Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List

Full text release has been delayed at the author's request until September 01, 2024

ETD Abstract Container

Abstract Header

An Eye Tracking Investigation of Classification Behavior on a Basic Family of Category Structures

Zhao, Li
http://orcid.org/0000-0002-5844-497X
http://rave.ohiolink.edu/etdc/view?acc_num=ohiou1564763671842416

Abstract Details

2019, Doctor of Philosophy (PhD), Ohio University, Experimental Psychology (Arts and Sciences).
Human categorization performance has been studied extensively during the past sixty years. Shepard, Hovland, and Jenkins (1961) conducted a seminal study in which they examined and established the learning difficulty ordering of the family of six structures with four positive examples defined over three binary dimensions (aka, the 32[4] structures family). This empirical ordering has proven robust whenever the experimental stimuli are defined over separable dimensions under both compact and distributed representations. Since the advent of this ordering, researchers utilizing this constraint have assumed that individuals allocate their attention to dimensions of stimuli optimally to categorize. In order to test this assumption, Rehder and Hoffman (2005a) performed an eye tracking version of the SHJ study with the 32[4] structure family and found evidence in support of the selective attention optimality assumption. However, in arriving to their conclusions, these researchers studied only four out of the six structure types in the family and only tested one instance of each structure type in the family. In this study, we comprehensively investigated all six structures for the first time from the standpoint of eye tracking behavior and did so more thoroughly by examining more than one instance per structure. First, we found that the learning difficulty ordering of the 32[4] structure family is: I < II < [III, IV, V] < VI, when the dimensions of experimental stimuli are separated with the distributed representation. There was a statistically significant improvement from the first to the second instance, regarding the classification behavior. Second, we found evidence that participants used the selective attention mechanism during the concept learning and categorization process by examining participants’ eye tracking performance (e.g., number of dimensions fixated, dwell time, saccades, similarity of scanpaths, etc.). In addition, we observed that individuals tend to use the decreasing pattern of attention allocation during the classification task. Third, we compared the degree of similarity of scanpaths in different structure types, and found that there were statistically significant differences among the types III, IV, and V structures, regarding participants’ scanpaths. Last, the fitness of two leading categorization models (i.e., the Generalized Context Model (GCM; Nosofsky, 1984, 1986) and Generalized Invariance Structure Theory Model (GISTM; Vigo, 2013a, 2014)) was tested. The results showed that GCM and GISTM can account for 70.51% and 95.36% of the variance of the degrees of learning difficulty, respectively. In addition, the attentional weights predicted by the GCM were not consistent with the attentional distribution in the empirical data. However, the dimensional binding cognitive mechanism on which GISTM is based, was presented in all six structure types.
Ronaldo Vigo (Advisor)
188 p.
Cognitive Psychology; Psychology
eye tracking; separable dimensions; selective attention; categorization; cognitive modeling

Recommended Citations

Citations

  • Zhao, L. (2019). An Eye Tracking Investigation of Classification Behavior on a Basic Family of Category Structures [Doctoral dissertation, Ohio University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ohiou1564763671842416

    APA Style (7th edition)

  • Zhao, Li. An Eye Tracking Investigation of Classification Behavior on a Basic Family of Category Structures. 2019. Ohio University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ohiou1564763671842416.

    MLA Style (8th edition)

  • Zhao, Li. "An Eye Tracking Investigation of Classification Behavior on a Basic Family of Category Structures." Doctoral dissertation, Ohio University, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=ohiou1564763671842416

    Chicago Manual of Style (17th edition)

ohiou1564763671842416
© 2019, all rights reserved.
This open access ETD is published by Ohio University and OhioLINK.