Author: David P. Nicoll
Nicoll, David P., 2024 Cueing Induced Inattentional Blindness, Flinders University, College of Education, Psychology and Social Work
Terms of Use: This electronic version is (or will be) made publicly available by Flinders University in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. You may use this material for uses permitted under the Copyright Act 1968. If you are the owner of any included third party copyright material and/or you believe that any material has been made available without permission of the copyright owner please contact copyright@flinders.edu.au with the details.
Augmented Reality (AR) technology is increasingly used as a visual search aid in operational search environments. In such contexts, visual search rarely involves a single well-defined target. Instead, target descriptors are necessarily vague, such as "search for any weapon", leading to both human and AR technology finding it challenging to identify all targets reliably. When AR technology cannot identify all targets reliably, the search task becomes a dual-search task. The first is for targets identified by the AR system, and the second is for targets not identified. In complex search environments, it would seem that providing any support to operators, such as cueing some targets, should aid their performance. However, prior research has demonstrated that performance improves for the cued search task, but performance on other concurrent tasks can suffer. In summary, making one task easier does not guarantee that other tasks will benefit.
This project adapted classic sustained inattentional blindness paradigms into a video game simulating a military overwatch scenario. During three experiments, participants performed a demanding primary task, where they were directed to click on any threats they saw. Depending on the experiment, trial and condition, people were aided by either an informative, uninformative, or no cue. Behavioural responses to cued and uncued threats were measured, with inattentional blindness measured as the miss-rate of uncued rare threats, which were never cued. Additionally, overt attention was recorded using eye-tracking for all experiments and a combination of both overt and covert attention was recorded using Steady-state Visually Evoked Potentials, using electroencephalography, for the last two experiments.
Experiment 1 was a within-subjects design with repeated onsets of the uncued rare threat character. Each participant experienced equal numbers of cued and uncued trials. When cues were provided, they were 100% informative in cueing frequent threat characters. Behavioural and eye-gaze measures were recorded. As expected, performance towards cued threats improved, but without the predicted cost to uncued threat search performance. However, there was a general trend toward reduced uncued threat search performance, and sizable differences in attention towards cued characters.
Experiment 2 moved to a between-subjects design with only a single rare threat character onset. Additionally, an uninformative cue condition was introduced. Participants were assigned to only one of the conditions experiencing uncued, informatively cued, or uninformatively cued frequent threats only. Both cued conditions showed increased inattentional blindness rates compared to the uncued condition but with differing attentional patterns.
Experiment 3 replicated Experiment 2 as closely as practical but using a virtual reality system. No changes were made to the experimental paradigm. Informative cue results closely replicated that seen in Experiments 1 and 2. However, unlike the Experiment 2 results the uninformative cue condition outcomes were more similar to the no-cue condition results.
These findings support prior research indicating that providing task-relevant visual cues leads to tunnelling of attention towards those cues, impairing the search for uncued targets. In contrast, task-irrelevant cues contribute to visual clutter and, depending on the search context, may increase inattentional blindness.
Keywords: cueing, inattentional blindness, augmented reality, virtual reality, visual search, operational search, attentional errors
Subject: Psychology thesis
Thesis type: Doctor of Philosophy
Completed: 2024
School: College of Education, Psychology and Social Work
Supervisor: Oren Griffiths