ArticlePDF Available

Keeping the Memory but Not the Possession: Memory Preservation Mitigates Identity Loss from Product Disposition

Authors:

Abstract and Figures

The reliance on donations to build inventory distinguishes nonprofits from traditional retailers. This reliance on consumer donations means these organization face an inherently more volatile supply chain than retailers who source inventory from manufacturers. The authors propose that consumer reluctance to part with possessions with sentimental value causes a specific bottleneck in the donation process. The goal of this research is therefore to provide nonprofits with tools to increase donations of used goods and provide a theoretical link between the literatures on prosocial behavior, disposition, memory, and identity. As such, the authors explore the effectiveness of memory preservation strategies (e.g., taking a photo of a good before donating it) in increasing donations to nonprofits. A field study using a donation drive demonstrates that encouraging consumers to take photos of sentimental possessions before donating them increases donations, while five laboratory experiments explicate this result by mapping the proposed psychological process behind the success of memory preservation techniques. Specifically, these techniques operate by ameliorating perceived identity loss when considering donation of sentimental goods.
Content may be subject to copyright.
1
Keeping the Memory but not the Possession:
Memory Preservation Mitigates Identity Loss from Product Disposition
KAREN PAGE WINTERICH
REBECCA WALKER RECZEK
JULIE R. IRWIN*
(pre-print; forthcoming at Journal of Marketing, May 2017)
*Karen Page Winterich is an Associate Professor of marketing and Frank and Mary Smeal
Research Fellow at the Smeal College of Business, Pennsylvania State University, University
Park, PA 16802, kpw2@psu.edu. Rebecca Walker Reczek is an Associate Professor of
marketing and Dean’s Faculty Fellow at the Fisher College of Business, The Ohio State
University, 2100 Neil Avenue, 538 Fisher Hall, Columbus, OH 43210, reczek.3@osu.edu. Julie
R. Irwin is Marlene and Morton Meyerson Centennial Professor of Business at the McCombs
School of Business, The University of Texas at Austin, 1 University Station B6700, Austin, TX
78712, julie.irwin@mccombs.utexas.edu. We are grateful to David Manos and John Duncan for
their collaboration to conduct the two field studies in the Penn State University residence halls
and also appreciate the cooperation of St. Vincent de Paul State College Thrift store for the third
field study. We thank Courtney Ironside, MacKenzie Ironside, Sarah Miller, Victoria Babb,
Gabriel Gonzales, and Daniel Shaffer for their research assistance in the field studies with a
special thank you to Stephanie Ironside for stepping up to assist with the replication field study.
We also appreciate helpful comments on this research from Rob Smith, Hari Sridhar, Chris
Summers, and Danny Zane. This research was supported in part by research grants from the
Smeal College of Business and the Smeal College of Business Sustainability Research Initiative.
2
Running Head: KEEPING THE MEMORY BUT NOT THE POSSESSION
Keeping the Memory but not the Possession: Memory Preservation Mitigates Identity Loss from
Product Disposition
The reliance on donations to build inventory distinguishes nonprofits from traditional
retailers. This reliance on consumer donations means these organization face an inherently more
volatile supply chain than retailers who source inventory from manufacturers. The authors
propose that consumer reluctance to part with possessions with sentimental value causes a
specific bottleneck in the donation process. The goal of this research is therefore to provide
nonprofits with tools to increase donations of used goods and provide a theoretical link between
the literatures on prosocial behavior, disposition, memory, and identity. As such, the authors
explore the effectiveness of memory preservation strategies (e.g., taking a photo of a good before
donating it) in increasing donations to nonprofits. A field study using a donation drive
demonstrates that encouraging consumers to take photos of sentimental possessions before
donating them increases donations, while five laboratory experiments explicate this result by
mapping the proposed psychological process behind the success of memory preservation
techniques. Specifically, these techniques operate by ameliorating perceived identity loss when
considering donation of sentimental goods.
Keywords: nonprofit marketing, donation, memory, identity, product disposition
3
Consumer product disposition determines the supply chain for organizations in the
second-hand goods marketplace, including nonprofits that rely on the donation of used goods.
For example, Goodwill Industries, the largest retail resale nonprofit, generated $5.37 billion in
retail sales in 2014 from its more than 3,000 stores and online auction site (Forbes 2015) and
used these sales to fund a number of social welfare programs. Other nonprofits such as Dress for
Success and the American Red Cross gather donations of clothing and household goods and
instead of selling them, directly distribute to people in need such as victims of natural disaster
and political refugees. In this research we explore a factor that distinguishes nonprofit resale
stores from traditional retail marketing, specifically the reliance on donations to build inventory.
Whereas a number of studies have examined relationships between retailers and manufacturers
for procurement (e.g., Basuroy, Mantrala, and Walters 2001; Mantrala and Raman 1999), little, if
any, research has considered the unique challenge of sourcing from consumers. Relying on
donations from consumers means these enterprises face an inherently more volatile supply chain
than traditional retailers who source inventory from manufacturers. In this research we address
how organizations can strategically increase donations of used but still usable physical
possessions. We do so by addressing a specific bottleneck in the disposition process, a reluctance
to part with possessions with sentimental value.
Many used goods sit in storage, in homes or in paid storage facilities, rather than entering
the secondary goods market where they can resume their useful life. For example, a survey
conducted by eBay and Nielsen revealed that the average American has 50 unused items in their
home, with some of the most common items including clothing, accessories, electronics, sporting
goods, and toys (Business Wire 2007). Some of these unused but still usable goods are
associated with treasured memories (e.g., toys children no longer play with, clothing worn on a
4
special occasion that no longer fits). These product-specific memories may be a key reason why
consumers are reluctant to even consider disposition of possessions they no longer use.
In this research, we therefore propose that actions designed to preserve consumers’
product-specific memories should increase consumers’ willingness to donate possessions,
particularly those with sentimental value. We further propose and find that memory preservation
increases donation likelihood through the link between product-specific memories and the
unique personal and social identities these product-specific memories help the consumer express
(Oyserman 2009). This research makes several theoretical contributions. The work adds new
elements to the growing knowledge base on product disposition (Haws et al. 2012; Jacoby,
Berning, and Dietvorst 1977; Okada 2001; Trudel and Argo 2013; Trudel, Argo, and Meng
2016). Much prior disposition research has focused on disposition when recipient identity is
known (Brough and Isaac 2012; Lastovicka and Fernandez 2005; Price, Arnould, and Curasi
2000). A contribution of the current research is identifying how to increase product disposition
when the recipient is unknown, which is the case for most donation contexts. This research also
contributes to the literature on donation behavior by focusing on factors that uniquely influence
donation of goods rather than time (Liu and Aaker 2008; Lee, Piliavin, and Call 1999) or money
(Anik, Norton, and Ariely 2014; Savary, Goldsmith, and Dhar 2016; Small and Verocchi 2009;
Smith, Faro, and Burson 2013), which have been explored more extensively. An additional
contribution of this research is the examination of the unique challenges for retailers sourcing
from consumers. Although literature has given due attention to factors in sourcing goods from
manufacturers (Basuroy, Mantrala, and Walters 2001; Mantrala and Raman 1999), the current
research considers personal factors that come into play when sourcing from consumers. As the
5
reuse and sharing economy continues to grow (Brosius, Fernandez and Cherrier 2013;
Lamberton and Rose 2012), understanding consumer sourcing will be of increasing importance.
This research also has practical implications for managers of nonprofits. By
demonstrating the effectiveness of specific memory preservation techniques in increasing
donation of used goods, we offer a relatively low cost, easily implementable intervention that
organizations can use to increase consumer donations, thereby increasing their supply of second-
hand goods. Overcoming the reluctance to part with sentimental goods is critical for charities to
spur the donation process.
Theoretical Background
Product-Specific Memories
Research on consumers’ memories suggests that consumers treat some particularly
positive memories (e.g., of an anniversary dinner or vacation) as assets to be protected
(Zauberman, Ratner, and Kim 2009). To protect these memories, consumers both avoid
situations they think will threaten the ability to retrieve these unique memories and engage in
acquisitive market behavior such as buying physical possessions that will serve as memory
markers and retrieval cues (e.g., a souvenir, Zauberman, Ratner, and Kim 2009). Recognizing
market goods may store memories for consumers, we consider how product-related memories
might explain the difficulty consumers face in relinquishing goods.
We focus on goods with sentimental value, which we define as goods that represent an
emotionally significant memory for the consumer. Sentimental value is emotional value that goes
beyond the functional or material value of the good because of the private meaning the good has
6
for the consumer (Holbrook 1994; Loewenstein and Issacharaoff 1994; Young 1991). The
sentimental value attached to a good may be the result of characteristics of the product itself (i.e.,
some product categories are inherently more attached to memories than others, like trophies) or
the result of experiences the owner has had with the product during ownership (e.g., Strahilevitz
and Loewenstein 1998) or during product creation, as goods that consumers have made
themselves can also come to have value beyond their functional or material value (Franke,
Schreier, and Kaiser 2010; Norton, Mochon, and Ariely 2012). Goods acquire sentimental value
when (1) they are from a class of goods considered inherently sentimental because they serve as
memory markers connected to emotionally significant events or people in one’s life (e.g.,
trophies, souvenirs, family heirlooms) and/or (2) consumers have become emotionally attached
to them during the course of product acquisition or consumption. Thus, all goods with
sentimental value are goods that are connected to an emotionally significant memory for the
consumer, although the degree of significance will of course vary across sentimental goods. This
connection to an emotionally significant memory is what distinguishes a good with sentimental
value from one that is valued more monetarily simply because one owns it (i.e., it is part of one’s
endowment; Kahneman, Knetsch, and Thaler 1991).
We propose it is the memories associated with sentimental goods that make it difficult to
dispose of possessions; consumers do not want to relinquish these memories. We further propose
that preserving product-specific memories for goods with sentimental value can overcome
barriers to donating, thereby increasing donations. We define memory preservation in the context
of physical goods as any action designed to retain product-specific memories, including writing a
note or journal entry describing the possession and its associated memories or taking a
photograph of the good, where the photograph can cue memories associated with the product
7
even after the product is gone. We propose memory preservation is effective at increasing
donation of goods with sentimental value because it reduces the fear of identity loss consumers
would otherwise feel when considering giving up possessions that are connected to the self (Belk
1988). We next turn to the link between product-specific memories and consumer identity.
Product-Specific Memories and Consumer Identity
Marketers frequently position and advertise their products as relevant to consumer
memories (Sujan, Bettman, and Baumgartner 1993) and identity (Chernev, Hamilton, and Gal
2011; Forehand and Deshpande 2001). Such marketing can be effective because “possessions are
a convenient means of storing the memories and feelings that attach our sense of past” (Belk
1988, p. 148). In other words, memories and identity are integrally linked: if a person remembers
nothing of her past, she has no identity (Kihlstrom, Beer, and Klein 2002). Indeed, the most
common explanations for valuing “treasured” possessions concerned the memories of other
people, occasions, and relationships (Csikszentmihalyi and Rochberg-Halton 1981), that is, the
owner’s various personal and social identities (e.g., their identity as an athlete, an alumnus of a
particular university, or as a parent; Oyserman 2009). Therefore, consumers seek to protect their
memories in part to help retain the identity the memories represent. Because possessions with
sentimental value are associated with identities, loss of these possessions can lead to strong
negative reactions (Burris and Rempel 2004; Ferraro, Escalas, and Bettman 2011) and a
diminished sense of self (Ahuvia 2005). Accordingly, possessions serve as an anchor for
identities such that the loss of self-linked possessions results in identity loss, making consumers
reluctant to donate possessions with sentimental value.
Much of the past work exploring how consumers give up possessions with sentimental
value has focused on the owner’s interest in what happens to the good after disposition. For
8
example, Brough and Isaac (2012) show that sellers are sensitive to the manner in which the
product will be used following a transaction, accepting lower selling prices for used goods if they
deem the buyer’s usage intentions for the product appropriate (e.g., intending to play a piano vs.
using it as furniture). Similarly, consumers giving up possessions with sentimental value prefer
recipients who know and appreciate a good’s meaning (Price, Arnould, and Curasi 2000).
Unknown recipients are considered acceptable only after a shared identity is uncovered that
allows the meaning of the possession to transfer along with the good (Lastovicka and Fernandez
2005). Although consumers may sometimes have control over the specific individual to whom
their possessions go when considering selling or giving to a friend or family member, this kind
of perceived control over the future fate of a possession is lacking in most donation contexts.
Large nonprofits, such as Goodwill, receive donations from millions of people (e.g., 87 million
people donated to Goodwill in 2013 alone; Goodwill 2014). For most donors, the recipient of
their good is someone with whom they have no relationship or, typically, even contact. Thus, a
key contribution of our work over that of Brough and Isaac (2012) and Lastovicka and
Fernandez (2005) is that we explore disposition when the sentimental value of the good cannot
transfer to the recipient because consumers have no knowledge of or control over the new owner.
Because our proposed mechanism to mitigate identity loss is focused on the memories
represented by the good, our work differs from previous research in both our proposed method
(i.e., memory preservation) and in our psychological explanation (i.e., the preservation of
memories to prevent identity loss). In doing so, this research offers insights for aiding consumer
disposition without relying on transfer of sentimental value to the recipient.
Moderating the Effectiveness of Memory Preservation
9
The essence of our theoretical argument is that memory preservation strategies are
effective at increasing donation of goods because preserving the memories preserves the
consumers’ sense of self (i.e., their identities). When concerns about identity loss are mitigated,
consumers are able to let go of physical possessions. This process implies that memory
preservation should be less likely to increase donation of goods lacking sentimental value. Such
goods are less likely to have treasured memories associated with them and instead are retained
for future usefulness or value (Haws et al. 2012; Okada 2001). We therefore predict that
preserving memories for goods will only increase donation likelihood for sentimental goods.
That is, the sentimental value of the good moderates the effect of memory preservation on
donation likelihood.
Our contention is that the effectiveness of memory preservation techniques is driven by
mitigating the identity loss consumers would otherwise feel when donating possessions with
sentimental value. Thus, one way to affirm this process is to show that if the identity most
relevant to the good in question is first reinforced in some other way, then consumers are more
willing to donate the good because they are not as concerned about identity loss. Any activity
which strengthens the association of an identity with the self-concept could serve as identity
reinforcement (Reed and Forehand 2016). For example, recalling experiences in which one had
confidence can reinforce the self (Gao, Wheeler, and Shiv 2009) as can recognition of an
identity-relevant behavior (Winterich, Mittal, and Aquino 2013). We therefore predict that the
effect of memory preservation on donation of a sentimental good should be mitigated when the
relevant identity is first reinforced in an alternate manner (i.e., by focusing on all of the other
non-product specific ways in which the consumer expresses that particular identity).
