United States

As children acquire the knowledge of their language&#39;s morphology, they invariably discover the productive processes that can generalize to new words. Morphological learning is made challenging by the fact that even fully productive rules have exceptions, as in the well-known case of English past tense verbs, which features the -ed rule against the irregular verbs. The Tolerance Principle is a recent proposal that provides a precise threshold of exceptions that a productive rule can withstand. Its empirical application so far, however, requires the researcher to fully specify rules defined over a set of words. We propose a greedy search model that automatically hypothesizes rules and evaluates their productivity over a vocabulary. When the search for broader productivity fails, the model recursively subdivides the vocabulary and continues the search for productivity over narrower rules. Trained on psychologically realistic data from child-directed input, our model displays developmental patterns observed in child morphology acquisition, including the notoriously complex case of German noun pluralization. It also produces responses to nonce words that, despite receiving only a fraction of the training data, are more similar to those of human subjects than current neural network models&#39; responses are.

CogSci 2021

The Greedy and Recursive Search for Morphological Productivity

### Welcome to IJCNLP-AACL 2023!
I am honored and happy to welcome you to IJCNLP-AACL 2023! The conference hosts both the 13th
International Joint Conference on Natural Language Processing (IJCNLP) and the 3rd Conference of
the Asia-Pacific Chapter of the Association for Computational Linguistics (AACL). 
The joint conferences of IJCNLP and AACL have been organized by alternating leadership in the
Asia-Pacific region, in odd years by Asian Federation of Natural Language Processing (AFNLP) and
in even years by AACL. The conference is organized by AFNLP this year. 

Considering the current situation that we are not yet fully over the coronavirus pandemic, the
conference is offered in a hybrid mode, off-line in Nusa Dua, Bali, Indonesia, and on-line with the help
of Underline, during November 1-4, 2023. For those attending off-line, I hope that you will find time to
explore the exciting local area, which remains quite warm and pleasant at this time of the year. For
those attending on-line, I hope that the attendance is not much hampered, sometimes even enhanced
and comfortable, though the on-line arrangements have been made without ample time. 

Message from Jong C. Park, Korea Advanced Institute of Science and Technology (KAIST) 
IJCNLP-AACL 2023 General Chair 
###### [Conference Handbook](https://drive.google.com/file/d/1S2JSXa105xJxIkNBfhYmGH20Om6wARBx/view?usp=sharing)

In order to acces this event page, you need to register [here](http://www.ijcnlp-aacl2023.org/)

IJCNLP-AACL 2023

The 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics

As children acquire the knowledge of their language's morphology, they invariably discover the productive processes that can generalize to new words. Morphological learning is made challenging by the fact that even fully productive rules have exceptions, as in the well-known case of English past tense verbs, which features the -ed rule against the irregular verbs. The Tolerance Principle is a recent proposal that provides a precise threshold of exceptions that a productive rule can withstand. Its empirical application so far, however, requires the researcher to fully specify rules defined over a set of words. We propose a greedy search model that automatically hypothesizes rules and evaluates their productivity over a vocabulary. When the search for broader productivity fails, the model recursively subdivides the vocabulary and continues the search for productivity over narrower rules. Trained on psychologically realistic data from child-directed input, our model displays developmental patterns observed in child morphology acquisition, including the notoriously complex case of German noun pluralization. It also produces responses to nonce words that, despite receiving only a fraction of the training data, are more similar to those of human subjects than current neural network models' responses are.

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! 

**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.

Bayesian experimental design (BED) is a methodology to identify designs that are expected to yield informative data. Recent work in cognitive science considered BED for cognitive models with tractable and known likelihood functions. However, as cognitive models have become more complex and richer, their likelihood functions are often intractable. In this work, we leverage recent advances in BED for intractable models and demonstrate their application on a set of multi-armed bandit tasks. We further propose a generalized latent state model that unifies two previously proposed models. Our experiments show that data gathered using optimal designs results in improved model discrimination and parameter estimation, as compared to naive designs. Furthermore, we find that increasing the number of bandit arms increases the expected information gain in experiments.

