Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/university-of-california-press/revisiting-the-temporal-pattern-of-regret-in-action-versus-inaction-uU00k4WTeI
References
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Publisher
- University of California Press
- eISSN
- 2474-7394
- DOI
- 10.1525/collabra.37122
- Publisher site
- See Article on Publisher Site
Abstract
Yeung, S. K., & Feldman, G. (2022). Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With Extensions Examining Responsibility. Collabra: Psychology, 8(1). https://doi.org/10.1525/collabra.37122 Social Psychology Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With Extensions Examining Responsibility 1 2 Siu Kit Yeung , Gilad Feldman 1 2 Psychology, Chinese University of Hong Kong, Hong Kong SAR, China, Psychology, University of Hong Kong, Hong Kong SAR, China Keywords: temporal pattern, judgment and decision-making, pre-registered replication, regret, action, inaction https://doi.org/10.1525/collabra.37122 Collabra: Psychology Vol. 8, Issue 1, 2022 The temporal pattern of regret is the phenomenon that people perceive or experience stronger regret over action compared to inaction in the short-term, yet stronger regret over inaction compared to action in the long term. Following mixed and null findings in the literature, we conducted replications and extension of Studies 1, 3, 4, and 5 in the classic Gilovich and Medvec (1994) which first demonstrated this phenomenon, with a single combined data collection in randomized display order with an online sample of Americans on MTurk (N = 988). We found support for the original findings using different designs in Studies 1, 3, and 4, yet with weaker effects. We failed to find support for such a pattern in Study 5. We discuss possible interpretations for these differences: our replication adjustments, the change in the meaning of action and inaction, or change in hypothetical versus real-life personal experiences. Extending the replications, we found support for stronger responsibility for action compared to inaction both in the short-term and the long-term. We conclude overall support for the effects, yet with follow-up work necessary to resolve the inconsistencies in the findings of the Study 5 replication. Pre-registration, materials, data, and code were made available on: https://osf.io/7m3q2/ goal was to conduct independent direct pre-registered well- Background powered replications of the temporal pattern of regret. Our The temporal pattern of regret regarding action and in- second goal was to use the same base methods to extend action was first demonstrated by Gilovich and Medvec these findings and examine whether a similar action-inac- (1994) who showed that whereas people tend to experience tion pattern of asymmetry would also be found regarding stronger regret for actions over inactions in the short term, perceptions and experiences of personal responsibility. they tend to experience stronger regret for things they did We begin by introducing the literature on the action-ef- not do over things they did when reflecting back on their fect and the temporal action-inaction effect. We then dis- lives. A large body of literature has found consistent sup- cuss the motivations for the current replication and outline port for an action-effect, the phenomenon that people as- replication hypotheses and designs, with an introduction of sociate stronger regret with action compared to inaction our extension to attributions of responsibility. (e.g. Gleicher et al., 1990; Kahneman & Tversky, 1982). Temporal Pattern of Regret However, Gilovich and Medvec (1994) suggested that this classic effect is moderated by temporal distance, such that Kahneman and Tversky (1982) were the first to demon- when retrospectively recalling their lifetime and long-term strate the action-effect, the stronger regret associated with regrets, people tend to associate stronger regret with inac- action over inaction, with many successful follow-up tion than with action. demonstrations (e.g. Feeney & Handley, 2006; Gleicher et Over the years the literature has seen many mixed find- al., 1990; Landman, 1987). The action-effect has been pre- ings on temporal patterns in regret (e.g. Bonnefon & Zhang, viously explained using several paradigms, such as the 2008; Byrne & McEleney, 2000; J. Feldman et al., 1999; higher perceived causality and responsibility associated Towers et al., 2016), possibly due to differences in methods with action (Kordes-de Vaal, 1996), and using norm theory and scenarios. This suggests the need for revisiting these suggesting that actions are perceived as an exception to the classic effects with pre-registered replications. norm of not acting in such situations, and exceptions are We conducted direct replications and extensions of Stud- ies 1, 3, 4, and 5 in Gilovich and Medvec (1994). Our first a Corresponding author: Department of Psychology, University of Hong Kong, Hong Kong SAR, China; gfeldman@hku.hk Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... more cognitively mutable than routines and therefore asso- views with several groups of participants. Study 1b examin- ciated with higher regret (Kahneman & Miller, 1986). ing greatest lifetime regrets overlapped with the more com- The classic experiments by Kahneman and Tversky prehensive Study 5, which manipulated temporal distance, (1982) demonstrated the action effect by presenting par- and included questions on lifetime and past week’s greatest ticipants with hypothetical short-term decision-making sit- regrets. Overall, the target studies for replication covered uations. Gilovich and Medvec (1994) tested whether these both scenario experiments (Studies 3 and 4) and surveys re- results would extend to evaluations of real-life long-term garding real-life experiences (Studies 1 and 5). experiences. In their Studies 1 and 5, they found that for the To the best of our knowledge, there have been no pub- retrospective and lifetime reflections the action-effect re- lished direct replications of these target studies. The article versed into an inaction-effect, in that participants tended to has been influential on research in social-cognitive psy- report stronger regret for their inaction compared to action. chology, emotions, and decision-making. At the time of In their Studies 3 and 4 they also demonstrated their find- writing (December 2021), there were 549 Google Scholar ci- ings using scenarios that were very similar to those of Kah- tations of the article and many important follow-up the- neman and Tversky (1982) when manipulating short-term oretical and empirical articles (e.g., Bonnefon & Zhang, versus long-term reflections. 2008; J. Feldman et al., 1999; Gilovich & Medvec, 1995; In follow-up work, Gilovich and Medvec (1995) proposed, Towers et al., 2016). investigated, and discussed several possible mechanisms for The original studies had small sample sizes (under 100 these effects, including mechanisms related to decrease in participants for 2-4 conditions), with similar samples in intensity of action regrets and increase in intensity of inac- conceptual replications, and revisiting underpowered clas- tion regrets overtime For example, they suggested that peo- sics is valuable in addressing possible concerns over false- ple engage in more compensatory behavior for action re- positive rates (Christley, 2010). In addition, findings were grets compared to inaction regrets. Meaning, that people not always consistent with some of the original hypotheses. tend to do more to try and rectify their action mistakes For example, Study 5 failed to find support for action-effect compared to inaction mistakes, explaining why inaction re- in the short-run, whereas Study 4 found support for action- grets may be stronger in the long-run compared to action effect in the short-run but failed to find support for a mean- regrets which may be weakened over time. Another possible ingful reversal to inaction-effect in the long run. explanation is that over time people seem to become more We aimed to revisit the original findings to try and ad- confident that they would have succeeded if they had taken dress mixed findings in follow-up literature identifying actions (Gilovich et al., 1993), as their memory and/or con- boundary conditions, and findings that were not in support cerns regarding outcome uncertainties or risks of action di- of the temporal pattern. Leach and Plaks (2009) conducted minished over time (Gilovich & Medvec, 1995). Further- a successful conceptual replication of Gilovich and Medvec more, Gilovich and Medvec (1995) proposed that intensity (1994) using scenario experiments, and found that the level of inaction regrets may increase overtime as people per- of abstraction mediated the temporal pattern of regret. Fur- ceive inaction mistakes seem to result in more negative thermore, Zeelenberg et al. (1998) found support for tem- consequences, perhaps in a wider range of areas overtime poral pattern of regret with a series of studies that coded (Rajagopal et al., 2006). Moreover, the debate with Kah- interpersonal regrets in the TV show “I Am Sorry” and real neman led to a coauthored adversarial collaboration with life regrets. Also, Bonnefon and Zhang (2008) asked partic- three studies, in which Gilovich et al. (1998) concluded that ipants to think of “something you personally regret” (p. 3, action regrets tend to primarily elicit hot emotions (e.g., one single regret only, did not specify whether the event anger) whereas inaction regrets tend to elicit feelings of is the most regretful or not) and found that the difference wistfulness (e.g., nostalgia) and despair (e.g., misery), for short-term regrets was minimal (48% inaction) whereas which may be the cause of the temporal differences. A fol- long-term regrets were more likely to be inactions. In con- low-up conceptual replication and extension by Leach and trast, several follow-up studies examining temporal pat- Plaks (2009) found support for the temporal pattern of re- terns of the action-effect indicated limited