United States

Human Theory of Mind enables us to attribute mental states like beliefs and desires based on how other people act. However, in many social interactions (particularly ones that lack observable action), people also directly think about other people&#39;s thinking. Here we present a computational framework, Bayesian inverse reasoning, for thinking about other people&#39;s thoughts. Our framework formalizes inferences about thinking by inferring a generative model of reasoning decisions and computational processes, structured around a principle of rational mental effort -- the idea that people expect other agents to allocate thinking rationally. We show that this model quantitatively predicts human judgements in a task where participants must infer the mental causes behind an agent&#39;s pauses as they navigate and solve a maze. Our results contribute to our understanding of the richness of the human ability to think about other minds, and to even think about thinking itself.

CogSci 2021

hinking about thinking through inverse reasoning

stages 

COLING, the International Conference on Computational Linguistics, is one of the premier conferences for natural language processing and computational linguistics. Often grouped within the field of artificial intelligence, but actually pre-dating the development of artificial intelligence, advances in computational linguistics and natural language processing are now some of the major drivers behind the use of artificial intelligence for commercial and social applications – for example, on-line search, machine translation and with voice-assisted conversational devices.

First established in 1965, the biennial COLING conference is held in diverse parts of the globe and attracts participants from both top-ranked research centers and emerging countries. Today, the most important developments in our field are taking place not only in universities and academic research institutes, but also in industrial research departments and in technological startups. COLING conferences provide opportunities for all these communities to showcase their exciting developments.

COLING 2020

COLING, the International Conference on Computational Linguistics, is one of the premier conferences for natural language processing and computational linguistics. Often grouped within the field of artificial intelligence, but actually pre-dating the development of artificial intelligence, advances in computational linguistics and natural language processing are now some of the major drivers behind the use of artificial intelligence for commercial and social applications – for example, on-line search, machine translation and with voice-assisted conversational devices.

Human Theory of Mind enables us to attribute mental states like beliefs and desires based on how other people act. However, in many social interactions (particularly ones that lack observable action), people also directly think about other people's thinking. Here we present a computational framework, Bayesian inverse reasoning, for thinking about other people's thoughts. Our framework formalizes inferences about thinking by inferring a generative model of reasoning decisions and computational processes, structured around a principle of rational mental effort -- the idea that people expect other agents to allocate thinking rationally. We show that this model quantitatively predicts human judgements in a task where participants must infer the mental causes behind an agent's pauses as they navigate and solve a maze. Our results contribute to our understanding of the richness of the human ability to think about other minds, and to even think about thinking itself.

poster

### Looking for Cogsci 2023? Please [click here](https://underline.io/events/405/reception) to switch to 2023 edition.


**Dear Colleagues,**

Welcome to the 43rd Annual Conference of the Cognitive Science Society! 

This year’s conference brings together the latest research from all Cognitive Science and also highlights this year’s theme **Comparative Cognition: Animal Minds.** 

The program features four plenary speakers: Friederike Range (University of Veterinary Medicine, Vienna), Onur Güntürkün (Ruhr University Bochum), Julia Fischer (University of Göttingen), and Erich Jarvis (The Rockefeller University). The keynotes and invited symposia showcase the broad spectrum of ideas that encompass the conference theme of comparative cognition. 

Further, the program includes the Rumelhart Prize keynote presentation by Susan Goldin-Meadow (University of Chicago), and symposia to celebrate the winner of the Elman Prize: Wei Ji Ma (New York University), the Heineken Award winner Robert Zatorre (McGill University), and the Robert J. Glushko Dissertation Prizes session.

The program committee for CogSci 2021 received 1,025 submissions, including 710 full papers, 275 member abstracts, 8 publication-based talks, as well as 15 proposals for symposia, and 17 for workshops and tutorials. After a rigorous review process, the committee selected 152 papers for oral presentation and inclusion in the conference proceedings (14.8%), 332 papers for poster presentation and inclusion in the proceedings (32.4%), and 175 papers for poster presentation with inclusion of abstracts in the proceedings (17.1%) We also selected 275 submitted member abstracts. In addition, we accepted 6 publication-based talks, 12 symposia, and 9 workshops and tutorials to make for an exciting and inclusive program.

