Revisiting the Differential Centrality of Experiential and Material Purchases to the Self: Replication and Extension of Carter and Gilovich (2012)

The self is closely associated with one’s memories (Klein, 2001). Memories of experiences play a major role in forming the self (Conway, 2005; Greenwald, 1981). People tend to access memories in a way that is consistent with their present image in order to maintain a consistent self (Conway et al., 2004), and the way that people make sense of experiences largely shapes their understanding of themselves (Oyserman, 2001). In contrast, material possessions are tangible objects that dwell outside of our memory (Carter & Gilovich, 2012). They are seen as an extension of the self in that they serve as an enhancement of who we are and as a reminder of who we were, therefore contributing to the central aspect of the self-concept to a lesser extent than experiences (Belk, 1988).

Carter and Gilovich (2012) proposed several reasons why relating experiences to the self could result in higher satisfaction with experiential purchases than with material purchases. First, people aim to maintain a positive self-impression, and may therefore associate experiences with more positive recollections than in reality (Dunning, 2005; Mitchell et al., 1997). Second, the intangible and ambiguous nature of experiences makes it easier for people to reconstruct experiences in one’s favor to enhance their self-image and boost their self-esteem (Carter & Gilovich, 2012; Dunning et al., 1995; Kunda, 1990). Third, the abstract nature of experiences also allows people to associate experiences with high-order purposes that they find important and valuable, and are, therefore, more closely linked to one’s self-concept (Trope & Liberman, 2003).

In addition, Carter and Gilovich (2012) highlighted individual differences in materialism as predicting people’s likelihood of attributing experiences to their self-image, as measured by the Material Values Scale (MVS; Richins & Dawson, 1992). According to Richins and Dawson (1992), materialists tend to define themselves through their possessions by considering possessions to be central in their lives, and attributing success and happiness to the material goods they own. Those who are more materialistic tend to indulge in more material possessions as they believe that possessions and money are essential for one’s happiness and social progress (Ward & Wackman, 1971). Therefore, individual differences in materialism may moderate the relationship between the two types of purchase (i.e., material and experiential purchases) and self.

We chose the article by Carter and Gilovich (2012) for replication based on its impact and the absence of direct replications.

Carter and Gilovich (2012)’s article has had an impact on scholarly research in the area of consumer behavior, social cognition, and judgment and decision-making. At the time of writing (November 2022), there were 476 Google Scholar citations of the article and many important follow-up theoretical and empirical articles. Some examples are Caprariello and Reis’s (2013) research showing that experiences are preferred to material possessions when experiences involve others, Zhang et al.’s (2014) study demonstrating that individual differences in buying tendencies affected the happiness gained from experiential and material purchases, and Lee et al.’s (2018) study on the role of social class in happiness in experiential purchases versus material possessions.

The phenomenon demonstrated has the potential for practical implications for both personal everyday life and commercial domains. With better knowledge of the associations between experiential and material purchases and self-identity, individuals may aim to overcome the cognitive and behavioral focus on material purchases as a way to affirm their self-identities and instead shift to exploring balancing with more experiential purchases. There are also potential marketing insights, to invest in consumer experience and integrate the understanding that experiences constitute an important aspect of consumer satisfaction with product consumption (Zauberman et al., 2009).

A meta-analysis by Weingarten and Goodman (2020) that summarized over 141 studies, including Carter and Gilovich’s (2012), found that consumers gained more happiness through experiential purchases than material possessions, d = 0.39, 95% CI [0.34, 0.43]. However, the meta-analysis also pointed out that many of the studies in the literature were underpowered, which raises the importance of well-powered independent replications of the phenomenon.

To the best of our knowledge, there are currently no published direct replications of this article. Given the impact of Carter and Gilovich’s research, and in response to the growing attention in psychological science for reproducibility and replicability, we revisited the classic phenomenon of Carter and Gilovich’s studies by conducting an independent, well-powered, pre-registered, and very close replication (e.g., Brandt et al., 2014; Open Science Collaboration, 2015; Van’t Veer & Giner-Sorolla, 2016; Zwaan et al., 2018).

