Change Detection Task

Background

The Visual Change Detection Task (CDT) is a classic psychological task used to measure Visual Working Memory (VWM) capacity. It quantifies the exact limit of visual information your brain can temporarily store, maintain, and manipulate after the physical objects disappear from view.

One of the earliest researchers to use a CDT-ish setup was W. A. Phillips. In his 1974 study, he presented participants with checkerboard-like patterns of square matrices and discovered that introducing a brief blank interval (a retention delay) between the first and second matrix caused a severe drop in a person's ability to spot changes. The test was further modified in 1997 by Steven J. Luck and Edward K. Vogel, who incorporated simple colored squares into their CDT procedure to strip away verbal memory and isolate visual capacity. Their work universally popularized the task across modern neuroscience to show that visual working memory is limited to about 3 to 4 items.

The Millisecond CDT is based on the published procedure by Anthony M. Harris and colleagues from 2020. While Luck and Vogel used a whole-display recognition layout (they presented 'clones' of the original layout or simply changed one color), Harris and colleagues used a partial-report single probe procedure: after presenting an array of colorful squares, they randomly presented only a single square and asked if that square was of the same color as the previous array square located in the same position. The advantage of the single-probe procedure is that it isolates the pure capacity of visual working memory storage.

Task Procedure

Millisecond's CDT tests 3 array sizes (4,6,8) in a blocked design. By default, each participant starts with the smallest array size (4) and moves up to the largest (8). Each block runs a sequence of 6 practice trials to familiarize participants with the procedure, followed by a fixed sequence of 60 test trials. By default, the procedure runs a fixed sequence of square positions. The square colors are sampled randomly for each trial from a list of 8 possible colors. All trials provide performance feedback. Each trial presents an array of colorful squares (each one has a different color) for 150ms on a gray background before the squares get erased. After a delay of 900ms (the retention interval), the single probe square appears. By default, the color of the probe is either the color of the square presented in the same position before (50% of trials) or has the color of an array square that was presented in a different position (50% of trials). Participants are asked to press one response key ("S") if the square color stayed the same and a different ("D") if the square color changed.

What it Measures

The Visual Change Detection Task (CDT) measurs Visual Working Memory (VWM) capacity.

Psychological Domains

• Visual Working Memory: Ability to keep all the relevant visual information in working memory
• Selective Attention: Ability to actively filter out background noise and focus purely on the target items during the brief initial flash.
• Feature Binding: Ability to bind sensory features (like a specific color and a specific shape) together into a single, cohesive mental object.
• Iconic and Sensory Memory: The retinal afterimage before internal mental representations take over.

Main Performance Metrics

• K-Value: Main measure of individual's Visual Working Memory capacity, calculated based on hit rate and correct rejections
• Signal Detection Measures: measures to assess sensitivity to change and general response style

Psychiatric Conditions

The following patient groups show impaired performance on visual CDT:

• Schizophrenia
• Attention-Deficit/Hyperactivity Disorder (ADHD)
• Major Depressive Disorder (MDD)
• Mild Cognitive Impairment (MCI) and Alzheimer's Disease (AD)
• Traumatic Brain Injury (TBI)
• Stroke

References

Search Google Scholar for peer-reviewed, published research using the Inquisit Change Detection Task.

---

Harris, A.M et al (2020). Behavioral and electrophysiological evidence for a dissociation between working memory capacity and feature-based attention. Cortex, 129, 158-174.