10
The effectiveness of memory preservation also hinges on the premise that consumers
want to preserve the identity reflected by the good with sentimental value. However, under some
circumstances, consumers may not have a strong desire to preserve their current identity. In other
words, their psychological connectedness to their future self (i.e., the perceived continuity
between their present and future selves; Parfit 1984) should determine the influence of memory
preservation on disposal. Although psychological connectedness has predominantly been
examined as a factor impacting intertemporal choice (Bartels and Urminksy 2011; Hershfield
2011), we propose that if the future self is believed to differ from the current self, then
consumers should be able to more easily dispose of identity-relevant possessions regardless of
memory preservation. Indeed, some research has found that when a good with sentimental value
no longer represents one’s current identity, it is moved to a less central location in one’s home
and only accessed when the identity is sought (Epp and Price 2010). Likewise, the effectiveness
of memory preservation should be mitigated when psychological connectedness to the future self
is low, as consumers should not fear identity loss from parting with goods when they do not
expect their current identity to be relevant in the future.
Finally, this research focuses on donation, which differs from another common disposal
method, selling, in the domains of memory and identity. We expect consumers to be reluctant to
dispose of goods with sentimental value via either donation or selling when no memory
preservation is present because both donation and selling typically result in the good going to a
stranger who cannot know and understand the good’s value. While memory preservation should
increase willingness to donate such possessions, we propose that it will not increase willingness
to sell sentimental goods. We base this prediction on a key distinction between selling and
donating: the focus on an economic transaction that offers personal gain and follows exchange
11
norms in the former versus a non-economic transaction with societal benefit that follows
communal norms in the latter (Aggarwal and Zhang 2006; Clark, Mills, and Powell 1986). The
economic focus of selling is problematic when trades are “sacred” and should not involve money
(McGraw and Tetlock 2005). Indeed there is a sizable literature on buying versus selling prices
and sentimentality (Boyce et al. 1992; Chatterjee, Irmak, and Rose 2013; Loewenstein and
Issacharoff 1994). The primary finding of this work is that sellers typically demand extremely
high prices that buyers are not willing to pay, and/or refuse to sell at all. This refusal to entertain
the idea of selling as a method of disposal due to sacredness is likely to apply to goods with
sentimental value. Therefore, we predict that even memory preservation techniques cannot
increase likelihood of selling a good with sentimental value because the economic exchange
associated with selling taints the sentimental value of the good regardless of whether the memory
is preserved. Thus, we expect a moderating effect of disposition type on the effectiveness of
memory preservation such that, when considering selling, consumers will be reluctant to dispose
regardless of memory preservation. In contrast, when considering donation, consumers will be
more likely to dispose after memory preservation.
Overview of Studies
We test our predictions in six studies. Study 1 demonstrates a meaningful increase in
actual donations in the field due to a promotional campaign encouraging consumers to use a
memory preservation strategy (i.e., taking a photo) before donating goods with sentimental
value. Studies 2-6 use laboratory and MTurk experiments to establish our hypothesized
mechanism for this effect. Study 2 supports the effectiveness of the memory preservation tactic
12
documented in the field study by showing that consumers are reluctant to donate possessions
with sentimental value because of the memories and identities linked to them. Study 3 then
demonstrates that the success of memory preservation occurs by mitigating identity loss. Studies
4 and 5 use moderators to affirm the effect of memory preservation on identity loss,
demonstrating that the effectiveness of memory preservation techniques is mitigated both when
the product-relevant identity is first reinforced in an alternate manner before memory
preservation (Study 4) and when consumers do not expect a strong connection between their
current and future identity (Study 5). Finally, Study 6 demonstrates that, as theorized, memory
preservation does not extend to the selling of goods.
Study 1
Before using laboratory experiments to test the psychological process we propose, we
conducted a field study to demonstrate the effectiveness of memory preservation techniques in
increasing overall donations. We implemented a promotional campaign encouraging memory
preservation (vs. a control campaign) in a way that can be easily mimicked by managers at
nonprofits. To do so we partnered with the housing office at Penn State University to conduct a
study of student donations.
Participants and Procedure
Participants were residents of one of six all-female sorority residence halls on a large
university campus. The six residence halls selected were located in the same area of campus and
housed upper-classman. Each hall was similar in size (i.e., same number of beds, ranging from
126-157 residents per building with a total 409 students across three buildings in the memory
13
preservation condition and 388 students across three buildings in the control condition for a total
of 797 residents). Because this study was conducted at the end of the fall semester, residents
were not required to move out. Given moving out is likely to increase donations, we also ensured
that there were an approximately equal number of residents moving out at the end of the fall
semester versus returning in the spring across the conditions. There were a total of 134 of 388
(35%) residents moving out in the three control condition residence halls and 147 of 409 (36%)
in the three memory preservation condition residence halls. Thus, neither the total number of
residents nor the number of residents moving out differed across the conditions.
To elicit donations, we advertised a holiday donation drive with all donations benefitting
the local Goodwill. In the memory preservation condition, the campaign stated “Don’t Pack up
Your Sentimental Clutter…Just Keep a Photo of it, Then Donate” whereas the control condition
campaign stated “Don’t Pack up Your Sentimental Clutter, Just Collect the Items, Then Donate”
(see Figure 1). The messages for both conditions asked students to think about all the items that
carry good memories but they do not use any more. They were then prompted to either take a
photo of the items and donate them or gather the items and donate them in the lobby. The
promotional flyers were hung in each resident bathroom as a “stall story” which is likely to
receive focal attention given the location and likely to be seen only by dorm residents. The flyers
were placed in the restrooms at the beginning of finals week because non-graduating students
tend to leave for the holiday break as soon as they finish their finals. On Monday after finals
week, four undergraduate research assistants who were unaware of the hypotheses or conditions
went to each residence hall. Together, they emptied the donation items out of the large plastic
donation bins and counted each item, which they recorded on a tally sheet. A total of 1,146 items
14
were counted across the six residence halls. Then the research assistants packed up the donation
items to be picked up by the local Goodwill.
---insert Figure 1 about here---
Results and Discussion
We conducted a logit model on the raw count data to examine whether number of
donations by residence halls in the memory presentation condition differed from that in the
control condition. Out of the total 1,146 items donated, there were 533 items donated in the three
halls in the control condition and 613 items donated in the three halls in the memory preservation
condition. This difference was significant as expected (Wald X2(1) = 7.81, p < .005). Put in
percentage terms, the 80 additional items donated in the memory preservation condition
amounted to a 15% (i.e., 80/533) increase in donations due to the manipulation. Because there
was slight variation in the number of residents between conditions, we also tested the weighted
counts of 544 and 601, with greater donations in the memory preservation condition, and the
effect was also significant in this adjusted logit test (Wald X2(1) = 4.64, p < .03).
The results of this field study indicate that a promotional campaign encouraging
consumers to engage in memory preservation for possessions with sentimental value increases
total donations. One limitation of this study is that because students were able to drop off
donations at any time, 24 hours a day, we were unable to staff research assistants at the donation
drop-offs and so were unable to survey participants directly to assess whether the items they
donated were indeed possessions with sentimental value. Further, since sentimental value is
based on memories with emotional significance for a good’s owner, research assistants could not
identify it with a visual inspection. We therefore turned to laboratory experiments in order to
provide psychological insights into what is driving the increase in donation rate observed in
15
Study 1. We note that we replicated the results of Study 1 in a larger field study conducted at the
end of the school year at Penn State University. In this study, reported in the Web Appendix, we
once again observe an increase in total donations in the dorms where the promotional campaign
encouraged consumers to engage in memory preservation.
Study 2
Study 2 begins the experimental exploration of the effectiveness of memory preservation
by investigating why consumers do not dispose of their possessions. The findings illuminate that
worries about memory and identity loss underlie reluctance to dispose of sentimental goods
providing insight into the success of the promotional campaign used in Study 1.
Participants and Procedure
A total of 81 U.S. adults from MTurk completed the study for a small payment (51%
female, 1 unspecified; Mage = 35.48, SD = 12.43). Participants were randomly assigned to one of
two conditions (sentimental value: yes or no). All participants were asked to think of a product
they currently own but no longer use that could be useful to someone else. Depending on
condition, we specified this product either should have special meaning to them or have no
special meaning (see Web Appendix).Goods with sentimental value that participants listed
included clothing, shoes, stuffed animals, and golf clubs, whereas goods without sentimental
value included a computer monitor, lamp, and dishes. Respondents first described the product,
including acquisition and use, and then provided reasons, in an open-ended format, why they had
not disposed of the product. Two coders blind to the hypotheses of the study coded these reasons
(r = .84, all disagreements were resolved through discussion) into the following categories: 1)
16
remaining value (e.g., still works, could be used again), 2) inconvenient disposal (e.g., don’t
know where or how to dispose, haven’t gotten around to it), 3) product-specific memories (i.e., it
brings back memories), and 4) identity-relevant (e.g., relates to type of person they are or were).
Sample thoughts for each category are included in Table 1.
Next, participants reported the approximate resale value of the item, “not what the item is
worth to you, but what someone else would pay for the item”. Product values ranged from $0-
$7,000, and had a log-normal distribution with a geometric mean of $20.55 (SE = $4.65).
Participants then responded to two items as a manipulation check for the sentimental value of the
product they imagined (“The item I thought about is special to me” and “I am emotionally
attached to this item,” α = .97, from 1-not at all to 9-extremely). Because sentimental value may
impact the difficulty of identifying an item, we also asked participants to respond to two items
regarding the difficulty of identifying a product (“It was difficult to recall such a product” and “I
could easily identify such a product,” α = .82, from 1-not at all to 7-very much so). Gender and
age were measured in all studies but did not impact results here or in subsequent studies and are
not discussed further.
Results and Discussion
Manipulation Checks. An ANOVA with possession type (sentimental value: yes vs. no)
predicting sentimental value (averaged across two items) revealed a main effect of possession
type (F(1, 79) = 73.96, p < .01; Msentimental value = 7.21, Mno sentimental value= 2.99), as expected. We
also conducted an ANOVA on difficulty of possession selection as the dependent variable. The
mean difficulties did not differ by possession type and were uniformly low (F(1, 79) = .001, p =
.98; Msentimental value = 2.07, Mno sentimental value = 2.08), suggesting participants could easily identify
both goods with and without sentimental value they no longer used and could dispose. Lastly, the
17
resale value did not differ for possessions with and without sentimental value for either the
logged (p > .70) or original (p > .30) values1.
Analysis of Thought Listings. The primary analyses were tests of the prevalence of each
thought listing across the two possession types. All results are reported in Table 1. Participants
had more thoughts regarding the remaining value of the good and the inconvenience of disposal
for goods without sentimental value than for goods with sentimental value. In contrast, those
considering a sentimental good had more thoughts about product-specific memories and relevant
identities than did those considering goods without sentimental value. This pattern suggests that
memory and identity relevance were more important for sentimental goods than for goods that
were not sentimental. A contrast code tested and affirmed this pattern (F(1, 79) = 42.84, p <
.0001).
---Insert Table 1 about here---
Discussion. This study supports our contention that consumers are reluctant to donate
possessions with sentimental value due to the memories and identities associated with the goods
rather than due to remaining product utility or the inconvenience of donating, which were more
likely to be the barriers to donation for goods without sentimental value. The unprompted
association of memories and identities with goods with sentimental value provides insight into
the effectiveness of the memory preservation campaign in the field study; taken together, the
studies suggest that addressing these potential losses through memory preservation likely spurred
consumers to engage in the donation process, increasing total number of items donated. To

1 We collected the same sentimental value, difficulty, and resale value measures in subsequent studies. Rated
sentimental value was always greater than the scale midpoint in the sentimental value condition. Difficulty and value
did not differ significantly between conditions. Additionally, controlling for value did not alter the significance of
the focal effect in any study other than as noted for Study 3 in footnote 3. See Web Appendix Tables 1 and 2 for
details on these analyses.
18
provide further evidence that suggesting memory preservation of sentimental goods in a donation
appeal may be critical for charities to overcome consumer reluctance to consider donation, we
showed female participants (n = 81 US MTurk workers; Mage = 35.54, SD = 11.27) one of two
donation appeals (memory preservation present or absent similar to those used in Study 1; see
Web Appendix for stimuli, measures and detailed results) and asked them to indicate how likely
they were to look for household items to donate using three items as well as how recently they
cleaned out unused goods in their home. When controlling for how recently consumers have
cleaned out their homes, consumers viewing the donation appeal suggesting memory
preservation for sentimental goods reported greater motivation to look for items in their home to
donate (Mpresent = 4.98 vs. Mabsent = 4.28, F(1, 78) = 4.13, p = .04). These results suggest that
donation appeals suggesting memory preservation for sentimental goods may aid in overcoming
consumer reluctance to begin the donation process, which may increase not only the number of
sentimental goods donated but total donations as well, simply by encouraging consumers to
initiate the process of identifying items to donate.
Study 3
The goal of study 3 is to directly demonstrate that the effectiveness of memory
preservation on donation likelihood arises specifically from the interplay between memory and
identity for goods with sentimental value. This study therefore examines the effect of memory
preservation for a specific good by asking participants to consider donating a good (either
sentimental or not) and giving some of the participants the opportunity to engage in a memory
preservation technique of their choice (e.g., writing about the object or taking a photo). We
19
expected memory preservation to increase donation likelihood for goods with sentimental value
only and to do so through a decrease in perceived identity loss.
Participants and Procedure
A total of 160 U.S. participants were recruited from MTurk for small payment. Of these,
four participants were excluded due to incomplete responses and an additional five participants
were excluded for listing items with resale value greater than $1000 when instructions specified
not to use items of high monetary value.2 Thus, we analyze 151 responses (40% female; Mage =
32.02, SD = 9.80, range = 18-61). The study had a 2 (sentimental value: yes vs. no) x 2 (memory
preservation: present vs. absent) between-subjects design. All participants were first asked to
think of a product they currently own but no longer use that could be useful to someone else.
Depending on condition, we specified this product should either have sentimental value or not.
Participants then described the product they were thinking of in one or two words for use later in
the study. Next all participants imagined they were cleaning out their home and came across this
item. In the memory preservation absent condition, participants received no further information.
In the memory preservation present condition, they were asked to imagine first preserving their
memory of the item by either writing about it or taking a picture (see Web Appendix). Two
coders blind to the hypotheses of the study coded each description (r = .86, all disagreements
resolved through discussion) for whether it involved taking a photo, writing a note or journal
entry, or taking some other action to preserve their product-specific memories. For example, two
participants wrote:

2When the five responses with extreme item values are included in analysis and value is used as a control, the
pattern remains the same with the interaction remaining significant for donation intentions (F(1, 151) = 4.39, p =
.04). The interaction for identity loss becomes marginal (F(1, 151) = 3.39, p = .07) with the effect of memory
preservation for sentimental goods on identity loss also reduced to marginal significance (Mpresent = 3.84 (1.78) vs.