We hope that you enjoy the conference this year!<br><br>

**Your Program Chairs,**
![](https://assets.underline.io/uploads/markdown_image/1/image/99984908caba5e6453e5fc5ba34f3507.jpg)

**Learn more about the Vienna Cognitive Science Hub at the University of Vienna, hosts of CogSci 2021**

<iframe width="560" height="315" src="https://www.youtube.com/embed/464h1m3iU68" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

To review the proceedings for the conference or review the searchable program, please visit the [website](https://cognitivesciencesociety.org/program/).


The Cognitive Science Society brings together researchers from around the world who hold a common goal: understanding the nature of the human mind. The mission of the Society is to promote Cognitive Science as a discipline, and to foster scientific interchange among researchers in various areas of study, including Artificial Intelligence, Linguistics, Anthropology, Psychology, Neuroscience, Philosophy, and Education. The Society is a non-profit professional organization and its activities include sponsoring an annual conference and publishing the journals Cognitive Science and TopiCS.

How should we represent social categories? Essentialism, which posits an internal category essence, has negative consequences such as group-based generalization and intolerance of nonconformity. Structural representations, which consider categories as situated in a larger context, could be more constructive. When considering a neutral context, adults who learned a structural (food availability) cause for a novel group’s diet, versus a biological (allergies, digestion) or cultural (taboo) cause, generalized in a context-sensitive manner, considered nonconformity to be more possible and acceptable, and suggested intervening on the structural context rather than the group to change the property. When considering social stratification, adults who learned a structural (discrimination) cause for a novel group working a low-status job, versus a biological (physically well-suited) or cultural (traditions/values) cause, showed more context-sensitive generalization, considered the present disparity to be more unacceptable, and suggested more structural interventions. Structural representations may consequently be a more constructive alternative to essentialism.

Let’s talk structure: the positive consequences of structural representations

Performance on cognitive tasks is typically affected by many factors simultaneously. For example, performance on a retrieval practice task is determined not only by memory encoding and retrieval processes, but also by motor processes, executive functions, affective state, etc. Often, these contributing processes also vary over time. This makes it challenging to attribute performance changes to a single mechanism, especially in real-world settings.

Here we analyse performance data from multiple retrieval practice sessions completed by special forces trainees before and after a high-intensity speed march. Learning outcomes improved from session to session, which suggests increased memory function. However, a linear ballistic accumulator fitted to the data showed that changes in non-memory processes—a decrease in non-retrieval time in particular—sufficed to explain the observed improvements.

This work demonstrates that even with a simple task, assuming that performance changes are attributable to a single cognitive process can lead to erroneous conclusions.

Memory Performance in Special Forces: Speedier Responses Explain Improved Retrieval Performance after Physical Exertion

The category variability effect is referred to as that the middle item between two categories is more similar to the low-variability category but tends to be classified as the high-variability category, which challenges the exemplar model. We however hypothesized that this effect can result from the use of the single-category strategy in a binary categorization task, specifically when only the low-variability category is referenced for categorization. One experiment was conducted with a recognition task inserted in the categorization task to selectively deepen the processing for the exemplars of the high-variability category, low-variability category, or both categories. The results showed that the strongest category variability effect occurred when the low-variability category was emphasized in the recognition task. The exemplar model SD-GCM provided a good account for the category variability effect, with a large weight for the low-variability category and a small weight for the high-variability category, hence verifying our hypothesis.

Exemplar Account for Category Variability Effect: Single Category based Categorization

We trust ingroups more than outgroups (Balliet et al. 2014, Psychological Bulletin), but to correctly identify group members we need reliable markers of group membership (Cohen 2012, Current Anthropology). Artificial language experiments show that linguistic tags can serve as such group markers (Roberts 2013, Language and Linguistics Compass). We now tested whether sharing a language also promotes trust, in comparison to other physical and cultural ingroup markers. We created an online alien game in which participants assumed the identity of an alien and played a simultaneous trust game (Berg et al. 1995, Games and Economic Behavior) with two other aliens exhibiting the ingroup versus outgroup marker types (artificial languages, body parts, costumes). GLM analyses showed that participants significantly entrusted aliens sharing their own social marker with more money, independently of marker type. Thus, a considerable number of people trusted those who spoke or looked like themselves more.

Do I trust you more if you speak like me?

Understanding how people mentally represent expenditures is crucial to understanding how they manage their money. In this paper, we report three studies that investigate people’s representation of budgeting categories by asking people to categorize common expenditures of money (e.g., rent, dining out, etc.). We then examine the implications of these taxonomic representations of expenditures for how people selectively restrict their uses of money (e.g., when overspending on one item, for which other items would people choose to spend less). We found that there is consensus in people’s representations of expenditures, and that both the category membership and taxonomic distance between items predict how people restrict their spending.

