Canada

Children&#39;s early language skill has been linked to later educational outcomes, making it important to measure early language accurately. Parent-reported instruments such as the Communicative Development Inventories (CDIs) have been shown to provide reliable and valid measures of children&#39;s aggregate early language skill. However, CDIs contain hundreds of vocabulary items, some of which may not be heard (and thus learned) equally often by children of varying backgrounds. This study used a database of American English CDIs to identify words demonstrating strong bias for particular demographic groups of children, on dimensions of sex (male vs. female), race (white vs. non-white), and maternal education (high vs. low). For each dimension, we identified dozens of strongly biased items, and showed that eliminating even some of these items can reduce the magnitude of differences between groups. Additionally, we investigated how well the relative frequency of words spoken to young girls vs. boys predicted sex-based word learning bias, and discuss possible sources of demographic differences in early word learning.

CogSci 2022

Estimating demographic bias on tests of children’s early vocabulary

Children's early language skill has been linked to later educational outcomes, making it important to measure early language accurately. Parent-reported instruments such as the Communicative Development Inventories (CDIs) have been shown to provide reliable and valid measures of children's aggregate early language skill. However, CDIs contain hundreds of vocabulary items, some of which may not be heard (and thus learned) equally often by children of varying backgrounds. This study used a database of American English CDIs to identify words demonstrating strong bias for particular demographic groups of children, on dimensions of sex (male vs. female), race (white vs. non-white), and maternal education (high vs. low). For each dimension, we identified dozens of strongly biased items, and showed that eliminating even some of these items can reduce the magnitude of differences between groups. Additionally, we investigated how well the relative frequency of words spoken to young girls vs. boys predicted sex-based word learning bias, and discuss possible sources of demographic differences in early word learning.

technical paper

Join us for CogSci2022! Our 44th Annual Conference will be centered around the theme of diversity, with an accompanying range of keynotes and invited speakers who themselves highlight the diversity of our field. Our three symposia are Cognition across Diverse Populations, Cognition within Diverse Societies, and Global Challenges for Cognitive Science. The conference is being offered in a fully hybrid environment for onsite and online attendees.
 
 [**The Program at a Glance can be found here**](https://cognitivesciencesociety.org/wp-content/uploads/2022/07/Cogsci-2022-PAAG-Detailed-July-12final-1.pdf)
 [**The Detailed Program can be found here**](https://cognitivesciencesociety.org/wp-content/uploads/2022/08/DeArmond-CogSci2022-program-v113-final.pdf)

This conference requires a registration to attend. Please use the link below to secure your ticket. The website will be available to all registered attendees on July 25th. 

[Register to Attend](https://cognitivesciencesociety.org/registration/)

Registration Is Required

The 44th Annual Meeting of the Cognitive Science Society highlighted research on the theme Cognitive Diversity in addition to the full breadth and diversity of research topics offered by the society’s membership. 





User-generated ratings — often elicited and presented as “star ratings” — have become a ubiquitous feature of the online consumer experience. While most research agrees that these user-generated ratings influence individual consumer decisions and overall consumer demand, there is less consensus as to whether user- generated ratings help consumers make better, welfare-enhancing decisions. In this manuscript, we expound on an intrinsic problem with the use of user-generated ratings in product choice decisions. Specifically, product ratings are typically given in an isolated (non-comparative) context, but are typically used in a comparative context, where relative differences in ratings may not reflect relative differences in quality. We provide a simple empirical demonstration of how this structural misalignment can lead consumers to choose suboptimal products and, ultimately, yield reduced consumer welfare.

User-Generated Star Ratings Are Not Inherently Comparable: How Star Ratings Structure Leads to Poor Choices

It may seem pointless to compare two products with the exact same average rating and total number of reviews without other review information. Now imagine a scenario in which the distribution of star ratings is also available to decision makers in addition to these two attributes. Will the decision still be uncertain as it is before or the distributions of stars will engender a preference towards one of the products? To answer this question, the current study used variability of star ratings as an approximation of a product’s distribution. The behavioral studies showed that participants exhibited distinctive choice patterns when the distribution of ratings was provided even when the average rating and total number of reviews were the same between two products involved in a comparison. A utility-based cognitive model was therefore developed to identify the underlying mechanism as to why people chose the way they did.

How Do People Use Star Rating Distributions?

We investigate the way people judge how helpful a review is in informing their decision as to whether to make a purchase. In particular, we are interested in how the summary statistics an individual sees influences judgments of a review’s helpfulness. We find perceived helpfulness of a given review decreases as the star rating of that review gets further from the mean rating. Additionally, participants were more likely to search for reviews close to the mean. Both of these findings are consistent with confirmation bias. We explore, but do not find support for, alternative possible explanations.

The Influence of Mean Product Ratings on Review Judgments and Search

Choice blindness is a puzzling phenomenon, showing how people often fail to detect a mismatch between their intentions and outcomes and then confabulate reasons for choices they never made. Despite extensive replications, little is known about the cognitive mechanisms underpinning it. To tackle this question, we report an online experiment showing for the first time that people still illusorily self-attribute unchosen outcomes despite being explicitly instructed to detect false feedbacks. Comparing standard implicit choice blindness with our new explicit task, we show how prior beliefs about manipulations and stimuli discriminability influence detection. Building on these results, we sketch a preliminary Bayesian model of choice blindness suggesting that choice blindness occurs when beliefs in the reliability of feedback override monitoring related cues indicative of a mismatched outcome. Finally, we discuss practical advantages of explicit task to study choice blindness and implications for the sense of agency and self-knowledge.


Choice blindness persists despite explicit instructions to detect false feedbacks: preliminary evidence for a Bayesian model of choice blindness

Expectations play a critical role in human perception, cognition, and decision-making. There has been a recent surge in modelling such expectations and the resulting behaviour when they are violated. One prominent recent proposal is the ViolEx model. To move the model & the literature forward, we identified three areas of concern and addressed two in this study - Lack of formalization and implementation. Specifically, we aim to provide the first implementation of ViolEx using the Active Inference formalism by leveraging the rich overlap between them. As a result, we successfully simulated all expectation violation coping strategies of ViolEx and analyzed their potential mediators. Through this analysis, we identified novel predictors of immunization & assimilation and documented the computational mechanism of their mediation. We see this as the first step in developing a formal research framework to study expectation violations and hope to serve as a base for future ViolEx studies.

Formalization and Implementation of ViolEx: An Active Inference perspective

We test whether people flexibly shift their sampling strategy for learning a functional relationship, based on the strategy’s perceived effectiveness. While general-purpose heuristics such as gathering information evenly across the environment may often approximate optimal sampling policies when opportunities for learning are sparse, these strategies may systematically fail when much of the environment is uninformative. Across several different classes of arbitrary smooth functions, participants (N = 89) made sampling choices that were initially consistent with a simple heuristic, but shifted their sampling strategy when this heuristic failed to be informative. People were subsequently more accurate at approximating the true function for smoother functions that can be more easily predicted by sampling evenly, t(645.3) = 6.803, p < .001; nevertheless, while individuals vary considerably, aggregate judgments across all functions are very accurate, suggesting that people show inductive biases in active learning, but effectively learn when to adjust their policies.

Dynamic Strategy Selection in Active Function Learning

Sequential sampling models have provided accurate accounts of people’s choice, response time, and preference strength in value-based decision-making tasks. Conventionally, these models are developed as Markov-type processes (such as random walks or diffusion processes) following the Kolmogorov axioms. Quantum probability theory has been proposed as an alternative framework upon which to develop models of cognition, including quantum random walk models. When modeling people's behavior during decision-making tasks, previous work has demonstrated that both the Markov and quantum models have their respective strengths. Recently, the open system model, which is a hybrid version of the Markov and quantum models, has been shown to provide a more accurate account of preference strength compared to the Markov and quantum models in isolation. In this work, we extend the open system model to make predictions on pairwise choice and response time and compare it to the Markov and quantum random walk models.


Open System Model of Choice and Response Time

The direction of our attention can influence our performance on a variety of tasks. For example, reading from left to right relates to people associating small numbers on the left and large numbers on the right. In contrast, reading from right to left relates to people associating small numbers on the right and large numbers on the left. The current study tests if this type of “attentional momentum” can be induced by storytelling based on pictures and whether it affects college students’ reaction time on an arithmetic verification task with equations in a traditional (e.g., 2+2=4) or non-traditional (e.g., 4=2+2) direction. Our results show that students were faster at verifying simple traditional math problems after telling stories based on pictures arranged from left to right, but faster at verifying simple non-traditional math problems after telling stories based on pictures arranged from right to left.


Premium content

Estimating demographic bias on tests of children’s early vocabulary

Downloads

Next from CogSci 2022

User-Generated Star Ratings Are Not Inherently Comparable: How Star Ratings Structure Leads to Poor Choices

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES