LATEST WORKS

For a full list see Google Scholar or Publons

  • Immoral professors and malfunctioning tools

    Abstract

    This paper sets a new standard for demonstrating that the effect of norm violations on causal selection is driven by changes in counterfactual relevance. We then outline a number important but unanswered questions about this approach to causal reasoning.

    Jonathan Kominsky, Jonathan Phillips

    Cognitive Science (2019)

    github

  • How We Know What Not To Think

    Abstract

    Humans often represent and reason about unrealized possible actions. But how do we select possible actions that are worth considering from the infinity of unrealized actions better left ignored? We propose that (1) across diverse cognitive tasks, the possible actions we consider are biased towards those of general practical utility and (2) a plausible primary function for this mechanism resides in decision making.

    Jonathan Phillips, Adam Morris, Fiery Cushman

    Trends in Cognitive Sciences (2019)

  • New horizons for a theory of epistemic modals

    Abstract

    Recent debate over the semantics and pragmatics of epistemic modals has focused on intuitions about cross-contextual truth-value assessments. In this paper, we advocate a different approach to evaluating theories of epistemic modals. Our strategy focuses on judgments of the incompatibility of two different epistemic possibility claims, or two different truth value assessments of a single epistemic possibility claim. We subject the predictions of existing theories to empirical scrutiny, and argue that existing contextualist and relativist theories are unable to account for the full pattern of observed judgments. As a way of illustrating the theoretical upshot of these results, we conclude by developing a novel theory of epistemic modals that is able to predict the results.

    Jonathan Phillips, Justin Khoo
    Australasian Journal of Philosophy, (2018)
    preprint     github   OSF    preprint    scholar   .bib    handout

  • Quantitative causal selection patterns in token causation

    Abstract

    This paper provides a quantitative exploration of the puzzling way that causal judgments are affected by the likelihood of the contributing causes, and uncovers some new patterns that are no current theories are well-suited to capture. These results provide a new goalpost for anyone modeling causal selection judgments.

    Adam Morris, Jonathan Phillips, Tobias Gerstenberg, Fiery Cushman

    PLoS ONE 14(8): e0219704 (2019)

    github

  • Sticky situations

    Abstract

    When do we judge that someone was forced to do what they did? One relatively well-established finding is that subjects tend to judge that agents were not forced to do actions when those actions violate norms. A surprising discovery of Young & Phillips 2011 is that this effect seems to disappear when we frame the relevant ‘force’-claim in the active rather than passive voice (pX forced Y to ϕq vs. pY was forced to ϕ by Xq). Young and Phillips found a similar contrast when the scenario itself shifts attention from Y (the forcee) to X (the forcer). We propose that these effects can be (at least partly) explained by way of the role of attention in the setting of quantifier domains which in turn play a role in the evaluation of ‘force’- claims.

    Jonathan Phillips, Matthew Mandelkern
    Proceedings of Semantics and Linguistic Theory (SALT), 28 (2018)
    OSF   github   pre-registration

  • Knowledge wh and False Beliefs

    Abstract

    We consider the phenomenon of false-belief sensitivity–a challenge to the common approach to knowledge wh. Across six experiments, our results provide evidence that truth judgments of knowledge wh ascriptions are affected by both the presence of false beliefs and the proportion of the subject’s beliefs that are false.

    Jonathan Phillips, B R George

    Journal of Semantics (2018)

    github   OSF   preprint   scholar    .bib   handout

  • The psychological representation of modality

    Abstract

    We argue that recent work on the impact of physical law, morality and probability on people’s judgments provides evidence for a more general hypothesis about the kind of cognition people use to think about possibilities. We suggest that this aspect of cognition is best understood using an idea developed within work in formal semantics: modality.

    Jonathan Phillips, Joshua Knobe

    Mind & Language (2018)

    For a quick and easy summary, check out this blog post!

    scholar  .bib

  • Factive theory of mind

    Abstract

    We offer an account on which tracking another agent’s understanding of the world and keeping that representation separate from one’s own are the essential features of a capacity for theory of mind. Using this, we demonstrate how to tell when factive representations, e.g., what others see, hear, or know, provide evidence for theory of mind.

    Jonathan Phillips, Aaron Norby

    Mind & Langauge (2018)

    preprint     supplement

  • Differentiating could from should

    Abstract

    We show that young children have difficulty distinguishing immoral events from impossible ones, and impossible events from immoral ones. The ability to differentiate between the possibiity and wrongness of different kinds of abnormal events is a developmental achievement.