Carter and Gilovich (2012) conducted seven studies on the differential centrality of experiential and material purchases to the self, with the overarching hypothesis that experiential purchases are more closely associated with the self than material purchases. Of those, we chose to replicate Studies 3A, 3B, 3C, and 5 given their simplified design, better methodological transparency and clarity, and higher feasibility for our intended target online sample. In contrast to the chosen studies, Studies 1, 2 and 4 required manual drawings that were difficult to be conducted on an online platform, required multiple raters for data analysis, and had less complete information about their process, which impeded reproduction and replication. We felt that these studies would best be examined following a successful replication of the baseline phenomenon.

Studies 3A, 3B, and 3C had within-subjects designs in which participants recalled both an experiential purchase and a material purchase and then compared the two types of purchases on how informative these were about a person’s true self. They found that people tend to view experiential purchases as providing greater insights into one’s true self than material purchases. Study 5 used a between-subjects design where participants imagined deleting the memory of either an experiential or a material purchase, and then rated the perceived change to their true self and their satisfaction with the purchase. They found that deletion of experiential purchases had a greater impact on one’s self-concept than deletion of material purchases, which was positively associated with greater satisfaction derived from experiential compared to material purchases. We summarized the deduced hypotheses from the studies in Table 1.

We added an extension to investigate changes in evaluations of material and experiential purchases over time. Past research has demonstrated evidence that different types of purchases elicit different extents of pleasure and regrets over time. On the one hand, satisfaction from experiential purchases tends to be more enduring than that from material purchases since the pleasure from material purchases can be easily compared with alternatives, thus undermining people’s satisfaction with them (Carter & Gilovich, 2010; Van Boven & Gilovich, 2003). On the other hand, material purchases are more likely associated with regrets of actions (buyer’s remorse) compared to experiential purchases that are more strongly associated with regrets of inaction (missed opportunity) (Rosenzweig & Gilovich, 2012). Since experiences tend to result in lasting satisfaction whereas material processions tend to elicit regrets on actions, we set out to include an exploratory comparison of the two on the differences between recalled evaluations of the purchases at the time of purchase compared to current evaluations. We hypothesized that over time, the evaluation of experiential purchases decreases to a less extent than that of material purchases.

We pre-registered the experiment on the Open Science Framework (OSF; https://osf.io/9vsxt/), and data collection was launched later that week. Pre-registrations, power analyses, and all materials used in these experiments are available on the OSF: https://osf.io/v2w5h/.

All measures, manipulations, exclusions conducted were reported; all studies were pre-registered and data collection was completed before analyses.

We conducted an a-priori power analysis with “pwr” package version 1.3-0 (Champely, 2020) to obtain the required sample size based on the reported t statistics and mean differences in the original article. We aimed for a statistical power of .95 and alpha of .05. Given our adjustment of the Study 5 design from a between-subjects to within-subjects design, we still used the between-subjects as a more conservative estimation.

Across the chosen studies, our power analysis indicated that we require 216 participants for Study 3A, 92 for Study 3B, 26 for Study 3C, and 162 for Study 5. Thus, the smallest sample sizes (that were also smaller than the original sample sizes) were from Studies 3B and 3C. As Study 3C had a higher sample size (i.e., 102) in the original paper than Study 3B (i.e., 101), multiplying the original sample size of Study 3C (102) by 2.5 times provided us with 255, which also exceeded the calculated sample sizes of the other studies. We rounded this number up to 300 in case of any exclusions to detect the smallest effect of replicated studies. Taking into consideration the possibility of weaker effects in replication, we aimed for high power given our allocated budget. Following Simonsohn’s (2015) general rule of thumb of multiplying by 2.5, we set out data collection goal to 750. We provided our calculations and more details in the Supplementary materials.

A total of 759 American Amazon Mechanical Turk (MTurk) participants completed the study on CloudResearch/TurkPrime (Litman et al., 2017) with USD 1.00 compensation, with 743 included in the analyses after pre-registered exclusions (M_age = 41.39, SD =12.90; 343 males, 393 females, 4 others; 3 rather not disclose). We include sample exclusion criteria and compare our sample with the sample of original article in the Supplementary materials (sections “Exclusion criteria” and “Pre-exclusions versus post-exclusions”).

Study 3A had two parts, purchase recalls (3A-1), and a single question comparison (3A-2). Because Studies 3A-2, 3B, and 3C from Carter and Gilovich (2012) had a very similar design, we combined the studies into a singular unified design of the studies presented in random order (Block Y). Studies 3A-1 and Study 5 also had a similar method prompting for recalls and were therefore combined into the survey with a unified within-subjects design, presented in random order (Block X). We note that Study 5 was originally a between-subjects design, yet we adjusted it to a within-subjects design with an order of appearance randomized and recorded, which allowed us to mirror the original article’s between-subjects analyses by focusing on the first displayed condition.