Mabsent = 4.46 (1.98), t(151) = 1.66, p = .09). 
20
“I would definitely take a few photos and tuck them away into a memories file. I
would also send the picture to my parents so they could remember my baby crib
that I spent time in and that my son spent time in.”
“I would definitely take pictures of me with this item. I spent many enjoyable nights
playing [my electric guitar] with my nondescript band at local clubs...now it sits and I
only really use my acoustic.”
Consistent with these examples, taking a photo was the most popular memory
preservation method. Participants were most likely to indicate they would take a photo (62%)
versus to write a journal entry or note (13%) or employ some other tactic (22%).
Next, donation likelihood was assessed with four items on a 7-point scale from 1-very
unlikely to 7-very likely. The items were: donate it to the local Goodwill, donate it to a local
charity for children and families in need, donate it to a national nonprofit organization, and
donate it to an international charity collecting items for those in poverty in other countries (α =
.88). Then, perceived identity loss from disposition was assessed with two 7-point scaled items
from 1-strongly disagree to 7-strongly agree (α = .96): “Thinking about getting rid of this item,
please indicate your agreement with each statement: ‘I will feel like I lost a piece of myself,’ and
‘I will feel like a part of me is gone.’” Finally, participants responded to the same items asking
them to rate sentimental value (α = .95), difficulty (α = .75), resale value, and demographics used
in Study 2. Product values ranged from $0-$800 with a geometric mean of $32.99 (SD = $4.79).
Results and Discussion
Donation Likelihood. We conducted an ANOVA with possession type (sentimental
value: yes vs. no), memory preservation (present vs. absent), and their interaction as the
independent variables and donation likelihood as the dependent variable. Possessions with
sentimental value were significantly less likely to be donated (F(1, 147) = 11.87, p > .01), and
memory preservation moderately increased donation likelihood overall (F(1, 147) = 3.25, p =
21
.07). Most importantly, the focal interaction was significant (F(1, 147) = 5.74, p = .02). For
possessions with sentimental value, participants had greater donation likelihood when memory
preservation was present than when it was absent (Mpresent = 3.90 (1.74) vs. Mabsent = 2.74 (1.59),
t(147) = 2.86, p > .01). For possessions without sentimental value, memory preservation did not
increase donation likelihood (Mpresent = 4.19 (1.85) vs. Mabsent = 4.35 (1.58), t(147) = -.44, p =
.66). See Figure 2.
---insert Figure 2 about here---
Identity Loss. The same ANOVA with identity loss as the dependent variable revealed a
significant main effect of possession type (F(1, 147) = 66.07, p < .0001): participants anticipated
greater identity loss when donating a possession with versus without sentimental value. There
was not a significant main effect of memory preservation condition (p > .30). The focal
interaction was significant (F(1, 147) = 5.19, p = .02): for products with sentimental value,
participants reported less identity loss in the memory preservation present than in the memory
preservation absent condition (Mpresent = 3.60 (1.52) vs. Mabsent = 4.44 (1.98), t(147) = 2.86, p >
.01). For products without sentimental value, memory preservation did not affect identity loss
(Mpresent = 2.08 (1.44) vs. Mabsent = 1.73 (1.22), t(147) = -.44, p = .66). See Figure 2.
Mediating Role of Identity Loss on Donation Likelihood. We tested for mediated
moderation using model 8 in Hayes’ (2012) Process macro. Mediated moderation was supported
because the indirect effect of the two-way interaction on donation likelihood through identity
loss was significant (indirect effect = .08 (.04), 95% CI: .0122 to .1927). When the item had
sentimental value, the conditional indirect effect of memory preservation was significant
(indirect effect = .11 (.07), 95% CI: .0054 to .2958). However, when the item lacked sentimental
value the conditional indirect effect of memory preservation was not significant (indirect effect =
22
-.05 (.04), 95% CI: -.1496 to .0207). Thus, the avoidance of identity loss seems to drive the
effectiveness of memory preservation for goods with sentimental value.
Discussion. Study 3 demonstrates that when considering donation of a specific item,
memory preservation increases donation likelihood for possessions with sentimental value (and
not for possessions that lack sentimental value) because actions designed to preserve memories
of the good before donating prevent the loss of product-relevant identities.
Having demonstrated the psychological process in this study with hypothetical donations,
we conducted a field study with St. Vincent de Paul’s State College Thrift Store to examine
whether memory preservation indeed reduces identity loss among actual donors. We give the
primary result here; details are available in the Web Appendix. In this study 64 participants
dropping off a donation to the nonprofit in person were randomly assigned to either a memory
preservation present condition or a control (memory preservation absent) condition. They were
asked to think of one specific item they were donating that was meaningful to them; we used
“meaningful” because we thought it would be easier for participants to assess than “sentimental”.
For participants in the memory preservation present condition, the researcher took a picture of
this item with an instant camera. The photo was given to participants to keep, and there were no
copies of the photos kept by the researchers. Photos were not mentioned in the memory
preservation absent condition. All participants then answered a series of ratings questions about
their item, all on a 1-7 point scale from “Strongly Disagree” to “Strongly Agree” with two items
for identity loss (from Study 3), one item for memory loss, and four product attachment items
(Mproduct attachment = 4.24, SD = 1.60, see Web Appendix for item wording).
Given our theorizing pertains to the loss of identity, we limit our primary analysis to
those who reported ownership of their donated item, leaving 39 participants for analysis.
23
Participants in the memory preservation condition reported less identity loss as well as memory
loss than those in the control condition (Identity loss: 1.95 vs. 3.21, F(1, 36) = 4.49, p = .04;
Memory loss: 1.24 vs. 2.27, F(1, 36) = 7.89, p < .01). A mediation analysis showed that the
indirect effect of memory preservation through memory loss was significant (indirect effect = -
.40 (.26), 95% CI: -1.00 to -.01). Thus, even among those who are already willing to donate their
possessions, memory preservation at the time of donation decreased the identity loss from
donating.
Study 4
Although Study 3 provided evidence of our proposed causal mechanism via mediation,
the goal of Study 4 was to do so using moderation (see Spencer, Zanna, and Fong 2005 for a
discussion of establishing this type of causal evidence) via a manipulation designed to address
the identity loss that underlies these effects. We therefore asked half of the participants, prior to
indicating donation likelihood for a possession with sentimental value, to engage in identity
reinforcement (i.e., by thinking about the various actions they take that express the identity to
which their possession with sentimental value is linked). We expected a significant interaction:
The effectiveness of memory preservation should be mitigated when identity reinforcement is
also present because there is no longer as much of a need to preserve identity.
Participants and Procedure
A total of 237 undergraduate students who were native English speakers at Penn State
University completed the study for extra course credit (39% female; Mage = 18.86, SD = .69).
The study employed a 2 (memory preservation: present vs. absent) x 2 (identity reinforcement:
24
present vs. absent) between-subjects design. All participants were asked to think of a possession
with sentimental value, following the same instructions used in Study 3. The only difference was
that we restricted the value of the item to $100 or less because we had high variation in product
value in Study 3. In addition, we asked all participants to indicate which identity they most
associated with this specific possession. As part of the instructions, they were given a definition
of identity and several examples (see Web Appendix for details).
Participants who were in the memory preservation present condition then completed a
similar memory preservation task to that used in Study 3. Once again, coding from two
independent coders (r = .84, all disagreements resolved through discussion) indicated taking a
photo was the most popular memory preservation method (68%). In the identity reinforcement
present condition, after being reminded of the identity they had previously indicated the product
represented, participants described the ways in which they express this identity that are not
related to a physical possession (see Web Appendix for complete stimuli).
Immediately after completing the memory preservation and/or identity reinforcement
writing (or indicating the product and relevant identity, for control participants), participants
responded to the same four donation likelihood items used in Study 3 (1 = very unlikely to 7 =
very likely; α = .91). Finally, participants responded to the two sentimental value items (α = .88),
two difficulty items (α = .74), measure of resale value, and demographic items used in prior
studies. Product value averaged $25.19 (SD = $29.54).
Results and Discussion
Donation Likelihood. An ANOVA with memory preservation (present vs. absent),
identity reinforcement (present vs. absent), and their interaction as the independent variables and
donation likelihood as the dependent variable revealed that memory preservation increased
25
overall donation likelihood (F(1, 233) = 6.48, p = .01). The main effect of identity reinforcement
was not significant (F(1, 233) = .68, p = .41).
Most importantly, the focal interaction was significant and in the expected direction (F(1,
233) = 4.48, p = .05). When there was no identity reinforcement, participants had greater
donation likelihood when memory preservation was present than when it was absent (Mmemory
preservation present= 4.35 (1.80) vs. Mmemory preservation absent = 3.20 (1.88), t(233) = 3.26, p > .01). This
effect mimics Study 3 where memory preservation reduces concerns about identity loss.
However, when participants engaged in identity reinforcement first, memory preservation did not
increase donation likelihood since the identity is not in danger of being lost (Mmemory preservation
present = 4.03 (1.90) vs. Mmemory preservation absent = 3.93 (1.98), t(233) = .31, p = .76). See Figure 3.
Additionally, when memory preservation is absent, identity reinforcement increases donation
likelihood (Midentity reinforcement present = 3.93 (1.98) vs. Midentity reinforcement absent = 3.20 (1.88), t(233) =
2.09, p = .04), consistent with our theorizing regarding the role of identity loss.
---Insert Figure 3 about here---
Discussion. Study 4 provides additional evidence that memory preservation is (1)
effective at increasing donation likelihood for possessions with sentimental value and (2) driven
by a decrease in concerns about identity loss. When identity loss is directly prevented with an
identity reinforcement manipulation, the effect of memory preservation is attenuated.
Study 5
26
In Study 5, we test the moderating role of psychological connectedness to the future self
on the effect of memory preservation on donation likelihood. The effectiveness of memory
preservation should be mitigated when psychological connectedness to the future self is low.
Participants and Procedure
A total of 204 undergraduate students at Penn State University completed the study for
extra course credit (50% female; Mage = 19.16, SD = .73). The study employed a 2 (memory
preservation: present vs. absent) x 2 (connectedness to future self: high vs. low) between-
subjects design. First, all participants read a short passage purportedly describing recent research
on the self that either argued that college students’ current identities are highly connected to their
future identities or are not at all connected (manipulation adapted from Bartels and Urminsky
2011).
Then, all participants were asked to think of a shirt they currently own but no longer use
that “has special meaning from an experience or event in your life when wearing the shirt”
Participants then indicated the shirt they were thinking of in one or two words. Afterwards, they
were asked to imagine cleaning out their closet, coming across this shirt, and realizing they have
not worn it in some time. In the memory preservation present conditions, they were told to
imagine they laid their shirt on their bed and took a picture of it. They then briefly described
what they will think about when they look at this picture of the shirt. In the memory preservation
absent condition, participants described what they think about when they see this shirt (see Web
Appendix for complete stimuli).
Participants then indicated, “How likely would you be to donate this shirt to a local
nonprofit organization?” on a seven-point scale (1 = very unlikely to 7 = very likely). They then
responded to the two sentimental value items (α = .87), two difficulty items (α = .69), and
27
measure of resale value used in prior studies. Product value averaged $27.41 (SD = $81.93). As a
manipulation check, participants indicated their perceived level of connectedness to their future
self from a set of six images of overlapping circles from Bartels and Urminsky (2011).
Results and Discussion
Manipulation Check. We conducted an ANOVA with memory preservation,
connectedness to future self condition (high vs. low), and their interaction as the independent
variables and the measure of perceived connectedness to one’s future self as the dependent
variable. Participants felt more connected to their future self in the connectedness condition (F(1,
190) = 20.76, p < .0001; Mhigh connectedness = 4.22, Mlow connectedness = 3.53). Neither memory
preservation nor the interaction were significant predictors of perceived connectedness to one’s
future self (ps > .20).
Donation Likelihood. An ANOVA with memory preservation (present vs. absent),
connectedness to future self condition (high vs. low), and their interaction as the independent
variables and donation likelihood as the dependent variable revealed no main effects (ps > .30).
The analysis of interest, however, was the interaction of memory preservation x connectedness to
future self, which was significant and in the predicted direction (F(1, 200) = 4.93, p = .03). When
connectedness to the future self was high, participants had greater donation likelihood when
memory preservation was present than when it was absent (Mmemory preservation present= 2.85 (2.02)
vs. Mmemory preservation absent = 2.10 (1.34), t(200) = 2.04, p = .05). In contrast, when connectedness
to the future self was low, memory preservation did not increase donation likelihood (Mmemory
preservation present = 2.55 (1.84) vs. Mmemory preservation absent = 2.94 (1.79), t(200) = -1.09, p = .28).
Likewise, when memory preservation was absent, participants in the high connectedness to
future self condition reported lower donation likelihood than those in the low connectedness to
28
future self condition (Mhigh connectedness = 2.10, Mlow connectedness = 2.94, t(200) = 2.31, p = .02).
When memory preservation was present, there was no significant difference in donation
likelihood between the high and low connectedness to the future self conditions (Mhigh connectedness
= 2.85 vs. Mlow connectedness = 2.55, t(200) = .83, p = .41). See Figure 4.
---insert Figure 4 about here---
Discussion. Study 5 underscores the role of memories and identity as the locus of the
memory preservation effect. When consumers perceive their future self will be different from
their current self, they are equally likely to donate regardless of memory preservation. This
pattern provides further support for our theorizing that memory preservation operates by
reinforcing the product-relevant identities one might otherwise lose when donating a possession
with sentimental value; low connectedness to the future self indicates the product-relevant
identity will not be as meaningful in the future and therefore does not require protecting prior to
disposition.
Study 6
This final study tests the moderating role of disposition type (selling vs. donating) on the
effect of memory preservation on disposition likelihood to establish that memory preservation
does not extend to selling because an economic transaction taints a sentimental good.
Participants and Procedure
A total of 110 U.S. adults from MTurk completed the study for a small payment (47%
female, 7 unspecified; Mage = 37.12, SD = 13.45). Participants were randomly assigned to one of
four conditions in a 2 (memory preservation: present vs. absent) x 2 (disposition type: donate vs.
