Visual Short-Term Memory Task

Version: v1 (current)

The Visual Short-Term Memory (VSTM) task measures the capacity and precision of visual working memory using a change detection paradigm.

Overview

Visual short-term memory (also called visual working memory) is the ability to briefly hold and manipulate visual information in mind. This task evaluates how many visual items participants can accurately remember and whether they can detect changes to those items.

Participants view a brief display containing multiple colored shapes, which then disappears. After a short retention interval, a test display appears, and participants must judge whether it matches the original display or if something has changed. The task varies the number of items (set size) to estimate the participant's memory capacity limit.

This paradigm is foundational for understanding individual differences in visual working memory capacity, which correlates with fluid intelligence, attention control, and academic achievement.

Scientific Background

The Capacity Question

Research using change detection paradigms has established that most people can accurately maintain 3-4 simple visual objects in working memory. This capacity limit appears constant whether remembering single features (colors only) or conjunctions (color and shape together), suggesting object-based storage.

Cowan's K Formula

The change detection paradigm allows estimation of memory capacity using Cowan's K:

K = S × (H - F)

Where:

• K = estimated capacity
• S = set size (number of items in display)
• H = hit rate (proportion of "different" trials correctly identified)
• F = false alarm rate (proportion of "same" trials incorrectly called "different")

For example, with set size 6, hit rate 0.75, and false alarm rate 0.10:

K = 6 × (0.75 - 0.10) = 3.9 items

Key Findings

• Capacity Limits: Most adults can maintain 3-4 items, with reliable individual differences
• Neural Correlates: Capacity correlates with sustained activity in parietal and occipital cortices
• Clinical Relevance: Deficits observed in ADHD, schizophrenia, and normal aging
• Educational Predictiveness: Correlates with STEM performance and reading comprehension

References

• Luck, S.J., & Vogel, E.K. (1997). The capacity of visual working memory for features and conjunctions. Nature, 390(6657), 279-281.
• Cowan, N. (2001). The magical number 4 in short-term memory. Behavioral and Brain Sciences, 24(1), 87-114.

Why Researchers Use This Task

1. Cognitive Capacity Assessment: Provides a quantitative estimate (K) of working memory capacity
2. Clinical Evaluation: Sensitive to attention deficits, cognitive decline, and neurological disorders
3. Developmental Research: Tracks capacity improvements from childhood to adulthood
4. Neuroscience Studies: Clean paradigm for EEG/ERP and fMRI with discrete events and clear retention period
5. Training Studies: Assesses improvements in visual memory from interventions

Configuration Options

Visual Parameters

Parameter	Default	Range	Description
Canvas size	600px	200-1200px	Size of square display area
Item size	60px	10-200px	Size of each colored shape

Practice Trials

Parameter	Default	Description
Enable practice	Off	Turn on to include practice trials with feedback

When practice is enabled:

• Participants see visual feedback after each practice response (green checkmark for correct, red X for incorrect)
• Practice trials use simpler set sizes to help participants learn the task
• Main trials begin after practice is complete (or skipped in optional mode)

Trial Configuration

All trials are configured manually in the spreadsheet. Each row defines one trial with the following columns:

Column	Description	Default	Example
set_size	Number of items in memory/test arrays	4	4
condition	Whether arrays match	"same"	"same" or "different"
block	Optional grouping label	""	"high_load"
stimulus_type	Type of visual stimuli	"colored_shapes"	"colored_shapes" or "colored_squares"
min_distance	Minimum spacing between item centers (px)	100	100
fixation_ms	Duration of fixation cross (ms)	500	500
memory_duration_ms	How long the memory array is displayed (ms)	500	500
retention_interval_ms	Blank delay between memory and test arrays (ms)	1000	1000
test_duration_ms	Maximum time for test array display (ms)	2000	2000

The task automatically generates for each trial:

• Random positions for items (within display area, respecting min_distance)
• Random colors (from a set of 10 distinct colors)
• Random shapes (circle, square, or triangle based on stimulus_type)
• For "different" trials: randomly selects one item to change

Timing Recommendations:

• Brief encoding (100-200ms): Prevents verbal recoding, forces visual memory
• Standard retention (900-1000ms): Typical delay in VSTM studies
• Generous test period (2000ms+): Allows self-paced responding

Example Configuration

Capacity Assessment Study (80 trials across 4 set sizes):

set_size	condition	block	stimulus_type	min_distance	fixation_ms	memory_duration_ms	retention_interval_ms	test_duration_ms
2	same		colored_shapes	100	500	100	900	2000
2	different		colored_shapes	100	500	100	900	2000
4	same		colored_shapes	100	500	100	900	2000
4	different		colored_shapes	100	500	100	900	2000
...	...		...	...	...	...	...	...

Repeat rows to reach the desired number of trials per set size (e.g., 20 per set size for 80 total).