    Andrew Shtulman, Jonathan Phillips

    Journal of Child Development (2018)

    scholar  OSF    poster  .bib

  • Morality constrains the default representation

    Abstract

    Three studies differentiate explicit reasoning about possibilities from default implicit representations, demonstrate that human adults often default to treating immoral and irrational events as impossible, and demonstrate that high-level cognitive judgments rely on default implicit representations of possibility rather than explicit reasoning.

    Jonathan Phillips, Fiery Cushman

    Proceedings of the National Academy of Sciences (2017)

    Check out some of the coverage in Aeon, Harvard Gazette, a radio interview, or Science Breakers

    Supporting Information   .bib   github   scholar     blog post

  • True Happiness

    Abstract

    Five studies demonstrate that the orindary concept of happiness deviates from the definition used by researchers studying happiness: the ordinary concept is sensitive to the perceived moral value of the life lived.

    Jonathan Phillips, Christian Mott, Julian De Freitas, June Gruber, Joshua Knobe

    Journal of Experimental Psychology: General (2017)

    See the discussion of this work at Psychology Today or PEA Soup

    preprint     github     poster    scholar   .bib

  • Causation and norms of proper functioning

    Abstract

    We demonstrate that causal judgments of inanimate objects are highly sensitive to whether the object violated a prescriptive norm by malfunctioning and that this effect is well-explained by changes in the relevance of counterfactual alternatives.

    Jonathan Phillips, Jonathan Kominsky

    Proceedings of the 39th Annual Conference of the Cognitive Science Society (2017)

    OSF repository   github   scholar    poster   .bib

  • Unifying morality's influence on non-moral judgments

    Abstract

    Past work has demonstrated that people’s moral judgments can influence their judgments about freedom, causation, the doing/allowing distinction, and intentional action. Nine studies show that the effect of morality in these four domains can be explained in a unified manner by changes in the relevance of alternative possibilities.

    Jonathan Phillips, Jamie Luguri, Joshua Knobe

    Cognition (2015)

    Here’s a summary blog post

    preprint   github    poster    scholar

  • Knowledge before belief

    Abstract

    Drawing on evidence from a wide range of fields in cognitive science that employ diverse methodologies, we find, across every field and method, robust evidence that knowledge is a more basic mental-state representation than belief, and that representations of knowledge do not depend on representations of belief.

    Jonathan Phillips, Wesley Buckwlater, Fiery Cushman, Ori Friedman, Alia Martin, John Turri, Laurie Santos, Joshua Knobe

    41st annual meeting of the Society for Philosophy and Psychology (2015/6)

  • A second look at automatic theory of mind

    Abstract

    Recent research by Kovács, Téglás, & Endress (2010) argued that human adults automatically represent other agents’ beliefs even when those beliefs were completely irrelevant to the task being performed. In a series of eight studies, we replicate the previous findings but demonstrate that the effects found in this work arose from artifacts in the experimental paradigm.

    Jonathan Phillips, Desmond Ong, Andrew Surtees, Jean Xin, Samantha Williams, Rebecca Saxe, Michael Frank

    Psychological Science (2015)

    Here’s a blog post about this work, and here’s an important update (both by Michael Frank)

    preprint     github     scholar     demo     .bib    markdown version

  • Causal superseding

    Abstract

    When agents violate norms, they are typically judged to be more of a cause of resulting outcomes. In this paper, we suggest that norm violations also affect the causality attributed to other agents, a phenomenon we refer to as ‘causal superseding’.

    Jonathan Kominsky, Jonathan Phillips, Tobias Gerstenberg, David Lagnado, Joshua Knobe

    Cognition (2015)

    preprint     scholar     CogSci proceeding     .bib

  • Manipulating Morality

    Abstract

    The present studies investigate how the intentions of third parties influence judgments of moral responsibility for other agents who commit immoral acts. Using cases in which an agent acts under some situational constraint brought about by a third party, we ask whether the agent is blamed less for the immoral act when the third party intended for that act to occur.

    Manipulating Morality: Third-Party Intentions Alters Moral Judgments by Changing Causal Reasoning
    Jonathan Phillips, Alex Shaw
    Cognitive Science (2014)
    preprint     github     osf    update     blog     media     scholar     .bib

  • The paradox of moral focus

    Abstract

    When we evaluate moral agents, we consider many factors, including whether the agent acted freely, or under duress or coercion. In turn, moral evaluations have been shown to influence our (non-moral) evaluations of these same factors. For example, when we judge an agent to have acted immorally, we are subsequently more likely to judge the agent to have acted freely, not under force. Here, we investigate the cognitive signatures of this effect in interpersonal situations, in which one agent (“forcer”) forces another agent (“forcee”) to act either immorally or morally. The structure of this relationship allowed us to ask questions about both the “forcer” and the “forcee.” Paradoxically, participants judged that the “forcer” forced the “forcee” to act immorally (i.e. X forced Y), but that the “forcee” was not forced to act immorally (i.e. Y was not forced by X).