29
sell) between-subjects design. First, all participants were asked to think of an item in their home
that is meaningful to them which they no longer use but could be of use to someone else. They
briefly indicated the item by completing the statement, “The special possession I am thinking of
is…” Then, participants in the memory preservation condition were instructed to do the
following: “Please take a moment to go get this item in your home and take a picture of it. You
can use your phone, a camera, or the web cam on your computer to take the picture -- just use
whatever is most convenient. We will ask you about the picture after you take it.” Those in the
memory preservation absent condition were instructed: “Please take a moment to go get this item
in your home.” This memory preservation manipulation is conservative because it asks
participants to take a picture but does not mention the purpose of the picture is to retain
memories. We also note that we were concerned the procedure had the potential to produce
noisier data than the typical study. We therefore conducted outlier analysis in which we
identified two outliers who were removed from the data set; all subsequent analysis is reported
with the 108 remaining participants.3
Consistent with Study 3, all participants then imagined cleaning out their home and
coming across this item and then were asked, “How likely would you be to donate (sell) this
item?” depending on disposition type condition (from 1-very unlikely to 7-very likely). They
then responded to the two sentimental value items (α = .90), two difficulty items (α = .62), and
measure of resale value used in prior studies. Product value averaged $52.17 (SD = $99.58).
Results and Discussion

3 We conducted outlier analysis using Cook’s D. Using the conservative cut-off value of 4/n (Bollen and Jackman
1985), we identified and removed two outliers. Retaining these participants reduces the interaction to non-
significance (p = .19), but the pattern remains. We conducted the same outlier analysis for other studies, but all other
studies either did not have outliers or results were not substantially altered by the removal of outliers. Thus, no
outliers are removed in the reported results for the other studies.
30
Disposition Likelihood. Neither memory preservation nor disposition type predicted
disposition likelihood (ps > .30). The analysis of interest, the interaction of memory preservation
x disposition type, was significant and in the predicted direction (F(1, 104) = 4.00, p = .05).
When considering donation, participants had greater disposition likelihood when memory
preservation was present than when it was absent (Mpresent= 2.93 (1.85) vs. Mabsent = 2.04 (1.36),
t(104) = 2.11, p = .04), mirroring our previous studies. In contrast, when considering selling,
memory preservation did not increase donation likelihood (Mpresent = 2.07 (1.28) vs. Mabsent = 2.35
(1.44), t(104) = -.69, p = .49). See Figure 5.
---insert Figure 5 about here---
Discussion. Study 6 affirms that, as expected, memory preservation is only effective at
increasing disposition of possessions with sentimental value when considering donating them,
not selling them. We discuss selling further in the General Discussion.
General Discussion
Second-hand goods play a large role in the marketplace for a variety of organizations
such as nonprofits. A primary problem with the business model of these organizations is the
volatility of the supply chain because they are dependent on consumer’s disposition of their
possessions. Aiding consumer disposition not only benefits organizations dependent on second-
hand goods but also benefits consumers who face financial and/or psychological costs from
retaining unused goods as well as society at large. In our work we provide insight on consumer
disposition, which should help managers dependent on second-hand goods better understand
their supply chain and, more specifically, use memory preservation to increase donation rates.
31
For example, we demonstrate that implementing a simple memory preservation strategy in the
field substantially increased the number of donations by 15%. When product disposition rates are
likely to be contingent on a number of factors such as effort/convenience, monetary value of the
item, and potential for reuse (Bayus 1991; Haws et al. 2012; Jacoby, Berning, and Dietvorst
1977; Lee et al. 2015; Okada 2001), being able to increase donations to this extent is notable. In
addition to providing an effective strategy nonprofit marketers can use to increase donations, we
also explicated why it worked, so that future researchers can expand upon it and perhaps use the
psychological drivers of the effect to develop other effective promotional tactics.
Theoretical Contributions
Although relevant to a number of market enterprises, including recycling, storage, resale,
and the product life cycle, product disposition has not been as widely researched as acquisition
and consumption. Though some disposition literature examines the transfer of meaning to the
new recipient (Brough and Isaac 2012; Lastovicka and Fernandez 2005; Price, Arnould, and
Curasi 2000), requiring knowledge of the recipient for disposition is not feasible in most
donation contexts. The current research contributes to the disposition literature by focusing on
overcoming the reluctance to part with possessions due to sentimental value without relying on
transferring the value to a known recipient during or after disposition. Our research therefore
contributes to the literature exploring product disposition by not only exploring the important
context of donation of used goods but also is the first that we know of to try to ameliorate the
identity loss that stems from donating a good with sentimental value through simple memory
preservation techniques aimed at the donor. By focusing on preserving the memories rather than
transferring the sentimental value, this disposition aid can be used more generally rather than in
limited contexts when the recipient is known.
32
Our focus is on memory preservation to aid donation though we also consider the extent
to which this applies to selling. In doing so, we contribute to disposition research exploring
disparities in willingness to accept and willingness to pay prices for sellers considering goods
with sentimental value and potential buyers. This literature finds consumers typically become
upset at the thought of selling and demand exorbitant prices for their possessions with
sentimental value that buyers may be unwilling to pay, potentially resulting in the owner not
giving up the good at all (Boyce et al. 1992; Chatterjee, Irmak, and Rose 2013). The reason for
this refusal appears to be a combination of loss aversion (Kahneman, Knetsch, and Thaler 1991),
which applies to all possessions, along with an aversion to thinking of goods with sentimental
value in monetary terms because of the taboo of applying money to certain sentiments (McGraw
and Tetlock 2005). Selling seems to be especially susceptible to emotional taboo responses to
monetary valuation (Boyce et al. 1992; Chapman and Johnson 1995; Irwin 1994). Thus, memory
preservation may not aid disposition via selling since the mitigated identity loss that allows for
donation does not overcome the taboo association of the good and memory with money that
arises from selling. Indeed, our last study affirms the difference in selling and donating for goods
with sentimental value. Even when memories represented by a good are preserved and identity
loss is mitigated such that the sentimental good can be disposed of, the economic exchange
associated with selling the sentimental goods limits disposition likelihood. While the current
work provides evidence of how barriers to donation due to sentimental value can be overcome,
future research can explore specific strategies that may encourage the selling of sentimental
goods to unknown others.
Our work also contributes to the literature on consumer memory. Only recently have
consumer psychologists started to empirically examine the link between memory and identity in
33
marketing contexts. Mercurio and Forehand (2011) found that consumer identity can influence
learning through memory by showing that, when ad content is at least moderately related to a
specific identity (i.e., gender) and that identity is activated at the time of content retrieval, there
is a significant increase in subsequent recognition of ad content. Dalton and Huang (2014) found
that viewing identity-relevant promotions and subsequently experiencing a threat to the relevant
identity results in strategically forgetting information regarding that identity (i.e., lower
recognition of previously viewed identity-relevant information) as a defense mechanism to cope
with the identity threat. Note that although this past research documents a general link between
memory and identity in consumption contexts, the memories explored were about marketing
information (i.e., advertisements, promotions). This work did not examine autobiographical
memories that were specific to an individual. The memories in these studies were therefore not
characterized by a sense of “mineness” (Klein and Nichols 2012), unlike memories that give
possessions sentimental value. Along with Zauberman, Ratner, and Kim’s (2009) work on
memory, ours is among the first research to document strategic protection of consumer-specific
consumption memories.
Relatedly, these findings contribute to the literature on consumer identity. Identity
research typically involves temporarily activating the salience of a particular identity. In
demonstrating a method to prevent identity loss without memory preservation (Study 5), we do
not alter the salience of the identity; instead, we have consumers consider other ways they “live”
this identity, thereby reinforcing the identity. Identity reinforcement may be an important factor
to consider in understanding consumer evaluations of identity-linked goods and other identity-
relevant consumer decisions (Reed 2004; Reed, Forehand, Puntoni, and Warlop 2012).
Additionally, we consider the role of a static versus dynamic identity by exploring connectedness
34
between one’s current and future self (Bartels and Urminksy 2011; Hershfield 2011). By
demonstrating that perceptions of how related one’s current self is to one’s future self can
influence disposition of identity-relevant goods, we provide insight into how connectedness to
one’s future self not only impacts intertemporal preferences (which is how it has typically been
studied in the marketing literature) but also identity-relevant behavior and disposition.
Lastly, this research contributes more generally to the retailing literature. Though there
has been due attention to factors in sourcing goods from manufacturers (Basuroy, Mantrala, and
Walters 2001; Mantrala and Raman 1999), the unique challenge that nonprofits face when
sourcing goods from consumers could benefit from more attention in the literature. The current
research takes a step in this direction by recognizing that a potential barrier to consumer
donations is their perceived identity loss when parting with possessions. When consumers realize
the emotionally significant memories tied to sentimental goods can be preserved, they are less
reluctant to begin the donation process. Understanding sourcing of used goods from consumers is
not only important to nonprofits but is likely to become increasingly important for for-profit
social ventures and other sustainable business models dependent on used goods for reuse and
repurposing.
Practical Implications
The current research has a managerial aim: to help nonprofits and other organizations
understand consumer reluctance to dispose of goods with sentimental value and, in doing so,
identify tactics managers can use to increase consumer donations. In Study 2, we found the two
main reasons consumers do not dispose of goods with sentimental value is due to the memories
and identities rather than the financial value or likelihood of reuse. Thus, We
Consistent with these reasons for not donating, we have shown that simple memory
35
preservation tactics significantly increase donations. Note that simply promoting a memory
preservation technique by suggesting consumers take pictures of sentimental goods prior to
donation effectively increased donations in Study 1. Thus, this tactic is an easy mechanism for
managers to use in promotional campaigns designed to elicit donations. The field study reported
in the discussion of Study 3 demonstrates that memory preservation strategies (i.e., taking a
photo of sentimental items) can also be used at the time of donation to reduce felt identity loss
when donating, potentially increasing the likelihood of future donations to the same nonprofit.
Given charity retailers are in competition for merchandise (Hibbert, Horne, and Tagg 2005;
Paden and Stell 2005), incorporating memory preservation into both donation appeals and/or the
actual donation process should substantially increase donations.
We therefore recommend at least two actions to nonprofits seeking to increase donations:
(1) Hold a donation drive in which all promotional material specifically recommends that
consumers take photos of possessions prior to donation so they can “keep the memories but lose
the clutter” (similar to that used in Study 1) and (2) Either in conjunction with a specific
donation drive or year-round, offer photo taking of donations at key drop-off centers during
business hours in order to decrease felt identity loss at time of donations. These two actions
should serve to increase donations during the campaign and train donors to engage in their own
memory preservation activities so that donation of possessions with sentimental value becomes
easier over time.
In addition, the links that this research provides should help managers develop other
mechanisms for increasing donation of goods with sentimental value in specific domains. For
example, the results of Study 4 suggest that shoring up consumers’ identities might make them
more immune to worrying about losing the memories that goods with sentimental value hold.
36
Our theoretical framework suggests that other memory preservation techniques than the ones we
tested would work as well, including industry- or category-specific techniques. Graduating
college seniors might donate sentimental college merchandise, for example, if their name was
added to a permanent wall on campus or if they were invited to participate in a special event
reinforcing their identity as soon to be alums. In addition, the results of Study 5 support the
notion that emphasizing a disconnect between current and future selves should reduce fear of
identity loss and increase donation likelihood. For instance, advertising could show empty
nesters enjoying travel and the simplicity of a downsized home as they consider donating items
used for their children.
This research also has practical implications for consumer welfare. By showing how to
decrease the identity loss associated with donating possessions with sentimental value, these
findings can aid consumers who are overcome with clutter, an increasing consumer problem
(Teitell 2012), and/or who may be forced to dispose of some possessions during life transitions
(e.g., downsizing to a smaller home, divorce, retirement, etc.; McAlexander 1991; Mehta and
Belk 1991; Young 1991). When disposition is necessary, consumer welfare can be enhanced by
engaging in memory preservation. Consumers struggling to part with goods with sentimental
value but wanting to minimize the clutter in their home or storage areas and give new life to their
possessions should take a photo, write a note, or otherwise document their memories associated
with the good so they can part with the good without losing their identity. The proliferation of
popular press books (e.g., the 2014 best seller, The Life-Changing Magic of Tidying Up: The
Japanese Art of Decluttering and Organizing by Marie Kondo) aimed at helping consumers to
de-clutter suggests that these recommendations could significantly enhance consumer welfare. In
addition, the companies tasked with helping people dispose of goods, such as estate sale
37
companies that sell an estate for charity, might profitably use these techniques to help their
consumers.
Limitations and Directions for Future Research
One limitation of the current research is that our laboratory studies involve hypothetical
donation. We do report the results of two field studies (one in the context of a donation drive
suggesting photos for memory preservation and the other exploring the effect of taking a photo at
the time of donation). However, we were not able to code the donations we received in these
field studies as having or lacking sentimental value. Another limitation of the present studies is
that the consumers were either asked to engage in the memory preservation themselves (e.g.,
take a photo) or were able to keep the memory preservation aid (e.g., they were given a photo
taken by the charity). Thus, these findings cannot speak to how important it is that the consumer
retains the memory preservation aid. In other words, we do not yet know if consumers need to be
able to look at the photo again after donation for it to be effective at increasing
donations/donation intentions. Additionally, the current research does not verify if the consumer
must engage in a physical act of memory preservation such as taking a photo or if it is enough
just to be cued to engage in memory preservation. Perhaps cueing memory preservation helps the
consumer remember that the good is a physical representation of an emotionally significant
memory even if no physical act of memory preservation is taken. Although it does not directly
address this question, Study 4 shows that reinforcing the identity related to a product with
sentimental value in an alternate way can also increase donation likelihood. Thus, thinking about
taking a picture to capture the memories the product represents may be enough to reinforce the
identity by encouraging consumers to actively rehearse their product-relevant memories. This
possibility is especially attractive for marketers because it implies that advertising memory
38
preservation techniques may prove effective regardless of whether a consumer physically
engages in the techniques.
Another concern, particularly for nonprofits seeking to resell donated goods, may be
whether the donated goods are of high resale value. In Study 2, which compared both
sentimental and non-sentimental goods, we did not find a difference in self-reported market
value. However, although high resale value and sentimental value are not always correlated,
many unused but undonated possessions may be both, as sentimental items may be kept
purposefully in good condition. Moreover, even if donated goods are not of high resale value,
increasing consumers’ likelihood of returning them to the second-hand marketplace is still
valuable since many goods such as textiles and electronics can be recycled, which increases the
organization’s revenues (Ewoldt 2015; Steeves 2016). Further, a reluctance to part with
possessions with sentimental value even if not of particularly high resale value may well be the
barrier that prevents consumers from starting the process of identifying items to donate
(including those with high resale value). As such, charities that might not benefit directly from
the donation of goods with high sentimental value but low resale value may still find themselves
benefitting from holding photo donation drives, as this type of donation drive may increase
overall donations, including non-sentimental donations with high resale value. Future research
could explore this avenue more directly, as well as explore whether the donation of other types
of goods can be increased using memory preservation strategies (e.g., goods that are perceived as
being “one of a kind” or part of a collection).
In terms of future research, it would also be interesting to explore whether consumers
experience more commitment to and an increased desire to donate to nonprofits explicitly
engaging in or encouraging memory preservation compared to those that do not. It is possible
39
that assuaging worries about identity loss endears the nonprofit to consumers, resulting in brand
loyalty and more repeat donations. Managers of nonprofits with a specific mission might also
wish to know whether an identity match between the consumer and the mission of the charity
might help mitigate identity loss in donating goods with sentimental value. For example, if a
particular good is linked with memories and identities associated with being a successful
business person, and the charity’s mission is specifically to help individuals develop business
acumen (e.g., Dress for Success), might this connection between identities (even though it is not
tied to a specific recipient) be enough to mitigate identity loss in a similar manner to the
operation of memory preservation, therefore increasing donations?
Finally, we speculate that memory preservation could be effective in increasing disposal
via other methods besides giving to nonprofits, as long as that method does not involve a
monetary exchange. For example, it would be interesting to know whether memory preservation
increases social recycling by giving it to strangers in one’s community (Donnelly et al. 2017).
We leave this to future researchers to explore.
40
REFERENCES
Aggarwal, Pankaj and Meng Zhang (2006), “The Moderating Effect of Relationship Norm
Salience on Consumers’ Loss Aversion,” Journal of Consumer Research, 33 (3), 413–19.
Ahuvia, Aaron C. (2005), “Beyond the Extended Self: Loved Objects and Consumers’ Identity
Narratives,” Journal of Consumer Research, 32 (1), 171–84.
Anik, Lalin, Michael I. Norton, and Dan Ariely (2014), “Contingent Match Incentives Increase
Donations,” Journal of Marketing Research, 51 (December), 790-801.
Bartels, Daniel M. and Oleg Urminsky (2011), “On Intertemporal Selfishness: How the
Perceived Instability of Identity Underlies Impatient Consumption,” Journal of Consumer
Research, 38 (1), 182–98.
Basuroy, Suman, Murali K. Mantrala, and Rockney G. Walters (2001), “The Impact of Category
Management on Retailer Prices and Performance: Theory and Evidence,” Journal of
Marketing, 65 (4), 16–32.
Bayus, Larry L. (1991), “The Consumer Durable Replacement Buyer,” Journal of Marketing, 55
(1), 42–51.
Belk, Russell W. (1988), “Possessions and the Extended Self,” Journal of Consumer Research,
15 (September), 139-68.
Bollen, Kenneth A. and Robert W. Jackman (1985), “Regression Diagnostics: An Expository
Treatment of Outliers and Influential Cases,” Sociological Methods & Research, 13 (4),
510–42.
Boyce, Rebecca R., Thomas C. Brown, Gary H. McClelland, George L. Peterson, and William
D. Schulze (1992), “An Experimental Examination of Intrinsic Values as a Source of the
WTA-WTP Disparity,” The American Economic Review, 82 (December), 1366-73.
Brosius, Nina, Karen V. Fernandez, and Hélène Cherrier (2013), “Reacquiring Consumer Waste:
Treasure in Our Trash?,” Journal of Public Policy & Marketing, 32 (Fall), 286-301.
Brough, Aaron R. and Mathew S. Isaac (2012), “Finding a Home for Products We Love: How
Buyer Usage Intent Affects the Pricing of Used Goods,” Journal of Marketing, 76 (July),
78-91.
Burris, Christopher T. and John K. Rempel (2004), “‘It’s the End of the World as We Know It’:
Threat and the Spatial-Symbolic Self,” Journal of Personality and Social Psychology, 86
(1), 19–42.
41
Business Wire (2007), “Average U.S. Household Has Over 50 Unused Items Worth $3100
According to eBay/Nielsen Survey,” (accessed November 1, 2014), [available at
http://www.businesswire.com/news/ home/20070426005614/en/Average-U.S.-
Household-50-Unused-Items-Worth#.VFJC46PD9kQ].
Chapman, Gretchen B. and Eric J. Johnson (1995), “Preference Reversals in Monetary and Life
Expectancy Evaluations,” Organizational Behavior and Human Decision Processes, 62
(3), 300–317.
Chatterjee, Promothesh, Caglar Irmak, and Randall L. Rose (2013), “The Endowment Effect as
Self-Enhancement in Response to Threat,” Journal of Consumer Research, 40 (3), 460–
76.
Chernev, Alexander, Ryan Hamilton, and David Gal (2011), “Competing for Consumer Identity:
Limits to Self-Expression and the Perils of Lifestyle Branding,” Journal of Marketing, 75
(3), 66–82.
Clark, Margaret S., Judson Mills, and Martha C. Powell (1986), “Keeping Track of Needs in
Communal and Exchange Relationships,” Journal of Personality and Social Psychology,
51 (2), 333–38.
Csikszentmihalyi, Mihaly and Eugene Halton (1981). The Meaning of Things: Domestic Symbols
and The Self. Cambridge University Press.
Dalton, Amy N. and Li Huang (2014), “Motivated Forgetting in Response to Social Identity
Threat,” Journal of Consumer Research, 40 (6), 1017–38.
Donnelly, Grant E., Cait Lamberton, Rebecca Walker Reczek, and Michael I. Norton (2016),
“Social Recycling Transforms Unwanted Goods into Happiness,” Journal of the
Association for Consumer Research, 2 (1), 48–63.
Epp, Amber M. and Linda L. Price (2010), “The Storied Life of Singularized Objects: Forces of
Agency and Network Transformation,” Journal of Consumer Research, 36 (5), 820–37.
Ewoldt, John (2015), “Thrift Stores Make More Revenue Being Less Choosy,” StarTribue,
(accessed March 31, 2017) [available at http://www.startribune.com/thrift-stores-make-
more-revenue-being-less-choosy/301292401/].
Ferraro, Rosellina, Jennifer Edson Escalas, and James R. Bettman (2011), “Our Possessions, Our
Selves: Domains of Self-worth and the Possession–Self link,” Journal of Consumer
Psychology, 21 (2), 169-77.
Forbes (2015), “The 50 Largest U.S. Charities,” edited by William Barrett and Janet Novack,
(accessed July 21, 2016), [available at http://www.forbes.com/top-charities/].
42
Forehand, Mark R. and Rohit Deshpande (2001), “What We See Makes Us Who We Are:
Priming Ethnic Self-awareness and Advertising Response,” Journal of Marketing
Research, 38 (3), 336–48.
Franke, Nikolaus, Martin Schreier, and Ulrike Kaiser (2010), “The ‘I Designed It Myself’ Effect
in Mass Customization,” Management Science, 56 (1), 125–40.
Gao, Leilei, S. Christian Wheeler, and Baba Shiv (2009), “The ‘Shaken Self’: Product Choices
as a Means of Restoring SelfView Confidence,” Journal of Consumer Research, 36 (1),
29–38.
Goodwill (2014), “About Us,” (accessed November 13, 2014), [available at
http://www.goodwill.org/about-us/].
Haws, Kelly L., Rebecca Walker Naylor, Robin A. Coulter, and William O. Bearden (2012),
“Keeping It All without Being Buried Alive: Understanding Product Retention
Tendency,” Journal of Consumer Psychology, 22 (April), 224-326.
Hayes, Andrew F. (2012), An Introduction to Mediation, Moderation, and Conditional Process
Analysis: A Regression-Based Approach, New York: Guilford Press.
Hershfield, Hal E. (2011), “Future self-continuity: How Conceptions of the Future Self
Transform Intertemporal Choice,” Annals of the New York Academy of Sciences, 1235 (1),
30–43.
Hibbert, Sally A., Suzanne Horne, and Stephen Tagg (2005), “Charity Retailers in Competition
for Merchandise: Examining How Consumers Dispose of Used Goods,” Journal of
Business Research, Special Section: The Nonprofit Marketing Landscape, 58 (6), 819–
28.
Holbrook, Morris B. (1994), “The Nature of Customer Value: An Axiology of Services in the
Consumption Experience,” in Service Quality: New Directions in Theory and Practice,
Roland Rust and Richard L. Oliver, eds. Newbury Park, CA: Sage, 21-71.
Irwin, Julie R. (1994), “Buying/Selling Price Preference Reversals: Preference for
Environmental Changes in Buying versus Selling Modes,” Organizational Behavior and
Human Decision Processes, 60 (3), 431–57.
Jacoby, Jacob, Carol K. Berning, and Thomas F. Dietvorst (1977), “What about Disposition?”
Journal of Marketing, 41 (April), 22-28.
Kahneman, Daniel, Jack L. Knetsch, and Richard H. Thaler (1991), “Anomalies: The
Endowment Effect, Loss Aversion, and Status Quo Bias,” The Journal of Economic
Perspectives (1986-1998); Nashville, 5 (1), 193.
43
Kihlstrom, John Fr. Jennifer S. Beer, and Stanley B. Klein (2002), “Self and Identity as
Memory,” in Handbook of Self and Identity, M. R. Leary and J. Tangney, eds., New
York: Guildford Press, 68-80.
Klein, Stanley B. and Shaun Nichols (2012), “Memory and the Sense of Personal Identity,”
Mind, 121 (483), 677–702.
Kondo, Marie (2014), The Life-Changing Magic of Tidying Up: The Japanese Art of
Decluttering and Organizing. Ten Speed Press.
Lamberton, Cait Poynor and Randall L. Rose (2012), “Yours, Mine and Ours: An Investigation
of Consumers’ Propensity to Participate in Commercial Sharing Systems,” Journal of
Marketing, 76 (July), 109-25.
Lastovicka, John L. and Karen V. Fernandez (2005), “Three Paths to Disposition: The
Movement of Possessions with sentimental value to Strangers,” Journal of Consumer
Research, 31, (March), 813-23.
Lee, Lichang, Jane Allyn Piliavin, and Vaughn R. A. Call (1999), “Giving Time, Money, and
Blood: Similarities and Differences,” Social Psychology Quarterly, 62(3), 276–90.
Lee, Miah, Eunju Ko, Seulgi Lee, and Kyulim Kim (2015), “Understanding Luxury
Disposition,” Psychology & Marketing, 32 (4), 467–80.
Liu, Wendy and Jennifer Aaker (2008), “The Happiness of Giving: The TimeAsk Effect,”
Journal of Consumer Research, 35(3), 543–57.
Loewenstein, George and Samuel Issacharoff (1994), “Source Dependence in the Valuation of
Objects,” Journal of Behavioral Decision Making, 7 (3), 157–68.
Mantrala, Murali K. and Kalyan Raman (1999), “Demand Uncertainty and Supplier’s Returns
Policies for a Multi-store Style-good Retailer,” European Journal of Operational
Research, 115 (2), 270–84.
McAlexander, James H. (1991), “Divorce, the Disposition of the Relationship, and Everything”,
in North American Advances in Consumer Research, Vol. 18, Rebecca H. Holman and
Michael R. Solomon, eds., Provo, UT: Association for Consumer Research, 43-48.
McGraw, A. Peter and Philip E. Tetlock (2005), “Taboo Trade-offs, Relational Framing, and the
Acceptability of Exchanges,” Journal of Consumer Psychology, 15 (1), 2–15.
Mehta, Raj and Russell W. Belk (1991), “Artifacts, Identity and Transition: Favorite Possessions
of Indians and Indian Immigrants to the United States,” Journal of Consumer Research,
17 (March), 398–411.
44
Mercurio, Kathryn R. and Mark R. Forehand (2011), “An Interpretive Frame Model of Identity-
Dependent Learning: The Moderating Role of Content-State Association,” Journal of
Consumer Research, 38 (October), 555–77.
Norton, Michael I., Daniel Mochon, and Dan Ariely (2012), “The IKEA Effect: When Labor
Leads to Love,” Journal of Consumer Psychology, 22 (3), 453–60.
Okada, Erica Mina (2001), “Trade-ins, Mental Accounting, and Product Replacement
Decisions,” Journal of Consumer Research, 27 (4), 433–46.
Oyserman, Daphna (2009), “Identity-based Motivation: Implications for Action-Readiness,
Procedural-readiness, and Consumer Behavior,” Journal of Consumer Psychology, 19
(3), 250–60.
Paden, Nita and Roxanne Stell (2005), “Consumer Product Redistribution,” Journal of
Marketing Channels, 12 (3), 105–23.
Parfit, Derek (1987), Reasons and Persons, Clarendon Press, Oxford.
Price, Linda L., Eric J. Arnould, and Carolyn Folkman Curasi (2000), “Older Consumers’
Disposition of Possessions with Sentimental Value,” Journal of Consumer Research, 27
(September), 179-201.
Reed II, Americus (2004), “Activating the SelfImportance of Consumer Selves: Exploring
Identity Salience Effects on Judgments,” Journal of Consumer Research, 31 (2), 286–95.
Reed II, Americus and Mark R Forehand (2016), “The Ebb and Flow of Consumer Identities:
The Role of Memory, Emotions and Threats,” Current Opinion in Psychology, Consumer
Behavior, 10, 94–100.
Reed II, Americus, Mark R. Forehand, Stefano Puntoni, and Luk Warlop (2012), “Identity-based
Consumer Behavior,” International Journal of Research in Marketing, Special Issue on
Consumer Identities, 29 (4), 310–21.
Savary, Jennifer, Kelly Goldsmith, and Ravi Dhar (2015), “Giving Against the Odds: When
Tempting Alternatives Increase Willingness to Donate,” Journal of Marketing Research,
52 (1), 27-38.
Sivadas, Eugene, and Ravi Venkatesh (1995), “An Examination of Individual and Object-
Specific Influences on the Extended Self and its Relation to Attachment and
Satisfaction,” in North American Advances in Consumer Research, Vol. 22, Frank R.
Kardes and Mita Sujan, eds., Provo, UT: Association for Consumer Research, 406-12.
Small, Deborah A. and Nicole M. Verrochi (2009), “The Face of Need: Facial Emotion
Expression on Charity Advertisements,” Journal of Marketing Research, 46 (December),
777-87.
45
Smith, Robert, David Faro, and Katherine Burson (2013) “More for the Many: The Influence of
Entitativity on Charitable Giving,” Journal of Consumer Research, 39 (5), 961-76.
Spencer, Steven J., Mark P. Zanna, and Geoffrey T. Fong (2005), “Establishing a Causal Chain:
Why Experiments Are Often More Effective Than Mediational Analyses in Examining
Psychological Processes,” Journal of Personality and Social Psychology, 89 (6), 845–51.