Participant Experience

Trial Sequence

1. Main Instructions: Explains the task (remember colored shapes, detect changes)
2. (Optional) Practice Instructions: If practice enabled
3. (Optional) Practice Trials: With visual feedback (checkmark/X)
4. (Optional) Trials Instructions: Before main trials begin
5. Main Trials: Each trial follows:
   • Fixation cross appears
   • Memory array appears (multiple colored shapes)
   • Blank screen (retention interval)
   • Test array appears
   • Participant responds "Same" or "Different"
   • Next trial begins

Response Methods

Buttons (on-screen):

• Click "Same" if test array matches memory array
• Click "Different" if anything changed

The task is self-paced unless strictly moderated. Participants can take time to decide, though faster responses within a few seconds are typical.

Data Output

Markers and Responses

The task records high-resolution timestamps in two separate collections. The stimulus_shown marker is emitted twice per trial: once for the memory array and once for the test array.

Markers (stimulus_shown):

{
  "type": "stimulus_shown",
  "ts": "2024-01-01T00:00:01.000Z",
  "hr": 1234.56,
  "data": {
    "trial_index": 1,
    "stimulus_id": "vstm_0_1",
    "set_size": 4,
    "condition": "different",
    "block": "high_load",
    "is_practice": false
  }
}

Response Data:

{
  "trial_index": 1,
  "stimulus_id": "vstm_0_1",
  "source": "button",
  "set_size": 4,
  "condition": "different",
  "response_value": "different",
  "response_correct": true,
  "is_practice": false,
  "block": "high_load",
  "latency_ms": 1234
}

Summary Artifact

A JSON file (vstm_summary_<taskIndex>.json) with capacity estimates and aggregated statistics:

{
  "task_kind": "vstm",
  "task_index": 0,
  "total_trials": 30,
  "overall": {
    "total": 30,
    "valid_responses": 28,
    "correct": 24,
    "accuracy": 0.857,
    "mean_rt_ms": 1234,
    "timeouts": 2
  },
  "by_set_size": {
    "2": { "total": 10, "accuracy": 0.95 },
    "4": { "total": 10, "accuracy": 0.85 },
    "6": { "total": 10, "accuracy": 0.70 }
  },
  "capacity_estimates_k": {
    "2": 1.90,
    "4": 3.20,
    "6": 3.40
  },
  "trials": [ /* per-trial data */ ]
}

Key metrics:

• capacity_estimates_k: Cowan's K at each set size, where K = set_size x (hit_rate - false_alarm_rate)
• accuracy: Proportion of correct responses by set size
• by_set_size: Performance broken down by number of items in the display

Instructions

The task uses four types of instructions you can customize:

1. Main Instructions: Shown on a dedicated page before the task begins. Explain what participants will see and how to respond.

2. Hint Instructions: Available via the "?" button during the task. Provide a quick reminder of the task rules.

3. Practice Instructions (if practice enabled): Shown before practice trials. Explain that this is practice with feedback.

4. Trials Instructions (if practice enabled): Shown after practice, before main trials. Remind participants that feedback will no longer be shown.

All instructions support rich text formatting and can be customized for your study.

Design Recommendations

For Standard Capacity Assessment

• Set sizes: Test 3-5 different set sizes (e.g., 2, 4, 6, 8) to capture the full capacity curve
• Trials per set size: 15-20 trials for stable K estimates
• Encoding duration: 100-200ms prevents verbal recoding
• Retention interval: 900-1000ms is standard
• Trial balance: Equal "same" and "different" trials

For Clinical or Developmental Populations

• Longer encoding: 250-500ms helps populations with processing speed differences
• Smaller set sizes: Focus on 2-6 items if expecting lower capacity
• Larger items: Increase item size (80-100px) for easier viewing
• More practice: 8-12 practice trials to ensure task understanding
• Breaks: Allow rest between blocks for attention maintenance

Common Set Size Configurations

Purpose	Set Sizes	Rationale
Quick screening	4, 6	Efficient, detects most individual differences
Full capacity curve	2, 4, 6, 8	Standard research protocol
Developmental	2, 3, 4	Avoids floor effects in children
High performers	6, 8, 10, 12	Avoids ceiling effects

Common Issues and Solutions

Issue	Possible Cause	Solution
K estimates too high (>7)	Verbal recoding, task too easy	Shorten encoding to 100ms, use more set sizes
High false alarm rate	Participants don't understand "same" vs "different"	Improve instructions, add more practice
Ceiling effects	Set sizes too easy	Test higher set sizes (8, 10, 12 items)
Floor effects	Set sizes too hard or timing too fast	Use smaller set sizes (2, 3, 4), increase encoding duration
No performance difference across set sizes	Not enough trials per condition	Increase trials per set size to at least 15-20

See Also

• N-back Task - Working memory task with updating demands
• Digit Span Task - Verbal working memory assessment