    Liane Young, Jonathan Phillips
    Cognition (2011)
    preprint     scholar     CogSci proceeding     .bib

Full List

How we know what not to think
Jonathan Phillips, Adam Morris, Fiery Cushman
Trends in Cognitive Sciences (2019)

Knowledge before belief
Jonathan Phillips, Wesley Buckwlater, Fiery Cushman, Ori Friedman, Alia Martin, John Turri, Laurie Santos, Joshua Knobe
41st annual meeting of the Society for Philosophy and Psychology (2015/6)

Immoral professors and malfunctioning tools: Counterfactual relevance accounts explain the effect of norm violations on causal selection
Jonathan Kominsky, Jonathan Phillips
Cognitive Science (2019)
github

Factive theory of mind
Jonathan Phillips, Aaron Norby
Mind & Langauge (2017)
preprint     supplement

Quantitative causal selection patterns in token causation
Adam Morris, Jonathan Phillips, Tobias Gerstenberg, Fiery Cushman
PLoS ONE 14(8): e0219704 (2019)
github

Sticky situations: Force and quantifier domains
Jonathan Phillips , Matthew Mandelkern
Proceedings of Semantics and Linguistic Theory (SALT), 28 (2018)
OSF   github   pre-registration

Estimating the Reproducibility of Experimental Philosophy
Florian Cova, Brent Strickland, … Jonathan Phillips , et al.
Review of Philosophy and Psychology (in press)
OSF

New horizons for a theory of epistemic modals
Jonathan Phillips , Justin Khoo
Australasian Journal of Philosophy, (2018)
preprint     github   OSF   preprint    scholar    .bib   handout

Knowledge wh and False Beliefs: Experimental Investigations
Jonathan Phillips , B R George
Journal of Semantics (2018)
github   OSF   preprint   scholar    .bib   handout

The psychological representation of modality
Jonathan Phillips , Joshua Knobe
Mind & Language (2018)
scholar   .bib

Differentiating could from should: Developmental changes in modal cognition
Andrew Shtulman, Jonathan Phillips
Journal of Child Development (2018)
scholar   OSF   poster  .bib

Causation and norms of proper functioning: Counterfactuals are (still) relevant
Jonathan Phillips , Jonathan Kominsky
Proceedings of the 39th Annual Conference of the Cognitive Science Society (2017)
OSF repository   github   scholar    poster   .bib

Morality constrains the default representation of what is possible
Jonathan Phillips , Fiery Cushman
Proceedings of the National Academy of Sciences (2017)
Supporting Information    .bib  github  scholar     blog post

True happiness: The role of morality in the folk concept of happiness
Jonathan Phillips , Christian Mott, Julian De Freitas, June Gruber, Joshua Knobe
Journal of Experimental Psychology: General (2017)
preprint     github     poster    scholar   .bib

Unifying morality’s influence on non-moral judgments: The relevance of alternative possibilities
Jonathan Phillips , Jamie Luguri, Joshua Knobe
Cognition (2015)
preprint   github    poster    scholar     .bib

A second look at automatic theory of mind: Reconsidering Kovács, Téglás, and Endress (2010)
Jonathan Phillips , Desmond Ong, Andrew Surtees, Jean Xin, Samantha Williams, Rebecca Saxe, Michael Frank
Psychological Science (2015)
preprint     github     scholar     demo     .bib    markdown version

Causal superseding
Jonathan Kominsky, Jonathan Phillips , Tobias Gerstenberg, David Lagnado, Joshua Knobe
Cognition (2015)
preprint     scholar     CogSci proceeding     .bib

The good in happiness
Jonathan Phillips , Sven Nyholm, Shen-yi Liao
Oxford Studies in Experimental Philosophy (2014)
preprint     interview     scholar     .bib

Manipulating morality: Third-party intentions alter moral judgments by changing causal reasoning
Jonathan Phillips , Alex Shaw
Cognitive Science (2014)
preprint     github     osf    update     blog     media     scholar     .bib

The paradox of moral focus
Liane Young, Jonathan Phillips
Cognition (2011)
preprint     scholar     CogSci proceeding     .bib

The ordinary concept of happiness (and others like it)
Jonathan Phillips , Luke Misenheimer, Joshua Knobe
Emotion Review (2011)
preprint     scholar     .bib    supplementary materials     YouTube

Moral judgments and intuitions about freedom
Jonathan Phillips , Joshua Knobe
Emotion Review (2009)
preprint     scholar     .bib     YouTube

Research

Our lab draws on the methods used across a number of disciplines (psychology, philosophy, linguistics, neuroscience, computer science) to answer central questions about how the human mind works. A primary topic we focus on is how humans represent and reason about non-actual possibilities—the vast infinity of things that didn’t (or have not yet) happened. Our understanding of what is possible is central to many of our most impressive, and uniquely human capacities: causal reasoning, theory of mind, planning, linguistic communication, moral judgment, and so on. Our work focus on these aspects of high-level cognition, both separately, and jointly by developing an understanding of the way that they all draw on a common underlying capacity to think about what is merely possible.

Possibilities

A main line of research has been to demonstrate the role that value plays in constraining which possibilities humans default to considering. Partly this work has just been an effort to uncover the empirical phenomenon: We’ve found that when adults are forced to decide what things are possibility when answering quickly (thus relying a default understanding of possibility), they begin to treat immoral actions as impossible. Moreover, children early in development judge immoral and imprudent actions to be impossible, even on reflection. The other part of this work has been to make sense of this phenomenon theoretically and formally. We’ve recently argued that value plays a similar role in which possible actions are considered in first-person decision making, and that this similarity is not merely a coincidence. We’ve also tried to provide a philosophically-oriented overview of how the resulting psychological picture of how we represent possibilities fits with theoretical work on judgments that require us to represent alternative possibilities.

Causation

A paradigm example of the kind of cognition that depend on representations of alternative possibilities is our judgments about what caused what. If you’re trying to figure out what the cause of a forest fire was, for example, part of what we do is consider what would have happened if things had happened differently: What if the person hadn’t made a campfire when there was a fire warning? Or what if there hadn’t been a drought? What is particularly interesting is that people share a strong tendency to consider certain kinds of counterfactuals and not others. We don’t tend to consider counterfactual possibilities in which there was no oxygen in the air and so the forest fire didn’t occur. Our work has investigated the way that that our ideas about what is normal shape our causal judgments because people tend to focus on counterfactuals that are more normal, ones in which people do the right thing, likely events occur, and things function normally.

Freedom and Responsibility

Another example of cognition involving alternative possibilities is how we judge whether someone was forced to do something and whether they were responsible for what they did. Typically, we only judge that someone was forced to do something when we don’t think it was possible for them to do something else instead. And we typically don’t hold other people response for their actions when they had no better alternative. Our work has demonstrated that judgments of force and responsibility depend on default representations of possibility, and that these judgments are surprisingly influenced by our moral judgments because we tend to not treat immoral alternative actions as possible. In fact, the impact of morality can lead to paradoxical patterns of judgments, where we judge that a person S was forced to do some action a by another person P, but that P did not force S to a.

Theory of Mind

Research on theory of mind has primarily focused on demonstrating and understanding the ability to represent others’ non-factive mental states, e.g., others’ beliefs in the false belief task. Our work has instead focused on how people represent others’ factive mental states, like what others know. We have developed a simple and theoretically motivated account on which tracking another agent’s understanding of the world and keeping that representation separate from one’s own are the essential features of a capacity for theory of mind. In ongoing work, we’ve review the theory of mind literature from across the cognitive sciences (research from comparative cognition, developmental psychology, cognitive psychology, linguistics, experimental philosophy, etc.) and find that across almost every measure, factive representations, like what others know are more basic than non-factive representations like what others believe.

Semantics

In our work on language, our lab largely focuses on developing formal semantic theories of modal terms—words like ‘might’ or ‘could’. Specifically, we’ve tried to provide an account of the meaning of these terms that integrates our emerging understanding of how people represent non-actual possibilities at a psychological level with standard semantics accounts that have been developed over the past half century in linguistics and philosophy. In pursuing this work, we have often tested the unique predictions of these formal models in experimental studies that form part of the growing tradition of work on experimental semantics. For a few examples, see our work on knowledge-wh, force, and epsitemic modals.

Team

We are looking for a new lab manager to join our team (see openings) !

