Document Type


Date of Degree

Spring 2015

Degree Name

PhD (Doctor of Philosophy)

Degree In


First Advisor

Larissa K. Samuelson


The similarity between objects is judged in a wide variety of contexts from visual search to categorization to face recognition. There is a correspondingly rich history of similarity research, including empirical work and theoretical models. However, the field lacks an account of the real time neural processing dynamics of different similarity judgment behaviors. Some accounts focus on the lower-level processes that support similarity judgments, but they do not capture a wide range of canonical behaviors, and they do not account for the moment-to-moment stability and interaction of realistic neural object representations. The goal of this dissertation is to address this need and present a broadly applicable and neurally implemented model of object similarity judgments. I accomplished this by adapting and expanding an existing neural process model of change detection to capture a set of canonical, task-general similarity judgment behaviors. Target behaviors to model were chosen by reviewing the similarity judgment literature and identifying prominent and consistent behavioral effects. I tested each behavior for task-generality across three experiments using three diverse similarity judgment tasks. The following behaviors observed across all three tasks served as modeling targets: the effect of feature value comparisons, attentional modulation of feature dimensions, sensitivity to patterns of objects encountered over time, violations of minimality and triangle equality, and a sensitivity to circular feature dimensions like color hue. The model captured each effect. The neural processes implied by capturing these behaviors are discussed, along with the broader theoretical implications of the model and possibilities for its future expansion.

Public Abstract

We compare objects and judge how similar or different they are throughout our daily lives. For example, we judge family relations from the similarity of faces, and we compare the similarity of products as part of our purchasing decisions. Similarity is also critical to specialist and industrial applications like measuring the uniformity of manufactured goods or comparing x-ray images to judge tumor growth.

Psychologists know a great deal about the exact rules that support similarity judgments and the resulting patterns of behavior. For example, we know how differences in features such as color, size, and shape interact with one another to influence similarity judgments. We also know how memories of other objects seen recently and how time pressure or different goals influence similarity judgments.

Less is known, however, about the neural processes behind these similarity judgments. The goal of this dissertation is to fill this gap in knowledge by creating a computer model that can predict and explain the most well-known similarity behaviors, using neurally realistic cognitive processes. This computer model links the neural activity that supports similarity judgments to those previously studied in the context of other cognitive tasks. These connections will allow psychologists to paint a more complete picture of how we process and understand objects in general.

The model also serves an important step toward direct applications of machine simulated artificial similarity judgments between objects. Machine-based similarity may lead to accurate automatic second opinions on medical images or more efficient satellite or surveillance image interpretation.


publicabstract, Modeling, Neural, Similarity


ix, 163




Copyright 2015 Gavin Jenkins

Included in

Psychology Commons