Date of Award

Spring 2020

Document Type

Thesis (Undergraduate)

Department or Program

Cognitive Science

First Advisor

Jonathan Phillips


The way we interact with the world is governed by a body of rules, many of which are unspoken. What makes these rules so compelling? How do they interact with our decision-making infrastructure? In a series of three studies, this article explores the shared adaptive sampling model’s ability to account for normative behaviors, using a time pressure paradigm in which subjects (N = 399) recruited via Amazon Mechanical Turk quickly judged whether actions of various degrees of moral permissibility and social acceptability were possible or impossible in a certain context (Phillips & Cushman, 2017; Phillips, Morris, & Cushman, 2019). When making intuitive judgements of possibility, participants regarded immoral actions as physically impossible, replicating findings from previous research (Phillips & Cushman, 2017). This effect was not found for actions that are simply abnormal, violating social norms rather than moral norms. These studies also provide evidence to suggest that beyond subjects’ assessments of moral permissibility or impermissibility, it is the general perceived value (good vs. bad) of the proposed actions that drives the morality deliberation effect. This is consistent with a two-staged decision-making process, as described in the shared adaptive sampling model, in which the mind samples generally good and generally likely possibilities before using contextual information to further select amongst them. Overall, these studies elucidate the similarities and differences between various normative judgements and general assessments of value and probability, corroborating the shared adaptive sampling model, providing reason to believe that moral norms are distinct from social norms, and offering insight into how various sorts of norms may interact with possibility sampling and decision-making.