Search for the Missing Dimensions: Building a Feature-Space Representation for a Natural-Science Category Domain

Nosofsky, Robert M.; Sanders, Craig; Meagher, Brian J.; Douglas, Bruce J.

doi:10.1007/s42113-019-00033-2

Cited by 13 publications

(12 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Being able to characterize mental representations of objects with a low-dimensional embedding is surprising, given their high degree of perceptual variability and our broad semantic knowledge of them 32 . Indeed, popular semantic feature production norms 15 , 16 have revealed thousands of binary features that participants name when asked about their explicit knowledge of objects.…”

Section: Discussionmentioning

confidence: 99%

Revealing the multidimensional mental representations of natural objects underlying human similarity judgements

et al. 2020

View full text Add to dashboard Cite

Objects can be characterized according to a vast number of possible criteria (e.g. animacy, shape, color, function), but some dimensions are more useful than others for making sense of the objects around us. To identify these “core dimensions” of object representations, we developed a data-driven computational model of similarity judgments for real-world images of 1,854 objects. The model captured most explainable variance in similarity judgments and produced 49 highly reproducible and meaningful object dimensions that reflect various conceptual and perceptual properties of those objects. These dimensions predicted external categorization behavior and reflected typicality judgments of those categories. Further, humans can accurately rate objects along these dimensions, highlighting their interpretability and opening up a way to generate similarity estimates from object dimensions alone. Collectively, these results demonstrate that human similarity judgments can be captured by a fairly low-dimensional, interpretable embedding that generalizes to external behavior.

show abstract

Section: Discussionmentioning

confidence: 99%

Revealing the multidimensional mental representations of natural objects underlying human similarity judgements

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Being able to characterize mental representations of objects with a low-dimensional embedding is surprising, given their high degree of perceptual variability and our broad semantic knowledge of them 32 . Indeed, popular semantic feature production norms 15,16 have revealed thousands of binary features that participants name when asked about their explicit knowledge of objects.…”

Section: Discussionmentioning

confidence: 99%

Revealing the multidimensional mental representations of natural objects underlying human similarity judgments

Hebart¹,

Zheng²,

Pereira³

et al. 2020

Preprint

View full text Add to dashboard Cite

Objects can be characterized according to a vast number of possible criteria (e.g. animacy, shape, color, function), but some dimensions are more useful than others for making sense of the objects around us. To identify these "core dimensions" of object representations, we developed a data-driven computational model of similarity judgments for real-world images of 1,854 objects. The model captured most explainable variance in similarity judgments and produced 49 highly reproducible and meaningful object dimensions that reflect various conceptual and perceptual properties of those objects. These dimensions predicted external categorization behavior and reflected typicality judgments of those categories. Further, humans can accurately rate objects along these dimensions, highlighting their interpretability and opening up a way to generate similarity estimates from object dimensions alone. Collectively, these results demonstrate that human similarity judgments can be captured by a fairly lowdimensional, interpretable embedding that generalizes to external behavior.

show abstract

“…The learning of real-world categories presents the observer with a number of fundamental challenges. First, the to-be-learned objects in such domains are generally composed of multiple complex dimensions, many of which may be difficult to describe or discern (e.g., Nosofsky, Sanders, Meagher, et al, 2018; Nosofsky et al, 2020). Learning which dimensions are most relevant to allowing correct classifications, and learning how to integrate information across the multiple dimensions, is likely even harder in complex, naturalistic domains than in artificially constructed ones.…”

Section: Feature Highlightingmentioning

confidence: 99%

Effects of feature highlighting and causal explanations on category learning in a natural-science domain.

Meagher¹,

McDaniel²,

Nosofsky³

2022

Journal of Experimental Psychology: Applied

Self Cite

View full text Add to dashboard Cite

Teaching natural-science categories is highly challenging because the objects in such categories are composed of numerous complex dimensions that need to be perceived, evaluated, and integrated. Furthermore, the boundaries separating such categories are often fuzzy. A technique that has been proposed and investigated for enhancing the teaching of natural-science categories is feature highlighting, in which diagnostic features for identifying category members are explicitly described and illustrated. Using rock classification in geology as an example target domain, the present study further investigated the potential benefits of feature highlighting and also of providing causal explanations for the highlighted features. The authors found that feature highlighting did not always lead to improved generalization to novel members of the taught categories. However, robust beneficial effects were seen when the categories were relatively confusable ones and the stated diagnostic features were highly valid for distinguishing among the categories. Finally, at least under the present conditions, supplementing the highlighted features with causal explanations of the reasons for their occurrence did not further enhance the participants' rockclassification learning and generalization. Although the teaching of causal explanations is fundamental to science education, clear evidence that causal explanations enhance classification-learning per se in this domain remains to be demonstrated. Public Significance StatementThe present study suggests that visual highlighting of diagnostic features can enhance novice participants' ability to learn fundamental rock classifications, which is a core goal in introductory undergraduate geoscience education. Although providing causal explanations for the presence of the features is also central to geoscience education, it does not further enhance the novices' ability to learn the classifications.

show abstract

Search for the Missing Dimensions: Building a Feature-Space Representation for a Natural-Science Category Domain

Cited by 13 publications

References 48 publications

Revealing the multidimensional mental representations of natural objects underlying human similarity judgements

Revealing the multidimensional mental representations of natural objects underlying human similarity judgements

Revealing the multidimensional mental representations of natural objects underlying human similarity judgments

Effects of feature highlighting and causal explanations on category learning in a natural-science domain.

Contact Info

Product

Resources

About