United States

Anchoring – the tendency for recently seen numbers to affect estimates – is a robust bias affecting expert and novice judgements across many fields. An anchoring task, in which people (N=301) estimated the number of circles in 10 stimulus figures after comparison to an anchoring value, was conducted within a larger study including numerous intelligence, personality, decision style and attention measures. Individual anchoring susceptibility was calculated and compared to potential predictor variables. Two of eight broad ability measures (from Catell-Horn-Carroll intelligence theory) correlated weakly but significantly with anchoring (Gq = 0.16, Gf = 0.12). No decision style or attention measures correlated significantly with anchoring, nor did the Big 5 personality traits, directly. Indirectly, however, as the anchoring task continued and fatigue increased, people relied more on anchors and higher neuroticism may have increased this tendency. Overall, results suggest our ability to predict anchoring is poor and implications of this are discussed.

CogSci 2021

Does anything predict anchoring bias?

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community."<br>

*Message from the General Chair, Dirk Hovy*





<br>

EMNLP 2025

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

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.

Human memory retains information related to survival more effectively, termed survival processing effect (SPE). Adaptive memory should also more effectively store reliable information than unreliable information. In an experiment (n=107) we asked whether the SPE depends on the reliability of information encoded by linguistic markers. In Turkish, evidentiality markers encode whether information is gathered firsthand (-miş) or not (-di), and provide cues on source reliability. We found that sentences processed for their relevance to survival were better recalled than those processed for their relevance to moving. This survival processing effect was significant for both direct and indirect evidentiality markers, although it was stronger for direct evidentiality. That SPE persisted in the indirect evidentiality condition suggests survival related information is privileged in encoding even when potentially marked as unreliable by linguistic markers. Thus, the effect of language on memory is not profound but rather computed online.

The effect of evidentiality markers on the survival processing effect

Confidence can be experienced for all kinds of decisions – evaluating the value of a piece of artwork, determining whether the lights are flickering, or remembering where we left our keys. These are fundamentally different kinds of decisions, but does that mean the confidence we feel is also fundamentally different for each one? Here, we test competing theories of domain-generality and domain-specificity in metacognitive ability by correlating individual differences in memory and perceptual confidence judgments in childhood. Children performed a recognition memory task and an area discrimination task followed by confidence judgments. Using 4 measures of metacognitive ability (indicated by higher confidence for accurate compared to inaccurate judgments: difference scores, meta-d’, MRatio, and HMeta-d’), we find no significant correlations between this ability in memory and perceptual tasks. These findings support an account of domain-specificity in children’s metacognitive abilities.

Perceptual and Memory Metacognition in Children

Widely recognized in differential psychology, but less so in cognitive science, the positive manifold is the phenomena of all cognitive tests inter-correlating positively. Frequently demonstrated in people, it can also be observed in non-human species. With 217 Ecuadorian adult participants, who performed 11 cognitive tests, we show that all 55 pairwise inter-correlations are positive, and of large magnitude. Additionally, factor analysis revealed a single underlying general, or g factor, often identified as general intelligence. This robustly replicates the positive manifold in a non- WEIRD (Western, Educated, Industrialized, Rich, Democratic) context. We further demonstrate that tests of lexical knowledge, such as word pronunciation, have particularly high loadings on g. We explore explanations for the positive manifold, and the implications for understanding the mind as being composed of independent cognitive processing modules. We propose that the positive manifold reveals a neglected but important role of lexical-conceptual knowledge in high-level, top-down, domain-general cognitive processing.

A Demonstration of The Positive Manifold of Cognitive Test Inter-correlations, and how it Relates to General Intelligence, Modularity, and Lexical Knowledge

Suppose humans exist in order to reproduce. Does it follow that an individual who chooses not to reproduce is committing a moral wrong? Past work suggests that, right or wrong, beliefs about species-level purpose are associated with moral condemnation of individuals who choose not to fulfill that purpose. Across two experiments we investigate why. Experiment 1 replicates a causal effect of species-level purpose on moral condemnation. Experiment 2 finds evidence that when a species is believed to exist to perform some action, people infer that the action is good for the species, and that this belief in turn supports moral condemnation of individuals who choose not to perform the action. Together, these findings shed light on how our descriptive understanding can sometimes shape our prescriptive judgments.

Why belief in species purpose prompts moral condemnation of individuals who fail to fulfill that purpose

Category production tasks (aka semantic fluency) typically concentrate on concrete categories, meaning little is known about abstract categories and potential differences in their structure. Using a category production task for 67 concrete (e.g., animals, tools) and 50 abstract (e.g., science, emotion) categories, we investigated differences in produced member concepts. Abstract categories were smaller than concrete categories, their member concepts were generated more slowly, and had longer and more phonologically-distinct names. Abstract category members were grounded in sensorimotor experience (i.e., with high sensorimotor strength), though overall to a lesser extent than concrete category members. Several ostensibly abstract categories appeared amongst the highest-rated in sensorimotor strength (e.g., sport) while some ostensibly concrete categories were amongst the lowest-rated (e.g., chemical element). The data highlight linguistic and semantic differences in concrete and abstract category structure, but also that they share a common sensorimotor grounding.