Steeves, Heather (2016), “9 Things You Didn’t Know You Could Donate to Goodwill,”
Goodwill Northern New England, (accessed March 31, 2017) [available at
https://goodwillnne.org/articles/9-things-you-didnt-know-you-could-donate-to-
goodwill/].
Strahilevitz, Michal A. and George Lowenstein (1998), “The Effect of Ownership History on the
Valuation of Objects,” Journal of Consumer Research, 25 (December), 276-98.
Sujan, Mita, James R. Bettman, and Hans Baumgartner (1993), “Influencing Consumer
Judgments Using Autobiographical Memories: A Self-Referencing Perspective,” Journal
of Marketing Research, 30 (4), 422–36.
Teitell, Beth (2012), “Boxed In, Wanting Out,” The Boston Globe, (accessed November 3, 2014)
[available at http://www.boston.com/community/moms/articles/2012/07/10
/new_Study_says_american_families_are_overwhelmed_by_clutter_rarely_eat_together_
and_are_generally_stressed_out_about_it_all/].
Trudel, Remi and Jennifer J. Argo (2013), “The Effect of Product Size and Form Distortion on
Consumer Recycling Behavior,” Journal of Consumer Research, 40 (4), 632–43.
———, ———, and Matthew D. Meng (2016), “The Recycled Self: Consumers’ Disposal
Decisions of Identity-Linked Products,” Journal of Consumer Research, forthcoming.
Winterich, Karen Page, Vikas Mittal, and Karl Aquino (2013), “When Does Recognition
Increase Charitable Behavior? Toward a Moral Identity-Based Model,” Journal of
Marketing, 77 (3), 121–34.
Young, Melissa Martin (1991), “Disposition of Possessions During Role Transitions,” in North
American Advances in Consumer Research, Vol.18, Rebecca H. Holman and Michael R.
Solomon, eds. Provo, UT: Association for Consumer Research, 33-39.
Zauberman, Gal, Rebecca K. Ratner, and B. Kyu Kim (2009), “Memories as Assets: Strategic
Memory Protection in Choice over Time,” Journal of Consumer Research, 35 (February),
715-28.
46
TABLE 1:
REASONS FOR NOT DISPOSING OF A POSSESSION BY POSSESSION TYPE (STUDY 2)
Thought type Thoughts
Sentimental
Value
No
Sentimental
Value t-value
Remaining value
(e.g., still works,
could be used
again)
“Because it still works perfectly so there is no reason to get
rid of it. Some younger nephews or cousins might enjoy
playing it if they are ever over, so I see no reason to dispose
of it.” .33
(.48)
.67
(.48) -3.14*
The shoes are pretty uncomfortable, but I did not dispose
them because I paid money for them and might use them
once again.”
Inconvenient
disposal
(e.g., don’t know
where/how to
dispose, haven’t
gotten around to it)
“I still have the product because out of convenience. I
haven't put any time or energy into selling or donating it.”
.05
(.22)
.26
(.45) -2.72*
“I still have it because I keep forgetting I have it. When I do
remember it is in the house, I cannot find a place that will
take it.”
Product memories
(i.e., brings back
memories)
“I keep it because I am holding onto the memories. I don't
want to throw it away because I would be tossing my
memories along with it.” .67
(.48)
.10
(.30) 6.41*
“It is sentimental and provides memories, but I no longer
use it. I can't bear the thought of getting rid of it because it
was meaningful to me.”
Identity relevance
(e.g., type of
person they are or
were)
“I have not disposed of it because it provides me insight as
to why I do what I do as a career as a chef. It brings me
back to a time of wonderment, which I still try to do today,
constantly learning and bettering myself in my culinary
trade.”
.49
(.51)
.14
(.35) 3.52*
“I did decide to stop using it this year. It is very heavy and
cumbersome to carry even though I love it and it has
sentimental value for me. I never wanted to part with it
because it was a very special gift given to me from my
mom and I cherish it. It is put away neatly in a bag in my
closet and I have never considered it parting with it since it
is so sentimental to me.” (references identity as a daughter)
Other “It's been buried in the cupboard, I actually haven't really
thought about it in quite some time.”
.26
(.44)
.26
(.45) .06
Total 3.10
(1.48)
3.02
(1.22) .26
Note: * p < .01, standard deviation is reported in parentheses.
47
FIGURE 1:
STUDY 1 PROMOTIONAL CAMPAIGN MESSAGE
MEMORY PRESERVATION ABSENT MEMORY PRESERVATION PRESENT
(CONTROL) CONDITION CONDITION
48
FIGURE 2:
POSSESSION TYPE AND MEMORY PRESERVATION ON DONATION LIKELIHOOD
AND IDENTITY LOSS (STUDY 3)
2.74
4.35
3.90
4.19
1
2
3
4
5
6
7
Sentimentalvalue Nosentimentalvalue
DonationLikelihood
PossessionType
MemoryPreservationAbsent
MemoryPreservationPresent
4.44
1.73
3.60
2.08
1
2
3
4
5
6
7
Sentimentalvalue Nosentimentalvalue
IdentityLoss
PossessionType
MemoryPreservationAbsent
MemoryPreservationPresent
49
FIGURE 3:
IDENTITY REINFORCEMENT AND MEMORY PRESERVATION ON DONATION
LIKELIHOOD (STUDY 4)
3.2
3.93
4.35
4.03
1
2
3
4
5
Absent Present
DonationLikelihood
IdentityReinforcement
MemoryPreservationAbsent
MemoryPreservationPresent
50
FIGURE 4:
CONNECTEDNESS TO FUTURE SELF AND MEMORY PRESERVATION ON DONATION
LIKELIHOOD (STUDY 5)
2.10
2.94
2.85
2.55
1
2
3
4
5
HighConnectedness LowConnectedness
DonationLikelihood
MemoryPreservationAbsent
MemoryPreservationPresent
51
FIGURE 5:
EFFECTIVENESS OF MEMORY PRESERVATION BY DISPOSITION TYPE (STUDY 6)
2.04
2.35
2.93
2.07
1
2
3
4
5
Donating Selling
DispositionLikelihood
MemoryPreservationAbsent
MemoryPreservationPresent
52
Web Appendix
Stimuli
Note: In all stimuli displayed here, underlining is only used here to highlight differences across
conditions. The text was not underlined when viewed by participants.
STUDY 2 STIMULI
Sentimental Value Condition: We all own products or items that are special to us. Often, these
products represent a memory of a specific experience or event in your life or they have acquired
a special meaning to you through repeated use. Many times these are products that we no longer
use in any way. However, the product may be of use to someone else. We'd like you to think
about one specific product that is meaningful to you that you no longer use, but that could still be
of use to someone else (e.g., clothing item, an item from your or your child's childhood,
household good, etc.). The item does not need to be of high monetary value. Just something that
has some special meaning to you, but could also be used by someone else. You should currently
own the item, and it should have special meaning to you. When you have thought of the specific
product you have in mind, please describe it in the space below using only a word or phrase (i.e.,
you don't need to describe it right now, just tell us what it is). Special item: [open-ended]
Thinking about your selected product, please begin by writing down a description of the
product. Then, please describe how you acquired the product, what the product means to you,
and how the product makes you feel. If you can, please write your description so that someone
reading this might be able to vividly visualize the product, imagine acquiring the product, and
the way in which you, specifically, feel about the product and understand what the product
means to you.
No Sentimental Value Condition: We all own many products or items that do not have any
specific meaning to us. Sometimes, you may not even recall how you acquired the product.
Many times these products include those that we no longer use in any way. However, the product
may be of use to someone else. We'd like you to think about one specific product that is not
meaningful to you that you no longer use, but that could still be of use to someone else (e.g.,
clothing item, a children's item, household good, etc.). The item does not need to be of high
monetary value. Just something that does not have any meaning to you, but could be used by
someone else. You should currently own the item, and it should NOT have special meaning to
you. When you have thought of the specific product you have in mind, please describe it in the
space below using only a word or phrase (i.e., you don't need to describe it right now, just tell us
what it is). Non-special item: [open-ended]
Thinking about your selected product, please begin by writing down a description of the
product. Then, please describe any information you can recall about the product such as how you
acquired the product and how you have used the product. If you can, please write your
description so that someone reading this might be able to vividly visualize the product, imagine
acquiring the product, and the way in which you, specifically, have used the product.
53
STUDY 3 STIMULI
Sentimental Value Condition: We all own products or items that are special to us because they
have some kind of special meaning.
Many times these are products that we no longer use in
any way. However, the product may be of use
to someone else. We'd like you to think about
one specific product that is meaningful to you that you no longer use, but
that could still be of
use to someone else (e.g., clothing item, an item from your or your child's
childhood, household
good, etc.). The item does not need to be of high monetary value. Just something
that has some
special meaning to you but could also be used by someone else. You should currently own the
item, and it should have special meaning to you. When you have thought of the specific product
you have in mind, please write it in the space below
using only a word or phrase (i.e., you
don't need to describe it in detail, just tell us what it is). The special possession I am thinking of
is….
No Sentimental Value Condition: We all own products or items that do not have any specific
meaning to us. Many times these are products that we no longer use in any way. However, the
product may be of use to someone else. We'd like you to think about one specific product that is
NOT meaningful to you that you no longer use, but that could still be of use to someone else (e.g.,
clothing item, a children's item, household good, etc.). The item does not need to be of high
monetary value. Just something that does not have any meaning to you, but could be used by
someone else. You should currently own the item, and it should NOT have special meaning to
you. When you have thought of the specific product you have in mind, please write it in the space
below using only a word or phrase (i.e., you don't need to describe it in detail, just tell us what it
is). The possession I am thinking of is….
All participants then read the following: Now, please imagine that you are cleaning out your
home and need to get rid of some things. You come across this item and realize you should get
rid of it.
Memory Preservation Present Condition: Since you are considering giving up this possession,
you decide you would like to preserve your memory of this item so you can still recall this item
even if you no longer own it. For example, you might do so by taking a photo of the item and/or
writing about the item. Doing so may help you remember what the item was, what it looked like,
how you acquired it, how and/or why you used it, etc. Below, please describe in detail what you
would do (photo, journal) to preserve your memory of this item. Keep in mind that this
photo/journal is so you personally can remember the item. If you can, please write your
description so that someone reading this might be able to imagine the way in which you,
specifically, are preserving your memory of the item.
Memory Preservation Absent Condition: [Received no additional information].
54
STUDY 4 STIMULI
SENTIMENTAL VALUE STIMULI: Same as study 3 with the monetary value specified to be
less than $100.
All participants then read the following: With what identity do you think the special item you
have been thinking of is most associated (you don't need to describe it, just one or two words for
the identity the product represents)? By identity, we mean what you consider to be important to
who you are as an individual or as a group
member and how you define yourself. For example,
the shirt you wore when winning a high school
sporting event or that you received for running a
5k may be associated with your athletic identity, a shirt
from college may be associated with
your student, college, academic, or fraternity/sorority identity, or a
shirt purchased while on
family vacation or received as a gift from a sibling or parent may be associated with your family
identity. Other examples of identities include: your identity as someone who enjoys the outdoors,
your identity as someone who likes the finer things in life,
your identity as someone who loves
to travel, your identity as a sibling
your identity as a friend, your identity as a son or daughter.
The identity associated with this
product is...
Now, please imagine that you are cleaning out your home and need to get rid of some things.
You come across this item and realize you should get rid of it.
Memory Preservation Present Condition: Since you are considering giving up this item, you
decide you would like to preserve the memories associated with your special item so you can still
recall these memories even if you no longer own the possession. For example, you might do so
by taking a photo of the item and/or journaling about your memories associated with the item.
Doing so may help you remember the meaning of the item, how special the item was to you, and
the identity associated with it. Below, please describe in detail what you would do (photo,
journal) to preserve your memory of this item. Keep in mind that this photo/journal to preserve
the item is so you personally can remember the item and its meaning to you. If you can, please
write your description so that someone reading this might be able to imagine the way in which
you, specifically, are preserving your memories associated with this item.
Memory Preservation Absent Condition: [Received no additional information].
Identity Reinforcement Present Condition: You told us earlier that the product you thought of
represented this part of you: [identity indicated earlier]. There are lots of ways to express this part
of you that are not tied to physical possessions but more about actions and how you spend your
time. Please take a moment in the space below to describe all of the ways that you express this
identity that are NOT tied to a physical possession. (e.g., things you say, things you do, how you
spend your time, etc.).
Identity Reinforcement Absent Condition: [Received no additional information].

55
STUDY 5 STIMULI
High Connectedness Condition (from Bartels and Urmnisky 2011): Recent research has
found that one's personal identity is far more stable than most people realize. The characteristics
that make you the person you are--your personality, temperament, major likes and dislikes,
beliefs, values, ambitions, life goals, and ideals--are established early in life and fixed by the end
of adolescence. These facts have led experts to conclude that the person you are now (while in
college) is fundamentally the same person you will be even very late in life, and the data seem to
support this argument. Longitudinal studies show that 1, 10, 30, and 50 years down the road,
even people who undergo major life events report (as do the people who know them best) that
the traits that make up their personal identity remain extremely stable. Please write a one
sentence summary of the paragraph you just read on the line below.
Low Connectedness Condition (from Bartels and Urminsky 2011): While it is generally true
that the person who wakes up is never exactly the same person who went to sleep the night
before, recent research suggests that one's identity is far less stable than most people realize, and
that this is especially true over the college years. The characteristics that make you the person
you are right now--your personality, temperament, major likes and dislikes, beliefs, values,
ambitions, life goals, and ideals--are likely to change radically, even over the course of a few
months. Several studies conducted with young adults have found large fluctuations in these
important characteristics, leading experts to conclude that the person you are now (while in
college) is quite a different person than the person you will be in just three months, much less a
year. Please write a one sentence summary of the paragraph you just read on the line below.
All participants then read the following: In this study we will ask you to think about an item
that you currently own. Your responses to all questions will be confidential and there are no right
or wrong answers. We all own items that are special to us. Often, these items represent a
memory of a specific experience or event in your life. Many times we no longer use these items
in any way. However, the item may be of use to someone else. We would like you to think about
a shirt (a t-shirt or sweatshirt) in your home that is special to you which you no longer wear but
could still be of use to someone else. You should currently own the shirt, and it should have
special meaning to you. When you have thought of the specific shirt you have in mind, please
write it in the space below using only a word or phrase (i.e., you don't need to describe it in
detail, just tell us what shirt it is). The shirt I am thinking of is….
All participants then read: Now, please take a moment to imagine that you are cleaning out
your closet. You come across this shirt and realize you haven’t worn it in some time.
Memory Preservation Present: You lay the shirt on your bed and take a picture of it. Below,
please briefly describe what you will think about when you look at this picture of the shirt.
