Visual search task
Find the T among the Ls. The more letters on screen, the longer it takes, and that slope shows how attention scans a scene.
- Measures
- Search efficiency (RT × set size slope)
- Template length
- about 5 minutes (30 trials)
A display of letters appears and you decide as quickly as you can whether it contains a T. The other letters are Ls, which share the same two lines arranged differently, so the T does not stand out.
The key measure is how much each extra letter adds to your response time. When the target differs by a single feature, like a red item among green ones, the slope is close to zero and the target seems to pop out. When it differs only in how features are combined, as with T and L, each extra item costs time.
Try it yourself
A shortened run of the same template researchers use. Nothing is recorded and you do not need an account.
How the task works
- 1A display with 4, 8 or 16 letters appears.
- 2Press F if there is a T, J if there is not. You have 5 seconds.
- 3The T is present on half of the trials.
What people usually find
Response time rises roughly linearly with the number of items, by tens of milliseconds per item. When the target is absent the slope is steeper, often about twice as steep, because people have to check more of the display before saying no.
How to score it
- Search slope: fit a line to mean correct RT against set size, separately for target-present and target-absent trials. The slope (ms per item) is the measure.
- The intercept reflects everything that is not search, such as encoding and response time.
Before you run it
- Display size depends on the participant's screen. Letters drawn in pixels vary in visual angle from one device to another; consider a virtual chinrest calibration if size matters for your question.
- The template uses 10 trials per set size. Slopes need more: 30 or more per cell for individual estimates.
Run it as a study
Use the template as it is
The demo above is the same code as the Visual Search template in the gallery. Add it to a project, change the settings in config.js, publish it and share the link. Reaction times and responses come back to your dashboard.
Create a free account to use itOr build a variant: compare pop-out with conjunction search
Paste this into the AI panel in the experiment editor. Read the code it writes before you collect anything.
Build a visual search task with two blocks. Feature block: find a red vertical bar among green vertical bars. Conjunction block: find a red vertical bar among green vertical and red horizontal bars. Use set sizes 4, 8, 16 and 32, target present on half of the trials, 20 trials per set size and presence in each block, with items placed at random positions on a jittered grid. Respond F for present and J for absent within 4000 ms. Record block, set size and target presence.
Common questions
What is a search slope?
The extra response time each additional item adds, in milliseconds per item. Flat slopes mean efficient, parallel search; steep slopes mean items are being checked one at a time or in small groups.
What is feature integration theory?
Treisman and Gelade's proposal that simple features are registered in parallel across the visual field, while combining features into objects needs focused attention, one location at a time.
Does screen size matter for online visual search?
It changes how large the display is in the visual field, which affects difficulty. Within-participant comparisons such as slopes are fairly robust, but record screen size and consider a calibration step.