Measures short-term/working memory
Sternberg task (ST)
How do people store and retrieve random information from short-term memory?
One of the primary goals of the Sternberg Task (also commonly referred to as a short term memory, memory search, or item recognition task) is to investigate how people store and retrieve items in a sequence and how increasing (or varying) the number of digits in the sequence affects their behavior. Robust findings from this task suggests a linear relationship between reaction times and the number of items studied (i.e., when memorizing larger sequences, we are slower to recognize the items, resulting in longer reaction times). This task emphasizes that outward behaviors (like reaction times) can be used to infer mental events (like memory), an underlying principle of cognitive psychology.
Resources:- Atkinson, R. C., Holmgren, J. E., & Juola, J. F. (1969). Processing time as influenced by the number of elements in a visual display. Perception & Psychophysics, 6, 321-327. https://link.springer.com/article/10.3758/BF03212784
- Background on Sternberg Scanning: https://www.neurobs.com/manager/content/docs/psychlab101_experiments/Sternberg%20Scanning/description.html
- Sternberg, S. (1966). High-speed scanning in human memory. Science, 153, 652-654.
Parameters
The following features can be tweaked (default values in parentheses):- Task instructions
- Number of trials (60)
- Whether or no to include full recall (full recall is where the participant must enter the sequence they saw in addition to identifying whether or not a digit was present)
- Color of digits
- Size of digits
- Language (chinese, english, spanish, french)
Note: trial sequence is randomized and balanced across conditions (1 digit/2 digit/3 digit/4 digit/5 digit/6 digit)
Note: trial sequence is also balanced by probe type (positive/negative). Positive means the digit was present; negative means it was not
Disclaimer
The following languages that are currently offered for this survey are not validated translations. This survey was translated using Google Translate and verified by members of the community.
What data is collected? How is it scored?
The following variables are recorded:
- ST_probe_sequence_length_1_true_percent
- ST_probe_sequence_length_2_true_percent
- ST_probe_sequence_length_3_true_percent
- ST_probe_sequence_length_4_true_percent
- ST_probe_sequence_length_5_true_percent
- ST_probe_sequence_length_6_true_percent
- ST_probe_type_negative_true_percent
- ST_probe_type_positive_true_percent
- ST_recall_sequence_length_1_true_percent
- ST_recall_sequence_length_2_true_percent
- ST_recall_sequence_length_3_true_percent
- ST_recall_sequence_length_4_true_percent
- ST_recall_sequence_length_5_true_percent
- ST_recall_sequence_length_6_true_percent
- ST_recall_type_negative_true_percent
- ST_recall_type_positive_true_percent
The main variables are the percent correct for full recall by probe type (positive/negative), percent correct probe by probe type (positive/negative), percent correct full recall by sequence length (1/2/3/4/5/6), and percent correct probe by sequence length (1/2/3/4/5/6)
Note: trial sequence is randomized and balanced across conditions (1 digit/2 digit/3 digit/4 digit/5 digit/6 digit)
Note: trial sequence is also balanced by probe type (positive/negative). Positive means the digit was present; negative means it was not
Background
How do people store and retrieve random information from short-term memory?
One of the primary goals of the Sternberg Task (also commonly referred to as a short term memory, memory search, or item recognition task) is to investigate how people store and retrieve items in a sequence and how increasing (or varying) the number of digits in the sequence affects their behavior. Robust findings from this task suggests a linear relationship between reaction times and the number of items studied (i.e., when memorizing larger sequences, we are slower to recognize the items, resulting in longer reaction times). This task emphasizes that outward behaviors (like reaction times) can be used to infer mental events (like memory), an underlying principle of cognitive psychology.
What participants see before taking the task
In this task, you are asked to memorize a sequence of digits and indicate whether a certain digit was included in the sequence.
What participants see after taking the task
The Sternberg Task measures your short-term memory (also called working memory) by assessing how you store and retrieve random information like a sequence of digits. By varying the complexity of the sequence, such as its size, researchers can harness people’s reaction times to learn about how our brain performs memory searches.
Aggregate Variables
These data are automatically written to a csv file upon completion of the task
more info
- ST_probe_sequence_length_1_true_percent
- ST_probe_sequence_length_2_true_percent
- ST_probe_sequence_length_3_true_percent
- ST_probe_sequence_length_4_true_percent
- ST_probe_sequence_length_5_true_percent
- ST_probe_sequence_length_6_true_percent
- ST_recall_type_negative_true_percent
- ST_recall_type_positive_true_percent
- ST_recall_sequence_length_1_true_percent
- ST_recall_sequence_length_2_true_percent
- ST_recall_sequence_length_3_true_percent
- ST_recall_sequence_length_4_true_percent
- ST_recall_sequence_length_5_true_percent
- ST_recall_sequence_length_6_true_percent
- ST_recall_type_negative_true_percent
- ST_recall_type_positive_true_percent
Scoring
In the Sternberg task, you (as the participant) are asked to identify whether a digit (the probe) was in a sequence digits you were asked to memorize. This sequence of digits can range in length (sequence length) from a 1-digit sequence (i.e. 3) to a 6-digit (i.e. 3, 4, 1, 6, 9, 7) sequence.
A probe can be positive, meaning that it was in the sequence you were asked to remember. Or the probe can be negative, meaning it was not in the sequence you were asked to remember.
You are then also asked to recall the sequence of digits (type the numbers you remembered into a fill in the blank).
True indicates that the participant answered correctly.
True percent indicates what percentage of this particular type of trial a participant answered correctly.
So, in the aggregated data file, the variable called "recall-sequence-length-4-true-percent" is telling you the percentage of correct responses a participant gave when they were asked to recall a sequence with 4 digits in it.
Duration
15 mins
Resources
- Atkinson, R. C., Holmgren, J. E., & Juola, J. F. (1969). Processing time as influenced by the number of elements in a visual display. Perception & Psychophysics, 6, 321-327. Read Full Article
- Background on Sternberg Scanning
- Sternberg, S. (1966). High-speed scanning in human memory. Science, 153, 652-654.