Exploring the Structure and Grounding of Concrete and Abstract Categories

Prior work demonstrates that humans represent the approximate number of items across multiple sets simultaneously. Here we investigated whether adults can approximately divide multiple distinct overlapping sets simultaneously. Participants viewed arrays with dots of one to four colors (potential dividends) and a non-symbolic divisor of 2-4. Participants then estimated the quotient of the division operation. On each trial a cue indicated whether to divide over the superset of all dots or a color subset. The cue was presented before or after the brief presentation of the dividend array. We found that the capacity to divide over a pre-selected subset was not affected by set size. By comparing the estimation error between the cue-before and cue-after conditions, we determined that a substantial proportion of participants can approximately divide three ensembles simultaneously (two subsets plus the superset). These findings demonstrate the computational efficiency and possible utility of ensemble representations.

Approximate division on multiple visual ensembles

A widespread tool in analogy research consists of multiple-choice tests that require identifying a relation between two situations and mapping it to another two situations, to find the correct response option. A key source of difficulty during such tests is attributed to the complexity of mapping. However, most people do not construct mappings purely in the mind, but also compare the emerging mapping with the existing response options, so their features may affect the reasoning process. This study examined the impact of relational match of error options with respect to the correct option (the proportion of correct elements present in a given error option) on the option selection. Results indicate that option selection depends almost linearly on the relational match. Moreover, the higher working memory capacity of the participants, the more relationally matching errors they select. The study suggests careful design of error options in multiple-choice reasoning tests, because the pattern of these options can affect the solution process.

Why people err on multiple-choice analogical reasoning tests

Studies in psycho-linguistics have provided compelling evidence that theoretical syntactic structures have cognitive correlates that inform and influence language perception. Generative grammar models also present a principled way to represent a plethora of hierarchical structures outside the domain of language. Hierarchical aspects of musical structure, in particular, are often described through grammar models. Whether such models carry perceptual relevance in music, however, requires further study. 
To address the descriptive adequacy of a grammar model in music, unfamiliar musical phrases consisting of chord progressions within the Jazz idiom were used, and zero to three chords were cut from the end of each phrase. A total of 150 participants were then presented with these stimuli and asked to provide a Closure Response, that is to predict how many more chords (0, 1, 2, or 3) were expected before the chord progression was complete. Simultaneously, a grammar model of hierarchical structure as well as a bigram model were trained over a corpus of 150 expert-annotated Jazz tunes. The models were then used to estimate probability distributions of Closure Responses in the stimuli presented to the participants. Bayesian mixed-effects models reveal that the models carry predictive value for the participants' response distributions and that the hierarchical model contains incremental predictive information over the bigram model. The present results suggest that -- akin to language -- hierarchical relationships between musical events have a cognitive correlate, which influences the perception and interpretation of music.

Hierarchical syntactic structure predicts listeners’ sequence completion in music

Laughter and gaze have an important role in managing and coordinating social interactions. We investigate whether laughs performing distinct pragmatic functions, either related to humour or to a potentially social discomforting utterance or situation, are accompanied by different gaze patterns. Using a multimodal corpus of dyadic taste-testing interactions, we show that people tend to not look at their partner while producing laughs related to humour, whereas laughs that relate to potential social discomfort are accompanied by gaze at the partner. With respect to the non-laughing partner's gaze at the laugher we observe the opposite pattern around the laughter offset. We show that gaze contributes to the synchronisation and alignment of laughter production, analogously to previously reported results for speech turn-taking. Our study also provides empirical evidence to the debate about gaze aversion, opposing the view that social stress is the main explanatory factor.

Looking at the Pragmatics of Laughter

Toddlers can learn cause-effect relationships between single actions and outcomes. However, real-world causality is often more complex. We investigated whether toddlers (12- to 35-month-olds) can learn that a sequence of two actions is causally necessary, from observing the actions of an adult demonstrator. In Experiment 1, toddlers saw evidence that performing a two-action sequence (AB) on a puzzle-box was necessary to produce a sticker, and evidence that B alone was not sufficient. Toddlers were then given the opportunity to interact with the box and retrieve up to five stickers. Toddlers had difficulty reproducing the required two-action sequence, with the ability to do so improving with age. In Experiment 2, toddlers saw evidence that performing a single action (B) was sufficient to produce an effect (i.e., a sequence was not causally necessary). Toddlers were more successful and performed fewer sequences in Experiment 2, suggesting some sensitivity to the sequential causal structure.

Downloads

Next from CogSci 2021

The effect of evidentiality markers on the survival processing effect

Similar lecture

How Does Cognitive Bias Affect Large Language Models? A Case Study on the Anchoring Effect in Price Negotiation Simulations