Human Category Learning: Toward a Broader Explanatory Account

Kurtz, Kenneth J.

doi:10.1016/bs.plm.2015.03.001

Cited by 20 publications

(33 citation statements)

References 82 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Typical categorization studies focus on gross measures of performance (Kurtz, 2015), usually using strategy analyses only as a secondary measure, or manipulation check (e.g., . Typical categorization studies focus on gross measures of performance (Kurtz, 2015), usually using strategy analyses only as a secondary measure, or manipulation check (e.g., .…”

Section: Simulation 3: Smith Et Al (2014)mentioning

confidence: 99%

Due Process in Dual Process: Model‐Recovery Simulations of Decision‐Bound Strategy Analysis in Category Learning

2018

View full text Add to dashboard Cite

Behavioral evidence for the COVIS dual-process model of category learning has been widely reported in over a hundred publications (Ashby & Valentin, ). It is generally accepted that the validity of such evidence depends on the accurate identification of individual participants' categorization strategies, a task that usually falls to Decision Bound analysis (Maddox & Ashby, ). Here, we examine the accuracy of this analysis in a series of model-recovery simulations. In Simulation 1, over a third of simulated participants using an Explicit (conjunctive) strategy were misidentified as using a Procedural strategy. In Simulation 2, nearly all simulated participants using a Procedural strategy were misidentified as using an Explicit strategy. In Simulation 3, we re-examined a recently reported COVIS-supporting dissociation (Smith et al., ) and found that these misidentification errors permit an alternative, single-process, explanation of the results. Implications for due process in the future evaluation of dual-process theories, including recommendations for future practice, are discussed.

show abstract

Section: Simulation 3: Smith Et Al (2014)mentioning

confidence: 99%

Due Process in Dual Process: Model‐Recovery Simulations of Decision‐Bound Strategy Analysis in Category Learning

2018

View full text Add to dashboard Cite

show abstract

“…DIVA (Kurtz, 2007(Kurtz, , 2015 offers a theoretical alternative to the reference point framework by representing each category as a generative model (Ng & Jordan, 2002) of the statistical regularities among its members. The DIVA model (see Fig.…”

Section: Our Approach: Testing the Explanatory Power Of Categorizatiomentioning

confidence: 99%

“…The design of the behavioral studies follows from a priori predictions made by two formal models: Nosofsky's (1984Nosofsky's ( , 1986) Generalized Context Model (GCM), and the DIVergent Autoencoder model (DIVA; Kurtz, 2007Kurtz, , 2015. The GCM is the canonical exemplar-based model for predicting the overall ease of learning a classification problem and classification performance after learning.…”

Section: Our Approach: Testing the Explanatory Power Of Categorizatiomentioning

confidence: 99%

“…To estimate the frequency of Reduced category generalization in DIVA, we conducted a 'grid-search' to generate predictions for partial-XOR generalization over a wide range of settings for DIVA's four parameters: number of hidden nodes, learning rate, initial weight range, and a focusing parameter, β (Conaway and Kurtz, 2014;Kurtz, 2015). At each point in the search, DIVA was initialized 2000 times (random initial weights and random presentation sequence) and trained for 12 blocks (12 observations of each of the training exemplars).…”

Section: A Priori Model Simulationsmentioning

confidence: 99%

See 1 more Smart Citation

Similar to the category, but not the exemplars: A study of generalization

Conaway

Kurtz

2016

Psychon Bull Rev

Self Cite

View full text Add to dashboard Cite

Reference point approaches have dominated the study of categorization for decades by explaining classification learning in terms of similarity to stored exemplars or averages of exemplars. The most successful reference point models are firmly grounded in the associative learning tradition-treating categorization as a stimulus generalization process based on inverse exponential distance in psychological space augmented by a dimensional selective attention mechanism. We present experiments that pose a significant challenge to popular reference point accounts which explain categorization in terms of stimulus generalization from exemplars, prototypes, or adaptive clusters. DIVA, a similarity-based alternative to the reference point framework, provides a successful account of the human data. These findings suggest that a successful psychology of categorization may need to look beyond stimulus generalization and toward a view of category learning as the induction of a richer model of the data.

show abstract

“…The vast majority of category learning studies adopt a discriminative modeling approach (Ashby & Maddox, 1993; Kruschke, 1992; Nosofsky; 1986; Nosofsky & Palmeri, 1997), which is appropriate given the heavy reliance on the artificial classification learning paradigm in this literature. A generative approach, however, is more appropriate when modeling data that do not reflect explicit classification decisions (Kurtz, 2015; Levering & Kurtz, 2014; see also Chin-Parker & Ross, 2004), such as the visual search data in the present study. Our position is that generative models better capture the features of a category used to construct visual-working-memory representations of search targets, similar to the features one might call to mind when forming a mental image of a target category.…”

Section: Introductionmentioning

confidence: 99%

Searching for Category-Consistent Features

Yu¹,

Maxfield

Zelinsky

2016

Psychol Sci

View full text Add to dashboard Cite

A generative model of category representation is introduced that uses computer vision methods to extract category-consistent features (CCFs) directly from images of category exemplars. The model was trained on 4800 images of common objects, and CCFs were obtained for 68 categories spanning subordinate, basic, and superordinate levels in a category hierarchy. Participants searched for these same categories. Targets cued at the subordinate level were preferentially fixated, but fixated targets were verified faster following a basic-level cue. The subordinate-level advantage in guidance is explained by the number of target category CCFs, a measure of category specificity that decreases with movement up the category hierarchy. The basic-level advantage in verification is explained by multiplying CCF number by sibling distance, a measure of category distinctiveness. With this model the visual representations of real-world object categories, each learned from the vast numbers of image exemplars accumulated throughout our everyday experience, can finally be studied.

show abstract

Human Category Learning: Toward a Broader Explanatory Account

Cited by 20 publications

References 82 publications

Due Process in Dual Process: Model‐Recovery Simulations of Decision‐Bound Strategy Analysis in Category Learning

Due Process in Dual Process: Model‐Recovery Simulations of Decision‐Bound Strategy Analysis in Category Learning

Similar to the category, but not the exemplars: A study of generalization

Searching for Category-Consistent Features

Contact Info

Product

Resources

About