics consumers use with material purchases. For instance, prior work finds that consumers prefer feature-rich products pre-con-sumption but feature-poor products post-conpre-con-sumption (Thompson, Hamilton, and Rust 2005). However, when it comes to experiences, which are intangible and have a finite usage time, will consumers still experience this “feature fatigue?”
In this research, we seek to understand how the number of fea-tures on an experience impacts pre- and post-consumption prefer-ences. Prior work in the task complexity literature has shown that
as the number of events in a given task increases, the outcome of the task becomes more difficult to predict (Campbell 1988). As a result, consumers may prefer feature-poor experiences because they are perceived to have less uncertainty than feature-rich experiences (Trope and Liberman 2003). Post-consumption, however, there is evidence that consumers should prefer feature-rich experiences. For instance, a feature-rich experience should provide more opportuni-ties for stories (Van Boven and Gilovich 2003; 2005), a better chance for a higher peak and end (Kahneman 1999), and less regret of inac-tion (Rosenzweig and Gilovich 2012).
In the current research, we show that consumers prefer feature-poor experiences pre-consumption, yet prefer feature-rich experi-ences post-consumption (the opposite of “feature fatigue;” studies 1 and 2) and that uncertainty mediates pre-consumption preferences.
We also show that pre-consumption preferences can be moderated by changing the temporal distance associated with the purchase (study 3). That is, when temporally close we expect the same effect but when temporally distant, we expect consumers to have no prefer-ence for feature-rich vs. feature-poor experiprefer-ences. Additionally, we suggest that Locus of Control (LOC; Rotter 1966) should impact how much consumers enjoy feature-rich (-poor) experiences (study 4) because it influences consumers’ initiation, effort, and persistence within an activity (Skinner 1996).
In our first study, we conducted a 2 (rich vs. feature-poor) x 2 (pre- vs. post-consumption) experiment. Participants (N=290) rated how much they liked, desired, and thought they would be satisfied with (α=.96) a feature-poor versus feature-rich set of comedy videos. The comedy videos told exactly the same jokes and lasted about 10 minutes. In the “feature-rich” condition there were 19 individual segments, while in the “feature-poor” condition, there were only 3 individual segments. As anticipated, pre-consumption, participants preferred the feature-poor experience (Mfeature-poor=6.57,
Mfeature-rich=5.22, F(1,289) = 15.02, p<.01), but post-consumption
they preferred the feature-rich (Mfeature-poor=6.35, Mfeature-rich=7.01, F(1,289)=4.18, p<.05). Thus, we show that contrary to the case of material goods, consumers prefer feature-poor experiences pre-con-sumption, but feature-rich experiences post-consumption.
In study 2, we use more typical experiential stimuli—a day long trip—to replicate our findings from study 1 and test whether uncer-tainty concerns mediate the relationship between feature-richness and experiential evaluations. Participants (N=181) were asked to imagine that they were considering (pre-consumption) or just return-ing from (post-consumption) a rich (6 activities) or feature-poor (2 activities) experience. We replicated our findings from study 1, where participants preferred the feature-poor experience pre-con-sumption (Mfeature-poor=7.34, Mfeature-rich=6.49, F(1,180) = 6.57, p<.05) and feature-rich experience post-consumption (Mfeature-poor=6.79, M
fea-ture-rich=7.49, F(1,180) = 3.93, p<.05). We also found that uncertainty mediates this relationship pre-consumption (b = -.66 [-1.10, -.29]), such that a feature-rich experience increases uncertainty which de-creases evaluations of the experience.
Using a feature-poor (7 activities) and feature-rich vacation (21 activities), in study 3 we tested whether the temporal distance of the experience would moderate preferences for feature-poor experienc-es. We asked participants (N=475) to imagine they were going on a feature-rich (vs. feature-poor) vacation either next week or next year.
Participants preferred the feature-poor experience when it was tem-porally proximate (Mfeature-poor=7.25, Mfeature-rich=6.34, F(1,474) = 6.31, p<.05), but feature-richness did not impact preferences when the experience was temporally distant (Mfeature-poor=6.64, Mfeature-rich=6.71, F(1,474) = .1, p=.75). We also found that uncertainty mediated this relationship in the temporally close condition (b = -.55, [-.86, -.26]).
In study 4, we test: (1) whether consumers’ pre- and post-con-sumption preferences flip when they actually choose their own ex-perience as well as (2) the moderating effect of locus of control on post-consumption evaluations. First, we asked participants to choose either a rich (4 activities for 5 minutes in total) or feature-poor experience (one activity for 5 minutes in total). Overwhelm-ingly, participants (N=190) chose the feature-poor experience (160, 84%) rather than the feature-rich (30, 16%). However, after doing the experience of their choice, those that picked the feature-rich perience rated it higher than those that chose the feature-poor ex-perience (Mfeature-poor=5.93, Mfeature-rich=6.45, F(1,189) = 3.86, p=.05).
Additionally, we found a marginally significant feature-richness x
Additionally, we found a marginally significant feature-richness x