Human morality is based on an early-emerging moral core

Woo, Brandon M.; Tan, Enda; Hamlin, Kiley

doi:10.31234/osf.io/98d36

Cited by 5 publications

(6 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Infants likely do not enter the world as moral 'blank slates,' as they often demonstrate a range of early-emerging proto-moral capacities (Hamlin, 2013;Hamlin et al, 2007;Woo et al, 2022). Newborns, not even two days old, show an early signature of empathy, crying along with the cries of other infants (but not when exposed to silence or simulated cries; Sagi & Hoffman, 1976).…”

Section: Proto-moral Cognitionmentioning

confidence: 99%

See 1 more Smart Citation

Artificial moral cognition: Learning from developmental psychology

Weidinger¹,

Reinecke²,

Haas³

2022

Preprint

View full text Add to dashboard Cite

An artificial system that successfully performs cognitive tasks may pass tests of 'intelligence' but not yet operate in ways that are morally appropriate. An important step towards developing moral artificial intelligence (AI) is to build robust methods for assessing moral capacities in these systems. Here, we present a framework for analysing and evaluating moral capacities in AI systems, which decomposes moral capacities into tractable analytical targets and produces tools for measuring artificial moral cognition. We show that decomposing moral cognition in this way can shed light on the presence, scaffolding, and interdependencies of amoral and moral capacities in AI systems. Our analysis framework produces a virtuous circle, whereby developmental psychology can enhance how AI systems are built, evaluated, and iterated on as moral agents; and analysis of moral capacities in AI can generate new hypotheses surrounding mechanisms within the human moral mind.

show abstract

Section: Proto-moral Cognitionmentioning

confidence: 99%

“…Even in humans, these initial core capacities merely serve as a 'blueprint' or 'first draft' for later-emerging social and moral cognition(Ting et al, 2020;Woo et al, 2022).3 That is, the helping behaviour is not extrinsically rewarded by, say, the experimenter or accompanying caregiver.…”

mentioning

confidence: 99%

Artificial moral cognition: Learning from developmental psychology

Weidinger¹,

Reinecke²,

Haas³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Causal models similarly can produce predictions about action trajectories, but these predictions rely on explicit representations of agents as noisy rational planners whose actions are caused by the expected utilities of different actions, conditioned on their knowledge and beliefs (Baker et al, 2009(Baker et al, , 2017Jara-Ettinger, 2019). In a head to head comparison of these classes of models on tasks and stimuli designed after the infant cognition literature, Shu et al (2021) and Tan et al (2022) found that models with the features described in Sections 3.1-3 (i.e. causal models of planning over utilities that connects physical and mental states) outperformed feedforward predictive models that lack one or more of these features.…”

Section: A Formal Computational Proposalmentioning

confidence: 99%

“…Does this understanding depend on inferred physical parameters (see Sosa et al, 2021;and Shu et al 2020, for evidence from adults)? Is this understanding an elaboration on intuitive psychology (Powell, 2022), or does it instead require additional, distinctively social, representations and computations (Woo et al, 2022)? The classic view, that these two domains of knowledge are informationally encapsulated modules, whose outputs are combined downstream, either in a module specialized for joint reasoning across the domains, or in domain-general reasoning.…”

Section: The Structure Of Domains In Infancy and Beyondmentioning

confidence: 99%

From infancy, we model people as minds, acting in a physical world

Liu

2022

Preprint

View full text Add to dashboard Cite

Since the founding of psychology as a discipline, one central question has concerned the developmental origins of the mind. What is the form and content of our earliest understanding of the social and physical world? Behavioral research from the past 30 years has shown that by their first birthdays, infants interpret the actions of other agents in terms of what they see and want, and how physical obstacles constrain and impose cost on actions. Here I argue that these findings cannot be understood without granting that infants have an integrated understanding of people and objects. Against 3 existing hypotheses about early psychological understanding (sensorimotor, teleological, and modular hypotheses), I submit that infants have an intuitive theory of psychology that operates over a world model. In Marr computational-level terms, this understanding is a computational framework of probabilistic forward and inverse action planning in which agents’ actions and mental states are constrained by, and can cause changes in, external world states. This proposal remains to be fully tested, and raises broader questions about the development of the mind and brain; thus, the paper ends with future interdisciplinary work across the disciplines of developmental psychology, computational cognitive science, and cognitive neuroscience.

show abstract

“…We build upon a body of experiments that have found that infants and toddlers show an early capacity for social evaluation: They preferentially look to and reach for agents who help others over agents who hinder others in the pursuit of their goals (13)(14)(15)(16)(17)(18)(19). Studies in this literature have provided evidence that infants and toddlers are sensitive to the intentions underlying social actions (20)(21)(22)(23)(24) and to the contexts in which social actions takes place (e.g., agents' group membership and history of social behavior (25)(26)(27)).…”

Section: Representations Of Action Cost and Evaluations Of Social Act...mentioning

confidence: 99%

Toddlers prefer agents who help those facing harder tasks

Woo¹,

Liu

Gweon³

et al. 2022

Preprint

View full text Add to dashboard Cite

Capacities to understand and evaluate others’ actions are fundamental to human social life. Infants and toddlers are sensitive to the costs of others’ actions, infer others’ values from the costs of the actions they take, and prefer those who help others to those who hinder them, but it is largely unknown whether and how cost considerations inform early understanding of third-party prosocial actions. In three experiments (N = 94), we asked whether 16-month-old toddlers value agents who selectively help those who need it most, because the individual they help faces a greater challenge or lacks the strength to succeed on their own. Toddlers preferred an agent who chose to help the individual attempting a high-cost action (climbing a steep ramp) over an agent who chose to help the individual facing a lower-cost action (climbing a shallow ramp), as evidenced by their selective reaching (Experiment 1) and looking (Experiment 2) to the former agent. This preference was specific to helping contexts: Toddlers looked equally at agents who chose to perform the harder vs. easier task while acting alone Experiment 2). Furthermore, when two individuals faced the same task but differed in strength, toddlers preferred an agent who helped the weaker individual to an agent who helped the stronger individual (Experiment 3). These results provide evidence that toddlers engage in need-based evaluations of helping, applying their understanding of action utilities to their social evaluations.

show abstract

Human morality is based on an early-emerging moral core

Cited by 5 publications

References 53 publications

Artificial moral cognition: Learning from developmental psychology

Artificial moral cognition: Learning from developmental psychology

From infancy, we model people as minds, acting in a physical world

Toddlers prefer agents who help those facing harder tasks

Contact Info

Product

Resources

About