Memory Preservation Absent: Below, please briefly describe what you think about when you
see this shirt.
56
WEB APPENDIX TABLE 1: MANIPULATION CHECK ON POSSESSION SPECIALNESS
AND DIFFICULTY OF IDENTIFYING POSSESSION
Study Possession
type
Rated
Sentimental
Value of
Possession
Difference from
scale midpoint
Difficulty of
identifying
possession
Difference from
scale midpoint
2
No Sentimental
Value 2.99
(2.57)
t(41) = -5.08,
p < .0001
2.08
(1.24)
t(41) = -9.98,
p < .0001
Sentimental
Value Present 7.21
(1.73)
t(38) = 7.94,
p < .0001
2.07
(1.16)
t(38) = -10.39,
p < .0001
3
No Sentimental
Value 2.18
(1.57)
t(80) = -10.41,
p < .0001
2.93
(1.86)
t(80) = -5.21,
p < .0001
Sentimental
Value Present 5.64
(1.47)
t(69) = 9.28,
p < .0001
2.41
(1.60)
t(69) = -8.34,
p < .0001
4 Sentimental
Value Present 5.21
(1.45)
t(235) = 12.75,
p < .0001
4.13
(1.75)
t(235) = 1.17,
p = .24
5 Sentimental
Value Present 5.09
(1.58)
t(201) = 9.86,
p < .0001
3.72
(1.73)
t(201) = -2.27,
p = .024
6 Sentimental
Value Present 5.61
(1.52)
t(104) = 10.87,
p < .0001
3.22
(1.81)
t(104) = -4.40,
p < .0001
Note: Standard deviation is reported in parentheses. Scale midpoint was four (1-7 scale) for all
measures, except sentimental value was measured on a 1-9 scale in Study 2 so results are
reported for difference from scale midpoint of five. In all studies, neither memory preservation
nor other factors impacted sentimental value or difficulty of selecting an item.
WEB APPENDIX TABLE 2:
RESULTS WHEN POSSESSION VALUE IS INCLUDED AS A CONTROL VARIABLE
Study
Effect of possession value
as a control variable
Significance of focal effect when controlling
for possession value
3 F(1, 146) = 9.84, p < .01
Possession type x
memory preservation condition:
F(1, 146) = 5.65, p = .02
4 F(1, 223) = .13, p = .72
Identity reinforcement x
memory preservation condition:
F(1, 223) = 4.42, p = .04
5 F(1, 196) = .05, p = .82
Connectedness to future self x
memory preservation condition:
F(1, 196) = 4.27, p = .04
6 F(1, 100) =.05, p = .83
Disposition type x
memory preservation condition:
F(1, 100) = 2.99, p = .08
Note: In Study 3, the pattern of the significant effect of value is such that higher value
possessions have lower donation likelihood.
57
WEB APPENDIX REPLICATION OF FIELD STUDY (STUDY 1)
Participants and Procedure
A field study was conducted in a set of Penn State University residence halls at the end of
the spring semester. The university runs an annual donation drive (“Trash to Treasure”) to collect
a variety of items in each of their more than 60 residence halls during move-out week to support
local nonprofits via the United Way. Eight residence halls, housing a total of 859 students, were
selected for the study. These residence halls selected were located in the same area of campus,
housed coed (38.6% male), primarily upper-classman students (5% freshmen, 38% sophomores,
31% juniors, 26% seniors), and were roughly the same size (i.e., same number of beds, ranging
from 81-126 residents per building with a total 427 students across four buildings in the memory
preservation condition and 432 students across four buildings in the control condition).
Each of the residence halls were exposed to a promotional campaign encouraging
students to donate; half received a control campaign and half received a campaign encouraging
students to engage in memory preservation (i.e., take a photo of a good with sentimental value)
before donating. The promotions had two components: 1) a poster was hung on each floor’s
bulletin board the week prior to move-out and 2) at the beginning of move-out week, the housing
office sent each student an email regarding donation of sentimental goods. In the memory
preservation condition, the campaign stated “Don’t Pack up Your Sentimental Clutter…Just
Keep a Photo of it, Then Donate” whereas the control condition campaign stated “Don’t Pack up
Your Sentimental Clutter, Just Collect the Items, Then Donate”. The messages for both
conditions asked students to think about all the items that carry good memories but they do not
use any more. They were then prompted to either take a photo of the items and donate them or
58
gather the items and donate them. The donation collection area was on the bottom floor of each
residence hall for which there were no bulletin boards with signs. During the Friday of move-out
week, a staff assistant who was unaware of the hypotheses went to each residence hall with one
of the authors to count the donations4. A total of 1,913 items were counted, averaging over two
items per resident and over 200 items per residence hall.
Promotional Campaign Message
Memory Preservation Absent Memory Preservation Present
(Control) Condition Condition

4 Though the author was aware of the hypotheses, she was unaware of the condition for each residence hall when
counting the donations as the signage was placed on each residence floor by housing staff. Thus, the first time the
author and research assistant entered the buildings was on the day of the counting and they entered each residence
hall with the assistance of the housing staff on the bottom floor where donations were collected. There were no signs
for this research study posted on the bottom floors since residents do not live on these floors.
59
Results
To test whether a memory preservation campaign increased donations, we examined
whether the number of donations by residence halls in the memory presentation condition
differed from those in the control condition. We conducted a logit model on the raw count data.
Out of the total of 1,913 items donated, there were 815 items donated in the four halls in the
control condition and 1,098 items donated in the four halls in the memory preservation condition.
This difference was significant as expected (Wald X2(1) = 41.56, p < .001). Put in percentage
terms, the 283 additional items donated in the memory preservation condition amounted to a
34.72% (i.e., 283/815) increase in donations due to the manipulation.
Because the data are count data, and it makes sense to ensure no confounds are inducing
the results in this field study, we also ran a Poisson model including a number of covariates. In
this model the eight residence halls served as repeated measures, and the percentage of female
residents in each hall, the percentage of seniors in each hall, and the number of residents in each
hall were control variables. Again, the primary test of interest was the effect of memory
preservation; that is, did the memory preservation manipulation increase the number of items
donated? Indeed there was an effect of memory preservation (Z = 2.94, p < .003), affirming a
significant overall increase in donations because of our manipulation. There were also main
effects of gender (Z = 2.34, p = .02; donations were greater in residence halls with a larger
proportion of female residents) and the number of residents (Z = 4.81, p < .001; donations were
greater in halls with more residents).
Discussion
This study demonstrates the effectiveness of a promotional campaign encouraging
consumers to use memory preservation before donating possessions with sentimental value.
60
Although we could not determine whether each donated item had sentimental value to the donor,
there was an increase in total donations for those receiving the memory preservation message.
These results indicate nonprofits can successfully use memory preservation in their promotional
material to increase donations.
While such a large-scale field study is important to demonstrate the generalizability of
the effect, it is not without its limitations. Because the study was conducted at the end of the
spring semester and was part of the university’s larger donation collection, one concern may be
the length of time during which donations were collected (two weeks) and the extent to which
residents actually read the messaging on each floor bulletin board which were hung
approximately two weeks prior to move-out. Additionally, a limitation of this study is that one of
the authors assisted with donation counting due to the short time frame in which counting needed
to be completed before donations were removed from the residence halls and the limited
availability of research assistants at this time of the semester (i.e., graduation weekend).
61
ADDITIONAL STUDY PRESENTED IN STUDY 2 DISCUSSION
This study was conducted to further test the idea that memory preservation of sentimental
goods can positively impact total donations rather than just donation of sentimental goods. We
propose that cueing memory preservation of sentimental goods overcomes a barrier to starting
the donation process. Once the perceived identity loss from donating sentimental goods is
mitigated, consumers should be more motivated to search for items that could be donated, which
should increase total donations, including both sentimental and non-sentimental goods.
We randomly assigned U.S. MTurk workers (81 females; Mage = 35.54, SD = 11.27) to
view one of two donation appeals (memory preservation: present vs. absent) similar to those
used in the field studies and asked them how likely they were to look for items in their home to
donate using three items: “I am likely to search through my belongings for items I could donate,”
“I am motivated to see what items I no longer use that I could donate,” and “I do not want to
gather items to donate” (reverse-coded; α = .94). We also asked participants “When was the last
time you cleaned out your home and got rid of items you no longer use?” with responses ranging
from 1 = More than 10 years ago to 7 = In the past week (M = 4.63, SD = 1.20, indicating the
average was within the past six months - 1 year).
As predicted, consumers viewing the memory preservation present appeal reported a
greater motivation to search for items to donate than did those viewing the memory preservation
absent appeal (Mpresent = 4.98 vs. Mabsent = 4.28, F(1, 78) = 4.13, p = .04) when controlling for
how recently consumers had cleaned out their homes (F(1, 78) = .40, p = .53). The effect of
memory preservation on motivation to search for donations remains significant when not
including the control (F(1, 79) = 3.88, p = .05). These results suggest that donation appeals
62
suggesting memory preservation for sentimental goods may aid in overcoming consumer
reluctance to begin the donation process, thereby increasing not just the number of sentimental
goods donated but total donations as well, simply by initiating the process of identifying items to
donate.
DONATION APPEAL
Memory Preservation Absent Memory Preservation Present
(Control) Condition Condition
63
FIELD STUDY PRESENTED IN STUDY 3 DISCUSSION
This field study demonstrates the psychological process uncovered in the studies with
hypothetical donations indeed operates in actual donation contexts. We partnered with a local
nonprofit, St. Vincent de Paul, that accepts clothing and housewares to resell in their thrift store
with proceeds benefiting those in need.
Participants and Procedure
For six days, one of three researchers (one of the authors or two research assistants) was
present at the donation drop-off door at the back of the thrift store. When a donor came to the
door, the researcher informed her that a research study on donated goods was being conducted by
Penn State University researchers, that the study involved a few minutes of their time answering
questions about their donated goods, participation was voluntary, and they would receive $2 as a
thank you for participating. Approximately 75% of donors agreed to participate, resulting in 64
participants (75% were female; 1 participant was less than 30 years old; 13 were between 30 and
49; 32 were between 50 and 69; and the remaining 18 were 70 or older). Those that declined
tended to do so because they did not have time (e.g., were dropping off over a lunch break at
work or needed to pick up children from school).
Participants were randomly assigned to either the memory preservation present condition
or a control (memory preservation absent) condition and the manipulations alternated, one
participant in the memory preservation (photo) condition and the next participant in the memory
preservation absent (control) condition. The only time the alternating pattern was changed was
when multiple donors arrived in the same period of time, in which case all participants in the
group were in the same condition. Once participants indicated their consent to participate, we
64
asked them to think of one specific item they were donating today that was meaningful to them;
we used “meaningful” because we thought it would be easier for participants to assess than
“sentimental”. For participants in the memory preservation present condition, the researcher took
a picture of this item with an instant camera.5 The photo was given to participants to keep, and
there were no copies of the photos kept by the researchers. Photos were not mentioned in the
memory preservation absent condition. All participants then completed the survey, indicating the
meaningful item they were thinking of in one or two words (e.g., “red sweater”) and responding
to a series of ratings questions, all on a 1-7 point scale from “Strongly Disagree” to “Strongly
Agree.”
Identity loss was assessed with two items (α = .95): “Thinking about donating this item,
please indicate your agreement with each statement: ‘I will feel like I lost a piece of myself,’ and
‘I will feel like a part of me is gone.’” Participants also rated whether: “I feel like I have lost the
memories associated with the item” and then responded to four items from Sivadas and
Venkatesh’s (1995) product attachment scale (α = .82) to indicate the degree to which they felt
attached to the product they were donating. The items were “I have no feelings for this product”
(reverse-coded), “I am emotionally attached to this product,” “I am sentimental about this
product,” and “This product reminds me of memories and experiences.” Finally, participants
indicated whether they were actually the owners of the item for which they filled out the survey,
their frequency of donating, their gender, and age. After survey completion, participants received
$2 and were thanked for their participation.

5Some of the photos included the donor holding the item whereas other photos included the item only without the
donor in the picture due to the natural behavior of participants when asked to take a photo of their item. We
conducted additional analysis within the memory preservation present condition to determine whether the presence
of the donor in the photo impacted identity loss. There was no effect of donor presence versus absence in the photo
for memory preservation on identity loss (F(1, 19) = .72, p = .41, Mdonor present in photo, = 2.35 vs. Mdonor absent in photo =
1.86).
65
Results and Discussion
Sample Characteristics. A total of 22 participants indicated they donated goods to a
nonprofit at least once per month, 30 donate once per season/4 times per year, 5 donate once per
year, and 6 donate less than once per year. Participants also indicated whether the item they were
thinking about when completing the survey belonged to themselves (39), immediate family (24),
or other (1). Given our theorizing pertains to the loss of identity, we limit our primary analysis to
those who reported ownership of their donated item, leaving 39 participants for analysis (82%
were female; 5 were between 30 and 49; 24 were between 50 and 69; and the remaining 10 were
70 or older).6 A majority (61%) of the meaningful items participants selected to think about
when completing the survey were clothing (e.g., jacket, dress, sweater, shoes) and this
distribution was consistent across both conditions. Other items included accessories (11%, e.g.,
purse or jewelry), housewares (8%, e.g., teapot, heating plate, candy dish), books (14%), or other
items (6%, e.g., toy, camping tent). The average product attachment was 4.24 (SD = 1.60).
Identity Loss. We conducted an ANOVA with memory preservation condition as the
independent variable, identity loss as the dependent variable, and product attachment as a control
variable. The effect of memory preservation was significant (F(1, 36) = 4.49, p = .04) and in the
predicted direction: participants in the memory preservation condition reported less identity loss
(M = 1.95) than those in the control condition (M = 3.21). Product attachment was also
significant (M = 4.24, SD = 1.60, F(1, 36) = 6.93, p = .01), such that identity loss increased with
product attachment, consistent with prior literature (Ferraro, Escalas, and Bettman 2011).

6We conducted additional analysis with all participants, owners and non-owners. The interaction of memory
preservation condition and ownership (coded as 1 if owner and -1 if not owner) was marginally significant (F(1, 59)
= 3.65, p = .06). Consistent with the main results, when participants owned the item, memory preservation decreased
identity loss (t = -2.06, p = .04). However, when participants were not the owners of the item, memory preservation
did not impact identity loss (t = .73, p = .47).
66
Mediating Role of Memory Loss. We next assessed whether participants’ memory loss
mediated identity loss. First, memory preservation condition predicted memory loss (F(1, 36) =
7.89, p < .01), such that participants in the memory preservation condition reported less memory
loss (M = 1.24) than those in the control condition (M = 2.27). When we include memory loss in
the model, it is a significant predictor of identity loss (t(35) = 4.25, p < .01), and memory
preservation condition is not significant (t(35) = -.10, p = .92). The indirect effect of memory
preservation through memory loss is significant (indirect effect = -.40 (.26), 95% CI: -1.00 to -
.01). Thus, these results provide support for our theorizing that the mechanism by which memory
preservation strategies reduce identity loss is preserving product-related memories.
