This experiment tests memory.

You may use a computer or tablet but will need to enter letters on a real or touchscreen keyboard.


If you will use a keyboard to respond, press any key to continue.

If you will tap a screen to respond, tap anywhere to continue.

Enter your name (optional)

Your name will appear in the results page and data file.

Press the enter key to continue with the instructions.

Tap the screen to continue with the instructions.

Introduction

This pair of experiments tests memory for briefly displayed letters.

After the display vanishes, you will enter the letters that you recall.

Although the displayed letters are all uppercase, you may enter lowercase letters in response.


There are 00 practice trials and 00 test trials, taking about 00 minutes in total.

Press any key to continue with the instructions.

Tap the screen to continue with the instructions.

Instructions – Whole Report

In each trial, you will see letters in a grid, visible for 1/? of a second.

When they vanish, use the keyboarduse the screen keys to enter all the letters that you recall. Press Enter or Return when you are done.


Press any keyTap the screen when you are ready for the first of 0000 practicetest stimuli.

Instructions – Partial Report

In each trial, you will see letters in a 3-row grid, visible for 1/? of a second.

When they vanish, you will hear a tone that tells which row to recall. Click the links below to listen to each tone. Be sure you know which row is cued by each sound. Tones:  ,  ,  . Set the volume to a comfortable level.

When you hear the tone, enter all the letters that you recall from the cued row.


Press any keyTap the screen when you are ready for the first of 0000 practicetest stimuli.

+

Press the enter key when you are ready for the next trial (00 remain).

This concludes the first set of practice trialsexperiment

Press the ← key to repeat the 00 practice trials.

Press any other key to start the 00 test trials.

Tap the left side of the screen to repeat the 00 practice trials.

Tap the middle of the screen to start the 00 test trials.

Press any key to start the second experiment.

Tap the screen start the second experiment.

This concludes the second set of practice trialsexperiment

Press the ← key to repeat the 00 practice trials.

Press any other key to start the 00 test trials.

Tap the left side of the screen to repeat the 00 practice trials.

Tap the middle of the screen to start the 00 test trials.

Press any key to see your results.

Tap the screen to see your results.

0 2 4 6 8 10 12 Whole Partial 0 2 4 6 8 10 12 0 0.2 0.4 0.6 0.8 1.0 Whole Partial Whole Partial Normal Masked Condition Letters in display Delay to cue (s) Number of letters
WholePartial
Trials
Correct
Available
Errors
Whole
ReportedErrorsN
3
4
6
8
9
12
Partial
AvailableErrorsN
6
9
12
Whole
ReportedErrorsN
0.00
Partial
AvailableErrorsN
0.00
0.15
0.50
1.00
Without Mask
WholePartial
Trials
Correct
Available
Errors
With Mask
WholePartial
Trials
Correct
Available
Errors

Explanation

When instructed to report all visible letters, you reported letters in 00 trials, as shown in the first bar. When instructed to report only one row, you reported letters in 00 trials, as shown in the dark part of the second bar. The cue to report a row came after the display ended and there were three rows. Thus you must have seen three times as many letters as you reported, or . The light part of the second bar shows this as the number of letters available. Lines on the bars show standard error of the mean, a measure of how much the report varied in each condition.

Reported letters that were not in the array (whole report) or not in the cued row (partial report) are shown as errors in the table (average per trial). If there are many errors, the results may be questionable.

The plot shows number of letters available, plotted against the number of letters displayed. For whole-report, this is simply the number of letters reported. For partial-report, it is the number of letters reported multiplied by 3 (the number of rows). Letters available is the number of letters reported multiplied by 3 (the number of rows). This is the number of letters that must have been seen in order to report an arbitrary row. Bars on the points show the standard error, a measure of how much the report varied in each condition.

Reported letters that were not in the array (whole report) or not in the cued row (partial report) are shown as errors in the table (average per trial). If there are many errors, the results may be questionable.

The plot shows number of letters available, plotted against delay between presentation and cue. Letters available is the number of letters reported multiplied by 3 (the number of rows). This is the number of letters that must have been seen in order to report an arbitrary row. Bars on the points show the standard error, a measure of how much the report varied in each condition.

Reported letters that were not in the array (whole report) or not in the cued row (partial report) are shown as errors in the table (average per trial). If there are many errors, the results may be questionable.

The first pair of bars shows the number of letters available when no mask appeared after the display. This should look like the results from Experiment A, with partial report having more available letters than whole report. The second pair of bars shows results when a mask appeared after the display but before any partial report cue. The mask disrupts the lingering visual memory, so the partial report cue is no longer helpful. Lines on the bars show the standard error, a measure of how much the report varied in each condition.

Reported letters that were not in the array (whole report) or not in the cued row (partial report) are shown as errors in the table (average per trial). If there are many errors, the results may be questionable.

The text area below shows raw data from all trials. You can copy it into a spreadsheet for further analysis or download it with the link below.

Raw data

T = trial; X = experiment; S = letter array; C = cued row (0 for none); M = mask (1 if masked); D = cue delay (ms); R = response; N = number correct; A = number available; E = errors