United States

Chinese texts are renowned for the lack of physical spaces between words in a sentence. Reading these sentences requires a stage of word segmentation, the mechanism of which may involve visual statistical learning. In three experiments employing the RSVP task along with the Saffran et al. (1997) paradigm, we provided evidence that foreign learners of Chinese could capture the statistical information embedded in a string of characters and use that information to tell apart a “word” from a “nonword”. The statistical learning effect (.57) was comparable to that observed previously in an auditory task using the same stimuli. The results of the experiments also suggested that significant visual statistical learning required a conscious level of processing that directed the participants’ attention at the characters as well as an unconscious level, at which the distributional information across the characters can be continuously computed and accumulated.

CogSci 2021

Visual Statistical Learning in the Reading of Unspaced Chinese Sentences

**Welcome to ACL-IJCNLP 2021!**

The great event is jointly organized by the Association for Computational Linguistics (ACL) and Asian Federation of Natural Language Processing (AFNLP). 

As in previous years, the program of the conference includes a poster session, tutorials, workshops and demonstrations in addition to the main conference.


We were able to keep the registration fees similar to those charged for the virtual ACL 2020. The one fee allows attendance at the main conference and any/all tutorials and workshops. These fees would be $125 Regular Early and $175 Regular Late; $50 Student Early and $75 Student Late. Early registration closes at midnight July 11, 2021 (Eastern Daylight Time).

**Reminder:** It is ACL’s policy that at least one author of each accepted paper (including ACL Finding papers) must register for the conference.

**Reminder2:** Underline site will open closer to the event. If you already registered you will receive access detail

Registration is now open

ACL-IJCNLP 2021

The great event is jointly organized by the Association for Computational Linguistics (ACL) and Asian Federation of Natural Language Processing (AFNLP).

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.

Folk-biological concepts are sensitive to both statistical information about feature prevalence (Hampton, 1995; Kim & Murphy, 2011; Rosch & Mervis, 1975) and teleological beliefs about function (Atran, 1995; Keil, 1994; Kelemen, Rottman, & Seston, 2013; Lombrozo & Rehder, 2012), but it is unknown how these two types of information interact to shape concepts. In three studies (N = 438) using novel animal kinds, we found that information about prevalence and teleology inform each other: People assume that common features are functional, and they assume that functional features are common. However, people use teleological information to predict the future distribution of features across the category, despite conflicting information about current prevalence. Thus, both information about prevalence and teleological beliefs serve important conceptual functions: Prevalence information encodes the current state of the category, while teleological functions provides a means of predicting future category change.

The function of function: People use teleological information to predict prevalence

In phonological form preparation, speakers are able to prepare in advance when to-be-spoken words share initial segments, despite not knowing which word they will be asked to produce. The standard account viewed preparation as partial production and therefore claimed that all possible words must share that segment (e.g. Roelofs, 1997). Alternatively, preparation occurs through a flexible sustained attention process that is external to production. Recently, O’Seaghdha and Frazer (2014) demonstrated preparation benefits in non-unanimous sets using picture naming and word reading versions of the blocked cyclic task, though not in the more attentionally-demanding paired associates version. Here, the current studies demonstrate small, but significant, preparation benefits using the paired-associates task, but only when cue-target pairs are highly associated (e.g. dog-cat, oreo-cookie, hot-cold, worm-bait). This finding supports the sustained attentional account of preparation in word production.

Sustained Attention in Phonological Form Preparation: Evidence from Highly Associated Word Pairs

Many events that humans and other organisms experience contain regularities in which certain elements within an event predict certain others. While some of these regularities involve tracking the co-occurrences between temporarily adjacent stimuli, others involve tracking the co-occurrences between temporarily distant stimuli (i.e., non-adjacent dependencies, NADs). Prior research shows robust learning of adjacent dependencies in humans and other species, whereas learning NADs is more difficult, and often requires support from properties of the stimulus to help learners notice the NADs. Here we report on four experiments that examined NAD learning from various types of visual stimuli. The results suggest that continuous movements aid the acquisition of NADs. We also found that human motion leads to more robust NAD learning compared to object motions, perhaps because of a richer representation. This richer representation could result in better memory and recall, and provide a stronger signal for NAD learning.

Dynamic Action Facilitates Learning of Non-Adjacent Dependencies in Visual Sequences

Process data recording students’ interactions with digital assessment items are available in digital educational assessments and have become a focus of cognitive scientists to analyze inquiry skills during problem solving. This study examines the inquiry behaviors of using tools (i.e., resource tabs) in short response construction items from the 2019 National Assessment of Educational Progress science assessment. We visualized the occurrence times and durations of response construction behaviors and tool use behaviors and conducted correlation analyses and mixed-effects regressions between the count (and duration) of tool use behaviors and item scores. The results reveal that tool use behaviors are significantly associated with item scores and probabilities of finishing the whole block the problem; increasing tool use durations or counts increases the chances of getting higher scores, but it also increases the chances of not finishing the block. This study exemplifies how to use process data to investigate scientific inquiry skills.

