Orange letters T and L in different orientations appear on a grey field, and the participant taps Present or Absent: is there an upright T among them? The number of letters varies from display to display, and the search slope, the extra time per letter, shows how efficient the search is. A port of the Open Lab library paradigm.
Configure parameters and run an interactive preview exactly as participants will experience it. No data is recorded.
Adjust parameters below, then start the preview on the right.
Include practice trials
Shown with feedback before the main task
Task parameters
Set sizes 5, 10, 15, 20: 48 trials, 6 with and 6 without the target per set size, shuffled. A display not answered within 10000 ms is an omission. Practice: 8 trials at set size 10, with feedback. About 2 min in all.
A port of the Open Lab library paradigm Visual Search Task, a conjunction search after Treisman & Gelade (1980): find an upright T among Ts turned by 90° or 270° and Ls at any of four angles, orange on grey. Defaults from the library paradigm: set sizes 5, 10, 15 and 20, 6 trials with and 6 without the target per set size, a 100 ms fixation cross, a 400 ms blank, and 8 practice trials at set size 10 with 800 ms of feedback. Changed for phones: Present and Absent buttons (Present on the left, as in the smartphone T-vs-L search of Gardasevic et al., 2022) instead of the Space bar for present and waiting for absent, so target-absent trials have response times too; a display not answered by the deadline is an omission, and the deadline is 10 s, as in Gardasevic et al., instead of the library's 4 s (whose end was the absent answer: with an Absent button, slow absent answers at the largest set sizes would time out and flatten the target-absent slope). The library's grid of 30 places is turned upright (5 × 6 cells) for a portrait screen, the letters are drawn as two bars of equal length, and the display has a fixed size of 300 × 360 pt with 24-pt letters, smaller only on a screen that cannot hold it (the size used is stored with the result).
Empty: every participant and session gets new displays, and the seed drawn is stored with the result, so the displays can be rebuilt. A number (0 to 4294967295) gives every participant the same displays in the same order.
Outputs: accuracy with and without the target, the false-alarm rate, d′ (log-linear correction, Hautus, 1995), the median response time of correct answers for each set size with and without the target (answers under 150 ms left out, as the library paradigm recommends), the search slopes with and without the target (the least-squares slope of those medians on the set size, in ms per item; Wolfe, 1998), the mean response time of hits and the number of answers under 150 ms. A T-vs-L search typically gives target-present slopes of about 20–40 ms per item and steeper target-absent slopes (Wolfe, 2021). Per trial: the set size, whether the target was there, every letter with its angle and place, the response, correct, omission and the response time. Duration estimate: 600 ms plus 30 ms per letter with the target and 60 ms without it (our estimate). Responses are timed from the drawn display to the touch-down, but this timing has not yet been validated on real devices.
This is a researcher preview. No participant data is recorded.
Simulated participant view
Orange letters T and L in different orientations appear on a grey field, and the participant taps Present or Absent: is there an upright T among them? The number of letters varies from display to display, and the search slope, the extra time per letter, shows how efficient the search is. A port of the Open Lab library paradigm.
No data is recorded
A grey field of 300 × 360 pt (5 × 6 places of 60 pt) with up to 30 orange letters, each drawn as two bars of equal length, 24 pt high; a white fixation cross before each display; two large buttons below, Present (left) and Absent (right), which count at touch-down. The field keeps this size on every phone and is smaller only on a screen that cannot hold it; the size used is stored with the result.
Useful when a study needs the efficiency of attentional search, the increase in response time per extra item, separately from overall speed: for example in studies of attention across the day, of fatigue, sleepiness or light exposure (a T-vs-L search ran in everyday life on participants' own phones in Gardasevic et al., 2022), or as practice for selection tests with attention and concentration tasks.
Not a standardised test and not a source of norms. A slope from 6 trials per set size and target condition is noisy for one person: use more trials, or aggregate across sessions, when individual slopes matter. Display density and letter size change how efficient the search is, and the viewing distance of a phone is not controlled, so compare slopes within a device rather than between devices. Not suitable for participants who cannot see small letters on a phone screen well.
Keep the defaults to run the library paradigm's design: set sizes 5, 10, 15 and 20, 6 trials with and 6 without the target per set size (48 trials), after 8 practice trials, with up to 10 s per display, as in the smartphone search with Present and Absent buttons of Gardasevic et al. (2022). Choose at least two set sizes for a slope; more trials per set size give steadier slopes. Leave the seed empty for new displays per participant and session, or set one to give everyone the same displays.
The search slope is the least-squares slope of the median response time of correct answers on the set size, separately for displays with and without the target; answers under 150 ms are left out, as the library paradigm recommends. In feature-integration theory (Treisman & Gelade, 1980), a target that differs from the distractors only in how its parts are arranged needs attention to move from item to item, so the response time rises with every added item. Wolfe (2021) gives target-present slopes of about 20–40 ms per item for a T-vs-L search, and steeper slopes without the target. Speed and accuracy trade off: read the slopes together with the accuracies, and check the omissions at the largest set size, because slow searches that reach the deadline drop out of the medians and flatten the slope. Differences from the library paradigm: two buttons instead of the Space bar for present and waiting for absent, so displays without the target have response times and a slope of their own; a display not answered by the deadline is an omission, and the deadline is 10 s, as in Gardasevic et al. (2022), instead of the library's 4 s, whose end was the absent answer (with an Absent button, slow absent answers at the largest set sizes would time out and flatten the target-absent slope); the grid is turned upright for a portrait screen, and the letters are drawn as two equal bars. Responses are timed from the drawn display to the touch-down, but this timing has not yet been validated on real devices.
Each run of this task records how its times were measured and the conditions of the session, such as screen brightness, battery saving and interruptions. Runs with possible problems are flagged on the Quality tab and in the export.
How SMAAT times cognitive tasksAdd Visual Search to your ESM study on SMAAT — free to get started.