After completing a consent form and verification checks, participants answered Block X and Block Y, in a randomized order. In Block X participants recalled both types of purchase, and for each indicated the cost, time, satisfaction (Study 3A-1), importance of purchase, willingness to exchange, happiness if exchanged, importance of memories, centrality to self, and past and current evaluations of the purchase (Study 5). In Block Y, participants compared material and experiential evaluations for self (Study 3A), strangers (Study 3B), and a new person (Study 3C).

We provided details of the measurements used in Studies 3A-2, 3B, and 3C in Table 2. In each sub-study of Study 3, participants were asked to imagine different perspectives and indicated whether experiential or material purchases provided greater insight into a person’s true self. The scenario in Study 3A-2 was related to a stranger’s perspective on the participant (1 = Material purchases; 9 = Experiential purchases). In Study 3B, the scenario shifted to the participant’s perspective of a stranger. Finally, in Study 3C, the scenario was based on the participant meeting a new person that could be potentially important in the participant’s life. The questions in 3B and 3C were on a 0 (material) to 100 (experiential) scale.

Additionally, we measured individual differences in materialism using the 15-item MVS (Richins, 2004). Participants rated statements related to materialism and material consumption on a 5-point scale (1 = Strongly disagree, 5 = Strongly agree). Higher scores indicate more personality variables related to materialism traits and a greater orientation towards materialism.

We provided details of the measurements used in Studies 3A-1 and 5 in Table 3.

In Study 5 we examined the impact of participants imagining the deletion of a past purchase memory on their self-concept. Participants recalled material and experiential purchases from the past five years and indicated satisfaction with (item shared with Study 3A-1) and importance of purchase (1 = Not at all important, 5 = Somewhat important, 9 = Extremely important).

They then imagined that they went back in time to the past and had an opportunity to change the recalled memory. They indicated their willingness to make such a change (1 = Absolutely not, 9 = Definitely), their happiness if they did so (1 = Much less happy, 9 = Much happier), the importance of the recalled memory (1 = Not at all important, 9 = Very important), and the perceived change to their self-concept if they made that change (1 = Not at all, 9 = A great deal).

We added an extension to Block X with two items per each purchase type aimed to investigate whether people’s evaluations of material and experiential purchases change differently over time.

Participants were asked about their feelings towards a recollected purchase. Participants reported their recalled feelings at the time of purchase (“Please rate how you felt about the material/experiential purchase you had recalled back then, at the time of purchase?”) and how they currently feel about their past purchases (“Please rate how you feel about the material/experiential purchase you had recalled now?”) (-5 = Very negative, 0 = Neutral, 5 = Very positive).

We summarized the replication as “close” using the criteria by LeBel et al. (2018; see Supplementary for details). The classification of the replications compared to the originals are summarized in Table 4.

We first conducted t-tests and Pearson’s r correlation analyses following the original article’s analyses. We provided descriptive statistics of all studies in Table 5.

We added several extensions to the original article. First, we extended the correlational analysis of MVS and type of purchase ratings to Study 3A, going beyond the original analyses applied to Study 3B and Study 3C. Second, both a between-subjects design and within-subjects design analysis were performed for Study 5 as a comparison to the between-subjects design analysis in the original article. Third, we further added correlation and partial correlation analyses to test whether greater satisfaction derived from experiential purchases was positively correlated with one’s self-concept affected by the willingness to exchange memories. Lastly, we compared current and past evaluations of material and experiential purchases to examine differences in change over time.

We conducted a one-sample t-test (two-tailed) against the scale midpoint of 5. Consistent with H1, we found that a person with experiential purchase knowledge is perceived to have a greater insight into one’s true self than a person with material purchase knowledge (M = 6.43, SD = 2.22, t(742) = 17.61, p < .001, d = 0.65, 95% CI [0.57, 0.73]).

We extended the analysis of the original article and conducted Pearson’s r correlation analysis between the MVS (α = .92) and ratings of the purchase type in providing insights into the participant’s true self. We found support for a negative correlation between MVS and Study 3A scores (r(741) = -0.20, p < .001, 95% CI [-0.27, -0.13]). That is, people who rated higher on trait materialism were less likely to believe that knowing one’s experiences would provide greater insight into one’s true nature.

