Inferring the intentional states of autonomous virtual agents

Pantelis, Peter C.; Baker, Chris; Cholewiak, Steven A.; Sanik, Kevin; Weinstein, Ari; Wu, Chia Chien; Tenenbaum, Joshua B.; Feldman, Jacob

doi:10.1016/j.cognition.2013.11.011

Cited by 28 publications

(28 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study differs in several ways from previous of studies of the accuracy of inferences of intention (Baker, Saxe, & Tenenbaum, 2009; Pantelis et al, 2014). First, other studies used simulated agents that did not necessarily produce ecologically valid examples of intentional behavior.…”

Section: Discussionmentioning

confidence: 65%

Control blindness: Why people can make incorrect inferences about the intentions of others

Willett

Marken

Parker

et al. 2017

Atten Percept Psychophys

View full text Add to dashboard Cite

There is limited evidence regarding the accuracy of inferences about intention. The research described in this article shows how perceptual control theory (PCT) can provide a “ground truth” for these judgments. In a series of 3 studies, participants were asked to identify a person’s intention in a tracking task where the person’s true intention was to control the position of a knot connecting a pair of rubber bands. Most participants failed to correctly infer the person’s intention, instead inferring complex but nonexistent goals (such as “tracing out two kangaroos boxing”) based on the actions taken to keep the knot under control. Therefore, most of our participants experienced what we call “control blindness.” The effect persisted with many participants even when their awareness was successfully directed at the knot whose position was under control. Beyond exploring the control blindness phenomenon in the context of our studies, we discuss its implications for psychological research and public policy.

show abstract

Section: Discussionmentioning

confidence: 65%

Control blindness: Why people can make incorrect inferences about the intentions of others

Willett

Marken

Parker

et al. 2017

Atten Percept Psychophys

View full text Add to dashboard Cite

show abstract

“…Previous studies have modeled the computational mechanisms of the human mind-reading ability as a Bayesian inference ( Fig. 2 left) [1,8,14]. In this paper, we collectively call these approaches the Bayesian Theory of Mind (BToM).…”

Section: Bayesian Modeling Of Theory Of Mindmentioning

confidence: 99%

Bayesian Inference of Self-intention Attributed by Observer

Fukuchi

Osawa

Yamakawa³

et al. 2018

Proceedings of the 6th International Conference on Human-Agent Interaction

View full text Add to dashboard Cite

Most of agents that learn policy for tasks with reinforcement learning (RL) lack the ability to communicate with people, which makes human-agent collaboration challenging. We believe that, in order for RL agents to comprehend utterances from human colleagues, RL agents must infer the mental states that people attribute to them because people sometimes infer an interlocutor's mental states and communicate on the basis of this mental inference. This paper proposes PublicSelf model, which is a model of a person who infers how the person's own behavior appears to their colleagues. We implemented the PublicSelf model for an RL agent in a simulated environment and examined the inference of the model by comparing it with people's judgment. The results showed that the agent's intention that people attributed to the agent's movement was correctly inferred by the model in scenes where people could find certain intentionality from the agent's behavior. CCS CONCEPTS• Computing methodologies → Theory of mind; KEYWORDS Reinforcement learning, Bayesian inference, Public self-awareness, Theory of mind, PublicSelf model, Human-agent interaction ACM Reference Format:

show abstract

“…Computational accounts of top–down reasoning about intentional agents in simple animated displays have modeled how the structure of agents' actions and of the situational context shape conceptually rich mental state inferences, such as attribution of goals to individual (Baker, Saxe, & Tenenbaum, ; see Fig. C) and interactive (Baker, Goodman, & Tenenbaum, ; Pantelis et al, ; Ullman et al, ) agents, or attribution of beliefs and desires (Baker, Saxe, & Tenenbaum, ). These accounts formalize the “principle of rationality” from philosophy (Dennett, ) and developmental psychology (Gergely, Nádasdy, Csibra, & Bíró, )—the assumption that intentional agents will act rationally to achieve their desires, given their beliefs about the world—in terms of probabilistic models of agents' rational belief‐, desire‐, goal‐, and context‐dependent action planning, based on accounts of rational utility‐theoretic planning from AI and economics.…”

Section: Introductionmentioning

confidence: 99%

The Cognitive Architecture of Perceived Animacy: Intention, Attention, and Memory

et al. 2019

Self Cite

View full text Add to dashboard Cite

Human vision supports social perception by efficiently detecting agents and extracting rich information about their actions, goals, and intentions. Here, we explore the cognitive architecture of perceived animacy by constructing Bayesian models that integrate domain‐specific hypotheses of social agency with domain‐general cognitive constraints on sensory, memory, and attentional processing. Our model posits that perceived animacy combines a bottom–up, feature‐based, parallel search for goal‐directed movements with a top–down selection process for intent inference. The interaction of these architecturally distinct processes makes perceived animacy fast, flexible, and yet cognitively efficient. In the context of chasing, in which a predator (the “wolf”) pursues a prey (the “sheep”), our model addresses the computational challenge of identifying target agents among varying numbers of distractor objects, despite a quadratic increase in the number of possible interactions as more objects appear in a scene. By comparing modeling results with human psychophysics in several studies, we show that the effectiveness and efficiency of human perceived animacy can be explained by a Bayesian ideal observer model with realistic cognitive constraints. These results provide an understanding of perceived animacy at the algorithmic level—how it is achieved by cognitive mechanisms such as attention and working memory, and how it can be integrated with higher‐level reasoning about social agency.

show abstract

Inferring the intentional states of autonomous virtual agents

Cited by 28 publications

References 27 publications

Control blindness: Why people can make incorrect inferences about the intentions of others

Control blindness: Why people can make incorrect inferences about the intentions of others

Bayesian Inference of Self-intention Attributed by Observer

The Cognitive Architecture of Perceived Animacy: Intention, Attention, and Memory

Contact Info

Product

Resources

About