<h1>Stimuli</h1>
<p>The task itself as well as its corresponding web pages may be found under the Materials node on OSF.</p>
<p><strong>Technologies used.</strong> The survey portion of the experiment was
programmed in PHP, HTML/CSS, and JavaScript. The sustained IB task was programmed in JavaScript
only. All data will be posted to and extracted from a MySQL database. Once the study is complete, the raw data
will be extracted from the MySQL database and posted under the Data node on OSF.</p>
<p><strong>Web Survey.</strong> For each page of the web survey, participants will be required to answer all questions before advancing to the next page. If they try to advance before answering all of the questions, an error message will appear on the website. In the page containing the sustained IB task, the button that allows participants to advance to the next page will not appear until the IB task is completed. </p>
<p><strong>Sustained Inattention Blindness Task.</strong> This task was based on the task used in <a href="http://www.theassc.org/files/assc/2471.pdf">Most et al. (2000)</a>. In this task, participants are presented with four L block letters (2 white, 2 black) and four T block letters (2 white, 2 black) with each letter having a width and height of 43px, and a thickness of 11px. During each trial, these letters move randomly about a 666px by 546px gray (#777777) window at a variable speed ranging from approximately 90px to 180px per second (~30 FPS), bouncing-off of the window edges once they come in contact. Both the speed and direction of each letter changes every 1000ms to 4000ms, making the trajectory for each one difficult to predict. (The change interval for each letter is different, making it so that the letters don’t change at the same time.) Also in the display is a blue 2px-thick line that spans the window horizontally and is vertically centered in the window. At the center of this horizontal line is an 11px by 11px blue fixation square. Participants are presented with a total of 6 15-second-long trials. During each trial, participants are instructed to count the number of times the white-colored letters completely cross the blue line while keeping their fixation on the blue square. Participants are asked to enter their counts after the completion of each trial. Five seconds into the final trial, an unexpected light gray (#B2B2B2) cross-shaped object enters the window on the right, travels horizontally toward the left-side of the window at 90px per second, and then exits on the left-side of the window, keeping the same position vertical position throughout its movement. Like the block letters, this cross has a height and width of 43px and a thickness of 11px. For each participant, the unexpected object will travel on 1 of 7 paths: on the blue line, 103px above or below the blue line, 206px above or below the blue line, or 254px above or below the blue line. The vertical position of the unexpected object is chosen randomly for each participant.</p>
<h1>Procedure</h1>
<p>Participants will first complete the sustained IB task, after which they will answer questions about the unexpected object (see below). Participants will then be asked questions regarding the quality of the displays (i.e., do they believe it worked correctly and and what they think went wrong if it didn't). Participants will then take a colorblindness test. Finally, participants will answer a number of demographic questions.</p>
<strong>Questions about the unexpected object that participants are asked to answer. The response alternatives are listed below each question.</strong>
<i>
<ol>
<li>
On that last trial of the task, did you notice anything that was not there on previous trials?
<ul>
<li>Yes</li>
<li>No</li>
</ul>
</li>
<li>
If you did notice something else on that last trial, was it moving? (If you are unsure or did not notice, please just guess.)
<ul>
<li>Yes</li>
<li>No</li>
</ul>
</li>
<li>
If it was moving, what direction was it moving in? (If you are unsure or did not notice, please just guess.)
<ul>
<li>Right</li>
<li>Left</li>
<li>Up</li>
<li>Down</li>
</ul>
</li>
<li>
If you did notice something else on that last trial, what color was it? (If you are unsure or did not notice, please just guess.)
<ul>
<li>Red</li>
<li>Green</li>
<li>Blue</li>
<li>Purple</li>
<li>Yellow</li>
<li>Gray</li>
<li>Black</li>
<li>White</li>
<li>Brown</li>
</ul>
</li>
<li>
If you did notice something else on that last trial, what shape was it? (If you are unsure or did not notice, please just guess.)
<ul>
<li>Rectangle</li>
<li>Triangle</li>
<li>Cross</li>
<li>Circle</li>
<li>T-Shaped</li>
<li>L-Shaped</li>
<li>B-Shaped</li>
<li>V-Shaped</li>
</ul>
</li>
</ol>
</i>
<h1>Planned Analyses</h1>
<p>When possible, effect sizes and their accompanying 95% confidence intervals will be presented. For data that do not appear to be normally distributed, transformations will be applied prior to any statistical tests. If a transformation is used, any differences between the analyses using transformed and non-transformed data will be reported.</p>
<p>To assess the impact that unexpected object distance from the line has on the relationship between inattention blindness and age while controlling for task quality and IB trial counting accuracy, age, distance, IB trial counting accuracy, mean FPS, the available y monitor resolution, as well as the interaction between age and distance will be submitted to a logistic regression. Overall, we have three predictions for this analysis. First, we predict that there will be a main effect for age such that IB increases with greater age. Second, we predict a main effect for distance such that IB increases with greater distance from the line. Third, we predict that there will be an age by distance interaction such that the strength of the relationship between age and inattention blindness will increase as distance from the line increases. In terms of sample size, we chose to collect data on 500 participants as sample sizes this large are usually easy to obtain in online studies and because with such a large sample size we will be able to calculate more precise confidence intervals for the observed effects. Assuming an overall detection rate of 34%, which was observed in the Most et al. (2000) study, and an <i>R</i><sup>2</sup> of 0.05 between the predictor and all other predictors, which was observed in our first experiment, with a sample size of 500, we have an 80% chance of detecting an odds ratio less than or equal to 0.760 or greater than or equal to 1.316 at the .05 significance level for each predictor.</p>
<p>All other analyses will be considered exploratory and will be described as such in any published reports of the data.</p>
<h1>Exclusion Criteria and Stopping Rule</h1>
<p><strong>Exclusion due to Impairments in Vision.</strong> Data from participants who are color-blind or
report not having normal or corrected-to-normal vision will be excluded.</p>
<p><strong>Exclusion due to Geographical Location.</strong> Participants located outside of the U.S. will be
excluded. As Mechanical Turk allows you to restrict a task to individuals living in a specific geographical region, this exclusion will occur as part of the recruitment phase (we will only recruit participants located in the USA.</p>
<p><strong>Exclusion by Identical IP Addresses.</strong> Participants sharing the same IP address or Mechanical
Turk Worker ID with a prior participant will be excluded (i.e., people who participate in more than one of the variants of the study). We will include data only from the first time a person completes any of our HITs on this topic. This exclusion will occur after recruitment.</p>
<p><strong>Exclusion due to Comprehension or Attention.</strong> One of the questions in the demographics survey asks participants to select and remember the middle item out of a list of 5 numbers. On the next page, participants are asked to enter the number they selected. If they type in the wrong number (i.e., a number other than the middle one), their data will be excluded from the analysis. In addition, any participants who enter a nonsense answer for any of the open response questions will be excluded. Exclusion in this case will be determined by the coding of two independent judges who are blind to the subject's age and condition. If either judge believes any open-ended response was nonsensical, data from that participant will be excluded from all analyses.</p>
<p><strong> Exclusion due to Task Quality.</strong> Participants will be excluded if they report in the open-response question regarding task quality that the task did not appear to function properly; that is, if it functioned in a way that was qualitatively different from how it should have functioned (e.g., if letters suddenly appeared/ disappeared, if the letters appeared in different sizes, etc.). Exclusion will be decided on by two independent coders who will go through the dataset and judge whether or not a participant should be excluded. A participant will be excluded from the analysis if either of the coders believes that he or she should be. Coders will be blind to the participant's age and condition when completing this coding.</p>
<p><strong>Exclusion due to Inconsistencies in Self-Reported Age</strong> Participants will be asked to provide their age in two ways. First, they will be asked to enter their age in years. Second, they will be asked to provide their birth year. These two questions will be separated by at least one screen. If we find that the age a participant provides and the age we calculate using their birth year is not equivalent, then that participant will be excluded from the analysis.</p>
<p><strong>Stopping Rule.</strong> Data collection will cease once we have data from a minimum of 500 useable subjects or when data collection has gone on for two months. If the latter occurs, this will be explicitly stated in any write-up of the results.</p>
<p>We will end data collection before looking at the IB data. Prior to stopping data collection, the only inspection of the responses will be by blind judges to determine exclusions (see above), and their judgments will be based on files that do not include IB data.</p>