Investigating Scientific Inquiry Skills from Process Data

An open problem in Libet task literature regards the relationship between the moment in which awareness of motor intention arises and inhibition efficiency in response to an external stimulus (taking into account both the reactive and proactive mechanisms). In this study, 112 volunteers performed the Libet’s clock task to evaluate motor intention awareness, a Stop Signal Task (SST) to evaluate the inhibitory efficiency in its mainly reactive component, and a Cued Go/No-Go to evaluate the inhibitory efficiency in its mainly proactive dimension. We observed that a delayed insurgence of the awareness of motor intent is related to a better reactive inhibitory efficiency. No relationship was observed with the proactive component.

Awareness motor intention and inhibitory control: the role of reactive and proactive components

Misconceptions in science are ubiquitous and difficult to revise, but refutation texts are one effective tool for prompting conceptual change (Tippett, 2010). However, refutation texts are effective only to the extent that students are forming coherent representations. Self-explanation is a well-studied method for increasing coherence of readers’ mental representations (Allen, McNamara & McCrudden, 2015). The current study examines the unexplored question of whether these common interventions enhance each other’s positive learning outcomes. Two-hundred fifteen UCSD undergraduates were randomly assigned to read either a refutation or an expository text about the phases of the moon, and were prompted to either self-explain or think aloud while reading. Students then took a post-test assessing knowledge of moon phases and related misconceptions. We measured both accuracy and explanatory qualities, such as causality and circularity, in order to assess the relative efficacy of refutation and self-explanation and their combination.

Effects of Combining Refutation and Self-Explanation on Student Learning

Most models of statistical learning do not consider the perceptual properties of the units that make up a sequence. Here, we manipulate the similarity of units in a modified version of classic statistical learning paradigm (Aslin et al., 1998). After a 3-minute familiarization stream, we asked the participants to rate their familiarity with words and part-words, with the addition of non-words (trisyllabic words without any previous exposure). We developed a simple recurrent neural network that used distributed representations for the syllables and produced similar results to the human data. We then explored how perceptual similarity impacted learning of different familiarization streams. Specifically, greater similarity relations among units were predicted to lead to poorer discrimination between words and part-words. Based on the model’s results, we created a new familiarization sequence of trisyllabic words and tested for impacts on perceptual similarity on human performance.

Perceptual similarity and learning from sequential statistics.

Prior research often finds increased altruism following natural disasters. One explanation is the social heuristic hypothesis: humans are prosocial by nature but become self-interested when they have the opportunity to deliberate. As the stress of a disaster lowers people’s ability to engage in effortful deliberation, their heuristic prosocial tendencies emerge. However, this link has often been explored with very simple tasks; here, we study the impact of COVID-related stress on outcomes in multi-issue negotiations with a computational virtual agent. In two experiments with a virtual negotiation partner, we explore two distinct pathways for how COVID-19 stress shapes prosocial behavior. Consistent with the social heuristic hypothesis, COVID-stress is correlated with giving, mediated by heuristic thinking. But COVID-stress also seems to enhance information-exchange and perspective taking, which allowed participants to grow more value which they could give away. Our results give new insights into the relationship between stress, cognition, and prosocial behavior.

Pandemic Panic: The Effect of Disaster-Related Stress on Negotiation Outcomes

Simple, non-pharmaceutical health interventions (e.g., masks) could prevent avoidable COVID-19 deaths (Reiner et al., 2020). Why do some still refuse to wear masks? People may find it easier to rely on implicit, prior knowledge to avoid needing to inhibit new, competing knowledge (e.g., heuristics; Tversky & Kahneman, 1974). Not only does knowledge impact decision making, but from a contextualized deficit framework, knowledge should interact with context to promote differential health decisions (Allum et al., 2008). A computer mouse-tracking paradigm evaluated how context (e.g., trust-in-experts; incidence rate) and germ knowledge impacted cognitive conflict driving mask-usage decisions. Results indicated that increased trust-in-experts, higher positive COVID-19 incidence rates, as well as accurate germ theories promoted mask-usage endorsement, which also reduced cognitive conflict between knowledge about new and old public-health mask guidelines. Contextual factors may help remediate the cognitive stress associated with inhibiting prior inaccuracies in favor of updated, scientific mask recommendations.

Implicit and Explicit Cognitive Processes Associated with COVID-19 Mask-Usage Decisions

Can mental simulation of alternative research outcomes reduce bias? We attempted to extend Hirt et al.’s (2004) finding of debiasing when alternative basketball standings were easy to simulate and participants were low in need for structure (NFS). Amazon Mechanical Turk (AMT) participants explained why taking notes by hand might improve test scores, then three groups of them explained different outcomes: Consider-opposite participants explained benefits of laptop notetaking, and two transfer groups explained either a plausible or an implausible outcome for unrelated research. None of the three groups differed from a baseline group in test score estimates or likelihood that taking notes by hand leads to higher test scores. However, low-NFS participants estimated marginally lower likelihood that notetaking by hand is superior, suggesting less bias to their initial explanation. We consider whether variation in participants’ psychology backgrounds might have overwhelmed effects, and discuss a replication with students taking introductory psychology.

Downloads

Next from CogSci 2021

The function of function: People use teleological information to predict prevalence

Similar lecture

MECT: Multi-Metadata Embedding based Cross-Transformer for Chinese Named Entity Recognition