Principal Investigator

Jonathan Phillips

  • Assistant Professor, Program in Cognitive Science
  • Graduate Advisor, Psychological and Brian Sciences
  • Affiliated Assistant Professor, Philosophy
  • jonathan.s.phillips@dartmouth.edu

Graduate Researchers

Catherine Holland PhD Student, starting Sep. 2019

Undergraduate Researchers

Eliza Jane Schaeffer

Hailey Scherer

Kyra Spaulding

Aneett Gawerc

Julian Grunauer

Donovan Fernandes

Clare Kennedy

Lidia Balanovich

Hannah LeBaron

Japser Meyer

Tracey Mills

Isabelle Morris

Darley Sackitey

Nathan Schneider

Collaborators and Affiliates

Liane Young

  • 2012-2020
  • Role: Affiliated researcher (Associate Professor, Boston College, Psychology)

Regan Bernhard

  • 2017-2020
  • Role: Affiliated post doc (Harvard University, Psychology)

Fiery Cushman

  • 2015-2020
  • Role: Affiliated researcher (Assistant Professor, Harvard University, Psychology)

Tobi Gerstenberg

  • 2015-2020
  • Role: Affiliated researcher (Assistant Professor, Stanford University, Psychology)

Thomas Icard

  • 2018-2020
  • Role: Affiliated researcher (Assistant Professor, Stanford University, Philosophy)

Justin Khoo

  • 2016-2020
  • Role: Affiliated researcher (Assistant Professor, MIT, Philosophy)

Joshua Knobe

  • 2011-2020
  • Role: Affiliated researcher (Professor, Yale University, Cognitive Science)

Jonathan Kominsky

  • 2014-2020
  • Role: Affiliated researcher (Post doc, Harvard University, Psychology)

Angelika Kratzer

  • 2016-2020
  • Role: Affiliated researcher (Emeritus Professor, University of Massachusetts Amherst, Linguistics)

Matthew Madelkern

  • 2016-2020
  • Role: Affiliated researcher (Post-Doctoral Research Fellow, University of Oxford, All Souls College)

Adam Morris

  • 2017-2020
  • Role: Affiliated graduate student (Harvard University, Psychology)

Adina Roskies

  • 2017-2020
  • Role: Affiliated researcher (Professor, Dartmouth College, Philosophy)
cute doge

Peanut

  • 2010-2020
  • Role: Cute pupper

Administrative Coordinator

Liz Cassell

News

New PhilLab Location

By Phillips / Jan 2020

Check out the new PhilLab space in Carpenter Hall! We will be hosting coffee hours 1-3pm every Friday, feel free to drop by and learn more about Cognitive Science!

Read More

New lab paper accepted for publication

By Phillips / Sept 2019

New lab paper with Adam Morris and Fiery Cushman has been accepted for publication at Trends in Cognitive Sciences!

Read More

Causation Paper @ PLoS One

By Phillips / July 2019

A new paper on causation with Adam Morris, Tobias Gerstenberg, and Fiery Cushman is out at PLoS One.

Read More

Factive ToM

By Phillips / April 2019

A new lab paper with Aaron Norby on theory of mind has been accepted for publication at Mind & Language.

Read More

New Grad Student

By Phillips / April 2019

Catherine Holland will be joining the lab as a PhD student in the Fall of 2019!!

Read More

Sticky Situations

By Phillips / October 2018

A new lab paper with Matthew Mandelkern on the semantics of force has been accepted for publication at Semantics and Linguistic Theory .

Read More

Join Us

We are always looking for folks who join us in interdisciplinary cognitive science research!

  • Open Positions

  • Lab Manager! If you’re interested in joining our lab as a lab manager, send me an email!

  • Applications for PhD and Postdoc positions

  • If you are interested in working with us as a PhD student or postdoc, please send me an email. State briefly why you are interested and attach a CV, including information about the research you’ve been engaged in recently. No need for a separate cover letter or certificates. Important: please insert “PhD Application” or “Postdoc Application” in the subject line. If you are applying to a specific advertisement, note this in your email.

  • Grad School Collaborators

  • If you are a graduate student in PBS or CS and would be interested in pursuing work collaboratively with our lab, email me (or any group member) or just stop by my office (Carpenter 002).

  • Undergrad Research

  • If you are interested in pursuing a research in cognitive science at Dartmouth, you're in good company! Undergraduate researchers typically commit to working roughly 10 hrs/week, and we have a strong preference for students to plan to work in the lab for more than one consecutive term, since this will allow you to contribute substantively to projects and perhaps even develop your own research ideas.

CONTACT US

We are part of the Program in Cognitive Science at Dartmouth College.