CoChat
Sign Up
Cited by 10Psychology of Moral and Emotional JudgmentEthics and Social Impacts of AIEthics in Business and Education

Exploring moral algorithm preferences in autonomous vehicle dilemmas: an empirical study

Open access
Sui, Tingting·2023·Frontiers in Psychology
DOI 10.3389/fpsyg.2023.1229245PMC PMC10713701
Open this paper in CoChat — free accountOpen PDF

Abstract

Introduction: This study delves into the ethical dimensions surrounding autonomous vehicles (AVs), with a specific focus on decision-making algorithms. Termed the “Trolley problem,” an ethical quandary arises, necessitating the formulation of moral algorithms grounded in ethical principles. To address this issue, an online survey was conducted with 460 participants in China, comprising 237 females and 223 males, spanning ages 18 to 70. Methods: Adapted from Joshua Greene’s trolley dilemma survey, our study employed Yes/No options to probe participants’ choices and Likert scales to gauge moral acceptance. The primary objective was to assess participants’ inclinations toward four distinct algorithmic strategies—Utilitarianism, Rawlsianism, Egoism, and a Hybrid approach—in scenarios involving AVs Results: Our findings revealed a significant disparity between participants’ preferences in scenarios related to AV design and those focused on purchase decisions. Notably, over half of the respondents expressed reluctance to purchase AVs equipped with an “egoism” algorithm, which prioritizes the car owner’s safety. Intriguingly, the rejection rate for “egoism” was similar to that of “utilitarianism,” which may necessitate self-sacrifice. Discussion: The hybrid approach, integrating “Utilitarianism” and “Egoism,” garnered the highest endorsement. This highlights the importance of balancing self-sacrifice and harm minimization in AV moral algorithms. The study’s insights are crucial for ethically and practically advancing AV technology in the continually evolving realm of autonomous vehicles.

AI snapshot

Written from the paper’s full text.

TL;DR

This online survey of 460 Chinese participants compared moral acceptance and design/purchase preferences for four autonomous-vehicle moral algorithms: Utilitarianism, Rawlsianism, Egoism, and a Hybrid approach. The Hybrid algorithm, balancing passenger protection with harm minimization, received the highest moral acceptability and purchase preference, while Egoism was least morally acceptable.

Key findings

  • Hybrid was the most preferred algorithm: 83.91% favored designing it and 77.39% intended to purchase it; among all four algorithms, 68.91% chose Hybrid as the one they would purchase.
  • Design and purchase preferences diverged: Utilitarianism had 55.65% design support but only 44.13% purchase intention; Rawlsianism had 58.26% design support and 55.65% purchase intention; Egoism had 39.13% design support and 41.52% purchase intention.
  • Moral acceptability was highest for Hybrid (mean = 5, SD = 1.62), followed by Rawlsianism (mean = 4.28, SD = 1.7), Utilitarianism (mean = 3.97, SD = 1.85), and Egoism (mean = 3.5, SD = 1.72); one-way ANOVA showed a significant main effect, F(3, 1836) = 58.92, p < 0.0001, η²p = 0.09.
  • Participants who chose to design or purchase a given algorithm reported higher moral acceptability for that algorithm, with substantial effect sizes (e.g., Hybrid purchase vs. not purchase: F(1, 459) = 464.97, p < 0.0001, η²p = 0.50).

Methods

  • An online survey was conducted via Credamo with 460 final participants (237 females, 223 males; ages 18–70) in China; from 480 original respondents, 20 were excluded for failing a control question (9) or completing in under 1 minute (11).
  • The design adapted Joshua Greene’s trolley-dilemma survey, using Yes/No choices for design and purchase and 7-point Likert scales for moral acceptability across four algorithms: Utilitarianism, Rawlsianism, Egoism, and Hybrid.
  • Analyses used SPSS 20 and included Chi-square tests, t-tests, ANOVAs, Bonferroni correction, and Kruskal–Wallis tests with post-hoc comparisons.

