An investigation of within-dimension stimuli in categorization and change detection

Abstract

This thesis examined the way in which information is combined in order to make decisions. It focused specifically on within-dimension decision-making with spatially separate stimuli. The primary aim was to characterize the underlying organization of processing over time (i.e., identify the processing architecture; whether information processing proceeds in serial, parallel, or is pooled into a single decision-making channel), stopping rules (i.e., whether the process is exhaustive or self-terminating), and the efficiency of processing (i.e., workload capacity). These attributes were assessed through the analysis of response times (RTs) using Systems Factorial Technology (SFT), the Logical-Rules paradigm, and relevant computational modeling. In the first part of the thesis, results from the categorization experiments showed that for the majority of participants, processing occurs coactively (i.e., is pooled into a single decision process). Workload capacity, however, was shown to be limited. This suggests that a violation of context invariance may have occurred and several theories are considered as potential explanations for this finding.

In the second part of the thesis, a novel modeling framework for characterizing the time course of change detection based on information held in visual short term memory was presented. Specifically, we sought to answer whether change detection is better captured by a first-order integration model, in which information is pooled from each location, or a second-order integration model, in which each location is processed independently. We conducted two experiments with both disjunctive OR rules and conjunctive AND rules (across locations) using a double factorial paradigm and a redundant target paradigm. These experiments showed that although capacity is generally limited in both tasks, architecture varies from parallel self-terminating in the OR task to serial self-terminating in the AND task. This novel framework allowed for model comparisons across a large set of models, ruling out several competing explanations of change detection.

As a whole, this thesis found both differences and similarities in decision-making using within-dimension stimuli across categorization and change detection tasks. The finding of differences in decision-making strategy across categorization and change detection tasks highlights that different perceptual operations can yield a variety of experimental results. It may be expected that other tasks such as visual search, identification, and detection might also diverge. A finding of limited capacity across both task types, however, points to a potential common bottle neck in processing efficiency at an earlier encoding stage.