Prevalence Effects in Visual Search: Theoretical and Practical Implications (J)

NOTE: This registration is linked to a Human Subjects registration in ASSIST. That, in turn, is part of an NCI Grant, R01EY017001. The grant describes many proposed experiments and notes that many others might be done as follow-up studies. At the suggestion of the NIH, the investigators grouped these studies into several "studies", each covering multiple experiments. The experiment described here is part of Aim 1 of R01EY017001. It is not possible to register a set of experiments through the PRS system in CT.gov and it is not possible to file an annual report for the grant (RPPR) without an NCT number for projects that have started collecting participants. Accordingly, the investigators are describing one experiment here that would be part of the "Project 3" bundle of studies.

When scientists study visual search in the lab, they usually ask people to do dozens of instances of the same type of task in a block. (e.g., Please find a vertical line….in the next 100 trials). However, in many (maybe most) visual searches in the real world, People are rarely looking for the same item (or same type of item) back-to-back. They do one search. Then they do another, and so on. (Find the jam. Now, find the bread. Now find a knife, etc). The investigators want to know if people would produce the same results if investigators mixed different searches together as they do when they test in long blocks of one specific type of search. This is important because the investigators want to know if the rules, established by years of blocked experiments, continue to apply in the more mixed real world.

In some conditions of this version of the mixed/blocked paradigm, observers get to choose the task for the current trial. There are five conditions, each run on a different group of observers.

1. Trial Choice condition: In this condition, participants are allowed to choose which task to perform on each trial. At the beginning of each trial, the message "Choose your task" appears in the center of the screen. Participants choose one of four tasks by clicking on the corresponding button on the left side of the patch. The button is deactivated when the number of trials completed in a task reached 50. If a deactivated button is clicked, the message "Choose another task" appears at the center again.
2. Block Choice condition: In this condition, participants begin by choosing one of the four tasks. Participants do not need to choose a task before each trial; the current trial is drawn automatically from the most recently chosen task until the participant actively chooses to switch tasks or until the full complement of trials for that task is exhausted. At the start of each trial, a "switch" button appears at the center of the display for 700 msec. If participants click on that button within the 700 msec, they are prompted to select the task. If participants does not click the switch button within the 700 msec window, the button vanished and a trial from the current task will be presented. When the number of trials of one task reaches its limit, participants are asked to select another task.
3. Random condition: In this condition, participants do not choose which task to perform on each trial. The order of the trials is randomized and assigned to the participant before each trial. Before each trial, the target task is highlighted and activated, while the other buttons remain deactivated. The trial begins only when the participant clicks on the correct task button.
4. Blocked condition: In this condition, participants also have no choice. Here, trials of each task type are presented in a blocked fashion. After 50 trials of one type, the task is changed to another task. To keep the motor demands similar to the choice conditions, at the beginning of each trial, participants need to click the highlighted button for the assigned task type. The order of blocks is counterbalanced across participants.
5. Yoked condition: Participants in this condition see trials in the order chosen by a participant in the Trial Choice condition. Again, to keep the motor demands similar to the choice conditions, at the beginning of each trial, participants need to click the highlighted task button for the assigned task type though they have no choice in the matter.

In all conditions, stimuli are placed in random locations in a jiggled 5 x 5 grid. The grid are placed within a square field that is 0.5 of the maximum height of the screen. Each stimulus item fit into an invisible box that is 0.11 of the screen height. If the viewing distance is about 60 cm, a standard computer screen would show a stimulus field of about 15 deg on a side and each item would be placed in a box of about 3 x 3 deg. The background is white. Trials are divided evenly between three set sizes: 8, 16, and 24 items and evenly between target-present and absent trials. Stimuli are presented until the observer responds by pressing either 'p' or 'q' to indicate target present or absent, respectively. Accuracy feedback is provided after each trial.