Discussion. This field study demonstrates that even among those who are already willing
to donate their possessions, memory preservation at the time of donation decreases the identity
loss from donating. Using an instant camera to provide donors with a photograph as we did in
this field study (or even taking a digital photograph for the donor on his or her phone or to email
to the donor) is a relatively easy and low cost intervention that nonprofits can employ at donation
centers to reduce felt identity loss at the time of donation.
... Second, in examining possession donation through the lens of MS, we also contribute to the growing body of work exploring factors that uniquely influence the giving of tangible possessions (Winterich et al. 2017) and product disposition more generally (Donnelly et al. 2017;Trudel and Argo 2013;Trudel, Argo, and Meng 2016;White, MacDonnell, and Dahl 2011). Importantly, by demonstrating that giving away self-connected possessions can be a form of strategic self-transcendence, we run counter to research suggesting that consumers are particularly reluctant to give up possessions that are closely connected to the self (Ferraro, Escalas, and Bettman 2011;Winterich et al. 2017). ...
... Second, in examining possession donation through the lens of MS, we also contribute to the growing body of work exploring factors that uniquely influence the giving of tangible possessions (Winterich et al. 2017) and product disposition more generally (Donnelly et al. 2017;Trudel and Argo 2013;Trudel, Argo, and Meng 2016;White, MacDonnell, and Dahl 2011). Importantly, by demonstrating that giving away self-connected possessions can be a form of strategic self-transcendence, we run counter to research suggesting that consumers are particularly reluctant to give up possessions that are closely connected to the self (Ferraro, Escalas, and Bettman 2011;Winterich et al. 2017). Indeed, we show that under the conditions outlined in our studies (i.e., when transcendence is possible), consumers are not motivated to keep a link to the selfconnected possession (Winterich et al. 2017) but instead are motivated to give that possession to others. ...
... Importantly, by demonstrating that giving away self-connected possessions can be a form of strategic self-transcendence, we run counter to research suggesting that consumers are particularly reluctant to give up possessions that are closely connected to the self (Ferraro, Escalas, and Bettman 2011;Winterich et al. 2017). Indeed, we show that under the conditions outlined in our studies (i.e., when transcendence is possible), consumers are not motivated to keep a link to the selfconnected possession (Winterich et al. 2017) but instead are motivated to give that possession to others. ...
Article
Past research demonstrates that reminders of one’s own mortality can lead to materialistic and self-serving consumer behaviors. In contrast, across five studies, we explore a condition under which mortality salience (MS) leads to increased tendency to give away one’s possessions—when the donation act is high in transcendence potential. We propose and find that consumers are more likely to donate their possessions to charity under MS (vs. comparison conditions) when the product is considered highly (vs. not highly) connected to the self. Moreover, we demonstrate that this tendency manifests only when transcendence is attainable through donation. In support of the proposition of transcendence as the underlying mechanism, the observed effects are attenuated under conditions where: (1) transcendence has already been satiated via alternative means or (2) the donated possession will not transcend the self (i.e., its physical integrity is lost by being broken down and recycled). The theoretical and practical implications of the work are discussed.
... Identity expulsion occurs when a consumer not only has a weak association with an identity, but begins to actively dissociate from the identity (i.e., seeing the identity as "not me"). In addition to dilution over time as described above, other sources of identity expulsion may arise from arguably more objective signals that trigger identity changes (Trudel et al., 2016;Winterich et al., 2017). For example, when a person graduates from secondary school, presumably they leave their "high school student" (or equivalent) identity behind and replace it with a professional or university student identity. ...
Article
Although it is well accepted that the self‐concept includes numerous identities, the preponderance of past consumer identity research has explored one identity at a time and this focus has limited new insights into the interplay between identities. We integrate across research streams to propose a Multiple‐Identity Network as a unifying framework to help inform and direct future research on multiple identities. This framework identifies three important areas of opportunity for research on multiple identities: (a) Identity Structure, (b) Identity Management, and (c) Identity Change processes that drive both structure and management.
... Importantly, consumers appear to identify with symbolic meanings of objects via psychological feelings of ownership. For example, individuals are less willing to part with possessions that are linked to the self because of a sense of identity loss (Winterich et al., 2017). ...
Full-text available
Book
APA Handbook of Consumer Psychology. Edited by Lynn R. Kahle, Tina M. Lowrey, & Joel Huber, American Psychological Association.
Article
Most consumers live surrounded by physical goods, some of which are used often and others that are largely neglected. In this article, we introduce the concept of a “possession portfolio” which we define as an individual’s holistic sense (vs. an objective listing) of the totality of the physical goods they own at a given point in time. We propose that there is an optimal possession portfolio for each individual where they feel they have achieved balance between the benefits of retaining the “right” number and type of possessions without incurring the downsides of disposing of too many possessions. In doing so, we highlight recent research relevant to the value assessment process that underlies the retention and disposition decisions made when individuals attempt to optimize their possession portfolio.
Article
This research puts forward the theoretical concept “print imprint,” articulating the connection between the printed newspaper and its reader’s “Self.” This paper contends that the printed newspaper draws out the meaningfulness of ownership, touch and nostalgia, all influential ingredients of the self. This research centers on interviews with 19 former readers of a weekly newspaper that shuttered. The findings illustrate the significance, usefulness and uniqueness of the printed newspaper. In particular, participants expressed a relationship with the printed newspaper, calling it “my paper.” Ultimately, this research argues that the loss of the weekly newspaper prompted a loss or lessening of self of the abandoned readers. Finally, this article argues this “print imprint” extends beyond printed newspapers and should be considered for all print products, including magazines and books, pointing to future research possibilities.
Article
Given the proliferation of consumer-to-consumer (C2C) rental platforms, it is important to understand the motivations of stakeholders to participate in these platforms. The viability of online C2C platforms depends on the willingness of lenders to list their possessions for rent. We suggest that psychological mechanisms related to emotional attachment may underlie this decision. Specifically, we show that emotional attachment predicts lenders’ concerns about property preservation, such that higher levels of attachment will be associated with a relatively stronger preference for policies focused on product protection (i.e., policies that provide assurance that the product will not be damaged) versus financial guarantees (i.e., policies that provide financial restitution in the event of product damage). Our results extend prior literature by providing insights into how emotional attachment influences lenders’ participation in C2C rental platforms. These findings further highlight practical interventions that platforms might use to increase lender participation. Our work informs practitioners about motivational and behavioral psychological phenomena underlying the emerging C2C business model.
Full-text available
Article
Consumption of used products has the potential to symbolically connect present and previous users of these products, something that would be appealing to lonely consumers. Accordingly, across seven studies, feeling lonely increased the preference for previously owned products. Specifically, the proportion of lone shoppers was higher in a used versus a regular bookstore, lone individuals (vs. those sitting in pairs) were more likely to select a used over new product, on Valentine’s Day, people without (vs. with) a date expressed stronger preference for used products and individual differences in loneliness during the COVID-19 pandemic predicted interest in used products. Other studies documented that the desire to symbolically connect underlies the effect of loneliness on consumption. At a time when loneliness is on the rise, the authors discuss implications for the marketing of used products and how feeling lonely might motivate reducing waste.
Article
Although an overabundance of possessions—called “clutter”—is both pervasive and maladaptive, understanding how possessions can accumulate remains understudied. In the present research, we suggest that our prior interactions with possessions—namely, a prior decision to forgo consumption—can reduce future usage intentions. Six studies demonstrate that forgoing using an item can make it seem more special, particularly when forgoing is attributed to waiting for a later occasion. As specialness increases, the item is restricted from future usage: it becomes less likely to be used in ordinary occasions and more likely to be reserved for a narrower set of extraordinary occasions. By transforming ordinary items into (perceived) treasures, we suggest that nonconsumption can encourage consumers to retain possessions indefinitely, waiting for future usage occasions that may never arise—ultimately, fueling the accumulation of clutter. These findings extend work on nonconsumption and special possessions and illuminate a novel driver of clutter. © 2021 the Association for Consumer Research. All rights reserved.
Article
The consequences of overconsumption and the recent popularity of simple living point to consumer interest in reducing belongings. They also raise an interesting question—what is a useful approach to downsizing and decluttering? We investigate how dis/order (messy vs. tidy items) affects downsizing and find, across nine focal studies, that (a) consumers retain fewer items when choosing from a disordered set because (b) order facilitates the comparisons within category that underlie the tendency to retain items. The impact of dis/order is altered by consumers’ comparison tendencies, waste aversion, and decision strategy (selection vs. rejection), which serve as theoretically and pragmatically relevant moderators. Though consumers’ lay beliefs favor rejecting from order (i.e., choosing what to get rid of from tidy items), our findings point to the usefulness of selecting from disorder (i.e., choosing what to keep from messy items) as a downsizing strategy. Together, this research has implications for consumer downsizing activities, the burgeoning home organization and storage industries, as well as sustainability.
Full-text available
Book
The Meaning of Things explores the meanings of household possessions for three generation families in the Chicago area, and the place of materialism in American culture. Now regarded as a keystone in material culture studies, Halton's first book is based on his dissertation and coauthored with Mihaly Csikszentmihalyi. First published by Cambridge University Press in 1981, it has been translated into German, Italian, Japanese, and Hungarian. The Meaning of Things is a study of the significance of material possessions in contemporary urban life, and of the ways people carve meaning out of their domestic environment. Drawing on a survey of eighty families in Chicago who were interviewed on the subject of their feelings about common household objects, Mihaly Csikszentmihalyi and Eugene Rochberg-Halton provide a unique perspective on materialism, American culture, and the self. They begin by reviewing what social scientists and philosophers have said about the transactions between people and things. In the model of 'personhood' that the authors develop, goal-directed action and the cultivation of meaning through signs assume central importance. They then relate theoretical issues to the results of their survey. An important finding is the distinction between objects valued for action and those valued for contemplation. The authors compare families who have warm emotional attachments to their homes with those in which a common set of positive meanings is lacking, and interpret the different patterns of involvement. They then trace the cultivation of meaning in case studies of four families. Finally, the authors address what they describe as the current crisis of environmental and material exploitation, and suggest that human capacities for the creation and redirection of meaning offer the only hope for survival. A wide range of scholars - urban and family sociologists, clinical, developmental and environmental psychologists, cultural anthropologists and philosophers, and many general readers - will find this book stimulating and compelling. Translations: Il significato degli oggetti. Italian translation. Rome: Edizione Kappa, 1986. Der Sinn der Dinge. German translation. Munich: Psychologie Verlags Union, 1989. Japanese translation 2007. Targyaink tukreben. Hungarian translation, 2011.
Full-text available
Article
Consumers are often surrounded by resources that once offered meaning or happiness but that have lost this subjective value over time—even as they retain their objective utility. We explore the potential for social recycling—disposing of used goods by allowing other consumers to acquire them at no cost—to transform unused physical resources into increased consumer happiness. Six studies suggest that social recycling increases positive affect relative to trash, recycling, and donations of goods to nonprofit organizations. Both perceptions of helping the environment and helping other people drive this increase in positive affect. We conclude that social recycling offers a scalable means for reengineering the end of the consumption cycle to transform unused resources into happiness. We suggest that further research should continue to enrich a general theory of disposition, such that we are able to maximize the ecological, interpersonal, and community utility of partially depleted resources.
Article
The authors investigate the affective nature of autobiographical memories and the conditions and mechanisms leading to transfer of this affect to ad and brand judgments. They find that when ads encourage the retrieval of autobiographical memories there is a higher level of felt affect and reduced processing of product attributes. Furthermore, this generally positive affect is readily transferred to the ad, thus enhancing ad evaluations. However, the extent of transfer of autobiographical affect to the brand depends on forging a link in the ad between the brand and the personal memory, as Study 1 demonstrates. Study 2 shows that when autobiographical memories are encouraged, brand evaluations are no different given strong versus weak product arguments, further evidence that brand evaluations are not based on an analysis of product claims when such memories are evoked. Study 2 also provides evidence for the dominance of affective inputs into brand judgments when both autobiographical affect and product arguments are relatively accessible from memory.
Article
What do consumers do with products once they have outlived their usefulness, and how does this relate to the purchase of replacement products?
Article
A study was conducted to develop a better understanding of the timing of consumer durable good replacement purchases. The demographic characteristics, attitudes and perceptions, and search behavior of consumers who replace a product during the early and late parts of its lifetime were examined. Results based on univariate and multivariate analyses of replacement buyers of new automobiles indicate that “early” replacement buyers are more concerned with styling and image and less concerned with costs than “late” replacement buyers. Further, early replacers have higher income, but lower levels of educational achievement and occupational status, than late replacers. Late replacers engage in more search activity. Findings also suggest that marketing efforts have differential effects on the replacement buyer segments.
Article
The authors examine how a reference to an unrelated product in the choice context impacts consumers’ likelihood of donating to charity. Building on research on self-signaling, the authors predict that consumers are more likely to give when the donation appeal references a hedonic product, as compared to when a utilitarian product is referenced or when no comparison is provided. They posit that this occurs because referencing a hedonic product during a charitable appeal changes the self-attributions, or self-signaling utility, associated with the choice to donate. A series of hypothetical and real choice experiments demonstrate the predicted effect, and show that the increase in donation rates occurs because the self-attributions signaled by a choice not to donate are more negative in the context of a hedonic reference product. Finally, consistent with these experimental findings, a field experiment shows that referencing a hedonic product during a charitable appeal increases real donation rates in a non-laboratory setting. The authors discuss theoretical implications for both consumer decision making and the self-signaling motives behind prosocial choice.
Article
It has been known for some time that consumers' identities influence purchasing decisions, and that people form strong identity connections or "links" with products and brands. However, research has yet to determine whether identity-linked products are differentially treated at disposal in comparison to products that are not identity-linked. Across seven studies, the current research shows that when an everyday product (e.g., paper, cups, aluminum cans) is linked to a consumer's identity it is less likely to be trashed and more likely to be recycled. Further, the tendency to recycle an identity-linked product increases with the strength and positivity of the connection between the consumer and product (or brand). Finally, the disposal behavior can be explained by consumers' motivation to avoid trashing a product that is linked to the self because it is viewed as an identity-threat. In sum, consumers will be more likely to recycle (rather than trash) a product if the product is linked to a consumer's identity. This occurs because placing an identity-linked product in the trash is symbolically similar to trashing a part of the self, a situation consumers are motivated to avoid.