Bayesian Experimental Design for Intractable Models of Cognition

The meaning of complex expressions (“two apples”) is computed by accessing and combining the concepts linked to their constituent words (“two”, “apples”). Across three eye-tracking experiments (N = 60), we demonstrate that preverbal infants can perform such computations and successfully derive the meaning of novel quantified noun phrases. Experiment 1 established that 12-month-olds can learn two distinct novel labels (pseudowords) denoting a singleton or a pair. Experiments 2-3 indicated that infants combine the meanings of the newly learnt quantity labels with those of familiar kind labels. When presented with four potential referents (e.g., 1duck, 2ducks, 1ball, 2balls) and asked to look at one ball, infants oriented to the target satisfying the meaning of both labels (1ball) over the distractors satisfying the meaning of the labels separately (2balls, 1duck). Conceptual combination skills that enable complex thought seem to be operational in infancy, and can be triggered by linguistic stimuli.

Unfants combine kind and quantity concepts

Immigration policy has been one of the top concerns of American voters over the last decade and has attracted some of the most heated rhetoric in politics and news media across the world. Much like other political language, talk about immigration is suffused with metaphor. To what extent does the language about immigration, and specifically the metaphors used, influence people’s views of the issues? How powerful are these metaphors? In our studies, we exposed participants to one of four versions of a passage about an increase in immigrants in one town. The four versions of the passage included all identical facts and figures and differed in only a single word at the beginning of the passage, describing the increase in immigrant labor as either an “increase,” a “boost,” an “invasion,” or a “flood.” Although the passages differed only in this one word, participants’ attitudes towards this increase and their predictions about its effects on the economy differed significantly depending on the metaphor. Of course, opinions on immigration differ across political affiliations. Remarkably, the single word metaphor was strong enough to mitigate much of the difference in opinion on immigration between Democrats and Republicans in our sample. Further analyses suggested that the results are not due simply to positive or negative lexical associations to the metaphorical words, and also that metaphors can act covertly in organizing people’s beliefs.

Linguistic Metaphors Shape Attitudes towards Immigration

We test for infants’ proposed altercentric learning bias (Southgate, 2020). By tracking others’ attention, infants build their own models of the environment based on the well developed models of the adults. If true, an event will be better remembered if witnessed together with someone else than if attended alone.
Eight-month-old-infants (6 conditions, n = 32/group) saw an object being hidden first in one, then in another location. At the end of each trial, one of the locations was revealed, always empty. Participants correctly remembered the location and looked longer if it was empty (conditions 1-2). When an agent attends the first hiding but not the second, infants misremember the object in the first location (condition 3), providing initial evidence for the altercentric bias. Contrary to our predictions, when the agent only attends to the last hiding location, or both, infants have no expectation of the object’s whereabouts (conditions 4-6).

Testing the Altercentrism Hypothesis in Young Infants

Our ability to perceive our own and other people’s bodies is critical to the success of social interactions. Research has shown that adults have a distorted perception of their own body and those of other adults. However, these studies ask perceivers to estimate for adults that have similar bodily make-up. This study explored the developmental progression in how children perceive their own body (5-12-yrs-olds) (Exp1) and whether children have similar distortions as adults when estimating the dimensions of adult bodies both unknown (Exp2) and familiar to them (Exp3). Overall, children showed similar distortions to those found in adult’s estimations for own body perception (i.e., limbs with a smaller density of sensory receptors showed a larger error than those with a higher density) and perception of adult’s bodies showed less distortion when perceiver and model were of the same gender, but not when the adult was familiar to the child.

Development of Self and Other’s Body Perception; Effects of Familiarity and Gender on How Children Perceive Adults.

Analyses of political discourse typically focus on the semantic content of politicians’ statements. The approach treats the meaning of a speaker’s words as independent from the speaker’s identity itself; however, there are reasons to believe that one might influence the other. Features of a speaker’s identity influence others’ judgements of their character (e.g., Kinzler & DeJesus, 2013), and thus speaker identity could influence listeners’ assessment of the semantics and validity of the statements themselves. Here, we collect U.S. participants’ judgements of the political orientation of different statements, from liberal to conservative, heard with one of three accents, a generic U.S. accent, a Southern U.S. accent or an Australian accent. In comparison to identical statements conveyed in the generic U.S. accent, participants tended to perceive the U.S. Southern accented statements as more conservative and the Australian accented statements as more liberal.

Judgement of political statements are influenced by speaker identity

Humans navigate daily decision-making by flexibly choosing appropriate approximations of what ought to be done. Which mental algorithms do people use, and when? We use behavioural experiments and modelling to investigate three computational principles known to influence decision making: noisy utility maximization, discounting, and the probability warping principle of Prospect Theory.
While these principles have been shown to separately influence human behaviour in simple laboratory tasks, such as bandits and gambles, we evaluate their combined use in the context of a naturalistic spatial search that required sequential decision-making. We found that while aggregate human behaviour can be reasonably well explained by an optimal planner with noisy utility maximization, individual-level behaviour exhibits consistent irregularities, that deviate from expected utility theory.
We show that model-based prediction of individual-level behaviours in our experiment is significantly improved by combining the three computational principles, and benefits particularly strongly from probability warping. Furthermore, our results suggest that probability warping may be a common factor of human decision making, that generalizes beyond the gambles explored in Prospect Theory, to natural human behaviours such as spatial search and navigation.

Unpacking the computations of human spatial search under uncertainty: noisy utility maximization, discounting, and probability warping

It is often assumed that language is strongly compositional, i.e., that the meaning of complex expressions is uniquely determined by the meanings of their constituents and their mode of composition (Fodor, 1987). Compositionality naturally connects to a broadly modular architecture of the language faculty, according to which our capacity for assigning meaning relies exclusively on lexical and syntactic knowledge (Baggio et al., 2015). Here, we discuss several arguments against strong compositionality. One such argument focuses on novel experimental data on the interpretation of privative adjectives (e.g., ‘fake’) (Partee, 2007). These data show that the interpretation of these adjectives is inexorably connected to the conceptual structure of the modified noun. We argue that lexical and syntactic information serve as important cues for, but do not uniquely determine, the process of meaning assignment (Martin, 2016). We discuss consequences for semantic theorising and the cognitive architecture of the language faculty.

Compositionality, modularity, and the architecture of the language faculty

The ability to plan under a variety of constraints, environments, and uncertainties is one of the greatest puzzles of human intelligence. Planning in human-like domains remains intractable for machine learning algorithms, where the greatest challenges is designing appropriate representations and state-evaluation heuristics. How do humans navigate their natural world so efficiently? Which computations drive them? Existing studies have investigated planning in simple tasks, such as gambles and multi-armed bandits, with little resemblance to natural tasks. We analyze human behavior in a search-and-rescue mission in a large 3D environment, designed to simulate a real-world context. We propose a hierarchical planning framework, which jointly solves an orienteering problem on the high level and a set of local Partially Observable Markov Decision Processes on the low level. Using this framework, we evaluate alternative computational models of human planning that captures core mental representations of human spatial planning.

Modeling human planning in a life-like search-and-rescue mission

Virtual laboratories that enable novice scientists to construct, evaluate and revise models of complex systems heavily involve parameter estimation tasks. We seek to understand novice strategies for parameter estimation in model exploration to design better cognitive supports for them. We conducted a study of 50 college students for a parameter estimation task in exploring an ecological model. We identified three types of behavioral patterns and their underlying cognitive strategies. Specifically, the students used systematic search, problem decomposition and reduction, and global search followed by local search as their cognitive strategies.

Downloads

Next from CogSci 2021

Bayesian Experimental Design for Intractable Models of Cognition

Similar lecture

Linguistic Productivity: the Case of Determiners in English