Name: Maugan R. D. Lloyd

Master's Thesis Proposal Meeting

Date: Monday, April 26th, 2021

Location: https://bluejeans.com/902155814

Advisor: Christopher K. Hertzog, Ph.D. (Georgia Tech)

Thesis Committee Members:

Christopher K. Hertzog, Ph.D. (Georgia Tech)

Paul Verhaeghen, Ph.D. (Georgia Tech)

Dobromir Rahnev, Ph.D. (Georgia Tech)

Title: Testing the Useful Field: Perceptual Learning Accounts for a Significant Proportion of Task Improvement in Visual Peripheral Detection Training 

Abstract:

Training improvements on the UFOV are correlated with improved driving behavior in older adults (Ball et al., 2010; Roenker et al., 2003), but the reason for the association remains opaque. Advocates for the efficacy of UFOV training suggest that global speed of processing or divided attention ability is improved by task training (Harvey et al., 2018; Edwards et al., 2018), whereas critics point to methodological flaws (Simons et al., 2016; Woutersen et al., 2017), or lack of theoretical grounding (Lintern & Boot, 2019). All have overlooked that the peripheral task of the UFOV is a consistently mapped (CM) visual search in which neither the target stimulus nor distractors change, and so can be expected to engender stimulus learning (Shiffrin & Schneider, 1977). 

The proposed study tests the hypothesis that large reductions in presentation time required for correct stimulus localization can occur in a peripheral task due to stimulus learning. This is tested by comparing adaptive-training perceptual-threshold functions, and transfer of training between experimental groups in a UFOV-like peripheral localization task. Training on a varied mapping (VM) group, in which any item in the stimulus set may be used as a target or distractor, is compared with a CM group, in which target and distractor sets are consistent. A simple discrimination task is simultaneously presented in the center of the array to ensure fixation, and to increase verisimilitude to the UFOV dual-task environment. The key prediction is that when transferred to new CM or VM stimuli in a second experimental phase, there will be a large performance cost for the CM-trained group relative to the VM-trained group due to the loss of familiar stimulus advantage. Smaller threshold changes in VM-trained participants will be due to task components unrelated to stimulus learning. 

Such findings would show that a major proportion of perceptual-threshold decrease following UFOV training cannot be accounted for by an improvement in underlying speed of processing or divided attention ability that then ameliorates older-adult driving behavior, as has been suggested (Edwards et al., 2018). Associations between the UFOV and older-adult driving may be due to a placebo effect (Foroughi et al., 2016), leading to new-found confidence in older adults in their cognitive abilities born from witnessing their own dramatic task-performance improvement in a psychology lab