generalizability gret, mediated by the higher level of abstraction of distant and identified various possible boundary conditions. Byrne inaction regret. We note that we did not set out to investi- and McEleney (2000) failed to conceptually replicate sce- gate the mechanisms of the proposed temporal pattern of nario experiments Studies 3 and 4 in Gilovich and Medvec regret and to first focus on revisiting and reassessing the (1994) adapting Kahneman and Tversky’s (1982) investor core phenomenon. scenario. A plausible explanation is that in Byrne and McE- leney (2000) scenario experiments, the factual and counter- Choice of Article for Replication: Gilovich and factual consequences were matched for the actor whereas Medvec (1994) the counterfactual consequences might be perceived to pos- sibly be better than factual consequences for the non-actor. We chose the Gilovich and Medvec (1994) article based Byrne and McEleney (2000) argued that the temporal pat- on several factors: the absence of direct replications, its im- tern of action-inaction effect only occurs in “situations pact, lack of statistical power in empirical evidence, and where the counterfactual consequences of mentally undone mixed or null findings. inactions are unknown, and possibly better than the factual We chose to replicate Studies 1a, 3, 4, and 5 as these consequences” (p. 1330). Moreover, Towers et al. (2016) studies focused on the intensity of regret rather than the asked participants about their single greatest regret in life number of action-inaction regrets, and were a better fit for and found action regrets were more intense than inaction our target sample, as Study 2 involved face-to-face inter- regrets, contradicting Gilovich and Medvec (1994) findings. Collabra: Psychology 2 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Towers et al. (2016) did not directly contrast actions versus based on the information provided, reported in Supplemen- inactions, but rather compared intensities of regret coded tary Table 4. as action or inaction in reports of regretful events and mea- Extension: Responsibility sured temporal distance continuously but not categorically (lifetime vs recent). Another highly cited article by Feldman We aimed to extend the replication to investigate the et al. (1999) asked participants about personal experiences generalizability of the temporal action-inaction effects in of regrets and found that participants reported higher num- regret findings to responsibility. Regret is associated with bers of long-term inaction regrets compared to action re- evaluations of self-agency and self-blame, key components grets but failed to find support for difference in intensity of of responsibility (Frijda et al., 1989; Zeelenberg et al., action-inaction regret, which is the focus of our replication 2002). Most of the evidence on the regret-responsibility link (Studies 1, 3, 4 and 5 of Gilovich & Medvec, 1994). in the context of action-inaction is based on hypothetical The above studies differ from Gilovich and Medvec scenario experiments. There are only few real-life expe- (1994) in methods or scenarios, reaching different conclu- rience surveys on regret-responsibility in the action-inac- sions. It is unclear if the failure to support Gilovich and tion literature, but there have been some real-life expe- Medvec (1994) is due to original results being unreplicable rience successful demonstrations on regret-responsibility or the differences in methods or scenarios. link outside the action-inaction literature (e.g. Breugel- Gilovich and Medvec (1994) inspired later work with im- mans et al., 2014). portant possible implications on regret. Later work pro- We note that regret and responsibility are positively cor- posed that inaction regrets may be more distressing and related yet separate constructs. Ordónez and Connolly depressing over a longer period (Broomhall et al., 2017; (2000) argued that some people experience some levels of Gilovich & Medvec, 1995), perhaps because negative feel- regret over outcomes that they have no agency over (e.g. the ings and senses of disquiet are stronger when one cannot outcome was reached by computer reassignment). There are fulfill the need for action (Roese et al., 1999). There appears situations in which the decision-maker experiences limited to be evidence that in the long run, people tend to ruminate responsibility but stronger regret, such as choosing a lesser- over inactions more compared to actions (Gilovich et al., known product brand (Simonson, 1992). Another plausible 1995; Savitsky et al., 1997), possibly because they perceive key difference is that regret tends to be associated with or imagine more possibilities of counterfactual outcomes counterfactual thoughts (Huang & Zeelenberg, 2012) com- from inactions compared to actions with clearer links to pared to responsibility, which is more strongly associated outcomes (Leach & Plaks, 2009; Rajagopal et al., 2006). with agency, causality, and morality (Connolly et al., 1997; However, before getting into mechanisms and practical im- Kordes-de Vaal, 1996). There may be discrepancies in ac- plications to sort out this literature, we believe it is essen- tion-inaction regret and responsibility findings, yet there tial to revisit the classic effects and assess their reliabil- are several studies reporting a positive regret-responsibility ity and replicability (e.g., Brandt et al., 2014; Zwaan et al., link (e.g. Ordónez & Connolly, 2000; Zeelenberg et al., 2018), with preregistered high-powered direct replications 2000, 2002). To the best of our knowledge, there are no and extensions. studies that compared long-term feelings of responsibility regarding action vs inaction. We expected findings for re- Methods, Hypotheses, and Findings of the Target gret to extend similarly to responsibility. See Table 2 for the Article extension hypotheses. The original Study 1 was conducted with adult partici- Method pants on the telephone, asking participants to compare in- tensity of their action and inaction regrets in general (Study Transparency and Openness 1A), and compare intensity of their greatest action regret We report the determination of sample size, all data ex- and greatest inaction regret (Study 1B). The original Studies clusions, all manipulations, and all measures in our studies 3 (within-subject) and 4 (temporal distance as the between- (Simmons et al., 2012). This manuscript is in line with Ap- subject factor) were conducted with undergraduates and us- pelbaum et al. (2018) Journal Article Reporting Standards ing scenario experiments that asked participants to com- (JARS) and Transparency and Openness Promotion (TOP) pare the intensity of regret of the decision-makers in the Guidelines (Nosek et al., 2015). We preregistered designs short run and the long run. The original Study 5 was con- and analysis plans of all studies before data collection. Pre- ducted with participants in public areas, asking participants registration, all data, code, and materials are available on to compare the intensity of regret of their greatest action regret and greatest inaction regret in the past week and in the Open Science Framework (OSF): https://osf.io/342td/ their lifetime. We did not include Study 1B in our replica- and https://osf.io/7m3q2/. Open-science details, disclo- tion, as it consisted of questions on greatest lifetime regret, sures, original effects calculations, power analyses, and pre- exclusion results are provided in the supplementary. We which overlapped with those of Study 5. analyzed data using RMarkdown (Xie et al., 2018, see We summarized the hypotheses in Table 1. The original RMarkdown output in site for analyses with the list of pack- authors hypothesized that there would be stronger regret ages) with RStudio version 1.3.1073 (RStudio Team, 2021) for inaction in the long run and stronger regret for action in and produced plots with the package ggplot2 version 3.3.3 the short run. We provide more details regarding the orig- (Wickham, 2016). inal article in the supplementary. We calculated Cramer V Collabra: Psychology 3 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Table 1. Gilovich and Medvec (1994): Summary of hypotheses Study Hypothesis Study 1 real-life Participants are more likely to report having experienced regret for life's inactions compared to Regret over past action- life's actions. inactions Study 3 (within-subject) Participants are more likely to associate stronger regret with recent actions than with recent Study 4 (between-subject) inactions. Hypothetical scenarios Participants are more likely to associate stronger regret with distant past inactions than with distant past actions. Study 5 real-life Participants are more likely to experience stronger regret over their most regrettable recent Regret over recent versus action than over their most regrettable recent inaction. distant past action-inactions Participants are more likely to experience stronger regret over their most regrettable distant past inaction than over their most regrettable distant past action. Table 2. Summary of extension hypotheses Study Hypothesis Study 1 real-life Participants are more likely to report feeling more responsible for life's inactions compared to Responsibility for past action- life's actions. inactions Study 3 (within-subject) Participants are more likely to associate stronger responsibility with recent actions than with Study 4 (between-subject) recent inactions. Hypothetical scenarios Participants are more likely to associate stronger responsibility with distant past inactions than with distant past actions. Study 5 real-life Participants are more likely to experience stronger feelings of responsibility for their most Responsibility for recent responsible recent past action compared to their recent most responsible inaction. versus distant past action- Participants are more likely to experience stronger feelings of responsibility for their most inactions responsible distant past inaction compared to their most responsible distant past action. making phenomena has shown MTurk, with Participants TurkPrime.com/CloudResearch to be a highly suitable plat- We recruited US-American participants from Amazon form for this research design (Collaborative Open-science Mechanical Turk (MTurk) with TurkPrime.com/ Research, 2022). Recently, Eyal et al. (2021) compared levels CloudResearch (Litman et al., 2017). Based on our extensive of attention, comprehension, and dishonesty of partici- experience of running similar replications on MTurk, to en- pants between several platforms and panels and found that sure high-quality data collection, we employed the follow- CloudResearch and Prolific provided higher quality com- ing CloudResearch options: Duplicate IP Block. Duplicate pared to other methods (Qualtrics, MTurk without Geocode Block, Suspicious Geocode Block, Verify Worker CloudResearch, Dynata). Country Location, Enhanced Privacy, CloudResearch Ap- A total of 1017 participants completed the study. We proved Participants, Block Low Quality Participants, etc. excluded 29 participants based on our pre-registered ex- We also employed the Qualtrics fraud and spam prevention clusion criteria (see supplementary for details), resulting measures: reCAPTCHA, prevent multiple submission, pre- in a total sample of 988 participants (M = 43.94, SD = age vent ballot stuffing, bot detection, security scan monitor, 13.62; 566 females, 408 males, 8 others, 6 prefer not dis- and relevant ID. MTurk has been shown to be a reliable plat- closing their gender). We report full results comparing pre- form for conducting studies in social psychology, judgment, exclusions versus post-exclusions in the supplementary. We and decision-making (Anderson et al., 2019; Buhrmester et provide a comparison of the target article samples and the al., 2011; Thomas & Clifford, 2017). Several recent stud- replication samples in Table 3. ies on the action-effect (e.g. G. Feldman, 2020; G. Feldman To estimate the required sample size, we used pwr pack- & Albarracín, 2017) and a recent large-scale collaborative age version 1.3 (Champely et al., 2018) and conducted an project with over 80 replications of judgment and decision- a-priori power analysis for chi-square goodness of fit 50-50 1 567 out of 1584 participants decided to drop out during the survey, likely because of the writing description task warnings. Collabra: Psychology 4 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Table 3. Differences and similarities between Gilovich and Medvec (1994) and our replication Gilovich and Medvec (1994) Replication and extension Sample size Study 1A: 60, Study 1B: 30, Study 3: 80, Study 4: 76, Combined sample: 988 after exclusion Study 5: 32 Geographic origin United States United States Gender Not reported 566 females, 408 males, 8 others, 6 prefer not disclosing their gender Median age (years) Not reported 42 Average age (years) Study 1A: 40.3, Study 1B: 40.1. Not reported for 43.94 other studies Standard deviation Not reported 13.62 age (years) Age range (years) Not reported 18-89 Medium (location) Telephone (Study 1), on the streets (Study 5), and Computer (online), Amazon Mechanical Turk lab (Study 3 and Study 4) Compensation Not reported Nominal payment: $0.8 USD/participant Year 1994 or before 2021 tests, comparing the proportion of action with stronger re- overall data quality. When some of the findings replicate gret versus inaction with stronger regret, and chi-square and others do not, combining the studies allows ruling out tests of association, testing the association between tempo- inattentiveness as a concern, adequacy of the target sample ral distance and action-inaction regret. We calculated and for these replications, or the adjustments to updated con- reported the original effect sizes in Supplementary Table 4. text (time, setting, etc.), so that we can instead focus on Aiming for a statistical power of 95% with an alpha of .05, the implications regarding specific designs and found ef- and based on the weakest meaningful effect (V = 0.24) that fects. This design has been tested and shown to be success- the original authors hypothesized and claimed to find sup- ful in several recent replications (Adelina & Feldman, 2021; port for, with more participants perceiving stronger regret Chandrashekar et al., 2021; Chen et al., 2021; Ziano et al., for inaction than for action in the long-term in Study 4 (but 2021). p > .05), the required sample was 920 participants. As we ex- Deviations from Original Studies pected some participants to be excluded, we aimed for 1000 participants. We provide more details in the supplementary. We provided detailed information of designs (type of study, sample, variables, exact wordings) of the original Design and Procedure studies in the Methods and Analyses of the original article We made adjustments to the design of the original stud- section of the Supplementary. We note several deviations ies. Extending the original studies and deviating from their from the original, summarized in Table 3 and Table 4. We procedures, we combined the replications of Studies 1A, 3, combined studies into a single survey, and we added re- 4, and 5 into a singular design in one data collection, with sponsibility questions as extensions. Finally, we recruited added extensions examining responsibility. First, partici- participants through MTurk online, instead of participants pants read the consent form. We first presented Study 5, from New York and Chicago, or Cornell University students. followed by Study 1. Both were personal experience stud- Replications Classification ies. We then randomized participants into either Study 3, a within-subject design, or Study 4, a between-subject de- We summarized Studies 1, 3, and 4 as “very close repli- sign, as Studies 3 and 4 consisted of the same hypothetical cations” and Study 5 as a “close replication” based on the scenario and questions. We placed Study 3/Study 4 at the criteria by LeBel et al. (2018) (see Figure S3 in the Supple- end to prevent the stimuli in the scenario from affecting mentary), with our classification analysis provided in Table personal responses. We then presented participants with funneling and demographic questions, followed by a de- briefing statement. See below sections for more specific and Study 5: Method detailed information about all studies. We note that the study numbers below are based on the study numbers of the First, in Study 5, participants answered questions regard- original article, but not the order of our replications. ing their greatest regrets. Temporal distance (the past week We decided on this design in order to address possible vs entire life, in counterbalanced order) was the indepen- concerns regarding the sample. Despite our ample expe- dent variable. We asked participants to think about and rience and accumulated evidence in support of validity of describe their greatest lifetime action-regret and greatest our chosen MTurk/CloudResearch sample for replications of lifetime inaction-regret, as well as greatest past week ac- classics in judgment and decision-making, reviewers often tion-regret and greatest past week inaction-regret. We then expressed concerns about online samples regarding inat- asked them which they regretted more. In the original tentiveness, suitability to context (time, setting, etc.), and Collabra: Psychology 5 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Table 4. Classification of the replication, based on LeBel et al. (2018) Design facet Replication Details of deviation Effect/hypothesis Same IV construct Same DV construct Same IV Same operationalization DV Same operationalization Population (e.g. Similar Both with American participants. However, our replications consist of age) participants from Amazon Mechanical Turk, instead of adults from New York and Chicago, or Cornell University undergraduate students in the original IV stimuli Different for Study 5; Study 5: Minor changes to ensure the wordings across conditions and action vs Same for Studies 1, 3, inaction are consistent and 4 DV stimuli Different for Study 5; Study 5: Minor wording change to Study 5 regret question. We removed “which Same for Studies 1, 3, one would you “undo” if you could”. We also asked participants to describe their and 4 regrets and responsibilities* in Study 5 (which was not required in the original), as this lowers the chance of quick irrelevant or random responses, ensuring participants are thinking about the task and responding seriously. We asked participants for brief descriptions and reminded them that they did not have to disclose information they did not feel comfortable with. Procedural details Different We combined Studies 1, 3, 4, and 5 into a single Qualtrics survey. The original article used separate samples. Physical settings Different Online data collection in our replication vs real-life and telephone data collection in the original Contextual Different The original authors conducted their studies in the early 1990s whereas we variables conducted our replications in 2021. Replication Studies 1, 3, 4: Very For our Study 5, the IV stimuli and the DV stimuli are different from that of the classification close replication; original study. Study 5: Close replication * Additionally, we checked descriptions of participants and conducted exploratory analyses excluding incorrect and irrelevant descriptions of action-inaction regrets or responsibili- ties. The findings are reported in the Supplementary Exploratory Analyses of Study 5 section. Such findings are very similar to findings reported in the main manuscript. study, participants were only required to think about but Study 5: Results not describe their regrets. However, in our replication we Replication: Regret asked participants to briefly write about their regrets. By having participants briefly describe their regrets, we felt For past week regrets, with the chi-square goodness-of- they would be more likely to engage in effortful reflections, fit test, we failed to find support for a deviation from a 50-50 and less likely to respond randomly, thereby ensuring better split in participants experiencing stronger regret for the ac- data quality. We reminded participants that they do not tion (52.34%) than inaction (47.66%), z = 1.04, χ2 (1, N = need to disclose any information they feel uncomfortable 535) = 1.17, p = .280, V = 0.05, 95% CI [0.00, 0.13] (see Figure sharing. We also went to great lengths to align expectations 1 top left plot). about the task – we made it clear in our study recruitment For lifetime regrets, with the chi-square goodness-of-fit and with a specific question in the consent screen that the test, we failed to find support for a deviation from 50-50 task involves brief writing and that the study is about life split in participants experiencing stronger regret for the ac- regrets (see “Study recruitment” and “Writing task expecta- tion (52.15%) than inaction (47.85%), z = 0.95, χ2 (1, N = tion alignment in consent” in the supplementary). 535) = 0.99, p = .320, V = 0.04, 95% CI [0.00, 0.13] (see Figure We randomized participants to either answer the repli- 1 top right plot). cation questions or extension questions. In the extension Comparing the proportion of participants experiencing condition, we asked participants about the action decision stronger action regret in the past week and experiencing that they felt most personally responsible for and the in- stronger inaction regret in lifetime (133/265, 265 is the total action decision that they felt most personally responsible number of participants who showed reversal, 50.19%), ver- for, in the past week and in their lifetime (in counterbal- sus the proportion of participants choosing inaction in the anced order). Similarly, we asked participants to very briefly past week and choosing action in lifetime (132/265, describe these events. They then answered which of those 49.81%), we failed to find support for a deviation from 50-50 they felt more responsible for. Full details are provided in distribution, z = 0.00, χ2 (1, N = 265) = 0.00, p = .951, V Table S5 in the supplementary. = 0.00, 95% CI [0.00, 0.14]. We also conducted a McNemar test, and failed to find support for the association between Collabra: Psychology 6 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Figure 1. Study 5: Action-inaction regret and responsibility - short-term (past week) and long-term (lifetime) Note. Error bars denote 95% confidence intervals. Short term = past week. Long term = lifetime. temporal distance and inaction-action regret, OR = 0.99, Study 1a: Method 95% CI [0.77, 1.27], p = 1. After Study 5, we presented Study 1. We asked partici- Extension: Responsibility pants about their action and inaction regretful experiences - “When you look back on your experiences in life and think For the responsibility over the week, with the chi-square of those things that you regret, what would you say you re- goodness-of-fit test, we found support for a deviation from gret more, those things that you did but wish you hadn’t, or 50-50 split in participants experiencing stronger respon- those things that you didn’t do but wish you had?” (Gilovich sibility for the action (56.29%) versus participants experi- & Medvec, 1994, p. 358). We asked another group of partic- encing stronger responsibility for the inaction (43.71%), z = ipants regarding their felt responsibility for life’s personally 2.63, χ2 (1, N = 453) = 7.17, p = .007, V = 0.13, 95% CI [0.03, responsible actions and inactions - “When you look back on 0.22]. More participants felt stronger responsibility for ac- your experiences in life and think of those things that you tion compared to inaction (see Figure 1 bottom left plot). feel personally responsible for, what would you say you feel For the responsibility over the lifetime, with the chi- personally responsible more, those things that you did but square goodness-of-fit test, we found support for a devia- wish you hadn’t, or those things that you didn’t do but wish tion from 50-50 split in participants experiencing stronger you had?”. responsibility for the action (56.51%) versus participants experiencing stronger responsibility for the inaction Study 1a: Results (43.49%), z = 2.73, χ2 (1, N = 453) = 7.68, p = .006, V = 0.13, Replication: Regret 95% CI [0.03, 0.22]. More participants felt stronger respon- sibility for action compared to inaction (see Figure 1 bottom We began by examining regret, we conducted a chi- right plot). square goodness of fit test against a 50-50 action-inaction Comparing the proportion of participants experiencing split , z = -5.79, χ2 (1, N = 535) = 34.07, p < .001, V = stronger action responsibility in the past week and experi- 0.25, 95% CI [0.17, 0.34], and found support for stronger re- encing stronger inaction responsibility in lifetime (100/201, gret for inactions; more participants reported experiencing 49.75%), versus the proportion of participants choosing in- stronger regret over lifelong inactions (62.62%) than those action in the past week and choosing action in lifetime (101/ reporting experiencing stronger regret over lifelong actions 201, 50.25%), we failed to find support for a deviation from (37.38%) (see Figure 2 left plot). 50-50, z = 0.00, χ2 (1, N = 201) = 0.00, p = .944, V = 0.00, 95% CI [0.00, 0.16]. We also conducted a McNemar test, and Extension: Responsibility failed to find support for the association between temporal To examine our responsibility extension, we conducted a distance and action-inaction responsibility, OR = 1.01, 95% chi-square goodness of fit test against a 50-50 action-inac- CI [0.76, 1.35], p = 1. tion split, z = 4.89, χ2 (1, N = 453) = 24.34, p < .001, V = 0.23, Collabra: Psychology 7 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Figure 2. Study 1A: Action-Inaction Regret and Responsibility Note. Error bars denote lower confidence intervals and upper confidence intervals (95%). 95% CI [0.14, 0.32], and found that more participants re- action Jim (42.39%), yet this did not meet our pre-defined ported stronger responsibility over lifelong regrettable ac- alpha leading us to conclude no support, z = -1.59, χ2 (1, N tions (61.59%) than those reporting stronger responsibil- = 248) = 2.73, p = .099, V = 0.10, 95% CI [0.01, 0.23] (see Fig- ity for lifelong regrettable inactions (38.41%) (see Figure 2 ure 3 top right for the plot). right plot). We conducted a chi-square test of independence and found support for the association between temporal dis- Studies 3 and 4: Method tance and action-inaction regret, χ2 (1, N = 495) = 14.68, p < .001, V = 0.17, 95% CI [0.08, 0.26]. We found that com- After completing Studies 5 and 1a, we presented par- pared to short-term, long-term perspective was associated ticipants with scenario experiments on college decisions: with stronger perceived regret for inaction. Inaction Dave stayed in the same college whereas Action Jim switched to another college, and both were unsatisfied. Extension: Responsibility We randomized participants into either a within-subject de- sign as in Study 3 (participants compared feelings of regret Examining responsibility in the short-term condition, we and responsibility of Dave vs. Jim in both short-term and conducted a chi-square goodness of fit test and found sup- long-term), or a between-subject design as in Study 4 (par- port for more participants perceiving stronger responsibil- ticipants answered short-term or long-term questions). In ity for action Jim (61.94%) than for inaction Dave (38.06%), Studies 3 and 4, participants answered both replication and z = 7.76, χ2 (1, N = 247) = 61.25, p < .001, V = 0.50, 95% CI extension questions. We provide more details on the de- [0.39, 0.60] (see Figure 3 bottom left for the plot). signs of Studies 3 and 4 in the supplementary’s Tables 7 and Examining responsibility in the long-term condition, we 8). conducted a chi-square goodness of fit test and found sup- port for more participants perceiving stronger responsibil- Study 4 (between-subject): Results ity for action Jim (74.90%) than for inaction Dave (25.10%), z = 5.78, χ2 (1, N = 248) = 34.13, p < .001, V = 0.37, 95% CI Replication: Regret [0.25, 0.49] (see Figure 3 bottom right for the plot). In the short-term condition of the between-subject de- We conducted a chi-square test of independence and sign study, we conducted a chi-square goodness of fit test failed to find support for the association between temporal and found support for more participants perceiving distance and action-inaction responsibility, χ2 (1, N = 495) stronger regret for action Jim (60.65%) than for inaction = 2.46, p = .117, V = 0.07, 95% CI [0.00, 0.16]. Dave (39.35%), z = 3.69, χ2 (1, N = 247) = 14.09, p < .001, V = 0.24, 95% CI [0.12, 0.36] (see Figure 3 top left for the plot). In the long-term condition, we conducted a chi-square goodness of fit test and found that more participants per- ceiving stronger regret for inaction Dave (57.61%) than for Collabra: Psychology 8 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Figure 3. Study 4: Short-Term and Long-Term Action-Inaction Regrets and Responsibilities Plots Note. Error bars denote lower confidence intervals and upper confidence intervals (95%). Study 3 (within-subject): Results Extension: Responsibility Replication: Regret Examining responsibility in the short-term condition, we conducted a chi-square goodness of fit test and found sup- In the short-term condition of the within-subject design port for more participants perceiving stronger responsibil- study, we conducted a chi-square goodness of fit test ity for action Jim (73.83%) than for inaction Dave (26.17%), (meant to mirror the analyses for Study 4 to allow for a com- z = 10.54, χ2 (1, N = 493) = 112.02, p < .001, V = 0.48, 95% CI parison) and found support for more participants perceiv- [0.40, 0.55] (see Figure 4 bottom left plot). ing stronger regret for action Jim (61.26%) than for inaction Examining responsibility in the long-term condition, we Dave (38.74%), z = 4.95, χ2 (1, N = 493) = 24.99, p < .001, V = conducted a chi-square goodness of fit test and found sup- 0.23, 95% CI [0.14, 0.31] (see Figure 4 top left plot). port for more participants perceiving stronger responsibil- In the long-term condition, we conducted a chi-square ity for action Jim (60.65%) than for inaction Dave (39.35%) goodness of fit test and found support for more participants , z = 4.68, χ2 (1, N = 493) = 22.36, p < .001, V = 0.21, 95% CI perceiving stronger regret for inaction Dave (57.61%) than [0.13, 0.29] (see Figure 4 bottom right plot). for action Jim (42.39%), z = -3.33, χ2 (1, N = 493) = 11.41, p Comparing the proportion of participants choosing ac- < .001, V = 0.15, 95% CI [0.06, 0.24] (see Figure 4 top right tion Jim in the short term and choosing inaction Dave in plot). the long term (94/123, 76.42%), versus the proportion of Comparing the proportion of participants choosing ac- participants choosing inaction Dave in the short term and tion Jim in the short term and choosing inaction Dave in the choosing action Jim in the long term (29/123, 23.58%), we long term (113/133, the total number of participants who found support for a deviation from 50-50, z = 5.77, χ2 (1, N showed reversal in answers, 84.96%), versus the proportion = 123) = 34.35, p < .001, V = 0.53, 95% CI [0.37, 0.67]. More of participants choosing inaction Dave in the short term participants chose stronger responsibility for action in the and choosing action Jim in the long term (20/133, 15.04%), short term and for inaction in the long term, compared to we found support for a deviation from 50-50, z = 7.98, χ2 (1, stronger responsibility for inaction in the short term and for N = 133) = 65.03, p < .001, V = 0.70, 95% CI [0.56, 0.80]. More action in the long term. We also conducted a McNemar test, participants chose action in the short term and inaction and found support for the association between temporal in the long term, compared to inaction in the short term distance and action-inaction responsibility, OR = 0.31, 95% and action in the long term. We also conducted a McNemar CI [0.20, 0.47], p < .001. Temporal distance had an impact test, and found support for the association between tempo- on the choice distribution. The difference in the proportion ral distance and action-inaction regret, OR = 0.18, 95% CI between action and inaction was weaker in the long-term [0.10, 0.29], p < .001. compared to that in the short-term. The effect was in the same direction as that in regret, but did not lead to a com- plete reversal of perceptions. Collabra: Psychology 9 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Figure 4. Study 3: Short-Term and Long-Term Action-Inaction Regrets and Responsibilities Plots Note. Error bars denote lower confidence intervals and upper confidence intervals (95%). for action versus inaction. More participants reported Overall Summary of Findings All Studies stronger regret for action in the short-run, but stronger re- We summarized descriptive statistics of all studies in gret for inaction in the long run. Table 5. We summarized the comparison of all effects of In Study 5, we failed to find support for an action-effect original studies versus our replications in Table 6, with an in the short-term, failed to replicate findings for the long- interpretation of the results based on the LeBel et al. (2019) term, and the proposed association between temporal dis- replication results evaluation criteria. tance and action-effect. Comparing Replication Findings to Original Possible Reasons behind the Discrepancy in Findings and Extensions Findings of Studies 1, 3, and 4 versus Study 5 We successfully replicated and found support for the Why did the replication of Study 5 fail while the replica- original findings in Studies 1, 3, and 4 (short-term regret tion of the other studies succeeded? and temporal effect) with smaller effect sizes. For our repli- Previous studies have shown MTurk to be a reliable plat- cation of Study 4’s long-term regret, the CIs of the replica- form for the study of action and inaction, and judgment tion did not cover the original effect size, and we failed to and decision-making more broadly (e.g. G. Feldman, 2020; find support for the effect (which was the case in both the G. Feldman & Albarracín, 2017). Our design and the other original and the replication). successful replications of Studies 1, 3, and 4 address con- We failed to successfully replicate Study 5. We failed to cerns regarding sample characteristics or time, given that find support for an action-effect in the short-term, and they were conducted using the same sample. Therefore, we failed to find support for an inaction-effect in the long- believe the more plausible explanations are the differences term. in methods and the likelihood of a false positive. We note Regarding our responsibility extensions, we conclude that the sample size in the original Study 5 was 32, with a stronger responsibility for action over inaction across all much higher likelihood of a false-positive. Our sample was studies, for both short-term and long-term. substantially larger and well-powered, yet we were unable to detect the inaction-effect in the long run. Discussion Why would the method used in Study 5 result in different findings? We believe this might have to do with the ways ac- We conducted a pre-registered replication of the tempo- tion and inaction are conceptualized in the different stud- ral pattern of action-effect by Gilovich and Medvec (1994), ies. In Study 5 action refers to “something they did” versus with a more diverse (Buhrmester et al., 2011) and high- inaction as “something they did not do”. In Study 3 and powered sample. We successfully replicated Study 1, which Study 4, action is conceptualized as a switching behavior, focused on general regrets, as well as Studies 3 and 4, which a change to the status quo, versus inaction, which is stick- were scenario studies asking participants to compare regret ing with the status quo. Unfortunately, these issues seem to Collabra: Psychology 10 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Table 5. Summary of descriptive statistics for all studies for both regret and responsibility Study Action Action Inaction Inaction Count Percentage Count Percentage Study 1 General Regret 200/ 37.38%, 95% CI [33.39%, 335/ 62.62%, 95% CI [58.44%, 535 41.56%] 535 66.61%] Study 1 General Responsibility 279/ 61.59%, 95% CI [57.03%, 174/ 38.41%, 95% CI [34.05%, 453 65.95%] 453 42.97%] Study 3 Short Term Regret 302/ 61.26%, 95% CI [56.89%, 191/ 38.74%, 95% CI [34.54%, 493 65.46%] 493 43.11%] Study 3 Long Term Regret 209/ 42.39%, 95% CI [38.11%, 284/ 57.61%, 95% CI [53.20%, 493 46.80%] 493 61.89%] Study 3 Short Term Responsibility 364/ 73.83%, 95% CI [69.78%, 129/ 26.17%, 95% CI [22.48%, 493 77.52%] 493 30.22%] Study 3 Long Term Responsibility 299/ 60.65%, 95% CI [56.27%, 194/ 39.35%, 95% CI [35.14%, 493 64.86%] 493 43.73%] Study 4 Short Term Regret 153/ 61.94%, 95% CI [55.75%, 94/ 38.06%, 95% CI [32.23%, 247 67.77%] 247 44.25%] Study 4 Long Term Regret 111/ 44.76%, 95% CI [38.70%, 137/ 55.24%, 95% CI [49.02%, 248 50.98%] 248 61.30%] Study 4 Short Term Responsibility 185/ 74.90%, 95% CI [69.14%, 62/ 25.10%, 95% CI [20.10%, 247 79.90%] 247 30.86%] Study 4 Long Term Responsibility 170/ 68.55%, 95% CI [62.52%, 78/ 31.45%, 95% CI [25.99%, 248 74.01%] 248 37.48%] Study 5 Greatest Past Week Regret 280/ 52.34%, 95% CI [48.10%, 255/ 47.66%, 95% CI [43.46%, 535 56.54%] 535 51.90%] Study 5 Greatest Lifetime Regret 279/ 52.15%, 95% CI [47.92%, 256/ 47.85%, 95% CI [43.65%, 535 56.35%] 535 52.08%] Study 5 Greatest Past Week 255/ 56.29%, 95% CI [51.69%, 198/ 43.71%, 95% CI [39.21%, Responsibility 453 60.79%] 453 48.31%] Study 5 Greatest Lifetime 256/ 56.51%, 95% CI [51.91%, 197/ 43.49%, 95% CI [39.00%, Responsibility 453 61.00%] 453 48.09%] be widespread in this literature, with recent reviews alert- Beyond differences in meaning, another possible expla- ing that action and inaction are often ill-defined terms and nation is regarding the differences between perceptions of in urgent need of clarifications (G. Feldman et al., 2021). others’ emotions in Studies 3 and 4 and the evaluation of Our findings are consistent with a large body of literature personal actual experiences of emotions elicited in Study 5. showing support for an action effect in high-risk recent sit- Perceptions of emotions in others tend to be less accurate uations that result in negative outcomes. The typical ac- and differ from actual personal experiences, especially if tion-effect scenarios refer to changing, switching, or devi- there is no personal relevance to the situation evaluated ating from a set reference point (e.g., past behavior, status in the presented scenario. The failed replication of Study 5 quo) (e.g. G. Feldman & Albarracín, 2017; Gleicher et al., seems consistent with most studies that directly ask partici- 1990; Landman, 1987), with norm theory (Kahneman & pants about their personal experiences (Bonnefon & Zhang, Miller, 1986) arguing that is likely due to action being per- 2008; J. Feldman et al., 1999 ; Towers et al., 2016), but in- ceived as more abnormal than inaction (G. Feldman, 2020), consistent with Zeelenberg et al. (1998) studies, in which and that exceptionality tends to elicit higher regret than they found support for temporal pattern of regret in real life normality (Kutscher & Feldman, 2019). The meaning of ac- experiences for interpersonal regrets (they did not test other tion and inaction in Studies 3 and 4 was closer to the typical kinds of regrets). action-effect and norm theory scenarios (Kahneman & Another possible explanation for our different findings Miller, 1986). However, in Study 5, the meaning of action in Study 5 is a change we made to the original study’s de- was far broader with no clear reference to a norm or a refer- sign. We required participants to describe their regrets, ence point. Therefore, the differences between action as in whereas Gilovich and Medvec (1994) only required partic- doing and inaction as in not doing seem less clear and with ipants to recall their regrets without writing those down. no clear indication of what to compare to. We, however, find this explanation unlikely. The target’s concern was that participants may be unwilling to describe 2 Feldman et al. (1999) found support for higher frequencies of inaction regrets compared to action regrets, but failed to find support for in- tensity differences between action and inaction regrets. Study 5 focuses on intensity but not frequency. Collabra: Psychology 11 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Table 6. Summary of statistical tests and comparison with the original effect sizes Chi- p Replication Original Interpretation square Cramer V and Cramer V CI and CI Study 1 - Chi-Square Goodness of Fit Test General Regret χ2 (1, N = < .001 V = 0.25, 95% V = 0.50, Signal, inconsistent, smaller (successful 535) = CI [0.17, 0.34] 95% CI replication) 34.07 [0.27, 0.70] Study 3 - Chi-Square Goodness of Fit Test Short Term χ2 (1, N = < .001 V = 0.23, 95% V = 0.53, Signal, inconsistent, smaller (successful Regret 493) = CI [0.14, 0.31] 95% CI replication) 24.99 [0.35, 0.70] Long Term Regret χ2 (1, N = < .001 V = 0.15, 95% V = 0.28, Signal, inconsistent, smaller (successful 493) = CI [0.06, 0.24] 95% CI replication) 11.41 [0.05, 0.48] Action-inaction χ2 (1, N = < .001 V = 0.70, 95% Insufficient Signal, the effect size cannot be directly vs. Temporal 133) = CI [0.56, 0.80] information compared, but successful replication Change 65.03 Study 4 - Chi-Square Goodness of Fit Test and Test of Independence Short Term χ2 (1, N = < .001 V = 0.24, 95% V = 0.53, Signal, inconsistent, smaller (successful Regret 247) = CI [0.12, 0.36] 95% CI replication) [0.24, 0.76] 14.09 Long Term Regret χ2 (1, N = = .099 V = 0.10, 95% V = 0.24, Unclear (see notes) (likely successful 248) = CI [0.01, 0.23] 95% CI replication) 2.73 [0.00, 0.52] Action-inaction χ2 (1, N = < .001 V = 0.17, 95% V = 0.38, Signal, inconsistent, smaller (successful vs. Temporal 495) = CI [0.08, 0.26] 95% CI replication) Change 14.68 [0.16, 0.61] Study 5 Short Term χ2 (1, N = = .280 V = 0.05, 95% V = 0.06, No signal, consistent (successful Regret 535) = CI [0.00, 0.13] 95% CI replication) 1.17 [0.00, 0.44] Long Term Regret χ2 (1, N = = .320 V = 0.04, 95% V = 0.56, No-signal, inconsistent (failed 535) = CI [0.00, 0.13] 95% CI replication) 0.99 [0.25, 0.81] Temporal χ2 (1, N = = .951 OR = 0.99, Insufficient No signal, likely failed replication Distance and 265) = 95% CI [0.77, information Action- 0.00 1.27] Inaction Regret Note. 1) We conducted Chi-Square goodness of fit tests for the above studies, except for Study 4 association between Temporal Distance and Action and Inaction Regret, in which we conducted a Chi-Square test of independence. The interpretation of outcome is based on LeBel et al. (2019). 2) For Study 4 long-term regret part, LeBel et al. (2019) Criteria B does not account that it is possible for a finding to not reach significance and for the effect size CIs to not cover the original effect size. This can be considered as a case of “no signal, inconsis- tent”. their very embarrassing and shameful regrets, yet taking a other platforms (Eyal et al., 2021). The successful repli- closer look at the responses (found in our dataset), we found cations of the other studies in our unified design address that many participants described highly personal, shame- concerns of attentiveness. We also addressed this concern ful, and somewhat tragic events, including events involving by conducting additional exploratory analyses (see the Ex- death, major career failures, major educational failures, ploratory Analyses of Study 5 section in the Supplementary) major relationship failures, etc. However, we cannot com- excluding responses in which participants seem to have pletely rule out the possibility of such differences having an misunderstood or confused action and inaction, or partic- impact on the findings, and future research can further test ipants reporting lifetime mistakes in past week mistakes this possibility by asking half of the participants to describe questions (or vice versa), and non-regret/non-responsibility a major regret and asking other participants to simply think responses. We found that only a very low % of participants about a major regret. (ranging from 1.68% for lifetime regret, to 7.28% for past Another possible explanation raised in the peer-review week responsibility) misclassified responses. Our results of was that Study 5 may not have been suitable for our online the exploratory analyses with those participants excluded MTurk target sample, given concerns of attentiveness and were consistent with results we reported above. seriousness. As we discussed earlier, MTurk with Studies 3 and 4 were also better controlled with specific CloudResearch/TurkPrime provides high-quality responses, scenarios, whereas Study 5 did not restrict the range of in which participants are as if not more attentive than on Collabra: Psychology 12 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... elicited regrets, which included many different domains in 2012), and more work is needed to contrast the two regard- life, such as education, work, relationships, finance, etc. ing action-effect and omission-bias. What is puzzling about the above explanations is that al- Possible Limitations and Future Directions though we failed to find support for lifetime recalls in Study 5, we did find support for the effects in Study 1. Both Study 1 We faced some challenges with the studies eliciting life and Study 5 asked participants about “things that they did” events. A small number of participants (1.68% to 7.28%) versus “things that they didn’t do”. Yet, one key difference wrote inaction events in the action description box or vice was that Study 1 elicited general regrets and asked partic- versa, with some participants reporting they did not expe- ipants to evaluate those together rather than contrasting rience any action or inaction regret or responsibility in the one most regrettable action against one most regrettable past week. In an online study we cannot rule out the pos- inaction. Therefore, it is possible that specificity is a mod- sibility that some participants may have copy-pasted, and erating factor of the effect. The autobiographical memory based on our experience there are indications that MTurk framework by Davison and Feeney (2008) suggested that re- participants typically dislike writing tasks. However, we gret is about remembering past events with different lev- tend to think that the likelihood of this being an issue in els of specificity and generality. They found that general re- our design is very low, as we adopted numerous quality con- grets were more likely to be for inactions over action, yet trol methods, aligned expectations in advance about the specific regrets were more likely to be for actions than inac- task, and only asked for brief descriptions in one or two tions. sentences. We checked all responses, and found that most The above proposed reasons for the discrepancies be- responses were of high-quality. We also conducted ex- tween studies in the same article are speculative, and we ploratory non-pre-registered analyses excluding possibly conclude that more work is needed to examine any of the irrelevant and incorrect responses (which were only a very proposed moderators with direct testing. small proportion of the entire sample) and the results (re- ported in the supplementary) were very similar to the re- Responsibility Extension sults after pre-registered exclusion or full results. There- Regret and responsibility are often positively related fore, we believe it is less likely that the null findings in (e.g. Zeelenberg et al., 2000, 2002), -yet are distinct con- Study 5 are due to this issue and find it more likely that such structs. We found consistent support for stronger respon- null findings are due to differences in meanings of action- sibility for action compared to inaction for the recent past, inaction. To resolve the discrepancy in findings between and the finding for responsibility in the recent past gener- Study 5 and Studies 1, 3, and 4, future studies can make ally aligned with that of regret. However, while there was adaptations to Study 5 by manipulating the definitions of some support for differences between recent past and dis- action and inaction (G. Feldman et al., 2021). Also, we can- tant past for responsibility, the effect seems much weaker not completely rule out the possibility that our adjustments than for regret, and we did not find a full reversal toward of adding brief writing to the Study 5 recall task may have stronger responsibility for inaction in the distant past. In- impacted the results, and so future studies may also com- stead, consistently across different designs, we also found pare findings of recall tasks that involve versus do not in- support for stronger responsibility for action for the distant volve writing the recalled memory, as well as test this phe- past events. Responsibility seems to be more strongly asso- nomenon with other non-MTurk/CloudResearch samples to ciated with morality, causality, and agency (Connolly et al., investigate if there are meaningful differences. 1997). Changes in perceived responsibility intensity per- We note that a single replication of a single article is in- haps fluctuate less over time compared to regret, which is a sufficient to answer all the questions in the literature with counterfactual emotion that may fluctuate over time more high certainty, and we call for more well-powered pre-reg- as people may feel more confident that they would have istered replications of work in this domain, preferably by made it if they had taken actions, thereby regretting inac- third-parties and in the form of Registered Reports. tion more in the long-run (Gilovich et al., 1993; Gilovich & We reported aggregated tendencies regarding temporal Medvec, 1995). That said, we note this explanation is spec- effects related to action and inaction yet there are individ- ulative and more work comparing mechanisms of regret and ual differences factors that may play a role in moderating responsibility in action-inaction studies is needed. these effects (e.g. action-state orientation, Diefendorff et These findings may hold important implications for the al., 2000; regulatory focus, Itzkin et al., 2016). link between the action-effect and omission-bias. Omission Many of the findings in this literature were conducted in bias extends the action-effect to reflect action-inaction mostly WEIRD settings (Western, Educated, Industrialized, asymmetries regarding responsibility and blame. More work Rich and Democratic; Henrich et al., 2010), and more re- is needed on the potential moderating effect of time with search is needed to study these effects in less WEIRD re- regret and responsibility examined together for both ac- gions, and/or include cultural dimensions as potential mod- tion-effect and omission-bias scenarios. erators of these effects. Our findings for responsibility in Study 3’s within-sub- In our extension, we found differences in findings re- ject design and Study 4’s between-subject design were garding regret and responsibility. Studies in the action-in- slightly different, with stronger effects for the within-sub- action literature rarely measure regret and responsibility ject design. There are quite a few judgment and decision- together, and more work is needed to investigate the associ- making effects that are stronger with within-subject design ations between the two constructs in the context of action- compared to between-subject design (Charness et al., inaction effects. Collabra: Psychology 13 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... We believe that more replications with extensions are Contributor Roles Taxonomy needed to better understand the robustness of the findings in this literature and examine new directions, together with Siu Kit Gilad meta-analyses of the action-inaction related literature Role Yeung Feldman (e.g., action-effect: Yeung & Feldman, 2022; omission bias: Conceptualization V Yeung et al., 2022), to examine possible moderating factors such as temporal distance, scenarios versus experience, be- Pre-registration V tween-subject versus within-subject study design compar- Data curation V ison, and the used meanings of action versus inaction. We Formal analysis V require a more comprehensive systematic aggregation of Funding acquisition V findings and insights to identify boundary conditions. Investigation V Conclusion Pre-registration peer review / verification V We conducted a replication and extension of Gilovich Data analysis peer review / verification and Medvec (1994) revisiting the temporal pattern of regret in action versus inaction and adding extensions examining Methodology V temporal pattern of responsibility. We found support for the Project administration V original findings on regret with different designs both ex- Resources amining lifelong experiences in Study 1 and hypothetical Software V scenarios in Studies 3 and 4, though with weaker effects. Supervision V However, we failed to find support for such a pattern in Study 5, and we discussed possible explanations. We called Validation for better conceptualizations of the terms action and in- Visualization V action in the literature, with more replications and exten- Writing-original draft V sions of classic studies of the action-inaction literature and Writing-review and editing V follow-up meta-analyses to help resolve inconsistencies in findings. We also reported the findings of an extension ex- In the table above, we employ CRediT (Contributor Roles Taxonomy) to identify the roles amining temporal pattern regarding attributions of respon- of the contributors. Check https://www.casrai.org/credit.html for more information about the roles. sibility, and discussed the regret-responsibility link. Competing Interests Acknowledgements The author(s) declared no potential conflicts of interest with respect to the authorship and/or publication of this ar- We would like to thank Qinyu Xiao, Adrien Fillon, Dr. ticle. Frances Jin. and Prof. Kin Fai Ellick Wong for their helpful comments. We thank the original author Thomas Gilovich Financial Disclosure/Funding for providing the original materials. The author(s) received no financial support for the re- Authorship Declaration search and/or authorship of this manuscript. Kit wrote the pre-registrations, conducted data analyses, Data Accessibility Statement and wrote the manuscript, as part of his mPhil thesis. Gilad Feldman supervised Siu Kit Yeung throughout, conducted Pre-registration, materials, data, and code are publicly the pre-registrations, and ran data collection. Gilad and Kit available on: https://osf.io/7m3q2/ jointly finalized the manuscript for journal submission. Submitted: June 07, 2022 PDT, Accepted: July 08, 2022 PDT This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CCBY-4.0). View this license’s legal deed at http://creativecommons.org/licenses/by/4.0 and legal code at http://creativecom- mons.org/licenses/by/4.0/legalcode for more information. Collabra: Psychology 14 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... References Adelina, N., & Feldman, G. (2021). Are past and future Chandrashekar, S. P., Yeung, S., Yau, K., Cheung, C., selves perceived differently from present self? Agarwal, T. K., Wong, C., Pillai, T., Thirlwell, T. N., Replication and extension of Pronin and Ross (2006) Leung, W., Li, Y., Tse, C., Cheng, B., Chan, H., & temporal differences in trait self-ascriptions. Feldman, G. (2021). Agency and self-other International Review of Social Psychology, 34(1), 29, asymmetries in perceived bias and shortcomings: 1–16. https://doi.org/10.5334/irsp.571 Replications of the Bias Blind Spot and extensions linking to free will beliefs. Judgment and Decision Anderson, C. A., Allen, J. J., Plante, C., Quigley- Making, 16(6), 1392–1413. https://sjdm.org/journal/2 McBride, A., Lovett, A., & Rokkum, J. N. (2019). The 0/201018/jdm201018.html MTurkification of social and personality psychology. Personality and Social Psychology Bulletin, 45(6), Charness, G., Gneezy, U., & Kuhn, M. A. (2012). 842–850. https://doi.org/10.1177/0146167218798821 Experimental methods: Between-subject and within- subject design. Journal of Economic Behavior & Appelbaum, M., Cooper, H., Kline, R. B., Mayo-Wilson, Organization, 81(1), 1–8. https://doi.org/10.1016/j.jeb E., Nezu, A. M., & Rao, S. M. (2018). Journal article o.2011.08.009 reporting standards for quantitative research in psychology: The APA Publications and Chen, J., Hui, L. S., Yu, T., Feldman, G., Zeng, S., Ching, Communications Board task force report. American T. L., Ng, C. H., Wu, K. W., Yuen, C. M., Lau, T. K., Psychologist, 73(1), 3. https://doi.org/10.1037/amp000 Cheng, B. L., & Ng, K. W. (2021). Foregone 0389 opportunities and choosing not to act: Replications of Inaction Inertia effect. Social Psychological and Bonnefon, J.-F., & Zhang, J. (2008). The intensity of Personality Science, 12(3), 333–345. https://doi.org/1 recent and distant life regrets: An integrated model 0.1177/1948550619900570 and a large scale survey. Applied Cognitive Psychology, 22(5), 653–662. https://doi.org/10.1002/acp.1386 Christley, R. M. (2010). Power and error: Increased risk of false positive results in underpowered studies. The Brandt, M. J., IJzerman, H., Dijksterhuis, A., Farach, F. Open Epidemiology Journal, 3(1), 16–19. https://doi.or J., Geller, J., Giner-Sorolla, R., Grange, J. A., Perugini, g/10.2174/1874297101003010016 M., Spies, J. R., & van ’t Veer, A. (2014). The replication recipe: What makes for a convincing Collaborative Open-science Research. (2022). replication? Journal of Experimental Social Psychology, Replications and extensions of classic findings in 50, 217–224. https://doi.org/10.1016/j.jesp.2013.10.00 Social Psychology and Judgment and Decision 5 Making. Open Science Framework. https://doi.org/10.1 7605/OSF.IO/5Z4A8 Breugelmans, S. M., Zeelenberg, M., Gilovich, T., Huang, W.-H., & Shani, Y. (2014). Generality and cultural Connolly, T., Ordóñez, L. D., & Coughlan, R. (1997). variation in the experience of regret. Emotion, 14(6), Regret and responsibility in the evaluation of decision 1037–1048. https://doi.org/10.1037/a0038221 outcomes. Organizational Behavior and Human Decision Processes, 70(1), 73–85. https://doi.org/10.10 Broomhall, A. G., Phillips, W. J., Hine, D. W., & Loi, N. 06/obhd.1997.2695 M. (2017). Upward counterfactual thinking and depression: A meta-analysis. Clinical Psychology Davison, I. M., & Feeney, A. (2008). Regret as Review, 55, 56–73. https://doi.org/10.1016/j.cpr.201 autobiographical memory. Cognitive Psychology, 57(4), 7.04.010 385–403. https://doi.org/10.1016/j.cogpsych.2008.0 3.001 Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon’s Mechanical Turk: A new source of Diefendorff, J. M., Hall, R. J., Lord, R. G., & Strean, M. L. inexpensive, yet high-quality, data? Perspectives on (2000). Action–state orientation: Construct validity Psychological Science, 6(1), 3–5. https://doi.org/10.117 of a revised measure and its relationship to work- 7/1745691610393980 related variables. Journal of Applied Psychology, 85(2), 250–263. https://doi.org/10.1037/0021-9010.85.2.250 Byrne, R. M. J., & McEleney, A. (2000). Counterfactual thinking about actions and failures to act. Journal of Eyal, P., Rothschild, D., Gordon, A., Evernden, Z., & Experimental Psychology: Learning, Memory, and Damer, E. (2021). Data quality of platforms and Cognition, 26(5), 1318–1331. https://doi.org/10.1037/ panels for online behavioral research. Behavior 0278-7393.26.5.1318 Research Methods, 1–20. https://doi.org/10.3758/s134 28-021-01694-3 Champely, S., Ekstrom, C., Dalgaard, P., Gill, J., Weibelzahl, S., Anandkumar, A., Ford, C., Volcic, R., Feeney, A., & Handley, S. J. (2006). Comparisons, & De Rosario, M. H. (2018). Package ‘pwr.’ R package mental models, and the action effect in judgments of version, 1. regret. Memory & Cognition, 34(7), 1422–1430. http s://doi.org/10.3758/bf03195907 Feldman, G. (2020). What is normal? Dimensions of action-inaction normality and their impact on regret in the action-effect. Cognition and Emotion, 34(4), 728–742. https://doi.org/10.1080/02699931.2019.167 Collabra: Psychology 15 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Feldman, G., & Albarracín, D. (2017). Norm theory and Kahneman, D., & Miller, D. T. (1986). Norm theory: the action-effect: The role of social norms in regret Comparing reality to its alternatives. Psychological following action and inaction. Journal of Experimental Review, 93(2), 136–153. https://doi.org/10.1037/003 Social Psychology, 69, 111–120. https://doi.org/10.101 3-295x.93.2.136 6/j.jesp.2016.07.009 Kahneman, D., & Tversky, A. (1982). The psychology of Feldman, G., Chandrashekar, S. P., Yeung, S. K., Xiao, preferences. Scientific American, 246(1), 160–173. http Q., Fillon, A., & Henne, P. (2021). What is action, what s://doi.org/10.1038/scientificamerican0182-160 is inaction? Clarifying action-inaction in judgment and Kordes-de Vaal, J. H. (1996). Intention and the omission decision making and recommendations for term use and bias: Omissions perceived as nondecisions. Acta typology [Manuscript in preparation]. Psychologica, 93(1–3), 161–172. https://doi.org/10.10 Feldman, J., Miyamoto, J., & Loftus, E. F. (1999). Are 16/0001-6918(96)00027-3 actions regretted more than inactions? Organizational Kutscher, L., & Feldman, G. (2019). The impact of past Behavior and Human Decision Processes, 78(3), behaviour normality on regret: Replication and 232–255. https://doi.org/10.1006/obhd.1999.2833 extension of three experiments of the exceptionality Frijda, N. H., Kuipers, P., & Ter Schure, E. (1989). effect. Cognition and Emotion, 33(5), 901–914. http Relations among emotion, appraisal, and emotional s://doi.org/10.1080/02699931.2018.1504747 action readiness. Journal of Personality and Social Landman, J. (1987). Regret and elation following action Psychology, 57(2), 212–228. https://doi.org/10.1037/0 and inaction: Affective responses to positive versus 022-3514.57.2.212 negative outcomes. Personality and Social Psychology Gilovich, T., Kerr, M., & Medvec, V. H. (1993). Effect of Bulletin, 13(4), 524–536. https://doi.org/10.1177/0146 temporal perspective on subjective confidence. 167287134009 Journal of Personality and Social Psychology, 64(4), Leach, F. R., & Plaks, J. E. (2009). Regret for errors of 552–560. https://doi.org/10.1037/0022-3514.64.4.552 commission and omission in the distant term versus Gilovich, T., & Medvec, V. H. (1994). The temporal near term: The role of level of abstraction. Personality pattern to the experience of regret. Journal of and Social Psychology Bulletin, 35(2), 221–229. http Personality and Social Psychology, 67(3), 357–365. http s://doi.org/10.1177/0146167208327001 s://doi.org/10.1037/0022-3514.67.3.357 LeBel, Etienne P., McCarthy, R. J., Earp, B. D., Elson, M., Gilovich, T., & Medvec, V. H. (1995). The experience of & Vanpaemel, W. (2018). A unified framework to regret: What, when, and why. Psychological Review, quantify the credibility of scientific findings. 102(2), 379–395. https://doi.org/10.1037/0033-295x.1 Advances in Methods and Practices in Psychological 02.2.379 Science, 1(3), 389–402. https://doi.org/10.1177/25152 Gilovich, T., Medvec, V. H., & Chen, S. (1995). Commission, omission, and dissonance reduction: LeBel, Etienne Philippe, Vanpaemel, W., Cheung, I., & Coping with regret in the" Monty Hall" problem. Campbell, L. (2019). A brief guide to evaluate Personality and Social Psychology Bulletin, 21(2), replications. Meta-Psychology, 3, 1–9. https://doi.org/ 182–190. https://doi.org/10.1177/0146167295212008 10.15626/mp.2018.843 Gilovich, T., Medvec, V. H., & Kahneman, D. (1998). Litman, L., Robinson, J., & Abberbock, T. (2017). Varieties of regret: A debate and partial resolution. TurkPrime.com: A versatile crowdsourcing data Psychological Review, 105(3), 602–605. https://doi.org/ acquisition platform for the behavioral sciences. 10.1037/0033-295x.105.3.602 Behavior Research Methods, 49(2), 433–442. https://do i.org/10.3758/s13428-016-0727-z Gleicher, F., Kost, K. A., Baker, S. M., Strathman, A. J., Richman, S. A., & Sherman, S. J. (1990). The role of Nosek, B. A., Alter, G., Banks, G. C., Borsboom, D., counterfactual thinking in judgments of affect. Bowman, S. D., Breckler, S. J., Buck, S., Chambers, C. Personality and Social Psychology Bulletin, 16(2), D., Chin, G., Christensen, G., Contestabile, M., Dafoe, 284–295. https://doi.org/10.1177/0146167290162009 A., Eich, E., Freese, J., Glennerster, R., Goroff, D., Green, D. P., Hesse, B., Humphreys, M., … Yarkoni, T. Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The (2015). Promoting an open research culture. Science, weirdest people in the world? Behavioral and Brain 348(6242), 1422–1425. https://doi.org/10.1126/scienc Sciences, 33(2–3), 61–83. https://doi.org/10.1017/s014 e.aab2374 0525x0999152x Ordónez, L. D., & Connolly, T. (2000). Regret and Huang, W. H., & Zeelenberg, M. (2012). Investor regret: responsibility: A reply to Zeelenberg et al. (1998). The role of expectation in comparing what is to what Organizational Behavior and Human Decision might have been. Judgment and Decision Making, 7(4), Processes, 81(1), 132–142. https://doi.org/10.1006/ob 441. http://sjdm.cybermango.org/journal/11/11303/jd hd.1999.2834 m11303.pdf Rajagopal, P., Raju, S., & Unnava, H. R. (2006). Itzkin, A., Van Dijk, D., & Azar, O. H. (2016). At least I Differences in the cognitive accessibility of action and tried: The relationship between regulatory focus and inaction regrets. Journal of Experimental Social regret following action vs. inaction. Frontiers in Psychology, 42(3), 302–313. https://doi.org/10.1016/j.j Psychology, 7, 1684. https://doi.org/10.3389/fpsyg.201 esp.2005.05.003 6.01684 Collabra: Psychology 16 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Roese, N. J., Hur, T., & Pennington, G. L. (1999). Yeung, S. K., & Feldman, G. (2022). Action-Inaction Counterfactual thinking and regulatory focus: Asymmetries in Emotions and Counterfactual Implications for action versus inaction and Thoughts: Meta-Analysis of the Action Effect sufficiency versus necessity. Journal of Personality and [Registered Report Stage 1]. Open Science Framework. Social Psychology, 77(6), 1109–1120. https://doi.org/1 https://doi.org/10.17605/OSF.IO/ACM24 0.1037/0022-3514.77.6.1109 Yeung, S. K., Yay, T., & Feldman, G. (2022). Action and RStudio Team. (2021). Integrated development for R. inaction in moral judgments and decisions: Meta- RStudio, Inc. analysis of Omission-Bias omission-commission asymmetries. Personality and Social Psychology Savitsky, K., Medvec, V. H., & Gilovich, T. (1997). Bulletin. https://doi.org/10.13140/RG.2.2.28468.58249 Remembering and regretting: The Zeigarnik effect and the cognitive availability of regrettable actions Zeelenberg, M., Van den Bos, K., Van Dijk, E., & Pieters, and inactions. Personality and Social Psychology R. (2002). The inaction effect in the psychology of Bulletin, 23(3), 248–257. https://doi.org/10.1177/0146 regret. Journal of Personality and Social Psychology, 167297233004 82(3), 314–327. https://doi.org/10.1037/0022-3514.8 2.3.314 Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2012). A 21 word solution. SSRN Electronic Journal. https://do Zeelenberg, M., van der Pligt, J., & Manstead, A. S. R. i.org/10.2139/ssrn.2160588 (1998). Undoing regret on Dutch television: Apologizing for interpersonal regrets involving Simonson, I. (1992). The influence of anticipating regret actions or inactions. Personality and Social Psychology and responsibility on purchase decisions. Journal of Bulletin, 24(10), 1113–1119. https://doi.org/10.1177/0 Consumer Research, 19(1), 105–118. https://doi.org/1 0.1086/209290 Zeelenberg, M., Van Dijk, W. W., & Manstead, A. S. R. Thomas, K. A., & Clifford, S. (2017). Validity and (2000). Regret and responsibility resolved? Evaluating Mechanical Turk: An assessment of exclusion Ordóñez and Connolly’s (2000) conclusions. methods and interactive experiments. Computers in Organizational Behavior and Human Decision Human Behavior, 77, 184–197. https://doi.org/10.101 Processes, 81(1), 143–154. https://doi.org/10.1006/ob 6/j.chb.2017.08.038 hd.1999.2865 Towers, A., Williams, M. N., Hill, S. R., Philipp, M. C., & Ziano, I., Kong, M. F., Kim, H. J., Liu, C. Y., Wong, S. C., Flett, R. (2016). What makes for the most intense Cheng, B. L., & Feldman, G. (2021). Revisiting regrets? Comparing the effects of several theoretical Disjunction Effect: Replication of Tversky and Shafir predictors of regret intensity. Frontiers in Psychology, (1992) and extension comparing between and within 7, 1941. https://doi.org/10.3389/fpsyg.2016.01941 subject designs. Journal of Economic Psychology, 83, Wickham, H. (2016). ggplot2: Elegant Graphics for Data 102350. https://doi.org/10.1016/j.joep.2020.102350 Analysis. Springer-Verlag New York. https://ggplot2.ti Zwaan, R. A., Etz, A., Lucas, R. E., & Donnellan, M. B. dyverse.org (2018). Making replication mainstream. Behavioral Xie, Y., Allaire, J. J., & Grolemund, G. (2018). R and Brain Sciences, 41. https://doi.org/10.1017/s01405 markdown: The definitive guide. CRC Press. https://do 25x17001972 i.org/10.1201/9781138359444 Collabra: Psychology 17 Revisiting the Temporal Pattern of Regret in Action Versus Inaction: Replication of Gilovich and Medvec (1994) With... Supplementary Materials Peer Review History Download: https://collabra.scholasticahq.com/article/37122-revisiting-the-temporal-pattern-of-regret-in-action- versus-inaction-replication-of-gilovich-and-medvec-1994-with-extensions-examining-responsibilit/attachment/ 94538.docx?auth_token=GY_rx_6kkuwT9pca1Ir7 Supplemental Material Download: https://collabra.scholasticahq.com/article/37122-revisiting-the-temporal-pattern-of-regret-in-action- versus-inaction-replication-of-gilovich-and-medvec-1994-with-extensions-examining-responsibilit/attachment/ 94539.docx?auth_token=GY_rx_6kkuwT9pca1Ir7 Collabra: Psychology
Journal
Collabra Psychology – University of California Press
Published: Aug 2, 2022
Keywords: temporal pattern; judgment and decision-making; pre-registered replication; regret; action; inaction