Our surroundings often appear stable and continuously available. We may feel that we can immediately notice when an object appears, disappears, or changes position. Research on visual attention, however, shows that this impression can be misleading.
When two versions of a scene are separated by even a brief visual interruption, people can fail to notice surprisingly large changes. This phenomenon is known as change blindness.
Pitch Glitch translates this phenomenon into a focused visual task. Several footballs first appear on the pitch. After a short interruption, the scene returns with an additional ball. The player must compare the two versions and identify which ball was added.
A sudden change normally produces a visual signal that automatically attracts attention. If an object appears while a person is looking at a continuous scene, the visual system can often detect the local movement or flicker created by that appearance.
A brief blank screen interrupts this signal. Instead of simply reacting to movement, the observer must remember the first scene and compare it with the scene shown afterward.
Successful detection therefore requires several processes:
A failure at any stage can make an obvious change difficult to notice.
In a foundational study, Ronald Rensink, Kevin O’Regan, and James Clark presented participants with alternating versions of real-world scenes. An original image and a modified version were separated by an 80-millisecond blank screen.
Each image contained one change involving the colour, position, appearance, or disappearance of an object. Participants had to identify what had changed.
Even though the images remained available through repeated alternations, some changes required many cycles to detect. Changes involving objects of marginal interest took an average of 17.1 alternations, or approximately 10.9 seconds, to identify. Changes involving central objects were found more quickly, requiring an average of 7.3 alternations, or about 4.7 seconds.
When the researchers removed the blank interruption, the changes were detected after only 1.4 alternations on average. This indicated that the changed objects were not simply too small or visually unclear. The interruption itself prevented the change from automatically capturing attention.
The study also demonstrated that not every part of a scene receives equal processing.
Objects that were important to the meaning of the scene were generally detected more quickly than marginal details. This suggests that attention helps determine which objects are encoded strongly enough to be compared after an interruption.
A person may look directly at a complex scene without creating a complete internal copy of it. Instead, the visual system prioritises selected objects and general structure. Details that receive less attention may be represented only partially or may not remain available long enough for a successful comparison.
Change detection therefore depends not only on eyesight but also on selective attention and visual memory.
Later research separated the change-detection process into two related components.
The first is the quality of the visual representation created before the interruption. If an object’s colour, position, or identity was not encoded clearly, the observer has little information with which to compare the second scene.
The second is the comparison process itself. Even when an object has been encoded, the observer must compare it with the object appearing in the corresponding position afterward.
A 2019 study asked participants to view alternating photographs of indoor scenes separated by blank screens. In one version of each photograph, an object such as a vase had been removed. Participants used a mouse to click on the object that repeatedly appeared and disappeared. The researchers found stable individual differences in people’s ability to detect these changes, connecting performance to the strength and stability of their visual representations.
An especially relevant experiment was conducted by Gaspar and colleagues in 2013.
In their change-detection training task, participants first saw a display containing three or five objects. A visual mask then appeared for 500 milliseconds. After the mask, participants saw a modified display in which one of the original objects had been replaced by a new object. They had to use the mouse to select the changed object.
The researchers tested 40 younger adults and 40 older adults. Participants in the training group completed 16 sessions involving objects from categories such as cars, signs, shapes, and letters. The duration of the first display was adjusted to maintain approximately 75% accuracy.
Both younger and older participants became faster at encoding the objects used during training. However, the improvement did not transfer reliably to tasks using unfamiliar objects or real-world driving scenes. The benefit was largely limited to the trained task and materials.
Pitch Glitch follows the central structure of a laboratory change-detection task:
The footballs act as individual visual objects distributed across the pitch. When the scene disappears, their locations must be temporarily maintained in visual working memory. Once the pitch returns, the player must compare the new arrangement with the remembered one and select the ball that does not match the original representation.
The interruption is therefore not merely a visual effect. It removes the motion signal that would otherwise reveal the added object automatically and turns the task into a comparison between memory and current perception.
Research on change blindness suggests several ways in which a Pitch Glitch round can become easier or more difficult.
A larger number of footballs increases the amount of visual information that must be encoded. Balls appearing close together may make individual locations more difficult to distinguish. A shorter initial display provides less time to inspect the arrangement, while a longer interruption may place greater demands on visual memory.
The position of the new ball can also matter. A ball appearing near the centre of attention may be detected more quickly than one added to a less noticeable part of the pitch.
Finally, similarity influences comparison. If all footballs look similar, location becomes the main identifying feature. If some balls differ in colour or design, these additional visual features may help organise the scene.
The cited studies did not investigate Pitch Glitch itself. They demonstrate how interruptions, selective attention, object familiarity, and visual memory influence change detection.
They do not prove that playing Pitch Glitch improves general visual attention or produces benefits that transfer to unrelated real-world activities. In fact, the findings of Gaspar and colleagues indicate that improvement in change-detection tasks can remain specific to the objects and task being practised.
A controlled study using Pitch Glitch would be required to establish whether repeated gameplay produces measurable training effects.
The scientifically supported connection is structural: Pitch Glitch uses the same sequence of visual encoding, interruption, comparison, and change localisation examined in experimental change-detection research.
Change blindness shows that seeing a scene does not mean that every object within it has been stored in detail. When visual continuity is interrupted, identifying a new or missing object requires attention, visual working memory, and a successful comparison between the scene before and after the interruption.
Pitch Glitch places this process at the centre of its gameplay. The player must remember the original distribution of footballs and identify the object introduced after a brief visual break.
What appears to be a simple search for a new ball is therefore a compact demonstration of how the brain detects change—and why even visible differences can sometimes remain unnoticed.
Rensink, R. A., O’Regan, J. K., & Clark, J. J. (1997). To See or Not to See: The Need for Attention to Perceive Changes in Scenes. Psychological Science, 8(5), 368–373.
https://www2.psych.ubc.ca/~rensink/publications/download/PsychSci97-RR.pdf
Andermane, N., Bosten, J. M., Seth, A. K., & Ward, J. (2019). Individual differences in change blindness are predicted by the strength and stability of visual representations. Neuroscience of Consciousness, 2019(1), niy010.
https://doi.org/10.1093/nc/niy010
Gaspar, J. G., Neider, M. B., Simons, D. J., McCarley, J. S., & Kramer, A. F. (2013). Change Detection: Training and Transfer. PLOS ONE, 8(6), e67781.