Limits...
Varying variation: the effects of within- versus across-feature differences on relational category learning.

Livins KA, Spivey MJ, Doumas LA - Front Psychol (2015)

Bottom Line: As a result, the way that they interact with feature variation is unclear.Both sets of results indicated that learning a relational category from a training set with a lower amount of variation is easier, but that learning from a training set with increased within-feature variation is significantly less challenging than learning from a set with increased across-feature variation.These results support the claim that, like feature-based category-learning, relational category-learning is sensitive to the type of feature variation in the training set.

View Article: PubMed Central - PubMed

Affiliation: Department of Cognitive Science, University of California, Merced, Merced, CA USA.

ABSTRACT
Learning of feature-based categories is known to interact with feature-variation in a variety of ways, depending on the type of variation (e.g., Markman and Maddox, 2003). However, relational categories are distinct from feature-based categories in that they determine membership based on structural similarities. As a result, the way that they interact with feature variation is unclear. This paper explores both experimental and computational data and argues that, despite its reliance on structural factors, relational category-learning should still be affected by the type of feature variation present during the learning process. It specifically suggests that within-feature and across-feature variation should produce different learning trajectories due to a difference in representational cost. The paper then uses the DORA model (Doumas et al., 2008) to discuss how this account might function in a cognitive system before presenting an experiment aimed at testing this account. The experiment was a relational category-learning task and was run on human participants and then simulated in DORA. Both sets of results indicated that learning a relational category from a training set with a lower amount of variation is easier, but that learning from a training set with increased within-feature variation is significantly less challenging than learning from a set with increased across-feature variation. These results support the claim that, like feature-based category-learning, relational category-learning is sensitive to the type of feature variation in the training set.

No MeSH data available.


Related in: MedlinePlus

An example of a possible training set and test phase exemplar. Imagine the participant was trained on A/C and B/D exemplars, where an “left-key” press was paired with A/C, and an “right-key” press was paired with B/D. If a participant were then shown an A/D exemplar during the test phase, then an “left-key” press would indicate that A/D was classified in the same way as an A/C exemplar, while a “right-key” press would indicate that it was classified in the same way as a B/D.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4321646&req=5

Figure 7: An example of a possible training set and test phase exemplar. Imagine the participant was trained on A/C and B/D exemplars, where an “left-key” press was paired with A/C, and an “right-key” press was paired with B/D. If a participant were then shown an A/D exemplar during the test phase, then an “left-key” press would indicate that A/D was classified in the same way as an A/C exemplar, while a “right-key” press would indicate that it was classified in the same way as a B/D.

Mentions: For example, suppose that a participant had been trained on A/C and B/D, where A/C had been associated with a left-key press, and B/D had been associated with a right-key press. For the generalization portion of the testing phase, novel A/D and B/C pairs could be used to determine which rule the participant had learned: if presented with an A/D pairing, then a left-key press would indicate that the participant was classifying the stimulus to be like the A/C pair. If A/C and A/D pairs are classified in the same way, then the participant must be attending to the above/below relation (since the left-key is the common relational value between them, representing cases where the occluder was above the occluded shape). Along the same lines, a right-key press would indicate that the participant was classifying by the “beside” rule (See Figure 7). While the study was not primarily concerned with which rule participants learned, the ability to determine which rule each participant learned was instrumental in determining generalization accuracy and ability (i.e., if they learned one rule, how accurately can they classify the novel exemplars by that rule?).


Varying variation: the effects of within- versus across-feature differences on relational category learning.

Livins KA, Spivey MJ, Doumas LA - Front Psychol (2015)

An example of a possible training set and test phase exemplar. Imagine the participant was trained on A/C and B/D exemplars, where an “left-key” press was paired with A/C, and an “right-key” press was paired with B/D. If a participant were then shown an A/D exemplar during the test phase, then an “left-key” press would indicate that A/D was classified in the same way as an A/C exemplar, while a “right-key” press would indicate that it was classified in the same way as a B/D.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4321646&req=5

Figure 7: An example of a possible training set and test phase exemplar. Imagine the participant was trained on A/C and B/D exemplars, where an “left-key” press was paired with A/C, and an “right-key” press was paired with B/D. If a participant were then shown an A/D exemplar during the test phase, then an “left-key” press would indicate that A/D was classified in the same way as an A/C exemplar, while a “right-key” press would indicate that it was classified in the same way as a B/D.
Mentions: For example, suppose that a participant had been trained on A/C and B/D, where A/C had been associated with a left-key press, and B/D had been associated with a right-key press. For the generalization portion of the testing phase, novel A/D and B/C pairs could be used to determine which rule the participant had learned: if presented with an A/D pairing, then a left-key press would indicate that the participant was classifying the stimulus to be like the A/C pair. If A/C and A/D pairs are classified in the same way, then the participant must be attending to the above/below relation (since the left-key is the common relational value between them, representing cases where the occluder was above the occluded shape). Along the same lines, a right-key press would indicate that the participant was classifying by the “beside” rule (See Figure 7). While the study was not primarily concerned with which rule participants learned, the ability to determine which rule each participant learned was instrumental in determining generalization accuracy and ability (i.e., if they learned one rule, how accurately can they classify the novel exemplars by that rule?).

Bottom Line: As a result, the way that they interact with feature variation is unclear.Both sets of results indicated that learning a relational category from a training set with a lower amount of variation is easier, but that learning from a training set with increased within-feature variation is significantly less challenging than learning from a set with increased across-feature variation.These results support the claim that, like feature-based category-learning, relational category-learning is sensitive to the type of feature variation in the training set.

View Article: PubMed Central - PubMed

Affiliation: Department of Cognitive Science, University of California, Merced, Merced, CA USA.

ABSTRACT
Learning of feature-based categories is known to interact with feature-variation in a variety of ways, depending on the type of variation (e.g., Markman and Maddox, 2003). However, relational categories are distinct from feature-based categories in that they determine membership based on structural similarities. As a result, the way that they interact with feature variation is unclear. This paper explores both experimental and computational data and argues that, despite its reliance on structural factors, relational category-learning should still be affected by the type of feature variation present during the learning process. It specifically suggests that within-feature and across-feature variation should produce different learning trajectories due to a difference in representational cost. The paper then uses the DORA model (Doumas et al., 2008) to discuss how this account might function in a cognitive system before presenting an experiment aimed at testing this account. The experiment was a relational category-learning task and was run on human participants and then simulated in DORA. Both sets of results indicated that learning a relational category from a training set with a lower amount of variation is easier, but that learning from a training set with increased within-feature variation is significantly less challenging than learning from a set with increased across-feature variation. These results support the claim that, like feature-based category-learning, relational category-learning is sensitive to the type of feature variation in the training set.

No MeSH data available.


Related in: MedlinePlus