Limitations

  • The paper states that vignette-based dilemmas may not fully capture real-world complexity and suggests future immersive technologies such as VR.
  • The paper states that the Rawlsian maximin procedure may have been too intricate for some participants to fully comprehend, potentially affecting responses.
  • The paper states that it focused only on moral dilemmas, while AVs involve broader legal, regulatory, safety, and societal issues; evident from the text, the sample was limited to China and the algorithms were presented in a fixed sequential order, which may limit generalizability or introduce order effects.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3389/fpsyg.2023.1229245

Cites

27 of 27 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

9 of 30 cited sources checked · 2 supported · 0 flagged · accuracy not yet measured

  • Premisecitation

    Moral acceptance of algorithms, evidenced by theoretical agreement, may not seamlessly translate into consumer behavior.

    “And it is worth noting that the moral acceptance of these algorithms, evidenced by theoretical agreement, may not seamlessly translate into consumer behavior (Bonnefon et al., 2016; Sui and Guo, 2020).”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Bonnefon et al., 2016 source

      The supplied passages and the additional indexed passages are labeled “Reprogramming of avian neural crest axial identity and cell fate” and discuss neural crest cells, not autonomous vehicles. Searches for moral acceptance, consumer behavior, and willingness to buy yielded no relevant passage from the cited paper in our copy, so the claim cannot be assessed from it. (LLM, conf=0.78)

      Experimental — accuracy not yet measured
    • cites Sui and Guo, 2020

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Premisecitation

    In AV dilemma scenarios, most participants showed a moral inclination toward utilitarianism, even when it entailed self-sacrifice.

    “In dilemma scenario of AVs, scholars also observed that most participants showed a moral inclination toward utilitarianism, even when it entailed self-sacrifice (Bonnefon et al., 2016; Gawronski et al., 2017; Bergmann et al., 2018; Faulhaber et al., 2019).”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites Bonnefon et al., 2016 source

      The supplied passages and the passages returned for this DOI are labeled “Reprogramming of avian neural crest axial identity and cell fate” (Abstract; page unavailable). They concern chick neural crest cells, not autonomous-vehicle dilemmas. Searches for utilitarian preferences and self-sacrifice yielded no relevant passage, so our copy cannot verify the claim. (LLM, conf=0.79)

      Experimental — accuracy not yet measured
    • cites Gawronski et al., 2017 source
    • cites Bergmann et al., 2018 source
  • Premisecitation

    Utilitarianism has garnered the highest support in previous surveys of trolley dilemma.

    “Among these approaches, utilitarianism, advocating for the preservation of more lives at the expense of fewer, has garnered the highest support in previous surveys of trolley dilemma (Hauser et al., 2007; Rehman and Dzionek Kozłowska, 2018; de Melo et al., 2021).”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites Hauser et al., 2007 source
    • cites Rehman and Dzionek Kozłowska, 2018 source

      In §3.2 (Data and results), the study reports that 87% of its respondents chose to sacrifice one person to save five in the switch scenario, and 61% chose to do so in the Fat Man scenario. That supports substantial preference for the life-saving choice in this survey, but the paper does not establish that utilitarianism garnered the highest support across previous surveys; it cautions that its small sample may not generalize. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
    • cites de Melo et al., 2021 source

      Experiment 4 (Method, §2.4; page unavailable) reports that participants often made utilitarian choices, and the General Discussion (§3; page unavailable) describes a preference for utilitarian choices under some risk conditions. The paper does not establish that utilitarianism received the highest support in previous trolley-dilemma surveys; its findings concern its own autonomous-vehicle scenarios. (LLM, conf=0.78)

      Experimental — accuracy not yet measured
  • Premisecitation

    People tend to avoid moral algorithms involving self-sacrifice.

    “Several investigations have delved into the specific perspectives of passengers, pedestrians, and observers in trolley dilemma, indicating people’s inclination to avoid moral algorithm involving self-sacrifice (Frank et al., 2019; Mayer et al., 2021).”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites Frank et al., 2019 source
    • cites Mayer et al., 2021 source

      The Introduction (page unavailable) reports reluctance to buy utilitarian autonomous vehicles and describes a self-protective bias, supporting an inclination to avoid self-sacrificing programming. But the Experiment 3 Discussion and General discussion (pages unavailable) report that a majority preferred self-sacrifice to save several others. The claim overstates avoidance as a general preference. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • Premisecitation

    A majority of participants indicated reluctance to purchase an AV programmed with a utilitarian moral algorithm.

    “Specifically, participants may agree, in principle, that it would be morally acceptable to sacrifice one passenger (including themselves) to save five passengers, a majority of participants indicated reluctance to purchase an AV programmed with a utilitarian moral algorithm (Bonnefon et al., 2016; Mayer et al., 2021).”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites Bonnefon et al., 2016 source

      The passages returned under this DOI are labeled “Reprogramming of avian neural crest axial identity and cell fate” (Abstract; no page given) and discuss neural crest cells, not autonomous vehicles. Searches for reluctance to purchase a utilitarian AV and for a utilitarian moral algorithm found no relevant passage in the available copy, so the attributed majority finding cannot be assessed from it. (LLM, conf=0.75)

      Experimental — accuracy not yet measured
    • cites Mayer et al., 2021 source

      The Introduction reports that Bonnefon et al.’s participants were unwilling to buy utilitarian autonomous vehicles for themselves despite judging that programming morally preferable. It does not say that a majority were reluctant to buy one. Mayer et al.’s majority findings in the General discussion concern choices in moral dilemmas, not purchasing. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • May combine two claims. 21
  • Classified as probably not a substantial claim. 47

  • May combine two claims.

    Findingown data

    The Hybrid approach received the highest moral acceptability rating, with a mean of 5 and SD of 1.62.

    “In terms of moral acceptability, the Hybrid received the highest rating on a scale ranging from 1 to 7, with a mean score of 5 and a standard deviation (SD) of 1.62.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The hypothetical vignette responses reflect how people would actually design or purchase AVs in real-world settings.
    • Rests on an assumption: The four moral algorithms (Utilitarianism, Rawlsianism, Egoism, Hybrid) adequately represent the range of possible AV moral algorithms.
    • Rests on an assumption: The sample of 460 Credamo participants is representative of the Chinese population or AV consumers.
    • Rests on an assumption: The fixed order of presenting the four algorithms did not affect participants’ responses.

  • May combine two claims.

    Findingown data

    Gender showed no significant differences in both design and purchase scenarios.

    “Gender showed no significant differences in both design and purchase scenarios.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    There was an asymmetrical relationship between design and purchase tendencies: greater inclination toward design for Utilitarianism, Rawlsianism, and Hybrid, and higher preference for purchasing for Egoism.

    “Participants exhibited a greater inclination toward design in the contexts of Utilitarianism, Rawlsianism, and Hybrid, while demonstrating a higher preference for purchasing in scenario of Egoism”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductcitation

    The survey was adapted from Greene's classic trolley dilemma survey, using Yes/No options and Likert scales.

    “The experimental design was adapted from Greene’s classic survey on trolley dilemma, which employed Yes/No options to exam people’s choices and Likert scales to measure moral acceptance Greene (2016).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Greene (2016) source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    The study conducted an online survey with 460 participants in China, comprising 237 females and 223 males, spanning ages 18 to 70.

    “To address this issue, an online survey was conducted with 460 participants in China, comprising 237 females and 223 males, spanning ages 18 to 70.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductreasoning

    The final dataset consisted of 460 participants after excluding 20 from an original 480: 9 failed the control question and 11 completed the survey in less than 1 minute.

    “In order to maintain the reliability of the survey results, 20 participants were excluded from the data set, because 9 of them failed to provide a correct answer to the control question,2 while 11 of them completed survey less than 1 min”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Participants were recruited through Credamo, which verified IP locations to ensure data covered a majority of provinces in China.

    “With the consent of anonymous participants, Credamo verified their location through their IP addresses to ensure that the data covered a majority of provinces in China.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    If companies design Egoistic AVs, they would face moral criticism and practical rejection.

    “if companies such as Mercedes-Benz design Egoistic AVs, they would not only face moral criticism but also practical rejection.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationreasoning

    The Egoistic algorithm may be seen as overly self-interested and potentially unfair, leading to less approval.

    “The Egoistic algorithm, by always prioritizing passenger safety, may be seen as overly self-interested and potentially unfair, leading to less approval.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    The perception of morality could contribute to the preference for the Hybrid algorithm.

    “The perception of morality indeed could contribute to the preference for the Hybrid algorithm.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    The context-aware decision-making of the Hybrid algorithm might be seen as more intelligent and morally sound.

    “This context-aware decision-making might be seen as more intelligent and morally sound, although the differences between the Hybrid and Egoism may be relatively small.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Interpretationreasoning

    If all AVs were programmed to protect their passengers at all costs, this could lead to a breakdown of trust and cooperation on the roads, resulting in more harm overall.

    “If all AVs were programmed to protect their passengers at all costs, this could lead to a breakdown of trust and cooperation on the roads, resulting in more harm overall.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Mercedes-Benz has remarked that their AVs would prioritize safeguarding the car's passengers in collision situations.

    “Mercedes-Benz, for instance, has remarked that their AVs would prioritize safeguarding the car’s passengers in collision situations (Leben, 2017, pp. 98–99).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Leben, 2017, pp. 98–99 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    As pedestrians, individuals might feel safer or more protected knowing that AVs would attempt to cause the least overall harm.

    “Therefore, as pedestrians, individuals might feel safer or more protected knowing that the AVs on the road would attempt to cause the least overall harm.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    The Hybrid method represents a balance between self-preservation and the welfare of others.

    “Hybrid method tries to represent a balance between self-preservation and the welfare of others.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    People may worry about the social implications of widespread adoption of Egoistic algorithms.

    “Secondly, people may worry about the social implications of widespread adoption of Egoistic algorithms.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    The perceived fairness of the Hybrid algorithm could lead to more acceptance among people.

    “This perceived fairness, even if it does not come into play in passenger-danger scenarios, could lead to more acceptance among people.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationreasoning

    Broad acceptance of Egoistic algorithms could potentially destabilize the cooperative equilibrium that currently exists.

    “Moreover, a broad acceptance of Egoistic algorithms could potentially destabilize the cooperative equilibrium that currently exists.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    The implementation of survival rate calculations cannot entirely eliminate the possibility of passenger self-sacrifice, which may lead individuals to hesitate in selecting such an algorithm.

    “Additionally, the implementation of survival rate calculations cannot entirely eliminate the possibility of passenger self-sacrifice, which may, in turn, lead individuals to hesitate in selecting such an algorithm.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Egoism conflicts with deeply held values of fairness and the equal value of all lives.

    “This may conflict with deeply held values of fairness and the equal value of all lives.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The most prominent finding was participants' preference for the Hybrid algorithm; it was preferred and perceived as the most morally acceptable choice.

    “The most prominent finding emerging from this statistical analysis was the preference of participants for the Hybrid algorithm. The data showed that this algorithm was not only preferred, but also perceived as the most morally acceptable choice”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationcitation

    The preference for the Hybrid algorithm might present a paradox of moral psychology that Philippa Foot indicated.

    “Moreover, the preference for the Hybrid algorithm might presented a paradox of moral psychology that Philippa Foot indicated.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Foot, 1969

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim.

    Interpretationown data

    People in general strive to balance competing moral considerations rather than subscribe to any single principle in its entirety.

    “People, in general, strive to balance competing moral considerations rather than subscribe to any single principle in its entirety.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Interpretationreasoning

    A moral algorithm based solely on "ought" principles proves inadequate, since people's tendency of purchasing AVs may closely resemble an "is" problem.

    “From the perspective of ethics, this distinction suggests that a moral algorithm based solely on “ought” principles prove inadequate, since people’s tendency of purchasing AVs may closely resemble an “is” problem, reflecting the actual preferences and behaviors of individuals, rather than a pure “ought” issue.1”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Individuals value broader ethical considerations such as fairness, societal harmony, moral consistency, and avoidance of direct moral culpability.

    “our findings suggest that individuals value broader ethical considerations. These considerations encompass fairness, societal harmony, moral consistency, and the avoidance of direct moral culpability”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationreasoning

    Increased moral responsibility could be a significant deterrent for many people.

    “This increased moral responsibility could be a significant deterrent for many people.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationreasoning

    Widespread adoption of Egoistic algorithms could engender a form of vehicular 'social Darwinism'.

    “The widespread adoption of Egoistic algorithms could engender a form of vehicular “social Darwinism””
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    Choosing an Egoistic AV could be viewed as endorsing a moral stance of disregarding others' wellbeing in favor of one's own safety.

    “the act of consciously choosing such a vehicle could be viewed as endorsing this moral stance, which essentially amounts to disregarding the wellbeing of others in favor of one’s own safety.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The egoistic approach met with significant criticism about prioritizing the lives of passengers over other vulnerable road users, such as children.

    “However, this approach met with significant criticism about prioritizing the lives of passengers over other vulnerable road users, such as children (Leben, 2017, p. 99).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Leben, 2017, p. 99 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The Rawlsian algorithm mainly relies on the Maximin procedure, a strategy of maximizing the minimum payoffs.

    “This algorithm mainly relies on Maximin procedure, a strategy of maximizing the minimum payoffs.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Leben (2017) source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    The preference for the Hybrid algorithm can be seen as a compromise between one's personal safety and the broader ethical concern for the wellbeing of others.

    “In this sense, the preference for the Hybrid algorithm can be seen as a compromise between one’s personal safety and the broader ethical concern for the wellbeing of others.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The Rawlsian approach received criticism about inequitable distribution of risks.

    “Nevertheless, this approach also received criticism about inequitable distribution of risks (Leben, 2017; Keeling, 2018).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Leben, 2017 source
    • cites Keeling, 2018 source

  • Probably not a substantial claim: may not pass S4 (based).

    Premisenone given

    There has been no survey to directly compare Utilitarianism, Egoism, and Rawlsianism in moral acceptance and purchase preference.

    “It is clear that Utilitarianism, Egoism, and Rawlsianism all faced their share of theoretical criticism, yet there has been no survey to directly compare their level of moral acceptance as well as preference of purchase, which is important in the previous research for discussing efficiency of moral algorithm.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationreasoning

    This endorsement could make the individual morally culpable for harm caused to others.

    “This endorsement could somehow make the individual morally culpable for the harm caused to others.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Greene's theory shows that people's deeds and deliberative moral judgments can sometimes conflict.

    “Briefly, Greene’s theory shows that people’s deed and deliberative moral judgments can be sometimes conflict with each other.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Greene, 2009 source

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    A skewed prioritization toward either extreme could undermine the acceptance and adoption of AVs, hampering benefits in safety, efficiency, and sustainability.

    “A skewed prioritization toward either extreme could undermine the acceptance and adoption of AVs, hampering the potential benefits they offer in terms of safety, efficiency, and sustainability.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Interpretationreasoning

    Neglecting the balance in moral algorithms could result in systems that are either overly utilitarian or excessively self-interested, alienating users and causing societal backlash.

    “Neglecting this crucial balance could result in systems that are either overly utilitarian or excessively self-interested, both of which could alienate users and result in societal backlash.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    The study concentrated on moral dilemmas within the context of AVs, which represent only a subset of broader challenges.

    “this article specifically concentrated on moral dilemmas within the context of AVs. Although these dilemmas are undoubtedly a crucial facet of the ethical landscape surrounding AVs, they represent only a subset of the broader challenges and considerations associated with this emerging technology.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    The Hybrid algorithm encapsulates a reconciliation of these conflicting impulses.

    “In this way, the Hybrid algorithm encapsulates a reconciliation of these conflicting impulses, offering a solution that seems to be both practically appealing and morally defensible.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Some people acknowledge the virtue of self-sacrifice for the greater good, yet desire assurance that their own safety will be prioritized when they are passengers in an AV.

    “some people are essentially acknowledging the virtue of self-sacrifice for the greater good, which is a key aspect of utilitarian philosophy. Yet, they also desire the assurance that their own safety will be prioritized when they are passengers in an AV.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    The automatic-emotional process is rapid and unconscious, leading to intuitive actions and judgments, while the conscious-controlled process is slower and involves deliberate reasoning.

    “The automatic-emotional process is rapid and unconscious, leading to intuitive actions and judgments, while the conscious-controlled process is slower and involves deliberate reasoning.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Greene, 2009 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The Rawlsian maximin procedure employed in the study can be quite intricate, potentially leading to challenges in participants' full comprehension.

    “we acknowledge that the Rawlsian maximin procedure employed in our study can be quite intricate, potentially leading to challenges in participants’ full comprehension.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

    Interpretationreasoning

    Understanding the balance in moral algorithms is a practical necessity for the successful integration of AVs into society.

    “Therefore, an understanding of this balance is not simply a philosophical or academic thought experiment, but a practical necessity for the successful integration of AVs into society.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Greene's theory suggests that moral decisions stem from automatic-emotional and conscious-controlled processes.

    “Greene’s theory suggested that our moral decisions stem from one of two distinct cognitive processes, namely, automatic-emotional process and conscious-controlled process (Greene, 2009).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Greene, 2009 source

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.

    Interpretationreasoning

    Such a balanced approach ensures that the societal benefits of autonomous technologies are realized while upholding fundamental ethical principles and accommodating diverse perspectives.

    “Such an approach ensures that the societal benefits of autonomous technologies are realized while upholding fundamental ethical principles and accommodating diverse perspectives.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    The preference for the Hybrid algorithm reflects a blend of the practical need for self-preservation with the moral imperative to minimize harm and promote the common good.

    “This reflects a blend of the practical need for self-preservation with the moral imperative to minimize harm and promote the common good.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Study design and conductreasoning

    Some individuals may not have completely grasped that the Rawlsian algorithm does not consider the number of individuals at risk but prioritizes minimizing the worst-case outcome based on death probability rates.

    “Specifically, some individuals may not have completely grasped that this algorithm does not consider the number of individuals at risk but instead prioritizes minimizing the worst-case outcome based on death probability rates.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    The preference for the Hybrid algorithm reveals a sort of moral compromise.

    “From a broader perspective, the preference for the Hybrid algorithm reveals a sort of moral compromise.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    The study's reliance on vignettes may not fully capture the complexity of real-world scenarios.

    “While this approach provides a controlled and systematic exploration of ethical decision-making, it may not fully capture the complexity of real-world scenarios.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    The effectiveness of AVs in enhancing road safety depends on their adoption and active use by individuals.

    “the effectiveness of AVs in enhancing road safety depends on their adoption and active use by individuals.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

    Interpretationreasoning

    By maintaining a balanced approach in moral algorithms, AVs can navigate complex moral dilemmas while promoting safety, fairness, and public trust.

    “By maintaining a balanced approach in moral algorithms, AVs can navigate complex moral dilemmas while promoting safety, fairness, and public trust.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    People understand the ethical virtues of minimizing overall harm and self-sacrifice, but retain the instinct for self-preservation.

    “People seem to understand the ethical virtues of minimizing overall harm and self-sacrifice, but at the same time, they retain the instinct for self-preservation.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    As the level of autonomy increases, the integration of moral and ethical considerations within operating algorithms becomes crucial.

    “As the level of autonomy increases, and assume greater responsibility for decision-making, the integration of moral and ethical considerations within their operating algorithms becomes crucial.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    According to Foot, people can appreciate and recognize moral virtues such as courage, generosity, or self-sacrifice, but recognizing a virtue does not necessarily mean embodying it themselves.

    “According to Foot, people can appreciate and recognize moral virtues such as courage, generosity, or self-sacrifice, acknowledging them as desirable traits. Yet, recognizing a virtue does not necessarily mean embodying it themselves”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Foot, 1969, p.209

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    AVs can truly contribute to a safer and more secure traffic environment only when readily embraced by consumers and integrated into daily routines.

    “It is when these vehicles are readily embraced by consumers and integrated into their daily routines that they can truly contribute to a safer and more secure traffic environment.”
    Quote in the paperNot a substantial claim
55 of 55

  • The paper relies on this without support.

    The hypothetical vignette responses reflect how people would actually design or purchase AVs in real-world settings.

    “Participants were asked to choose whether they would design (“Yes” or “No”) an AV according to the corresponding moral algorithm after reading the vignettes.”
    “they were asked to indicate whether they would like to purchase the AV with this algorithm (“Yes” or “No”).”

  • The paper relies on this without support.

    The four moral algorithms (Utilitarianism, Rawlsianism, Egoism, Hybrid) adequately represent the range of possible AV moral algorithms.

    “The survey aimed to assess participants’ tendency (design, purchase and moral acceptability) toward four distinctive moral algorithms (Utilitarianism, Rawlsianism, Egoism, and Hybrid approach) in scenarios of passenger-involved dilemma.”

  • The paper relies on this without support.

    The sample of 460 Credamo participants is representative of the Chinese population or AV consumers.

    “The final dataset consisted of 460 participants, including 237 females and 223 males.”
    “Credamo verified their location through their IP addresses to ensure that the data covered a majority of provinces in China.”

  • The paper relies on this without support.

    There is not enough empirical evidence that people would be willing to buy Egoistic AVs.

    “Moreover, there is also not enough empirical evidence that people would be willing to buy Egoistic AVs.”

  • The paper relies on this without support.

    The fixed order of presenting the four algorithms did not affect participants’ responses.

    “Throughout the course of the experiment, we sequentially presented four different moral algorithms—Utilitarianism, Rawlsianism, Egoism, and Hybrid approach—replicating the order of occurrence in previous discussions and studies.”

  • The paper relies on this without support.

    The study's sample is representative of the general public.

    “The intricate equilibrium seems to resonate with the general public, providing a morally robust framework for decision-making in AVs.”
6 of 6

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Consistentχ²(3) = 195.65reported p < .0001 · recomputed p < .001

    “χ2 (3, N = 1840) = 195.65, p < .0001”

  • Consistentχ²(3) = 148.72reported p < .0001 · recomputed p < .001

    “χ2 (3, N = 1840) = 148.72, p < .0001”

  • ConsistentF(3, 1836) = 58.92reported p < .0001 · recomputed p < .001

    “F(3, 1836) = 58.92, p < .0001”

3 of 3

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Self-check results appear here once an independent check has confirmed them.

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. Sui, 2020, Approaches to the trolley problem in autonomous Vehicles' ethical algorithm for decision making, J. Dialect. Nature, 42, 85, 10.15994/j.1000-0763.2020.10.011 · DOI

Cited by

Sorted by citations

Related

Authors
Sui, T.
Affiliations
Peking University (CN)
Corresponding author
Sui, Tingting
ORCID
Sui, Tingting
Venue
Frontiers in Psychology
Publisher
Frontiers Media SA
Published
2023-11-28
Volume
14
Pages
1229245-1229245
ISSN
1664-1078 (ISSN-L)
DOI
10.3389/fpsyg.2023.1229245
PMC
PMC10713701
OpenAlex
W4389097946
Semantic Scholar
adda4785dc9cb4ac67e64bfa451c34d444d600f1
Cited by
10
References
27
Citations by year
2026: 3 · 2025: 3 · 2024: 4
FWCI
1.40
Type
journal article
Language
en
Open access
Licence
Locations
  • Frontiers in Psychology (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PhilPapers (PhilPapers Foundation) (submittedVersion)
  • PubMed Central (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
National Social Science Fund of China (23CZX048)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Psychology of Moral and Emotional Judgment
  • Social Sciences › Safety Research › Ethics and Social Impacts of AI
  • Social Sciences › Decision Sciences › Information Systems and Management › Ethics in Business and Education
Keywords
autonomous vehicles, trolley problem, utilitarianism, egoism, algorithmic decision-making, public acceptance of autonomous vehicles, self-sacrifice, harm reduction

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-19.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-19.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review