We conducted a one-sample t-test (two-tailed) against the scale midpoint of 50. Since the rating was heavily left-skewed (skewness = -1.12, SE = .09), we also ran the analysis on the data with square transformation. However, as we found support for the hypotheses regardless of whether the transformation was conducted, we reported untransformed data for the convenience of comparison with the original study. Analyses for the transformed data analysis can be found in the Supplementary.

We found support for H2 that knowledge of a stranger’s experiences was perceived to give participants greater insight into the stranger’s true self than knowledge of possessions (M = 72.08, SD = 25.08), t(742) = 24.00, p < .001, d = 0.88, 95% CI [0.80, 0.96].

In addition, a Pearson’s r correlation analysis showed that responses to the MVS were negatively correlated with the scores from Study 3B, r(741) = -0.19, p < .001, 95% CI [-0.26, -0.12]. That is, people who rated higher on trait materialism were less likely to believe that knowing the experiences of a person would provide greater insight into the person’s true self.

We conducted a one-sample t-test against the scale midpoint of 50 for each of the sub-questions in Study 3C. Since the scores were heavily left-skewed (skewness_insight = -1.17, SE = .09; skewness_useful = -1.23, SE = .09, skewness_fun = -2.03; SE = .09), a cubic transformation was performed. However, as we found support for the hypotheses regardless of whether the transformation was conducted, we reported untransformed data for the convenience of comparison with the original study. Analyses for transformed data can be found in the Supplementary.

In the case of meeting someone new, we found that a person’s experience was perceived to provide greater insight into a person’s true self, (M = 76.35, SD = 23.37), t(742) = 30.74, p < .001, d = 1.13, 95% CI [1.04, 1.22], that a person’s experience was perceived to be more useful, (M = 76.27, SD = 23.04), t(742) = 31.08, p < .001, d =1.14, 95% CI [1.05, 1.23], and that a person’s experience was perceived to be more fun to talk about, (M = 85.10, SD = 17.95), t(742) = 53.32, p < .001, d = 1.96, 95% CI [1.83, 2.08] than their material purchases. Thus, the H3 was supported.

We also conducted Pearson’s r correlation analyses in the case of meeting a new person and found that responses to the MVS were negatively correlated with knowledge of experiences providing greater insight into a person, r(741) = -0.15, p <.001, 95% CI [-0.22, -0.08], knowledge of experiences being more useful, r(741) = -0.20, p < .001, 95% CI [-0.26, -0.13], or that knowledge of experiences would be more fun to talk about, r(741) = -0.22, p < .001, 95% CI [-0.29, -0.15]. That is, when meeting someone new, people who rated higher on trait materialism were less likely to believe that knowing the experiences of a person would provide greater insight into the person’s true self, be more useful to know the person and be more fun to talk about. We summarized the correlational analyses in Table 6.

To replicate Study 3A examining satisfaction of purchases based on the recalls of both experiential and material purchases, we ran a two-tailed paired t-test and found that participants were more satisfied with experiential purchases (N = 743, M = 8.10, SD = 1.31) than material purchases (M = 7.92, SD = 1.28), t(742) = 3.15, p = .002, d = 0.15, 95% CI [0.07, 0.22].

To replicate Study 5’s within-subjects design, we first separated the data into experiential and material conditions based on the first condition seen by participants to replicate the between-subjects design conducted by the original article. After averaging the scores of four memory exchange items, we created a composite “memory exchange index” (please refer to the Supplementary for Cronbach Alpha of different conditions) to represent the willingness to replace memories of past purchases with new memories.

We ran an independent Welch’s t-test (two-tailed) comparing the material and experiential exchange indexes and found support for H4. That is, deleting experiential purchase memories was believed to have a greater impact on one’s self-concept, such that participants in the material condition (M = 4.75, SD = 1.36, N = 393) were more willing to exchange their memories than participants in the experiential condition (M = 4.22, SD = 1.33, N = 350), t(741) = 5.36, p < .001, d = 0.39, 95% CI [0.25, 0.54]. In addition, participants in the experiential condition were more satisfied with their purchase (M = 8.21, SD = 1.21, N = 350), than participants in the material condition, (M = 8.00, SD = 1.11, N = 393), t(741) = 2.51, p = .01, d = 0.18, 95% CI [0.04, 0.33].