Mental Representation of Budgeting Categories

COVID-19 and the 2021 U.S. Capitol attacks have highlighted the potential dangers of pseudoscientific and conspiratorial belief adoption. Approaches to combating misinformed beliefs have tried to "pre-bunk" or "inoculate" people against misinformation adoption and have yielded only modest results. These approaches presume that some citizens may be more gullible than others and thus susceptible to multiple misinformed beliefs. We provide evidence of an alternative account: it's simply too hard for all people to be accurate in all domains of belief, but most individuals are trying. We collected data on a constellation of human beliefs across domains from more than 1,700 people on Amazon Mechanical Turk, and find misinformed beliefs to be broadly, but thinly, spread among the population. Further, we do not find that individuals who adopt one misinformed belief are more likely to engage in pseudoscientific or conspiratorial thinking across the board, in opposition to "slippery slope” notions of misinformation adoption.

"Fringe" beliefs aren't fringe

Language emergence is characterized by a high degree of lexical variation. It has been suggested that the speed at which lexical conventionalization occurs depends partially on social structure. In large communities, individuals receive input from many sources, creating a pressure for lexical convergence. In small, insular communities, individuals can remember idiolects and share common ground with interlocuters, allowing these communities to retain a high degree of lexical variation. We look at lexical variation in Kata Kolok, a sign language which emerged six generations ago in a Balinese village, where women tend to have more tightly-knit social networks than men. We test if there are differing degrees of lexical uniformity between women and men by reanalyzing a picture description task in Kata Kolok. We find that women’s productions exhibit less lexical uniformity than men’s. One possible explanation of this finding is that women's more tightly-knit social networks allow for remembering idiolects, alleviating the pressure for lexical convergence, but social network data from the Kata Kolok community is needed to support this explanation.

Social structure and lexical uniformity: a case study of gender differences in the Kata Kolok community

How do people acquire information and make decisions in an inherently complex world? We use the game Entropy Mastermind to investigate the cognitive strategies people adopt under various task conditions in situations characterized by high combinatorial complexity. N = 42 participants completed a total of 271 games, varying in incentive structure: In the speed condition, participants were incentivised for solving games quickly; in the efficiency condition, incentives were given for solutions requiring few problem solving steps; in the mixed condition, both speed and efficiency were incentivised.
We found that participants adapted their problem solving strategies to the imposed constraints: In the speed condition simpler strategies were used, making feedback easy to interpret. Our results support the hypothesis that in complex environments people flexibly adjust their cognitive strategies to optimize the fit between situational requirements and their cognitive resources.

Fast or efficient? Strategy selection in the game Entropy Mastermind

In addition to saliency and goal-based factors, a scene’s semantic content has been shown to guide attention in visual search tasks. Here, we ask if this rapidly available guidance signal can be leveraged to learn new attentional strategies. In a variant of the scene preview paradigm (Castelhano & Heaven, 2010), participants searched for targets embedded in real-world scenes with target locations linked to scene gist. We found that activating gist with scene previews significantly increased search efficiency over time in a manner consistent with formal theories of skill acquisition. We combine VGG16 and EBRW to provide a biologically inspired account of the gist preview advantage and its effects on learning in gist-guided attention. Preliminary model results suggest that, when preview information is useful, stimulus features may amplify the similarities and differences between exemplars.

Leveraging rapid scene perception in attentional learning

People's visual experiences are easy to examine along natural language boundaries, e.g., by categories or attributes. However, it is more difficult to elicit detailed visuospatial information about what a person attends to, e.g., the specific shape of a tree. Paying attention to the shapes of things not only feeds into tasks like visual category learning, but also enables us to differentiate similarly named objects and to take on creative visual pursuits, like poetically describing the shape of a thing, or finding shapes in the clouds or stars. We use a new data collection method that elicits people's prioritized attention to shapes during visual photo inspection by asking them to trace important parts of the image under varying time constraints. Using data collected via crowdsourcing over a set of 187 photographs, we examine changes in patterns of visual attention across individuals, across image types, and across time constraints.

hinking about thinking through inverse reasoning

Downloads

Next from CogSci 2021

Let’s talk structure: the positive consequences of structural representations

Similar lecture

Bi-directional Cognitive Thinking Network for Machine Reading Comprehension

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES