United States

Conversational partners expect each other to communicate rationally and cooperatively and to contribute relevant and informative utterances. Often however, speakers produce trivial utterances which may violate our expectations of informativity which can prompt listeners to infer about a speaker’s goals for producing such an utterance. The present work investigates how and when listeners derive informativity-based inferences and demonstrates that speaker knowledge plays an important role in the computation of inferences. Furthermore the data shows that these inferences do not always arise automatically and their computation is costly.

CogSci 2023

Availability and Timing of Informativity Inferences

The 45th Annual Meeting of the Cognitive Science Society will take place in Sydney, Australia as a modified hybrid conference. Presentations from the main-session room will be live-streamed every day for all events. This will include all invited talks and award sessions. Recordings of all CogSci 2023 sessions will be made available for on-demand viewing after the conference.<br>
The in-person component will be held at the International Convention Centre Sydney at Darling Harbour. The Conference will take place Wednesday July 26th – Saturday July 29th, 2023 (Wednesday, July 26th is the Workshops Day).

Join us!

<br>[**The Program at a Glance can be found here**](https://cognitivesciencesociety.org/wp-content/uploads/2023/07/COgSci-PAAG-V2.pdf)
<br>[**The Detailed Program can be found here**](https://online.fliphtml5.com/ebtyf/mswz/)
<br><br>

#### **CogSci 2023 Program Book**
<iframe style="width:700px;height:400px" src="https://online.fliphtml5.com/ebtyf/tzgk/"  seamless="seamless" scrolling="no" frameborder="0" allowtransparency="true" allowfullscreen="true" ></iframe>
<br>

The 45th Annual Meeting of the Cognitive Science Society presents the latest research across cognitive science and highlights the theme of Cognition in Context.

poster

Intransitive verbs in human language can be subcategorized into at least two linguistic categories, unaccusative and unergative verbs. The definitive factor of this categorization is argued to be the meaning of the verb, while the Constructional approach suggests the sentence a verb appears in plays a role in categorization. In this study, we develop a word2vec model which leverages the context in which a verb appears in to represent the semantics of verbs and this basis of distributional semantics serves as a method to examine the Constructional approach. Our results suggest categories of Unaccusativity of new verbs can be formed on the basis of the contexts in which verbs appear, and the frequency of verbs in certain contexts can be a significant factor. We administered a child acquisition experiment to further demonstrate that, given the meaning of a new verb, sentences in which new verbs appear do significantly affect the categorization of Unaccusativity of new verbs. We conclude that, in addition to the meaning of intransitive verbs, the contextual environment is indeed a crucial factor in Mandarin Unaccusativity.

The categorization of intransitive verbs in Mandarin: Evidence from word2vec modeling and behavioral experiment

A recent proposal argued a structure like X1’s-X2 is recursive if X1 and X2 are substitutable as the head; the head requirement avoids wrong generalizations like ‘*dogs chase cats chase rats’ since NP is not the head of NP1-V-NP2 although the NPs can be substitutable. To test the role of headedness in the distributional learning of recursion, we exposed 50 adults to one of two artificial languages: both allowed A1-B-A2 strings, with A1 and A2 substitutable; but the head was A2 in one language and B in the other. Non-head elements were optional, so the two languages allowed different linear strings. In the rating test, at predicted, participants from both conditions learned the head; and while there was no difference for ABA strings between conditions, suggesting they all learned linear substitutability, participants exposed to head substitutability (A-head) were more willing to allow recursion (ABABA) with both attested and unattested words.

Distributional learning of recursive structures: The role of the structural representation

The development of large language models (LLMs) enables the investigation of cognitive phenomena at an unprecedented scale. We applied LLM-derived measures on large narrative datasets to characterize the structure and dynamics of memory retrieval. Specifically, we found that autobiographical narratives flow less linearly from sentence to sentence than biographical narratives. Furthermore, the treatment of topics within biographies tends to be more coherent and are also written at a higher level of complexity than autobiographies. In summary, the narrative flow differences suggest that when authors rely on their own memory, retrieval proceeds in a less organized manner likely reflecting spontaneous cueing of associated memories. Our results demonstrate the utility of applying LLMs to narrative text to study cognitive phenomena.

Narrative flow quantification of autobiographies reveals signatures of memory retrieval

When choosing how to describe what happened, we have a number of causal verbs at our disposal. In this paper, we develop a model-theoretic formal semantics for nine causal verbs that span the categories of CAUSE, ENABLE, and PREVENT. Our model uses structural causal models (SCMs) to represent participants' mental construction of a scene when assessing the correctness of causal expressions within the presented context. Furthermore, SCMs enable us to model events relating both the physical world as well as agents' mental states. In experimental evaluations, we find that the proposed semantics exhibits a closer alignment with human evaluations in comparison to prior accounts of the verb families.

A Semantics for Causing, Enabling, and Preventing Verbs Grounded in Structural Causal Models

We present a computational (LSTM) model that learns to produce English (3sg and -bare) verb inflection when trained on English child-directed speech (CDS). The model is trained on input containing morphemized verbs and learns to predict the next word given a preceding sequence of words. The model produces the type of error (-bare for -3s) shown by English-learning children while avoiding errors that children do not often make (-3s for -bare). The model also shows the same type of sensitivity to input statistics that has been reported in English-learning children. Finally, we manipulated the length of the sequences the model is trained on, and show that this results in the delayed acquisition of -3sg forms that is characteristic of English-learning children with Developmental Language Disorder (DLD). Taken together these results suggest that input-driven learning is a major determinant of both typical and impaired acquisition of English verb inflection.

Simulating children’s verb inflection errors in English using an LSTM language model

A central challenge of cross-situational word learning is retaining word-referent mappings across exposures. We evaluate Memory-Bound Pursuit (MBP), a hypothesis-testing model of cross-situational word-learning which aims to account for learners’ memory constraints via a single parameter targeting the number of words that can be learned concurrently. Here, we show that by varying this parameter with age, MBP can capture both children’s and adults’ cross-situational word-learning success under varying levels of ambiguity. We also present new experimental findings supporting novel predictions made by MBP about the retention of word-referent mappings across intervening exposures. These findings suggest that MBP provides a strong baseline model of cross-situational word learning, capturing both developmental trends and experimental evidence of memory limitations for word learning.

Memory as a computational constraint in cross-situational word learning

The human lexicon expresses a wide array of objects with a limited set of named categories. Previous work has suggested that semantic categories are structured compactly to enable informative communication. Informativeness is typically quantified with respect to the entirety of a semantic domain and not at the level of individual words. We develop a measure of name informativeness in an information-theoretic framework grounded in visual object representations derived from natural images. We use computer vision models to characterize informativeness of individual names with respect to large-scale data in a naturalistic setting. We show that our informativeness measure predicts degrees of specificity in lexical categories more precisely than alternative measures based on entropy and frequency. We also show that name informativeness jointly captures within-category similarity and distinctiveness across categories. Our analyses suggest how the variability of names in the broad lexicon may be understood through the lens of information theory.

Quantifying informativeness of names in visual space

Computational models of morphology acquisition have played a central role in debates over the nature of morphological representations. Recent advances in natural language processing have renewed these debates with the advent of powerful artificial neural network architectures of morphological inflection. However, the actual suitability of these advanced neural models as models of human morphology acquisition remains uncertain. We argue that much of this confusion stems from inconsistent methods of training and evaluation. In this work, we demonstrate that more careful data set creation and an evaluation combining quantitative analysis and comparison with human development will put the evaluation of neural models on firmer ground.

Re-Evaluating the Evaluation of Neural Morphological Inflection Models

This study examines the reliability and validity of using two types of crowd-sourced judgments to collect lexical diversity scores. Scaled and pairwise comparison approaches were used to collect data from non-expert Amazon Mechanical Turk workers. The reliability of the lexical diversity ratings for the crowd-sourced raters was assessed along with those from trained raters using a variety of reliability statistics. The validity of the ratings was examined by 1) comparing crowd-sourced and trained ratings, 2) comparing crowd-sourced and trained ratings to ratings of language proficiency, and 3) by using an objective measure of lexical diversity to predict the crowd-sourced and trained ratings. The results indicate that scaled crowd-sourced ratings showed strong reliability in terms of text and rater strata and showed fewer misfitted texts than the trained raters. The scaled crowd-sourced ratings were also strongly predicted by lexical diversity features derived from the texts themselves.

Crowd-Sourcing Human Ratings of Linguistic Production

Words that regularly fill the same sentential slots are said to be paradigmatically related. Paradigmatic relations may be retained through a direct association or a latent representation at encoding, or by reinstating context during retrieval. We paired proper names by embedding them into two instances of the same sentence frame, each in a separate list, yielding blocks of two study-cloze sessions. The pairing between proper names was fixed across twelve blocks. In the static condition, the same sentence frames were used across blocks, while in the dynamic condition sentence frames changed for each block. Interference should accrue in both conditions if paradigmatic relations are based on a direct association or overlap in a latent representation, however, if paradigmatic relations are mediated by retrieved context then changing the sentence frame should release interference. Our results are consistent with a context-mediation account of paradigmatic relations.

Premium content

Downloads

Next from CogSci 2023

The categorization of intransitive verbs in Mandarin: Evidence from word2vec modeling and behavioral experiment

Similar lecture

What inferences do people actually make upon encountering informationally redundant utterances? An individual differences study