We extended Study 5’s analysis of the original article and conducted two-tailed paired t-tests based on a within-subjects design. Consistent with the results of between-subjects analyses, participants were more willing to exchange their memories of material purchases (M = 4.90, SD = 1.42, N = 743) than their memories of experiences (M = 4.11, SD = 1.44, N = 743; t(742) = 12.24, p < .001, d = 0.55, 95% CI [0.47, 0.62]). Additionally, as reported in Study 3A, participants were more satisfied with experiential purchases (M = 8.10, SD = 1.31, N = 743) than material purchases (M = 7.92, SD = 1.28, N = 743), t(742) = 3.15, p = .002, d = 0.15, 95% CI [0.07, 0.22].

We conducted two correlational analyses and found support for H5 that the greater satisfaction derived from experiential purchases was negatively associated with the willingness to exchange memory. First, a Pearson’s r correlation analysis between the exchange index of experiential purchases and satisfaction of experience purchases revealed that the exchange index was negatively correlated with satisfaction, r(348) = -0.34, p < .001, 95% CI [-0.43, -0.24]. Second, we further conducted a Pearson’s r partial correlation analysis between the two items while controlling for the satisfaction of material purchases. The result also showed a negative correlation, r(348) = -0.36, p < .001, 95% CI [-0.45, -0.26], which confirmed findings of Pearson’s r correlation analysis.

We added an exploratory extension comparing changes in evaluations of material purchases and experiential purchases over time. We conducted a 2 (purchase type: material vs. experiential) × 2 (feeling time: current vs. past) within-subjects two-way ANOVA and found support for the main effects of purchase type and time. Overall, participants rated past feelings (M = 3.24, SD = 1.95) as more positive than current feelings (M = 2.66, SD = 2.24; F(1, 742) = 101.10, p < .001, η²p = 0.12, CI[0.08, 0.17]), and perceived experiential purchases (M = 3.21, SD = 2.09) as more positive than material purchases (M = 2.69, SD = 2.12; F(1, 742) = 54.70, p < .001, η²p = 0.07, CI[0.04, 0.11])

We also found support for an interaction between purchase type and feeling type (F(1, 742) = 5.54, p = .019, η²p = 0.01, CI[0.00, 0.02]). Participants’ current evaluation of experiential purchases (M_now = 2.97, SD_now = 2.21; M_past = 3.45, SD_past = 1.93) decreased to a lesser extent than that of material purchases (M_now = 2.35, SD_now = 2.23; M_past = 3.04, SD_past = 1.95). We summarized descriptive statistics and statistical tests in Table 5 and Table 7, respectively.

In sum, both between-subjects design and within-subjects design support our hypotheses. Participants were less willing to exchange their experiential purchase memories than material purchase memories. Also, deleting memories of experiential purchases (vs. material purchases) had a greater impact on their self-concept, which was positively associated with greater satisfaction derived from experiential purchases. The effect size of between-subjects design on willingness to exchange memories, d = 0.39, was smaller and below the range of confidence intervals of that of within-subjects design, d = 0.55, 95% CI [0.47, 0.62]. Similarly, the effect size of between-subjects design on the association between the willingness to exchange experiential memories and satisfaction with experiential purchases, r = -0.34, was smaller and below the range of confidence intervals in the correlation analysis compared to that of within-subjects design (r(741) = -0.43, 95% CI [-0.49, -0.37]), as well as in the partial correlation analysis, r = -0.36, compared to the within-subjects design’s effect size (r(741) = -0.44, 95% CI [-0.50, -0.38]).

For all our chosen studies, we were able to successfully replicate the original findings. For Studies 3A, 3B, and 3C, the replication effect sizes were larger than the original findings, and the replication range of confidence intervals did not cover the original effect size point. For Study 5, the replication effect sizes were smaller than the original findings, and the replication range of confidence intervals did not cover the original effect size point. A comparison of the statistical tests of replication and original effects is in Table 8.

In a unified design replication of four studies reported in Carter and Gilovich (2012), we found strong support for the core hypotheses that experiential purchases were more closely associated with the self, which leads to higher satisfaction in experiential purchases than material purchases. In Studies 3A, 3B, and 3C, we found that knowledge of experiential purchase was perceived to provide greater insight into a person’s true self, a stranger’s true self, and as more useful, insightful, and fun to know when meeting someone new. With both between-subjects analyses and the extended within-subjects analyses in Study 5, we found that deleting experiential purchase memories would have a greater impact on self-concept than material purchase memories, and that the willingness to exchange memories was negatively associated with experiential purchase satisfaction. We also added an extension and found that experiential purchases were evaluated more positively than material purchases, that evaluations of material and experiential purchases declined over time, and that the decline for experiential purchases was weaker than that of material purchases.

Our findings contribute in several ways to the growing research about purchase types, the self, and satisfaction levels (Caprariello & Reis, 2013; Carter & Gilovich, 2010; Van Boven & Gilovich, 2003). First, our study supported the original research with a high-powered diverse sample and mostly found effects consistent with and larger than in the original (above d = 0.8 for Studies 3B and 3C, exceeding 0.38 and 0.74 reported in the original research). Previous research demonstrated that the disappointment that people feel during an event was short-lived and dissipated quickly, leaving one with a rosy retrospection (Mitchell et al., 1997). In contrast, people often adapt to features of material possessions, and thus, the pleasure derived from them decreases over time (Frederick & Loewenstein, 1999; Van Boven, 2005). Our research replicated Carter and Gilovich’s findings and pointed out that the reason behind the positive association between experiential purchases and satisfaction levels is related to one’s self-concept. We demonstrated that experiential purchases have a greater impact on self-concept than material purchases, which is positively associated with satisfaction of purchases in the experiential condition.

We also found that experiential purchases were perceived as more positive and that these positive evaluations declined less over time than material purchases. This aligns with prior research that satisfaction from experiential purchases tends to be more enduring than that from material purchases (Carter & Gilovich, 2010; Rosenzweig & Gilovich, 2012; Van Boven & Gilovich, 2003).

Our replication had some limitations which suggest several directions for future research. First, one limitation of the current investigation is related to the generalizability of our sample, where the majority classified themselves to be from working to middle-class income families in the United States. We asked participants to recall purchases of at least USD 50, and participants were expected to have discretionary income that allows them to afford such purchases. Therefore, it is possible that findings would differ for more diverse and non-WEIRD populations (Western, Educated, Industrialized, Rich, and Democratic; Henrich et al., 2010). Future research could examine the effect in other countries and broader demographics to examine purchases of lower monetary amounts.

Second, there is an implicit assumption of a shared clear interpretation of the definition of material or experiential purchases, yet the boundaries between experiential and material purchases are not always clear. For example, certain goods may lie in between these two categories due to the experience they provide despite being physical objects (Carter & Gilovich, 2012; Van Boven & Gilovich, 2003). Hence, it is possible that participants have different categorizations of the purchase type for the same goods, which we have observed in our experiment. Future research could counter the issue by using specific, well-defined stimuli to test the effect.

We asked participants to recall their purchases and feelings, relying on participants’ ability to reconstruct their memories. Retrospective reports and evaluations may be affected by memory and feelings regarding the experiences, and thus might not accurately reflect true behaviors, cognitive and affect feelings in the past (Kahneman & Krueger, 2006; Kahneman & Riis, 2005). Future research may aim to adopt more direct measures, such as the Experiencing Sampling Method and Day Reconstruction Method, in which participants would be prompted to record their experiences, evaluations, and feelings as close as possible to the occurrence (Kahneman et al., 2004; Kahneman & Krueger, 2006; Stone & Shiffman, 2002).

Ten years after Carter and Gilovich (2012), we found strong support for their findings with several successful extensions highlighting differences between material and experiential purchases. We conclude these findings as a consistent reliable phenomenon and laid out several promising directions for future research.

The authors declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

The authors received no financial support for the research and/or authorship of this article.

Michelle Chee conducted the replication as her thesis. Yiyu Chen verified, reviewed, and drafted the manuscript for submission. Gilad advised and supervised each step, conducted the pre-registrations, ran data collection, and edited the manuscript for submission.

In the table below, employ CRediT (Contributor Roles Taxonomy) to identify the contribution and roles played by the contributors in the current replication effort. Please refer to https://www.casrai.org/credit.html for details and definitions of each of the roles listed below.

Materials, data, and code are available on: https://osf.io/v2w5h/

Citation of the target research article: