Canada

Object interactions – collisions, scraping and rolling – create many of the sounds that we hear in the world around us. These sounds are generated via lawful physical dynamics. Anecdotally, humans possess some intuitive knowledge of the physical generative processes underlying sound production, but little is known about the extent and nature of this knowledge. In this work, we study the human ability to make inferences about physical properties like mass, size, velocity and surface roughness from contact sounds, and test whether it is consistent with the use of a generative model of the underlying physics. Additionally, we propose a computational model that uses analysis-by-synthesis to infer object properties and their physical configurations using contact sounds as input, and compare it to human behavior.

CogSci 2022

Inferring object interactions from scraping and rolling sounds

poster

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. 





Detecting analogy is an important high-level cognitive skill that is involved in many aspects of human reasoning. While Structure Mapping Theory (Gentner, 1983) is a well-recognized high-level theory of analogy, it lacks a neural process implementation that links to perception and attention. Avoiding algorithmic computation on ungrounded symbols, we present a dynamic neural architecture built from interacting neural populations that establishes analogy between objects in two visually presented scenes. Consistent with SMT, it accounts for how humans find such analogies.


A Perceptually Grounded Neural Dynamic Architecture Establishes Analogy Between Visual Object Pairs

Analogy problems involving multiple ordered relations of the same type create mapping ambiguity, requiring some mechanism for relational integration to achieve mapping accuracy. We address the question of whether the integration of ordered relations depends on their logical form alone, or on semantic representations that differ across relation types. We developed a triplet mapping task that provides a basic paradigm to investigate analogical reasoning with simple relational structures. Experimental results showed that mapping performance differed across orderings based on category, linear order, and causal relations, providing evidence that each transitive relation has its own semantic representation. Hence, human analogical mapping of ordered relations does not depend solely on their formal property of transitivity. Instead, human ability to solve mapping problems by integrating relations relies on the semantics of relation representations. We also compared human performance to the performance of several vector-based computational models of analogy. These models performed above chance but fell short of human performance for some relations, highlighting the need for further model development.

Impact of Semantic Representations on Analogical Mapping with Transitive Relations

In this paper we present a novel method for a naive agent to detect novel objects it encounters in an interaction. We train a reinforcement learning policy on a stacking task given a known object type, and then observe the results of the agent attempting to stack various other objects based on the same trained policy. By extracting embedding vectors from a convolutional neural net trained over the results of the aforementioned stacking “play,” we can determine the similarity of a given object to known object types, and determine if the given object is likely dissimilar enough to the known types to be considered a novel class of object. We present the results of this method on two datasets gathered using two different policies and demonstrate what information the agent needs to extract from its environ- ment to make these novelty judgments.

Exploiting Embodied Simulation to Detect Novel Object Classes Through Interaction

Mental simulation is a powerful cognitive capacity that underlies people's ability to draw inferences about what happened in the past from the present. Recent work suggests that eye-tracking can be used as a window through which one can study the process of mental simulation in intuitive physics tasks. In our experiment, participants have to figure out in which of three holes a ball was dropped in a virtual Plinko box. We develop a computational model of human intuitive physical reasoning in Plinko that runs repeated simulations in a noisy physics simulator in order to infer in which hole the ball was dropped. We evaluate our model's behavior against multiple human data signals: trial judgments, response times, and eye-movement data. We find that a model that sequentially samples simulations while balancing uncertainty and reward best explains the patterns of participant behavior we observe in these three signals.

Looking into the past: Eye-tracking mental simulation in physical inference

Previous research using motor dual tasks and TMS has suggested that interference with cortical motor information impacts word reading. This research offers support for theories of grounded cognition by suggesting a functional role of sensorimotor information in conceptual representation. However, motor dual tasks and TMS are limited to partial interference of body representations. To address this, the current electroencephalography (EEG) study induces body illusions to interfere with the broader representation of body information during verb reading. During this task, participants categorized words related to hands or feet while we measured EEG. With spatiotemporal representational similarity analysis (RSA), we demonstrate that sensorimotor information is decodable in normal conditions and is delayed during illusion conditions. Our results suggest that during illusion conditions, sensorimotor information is disrupted by body illusions. This supports theories of grounded cognition for word reading.

The role of body representations in higher order cognition

Recent work suggests that learning perceptual classifications can
be enhanced by combining single item classifications with adaptive
comparisons triggered by each learner’s confusions. Here, we
asked whether learning might work equally well using all
comparison trials. In a face identification paradigm, we tested
single item classifications, paired comparisons, and dual instance
classifications that resembled comparisons but required two
identification responses. In initial results, the comparisons
condition showed evidence of greater efficiency (learning gain
divided by trials or time invested). We suspected that this effect
may have been driven by easier attainment of mastery criteria in
the comparisons condition, and a negatively accelerated learning
curve. To test this idea, we fit learning curves and found data
consistent with the same underlying learning rate in all conditions.
These results suggest that paired comparison trials may be as
effective in driving learning of multiple perceptual classifications
as more demanding single item classifications.

Comparisons in Adaptive Perceptual Category Learning

To explain how perception processes are performed, understanding how continuous sensory streams are temporally segmented into discrete units is central. This is particularly the case in speech perception where temporal segmentation is key for identifying linguistic units contained between consecutive events in time. We propose an original probabilistic construct, that we call "Bayesian gates", to segment temporally continuous streams of sensory stimuli into sequences of decoders. We first define Bayesian gates mathematically and describe their properties. We then illustrate their behavior in the context of a model of word recognition in speech perception. We show that, based on an event detection module, they sequentially parse the acoustic stimulus, so that each syllable decoder only processes a segment of the sensory signal.

Bayesian gates: a probabilistic modeling tool for temporal segmentation of sensory streams into sequences of perceptual accumulators

Categorization is a fundamental cognitive strategy employed to ease information processing and to aid memory formation. Past research on how humans categorize objects has used images of objects as experimental stimuli. Concurrently, studies in the past 10 years have found differences in the processing of images as compared to real-world objects. One proposed explanation is that these results are due to differences in the affordances of images versus objects. Using a similarity judgement paradigm, we explored the effect of affordances in a categorization task including words (object names), images, and objects. Consistent with previous research, we found significant differences in how participants made similarity judgements of images and objects. Moreover, we found that similarity judgments using object names were much more similar to the judgments of pictures than of objects. An exploratory cluster analysis opens the possibility of framing such differences as affordance driven. These results suggest a need for more ecologically valid categorization tasks, more conservative inferences when using images as stimuli in these tasks, and the need for further exploring the role of affordances in categorization.

The Treachery of Images: Objects, Pictures, Words and the Role of Affordances in Similarity Judgements

Biodiversity is a complex concept entailing scientific and political aspects. The usage of analogies, especially metaphors, that have positive influences on the understanding of complex concepts, on attitudes and behaviors, seems an interesting strategy to achieve this goal. Based on biodiversity analogies elaborated by 259 participants, a first study aims to identify two important protective approaches: preservationism that encourages humankind to limit their intervention on nature and conservationism that allows humankind to exploit nature with parsimony. We analyzed their analogies and results highlight three major groups: a scientific, a conservationist and a preservationist dimension. A second study investigates the effects of metaphorical framing on environmental attitudes and behaviors. 277 University students read a short text framing biodiversity with a preservationist or conservationist metaphor or without metaphor framing. A decision-making task and an environmental concern scale were completed. Results showed an effect of the conservationist metaphor on the decision-making task.

Framing biodiversity Conceptions

The way we think about social issues may be significantly affected by the language used to speak about them. Thibodeau & Boroditsky (2011) found that significant differences in reasoning were noted after exposure to even a single metaphor. In four experiments, we suggest similar findings through testing the effects of metaphorical framings of the crisis of homelessness in Los Angeles County where either 1) HOMELESSNESS IS AN ILLNESS, 2) HOMELESSNESS IS A HOSTILE ENCOUNTER or 3) a non-metaphorical framing. Participants read a brief passage or were given visual stimuli (a novel approach) designed to evoke one of the frames or the control and then responded to a series of questions on how to best address the issue. Current analysis of the data in the ongoing survey suggests that use of conceptual metaphor leads participants to recruit frame-semantic knowledge of the source domain to draw inferences that lead to differences in reasoning.

Premium content

Inferring object interactions from scraping and rolling sounds

Downloads

Next from CogSci 2022

A Perceptually Grounded Neural Dynamic Architecture Establishes Analogy Between Visual Object Pairs

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES