define change blindness

• Social Anxiety Disorder
• Bipolar Disorder
• Kids Mental Health
• Therapy Center
• When To See a Therapist
• Types of Therapy
• Best Online Therapy
• Best Couples Therapy
• Best Family Therapy
• Managing Stress
• Sleep and Dreaming
• Understanding Emotions
• Self-Improvement
• Healthy Relationships
• Relationships in 2023
• Student Resources
• Personality Types
• Verywell Mind Insights
• 2023 Verywell Mind 25
• Mental Health in the Classroom
• Editorial Process
• Meet Our Review Board
• Crisis Support

Why Change Blindness Happens

Examples and Coping Strategies

Kendra Cherry, MS, is a psychosocial rehabilitation specialist, psychology educator, and author of the "Everything Psychology Book."

Emily is a board-certified science editor who has worked with top digital publishing brands like Voices for Biodiversity, Study.com, GoodTherapy, Vox, and Verywell.

Tetra Images/Getty Images

• In the Real World

If something in your visual field were to change dramatically before your eyes, you would notice it immediately, right? Not necessarily. Your surroundings simply offer too much information for your brain to fully process, so you might miss even large changes. This phenomenon is known as change blindness. Here's why change blindness happens.

Change blindness is "a phenomenon of visual attention in which changes to a visual scene may go unnoticed under certain circumstances, despite being clearly visible and possibly even in an attended location."

Research on Change Blindness

The idea of change blindness isn't new; in fact, researchers have studied it for decades. The ability to detect change plays an important role in daily life—helping you notice when a car drifts into your lane or a person enters a room. Many fascinating experiments have explored aspects of this phenomenon, such as:

• How changing the direction of a stimulus induces change blindness, which might explain how many magic tricks work.
• Whether people noticed—or did not notice—changes in type, distance, complexity, and field of view in an immersive environment. They found that people with good working memories were more apt to notice them than were their more forgetful counterparts.
• How strength and stability of a stimulus can help you perceive change
• Whether a brief break between one version of an image and another would make changes more difficult to detect.
• Whether participants would notice that their conversation partners had been swapped out for others after a period of distraction.

Causes of Change Blindness

If the ability to perceive change is so important, why do humans often fail to notice even major ones? Researchers have a few ideas.

Focused Attention and Limited Resources

At this moment, your attention is focused on the words you are reading. Are you giving any attention to the color of the nearby wall? Are you aware of the position of your feet? Until now, you probably weren't paying attention to either of those things.

According to researchers Daniel Simons and Daniel Levin, that's because your limited capacity for attention forces you to choose what to focus on. Lots of information simply passes us by because we lack the resources to attend to it.

Expectations and Experiences

Certain changes—particularly those that are artificially produced in an experimental lab—tend to go unnoticed because they're unexpected. How often does a person suddenly turn into someone else, an object suddenly blink into existence, or a person's shirt change color? These things simply don't happen, so they're typically overlooked when they're staged for an experiment.

One reason people think they would see the changes may be that they know from past experiences that changes that occur in real life are usually easy to see. But there is an important difference between changes that occur in real life and those that occur in change detection experiments. Changes that occur in real life are often accompanied by a motion, which provides a clue that indicates a change is occurring.

Other Factors

Other factors that can influence change blindness include attention , age, presentation, and the use of psychoactive drugs . Researchers have also found that distraction increases change blindness.

Age can also play a role: studies have found that older people are less likely to detect changes in a visual scene than younger people. Again, the ability to take in visual information is constrained by limited resources.

The basic problem is that far more information lands on your eyes than you can possibly analyze and still end up with a reasonable-sized brain .

To cope with an overwhelming amount of data, you focus on a single part of the environment that you deem important enough to process.

Change Blindness in the Real World

Change blindness might cause problems in real-world situations, such as:

• Air Traffic Control. A disaster could result if an air traffic controller failed to detect changes when monitoring takeoffs, landings, and flight paths.
• Driving. Failure to detect changes in the environment while you are driving can lead to dire consequences. Distractions such as talking on the phone or texting while you drive can decrease attention and increase change blindness.
• Eyewitness Testimony. Change blindness can affect an eyewitness's ability to recount the details of a crime or to correctly identify the perpetrator.
• Social Interactions. Change blindness can affect day-to-day social interactions—for example, asking the wrong waiter for the check when you're dining out.

A Word From Verywell

Humans rely on their ability to detect change. Yet the human brain sometimes lacks sufficient resources to focus on all the details when so much information floods it at any given moment. So, it directs attention to the most important stimuli and lets the rest go. The result: Change blindness.

Herbranson WT. Change blindness . In: Vonk J, Shackelford T, eds.  Encyclopedia of Animal Cognition and Behavior . Springer International Publishing; 2019:1-4.

Yao R, Wood K, Simons DJ. As if by magic: An abrupt change in motion direction induces change blindness .  Psychol Sci . 2019;30(3):436-443. doi:10.1177/0956797618822969

Martin D, Sun X, Gutierrez D, Masia B. A study of change blindness in immersive environments .  IEEE Trans Visual Comput Graphics . 2023;29(5):2446-2455. doi:10.1109/TVCG.2023.3247102

Andermane N, Bosten JM, Seth AK, Ward J. Individual differences in change blindness are predicted by the strength and stability of visual representations .  Neuroscience of Consciousness . 2019;2019(1).

Blackmore SJ, Brelstaff G, Nelson K, Trościanko T. Is the richness of our visual world an illusion? Transsaccadic memory for complex scenes . Perception . 1995;24(9):1075-1081. doi:10.1068/p241075

Simons D, Levin D. Failure to detect changes to people during a real-world interaction .  Psychon Bull Rev . 1998;5(4):644-649. doi:10.3758/bf03208840

Goldstein E, Brockmole J.  Sensation and Perception . 10th ed. Independence: Cengage; 2017.

Costello MC, Madden DJ, Mitroff SR, Whiting WL. Age-related decline of visual processing components in change detection . Psychol Aging . 2010;25(2):356-368. doi:10.1037/a0017625

Angier N. Blind to change, even as it stares us in the face .  The New York Times . Published April 1, 2008.

Romer D, Lee Y, McDonald C, Winston F. Adolescence, attention allocation, and driving safely . Journal of Adolescent Health . 2014;54(S5):S6-S15. doi:10.1016/j.jadohealth.2013.10.202

Nellson KJ, Laney C, Fowler NB, Knowles ED, Davis D, Loftus EF. Change blindness can cause mistaken eyewitness identification . Legal and Criminological Psychology. 2011;16(1):62-74. doi:10.1348/135532509X482625

By Kendra Cherry, MSEd Kendra Cherry, MS, is a psychosocial rehabilitation specialist, psychology educator, and author of the "Everything Psychology Book."

By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts.

• Reference Manager
• Simple TEXT file

People also looked at

Original research article, a comparison of change blindness in real-world and on-screen viewing of museum artefacts.

• 1 NeuroMetrology Group, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom
• 2 Medical Sciences Division, University of Oxford, Oxford, United Kingdom
• 3 Ashmolean Museum Engagement Programme, Ashmolean Museum of Art and Archaeology, University of Oxford, Oxford, United Kingdom
• 4 Department of Experimental Psychology, University of Oxford, Oxford, United Kingdom

Change blindness is a phenomenon of visual perception that occurs when a stimulus undergoes a change without this being noticed by its observer. To date, the effect has been produced by changing images displayed on screen as well as changing people and objects in an individual’s environment. In this experiment, we combine these two approaches to directly compare the levels of change blindness produced in real-world vs. on-screen viewing of museum artefacts. In the real-world viewing condition, one group of participants viewed a series of pairs of similar but slightly different artefacts across eye saccades, while in the on-screen viewing condition, a second group of participants viewed the same artefacts across camera pans on video captured from a head-mounted camera worn by the first set of participants. We present three main findings. First, that change blindness does occur in a museum setting when similar ancient artefacts are viewed briefly one after another in both real-world and on-screen viewing conditions. We discuss this finding in relation to the notion that visual perceptual performance may be enhanced within museums. Second, we found that there was no statistically significant difference between the mean levels of change blindness produced in real-world and on-screen viewing conditions (real-world 42.62%, on-screen 47.35%, X 2 = 1.626, p > 0.05 1 d.f.). We discuss possible implications of these results for understanding change blindness, such as the role of binocular vs. monocular vision and that of head and eye movements, as well as reflecting on the evolution of change detection systems, and the impact of the experimental design itself on our results. Third, we combined the data from both viewing conditions to identify groups of artefacts that were independently associated with high and low levels of change blindness, and show that change detection rates were influenced mainly by bottom-up factors, including the visible area and contrast of changes. Finally, we discuss the limitations of this experiment and look to future directions for research into museum perception, change blindness, real-world and on-screen comparisons, and the role of bottom-up and top-down factors in the perception of change.

Introduction

Change blindness is defined as the failure to detect when a change is made to a visual stimulus ( Simons and Levin, 1997 ). It occurs when the local visual transient produced by a change is obscured by a larger visual transient, such as an eye blink ( O’Regan et al., 2000 ), saccadic eye movement ( Grimes, 1996 ; McConkie and Currie, 1996 ), screen flicker ( Rensink et al., 1997 ), or a cut or pan in a motion picture ( Simons, 1996 ; Levin and Simons, 1997 ); or when the local visual transient produced by a change coincides with multiple local transients at other locations, known as mud-splashes, which act as distractions, causing the change to be disregarded ( O’Regan et al., 1999 ). Change blindness is distinct from inattentional blindness, which occurs when an individual is blind to the presence of an entire object while performing a distracting task [as in the well-known “gorilla in the room” experiment ( Simons and Chabris, 1999 )]. In contrast, change blindness occurs when an individual is blind to changes occurring to an object with which they are actively engaged. Because of this, when missed changes are later pointed out to the observer, they are usually met with a sense of disbelief at how something could ever have been missed. The surprising nature of change blindness results from a disconnect between the assumption that our visual perceptions are so detailed as to be virtually complete, and the actual ability of the visual system to represent and compare scenes moment-to-moment. In this way, change blindness is a testable phenomenon that can be used to investigate the nature of visual representations in different conditions ( Simons and Rensink, 2005 ).

In most of the studies published to date, change blindness has been produced using altered photographs or videos of natural scenes displayed on computer screens. More recently, change blindness has also been shown to take place in more naturalistic scenarios. For example, in one real-world experiment, more than half of participants failed to notice the changing of a conversation partner in front of them ( Simons and Levin, 1998 ; Levin et al., 2002 ), and in another, more than half of participants were blind to the changing of an object’s colour or a printed word’s font ( Varakin et al., 2007 ).

In the current experiment, we sought first to demonstrate whether change blindness could be produced inside a museum, using ancient museum artefacts as visual stimuli. It has been suggested that the visual interactions taking placed within museums involve enhanced perception compared to everyday visual interactions ( O’Neill and Dufresne-Tassé, 1997 ), raising the question of whether change blindness is still a demonstrable phenomenon under such conditions. Inattentional blindness has been previously investigated in a museum setting ( Levy, 2011 ), but as far as we are aware this is the first attempt to produce change blindness inside a museum.

Once it has been produced, we will directly compare the levels of change blindness produced by a single set of visual stimuli viewed in both on-screen and real-world conditions. In the real-world condition, one group of participants viewed a series of pairs of similar but slightly different artefacts across eye saccades, while in the on-screen condition, a second group of participants viewed the same series of artefacts across camera pans on video captured from a head-mounted camera worn by the first set of participants. It is important to know whether or not this shift to more on-screen interaction has negative consequences such as increased change blindness. To the best of our knowledge, this is the first attempt to directly compare change blindness levels produced in on-screen and real-world viewing conditions.

Our motivation for making this comparison was twofold. First, as a response to the relative lack of comparisons between on-screen and real-world perception made to date, despite the extensive use of both conditions across human visual perception research. Because non-stereoscopic cameras capture and display light from a single perspective, on-screen viewing conditions provide only monocular cues to visual depth. These depth cues include linear perspective, object occlusion, and motion parallax ( Cutting, 1997 ; Albertazzi et al., 2010 ). By contrast, because in real-world viewing conditions light reflected from the three-dimensional environment is captured from the perspective of both eyes without passing through a camera, binocular depth cues, including binocular disparity and ocular convergence, become available in addition to the monocular cues. There is evidence to suggest that binocular stereoscopic vision confers an advantage over monocular vision in certain perception performance tasks, including the analysis of complex visual scenes ( Jones and Lee, 1981 ), surface visualisation ( Wickens et al., 1994 ), and the programming of prehensile movements ( Servos et al., 1992 ). However, evidence of preserved function without stereopsis also exists, most notably amongst pilots ( Snyder and Lezotte, 1993 ), and the overall functional significance of binocular stereopsis remains unclear ( Fielder and Moseley, 1996 ). Based on this evidence and our own observations, our hypothesis is that change blindness levels will be lower in the real-world condition than in the on-screen condition, because the perceptual advantages of binocular over monocular vision will produce a greater rate of change detection in the real-world scenario.

We were also motivated to make this comparison by the increasing frequency and importance of on-screen visual interactions alongside real-world interactions in modern working and social life. The growing accessibility of high-speed internet and the capability of smart portable devices has already significantly changed the way that many people exchange visual information. A recent report found that adults in the United States spend an average of more than 8 h a day accessing media through a device with a screen ( The Nielsen Total Audience Report - Q1 2016, 2016 ). For many people, this amount of time will account for the majority of their waking day and such a significant shift in behaviour warrants further investigation in its own right.

Materials and Methods

We recruited 62 participants through an advertisement describing a neuropsychological experiment taking place at the Ashmolean Museum in Oxford. The group of participants consisted of students and employees of the University of Oxford, covering a wide range of disciplines from Art History and Fine Art to Law and Medicine. While none of the participants were artists, they might all be considered to hold some form of interest in art, or art history, given that they responded to our advert. The participants were allocated using a random number generator to either real-world or on-screen viewing conditions. 31 participants were allocated to each group. The mean age of participants in the real-world group was 22.8 years (SD ± 5.3 years) and 58.1% were female. The mean age of participants in the on-screen group was also 22.8 years (SD ± 5.9 years) and 58.1% were female. No attempt was made to match the groups. The exact sex matching occurred by chance. The close age matching results from the participants predominantly being university students. This study was carried out with permission from the Central University Research Ethics Committee (CUREC), and all subjects gave their written informed consent after the experimental procedures had been explained to them, in accordance with the Declaration of Helsinki.

Museum Setting

The experiment was conducted in the Ashmolean Museum of Art and Archaeology, part of the University of Oxford. Twelve pairs of artefacts from the museum’s collection were used, including three pairs of Japanese woodblock prints, one pair of Chinese porcelain bowls, two pairs of Iranian tiles, one pair of Athenian lekythoi, one pair of Renaissance bronze medals, two pairs of Anglo-Saxon brooches, and two pairs of English silverware. These artefacts were chosen because although they had originally been designed to appear identical in their pairs, through their individual manufacture and subsequent usage they had all come to exhibit differences, ranging from relatively subtle to more major differences in appearance, including differences in colour, shape, and design. There were differences between all 12 pairs of artefacts used in the experiment.

Change Blindness Paradigm

Twelve pairs of artefacts were displayed in a fixed order before each participant. For each pair of artefacts, a participant observed one item for a short period of time before looking to the second item and observing it for the same length of time as the first. As participants looked from one item to the next, the differences between their appearances generated local visual transients. However, the transition of looking from one item to the other generated a larger visual transient which would to a certain extent obscure the local transients, and thus produce a corresponding degree of change blindness. This degree was measured by participants responding to the question: Did you notice any differences between the two objects? They were then required to describe any differences they did notice in writing after viewing each pair of artefacts. Subsequently, the participants’ descriptions were marked as either correct or incorrect according to the actual differences manifest between the objects. If none of the changes existing between a pair of artefacts were correctly identified, the participant was recorded as being change blind with respect to that pair. If a single change was correctly identified, they were recorded as not being change blind. The degree of change blindness recorded was therefore a reflection of the balance of local and large visual transients that were produced by observing these pairs of museum artefacts in real-world and on-screen viewing conditions.

The length of time for which participants observed each artefact was set at a duration that would produce a change blindness effect appropriate to allow for a comparison to be made between the two conditions. The requisite duration was determined through a series of trials in which photographs of the pairs of artefacts were observed in series on a monitor for different lengths of time. An observation time of 5 s per artefact separated by an interval of 2 s resulted in change blindness in 15% of the pairs. Observation time of 2 s with an interval of 0.5 s produced change blindness in 20%, and an observation time of 0.25 s with an interval of 0.25 s produced 57% change blindness. Given that the motion of turning to look from one artefact to another would produce an interval between fixations of less than 100 ms ( Grossman et al., 1988 ), an observation time of 1 s was chosen in order to achieve approximately 50% mean change blindness in the on-screen condition. This was thought to be optimal in allowing for a comparison to be made between this and the real-world condition.

Both viewing conditions were similarly controlled to standardise the nature and duration of the periods of observation, and the transition from one artefact to another. The artefacts were placed in their pairs on a table in a room within the museum (Figure 1A ). They remained covered for the majority of the experiment, and members of museum staff were present to ensure their safekeeping throughout. The items in each pair were placed 40 cm apart, and a chair was placed in front of each pair of artefacts to provide a viewing distance of 75 cm. A high definition 32-inch LCD screen was also present in the room with a chair placed in front of it. The real-world viewing condition consisted of participants sitting in front of and viewing the artefacts on the table before them (Figure 1B ). The on-screen condition consisted of a separate group of participants sitting in front of the screen and viewing the artefacts on its display (Figure 1C ). Both participants were aware of each other and their roles throughout the course of the experiment.

FIGURE 1. (A) The experimental setup within the museum, showing the artefacts (covered), two participants, and an experimenter. (B) The real-world viewing condition: the participant is sat in front of a pair of artefacts, wearing a pair of modified goggles and head-mounted camera. (C) The on-screen viewing condition, the participant is sat in front of a monitor, wearing a pair of modified goggles and watching a live feed from the head-mounted camera. Images reproduced with permission from Ashmolean Museum, University of Oxford. All the persons depicted on this picture gave their consent for publication.

All participants’ visual fields were restricted by wearing a pair of goggles that were modified for the purposes of this study. Opaque inserts were fitted to the inside of the goggles to leave a window of 3 cm diameter in front of each eye. This restricted the binocular field of view to 45.56° (0.79°rad) horizontally and 48.14° (0.84 rad) vertically at the 75 cm viewing distance. The field was sufficient to contain the full surface of the largest artefact while also not allowing both of the smallest artefacts to be viewed when the visual field was centred on one of them, in both the real-world (Figure 2A ) and on-screen conditions (Figure 2B ). These steps were taken to ensure that participants would not be able to make multiple eye saccades between the items in front of them, which would have added a significant uncontrolled variable.

FIGURE 2. (A) The real-world participants’ views of the largest (top) and smallest (bottom) artefacts through the modified goggles. The whole surface of the largest artefact was visible, but both items of the smallest pair of artefacts were not visible at the same time (to scale). (B) The on-screen participants’ views of the largest (top) and smallest (bottom) artefacts through the modified goggles on the screen (to scale). Images reproduced with permission from Ashmolean Museum, University of Oxford.

Real-World Viewing Condition

Once sat in front of the first pair of artefacts, the real-world participant was instructed to start with their head toward the item on their left, so that their visual field would be centred on the first artefact. The artefact was initially obscured by a small screen. On an auditory cue the screen was manually removed by an experimenter so that the participant could view the first artefact. This period of observation lasted for 1 s, after which another cue sound signalled for the participant to turn their head and eyes to look at the second item to their right, so that their visual field would now be centred on the second artefact. This period of observation lasted for a further 1 s, after which a small screen was placed between the participant and the second item by an experimenter so that it could no longer be seen. In this way, both artefacts were viewed for a duration of 1 s, with a brief visual transition interrupting the viewings.

The visual transition which occurred in the real-world viewing condition consisted of a combination of a head rotation and a saccadic eye movement. This combination has been defined elsewhere as a gaze shift ( Binder et al., 2009 ), where gaze is defined as the sum of eye position with respect to the head and head position with respect to the body. When the visual field shifts more than 15–20°, an eye saccade is normally accompanied by a head rotation in order to return the eyes to a neutral position within the orbits and allow the extra-ocular muscles to relax. In this case, the shift was 28.1° (0.49 rad), and participants in the real-world condition were specifically instructed to turn both their head and eyes to view the second artefact in each pair.

The coordination of gaze shifts is complex but the basic elements are well-understood ( Pelisson and Guillaume, 2009 ). As the head initially rotates and the eyes stay fixed on the first target, eye movement is under the control of the vestibulo-ocular reflex (VOR). Once head rotation has brought the new target into the visual field, an endogenous eye saccade occurs to move the point of foveation from the first target to the second. Following this, though the second target is now foveated, there is still residual head rotation due to a lag in the control of head movement relative to that of the eyes, and this is compensated for by a further period of VOR eye movement. The components of the gaze shift are therefore an initial period of VOR, an exogenous eye saccade, followed by a further period of VOR. It is not yet known whether VOR eye movements are able to induce change blindness by themselves, but that eye saccades are able to is well-established ( Grimes, 1996 ; McConkie and Currie, 1996 ). Thus, in the real-world viewing condition in this experiment, the large visual transient consisted of an eye saccade which was preceded and followed by a period of VOR eye movement.

On-Screen Viewing Condition

While the above processes were taking place, a small head-mounted high definition video camera was attached to the goggle strap of the participant in the real-world viewing condition. The camera used was a Contour+2 HD with 170° wide-angle lens, operating at a frame rate of 30 fps and 1920 × 1080 resolution, weighing 156 g, and measuring 98 mm × 60 mm × 34 mm. It was connected by an HDMI cable to 1080p high definition 32-inch LCD screen, producing a live video feed on the screen in front of the participant in the on-screen condition. The acuity achievable when viewing this screen was 20/70, which, although inferior to 20/20 vision, was significantly greater than the level required to resolve the smallest change detected by any participant in the real-world condition, which was measured to be 20/180 (a change of 2 mm diameter viewed at 75 cm). The on-screen participant wore an identical pair of modified goggles to their counterpart in the real-world group (except without a camera attached to the goggle strap), which, as in the real-world group, prevented multiple eye saccades being made between artefacts.

Unlike participants in the real-world viewing condition, however, on-screen participants did not have to follow instructions to move their head or eyes on auditory cues. Instead, as the real-world participant rotated their head to look from the first item to the second, the head-mounted camera also rotated and the footage on the screen panned across to reveal the second artefact to the on-screen participant. An equivalent change to the contents of the visual field was therefore produced without an equivalent gaze shift taking place. Thus, in the on-screen viewing condition, the large visual transient consisted of a camera pan rather than an eye saccade preceded and followed by a period of VOR. Of course, the other difference between viewing conditions was that artefacts were viewed directly by participants in the real-world group, while they were viewed on an LCD display in the on-screen group. On-screen participants viewed the screen from a distance of 75 cm, and the camera and screen were calibrated so that the representations of the artefacts were displayed at life-size in order to match conditions in the real-world conditions. In both real-world and on-screen viewing conditions, therefore, the artefacts subtended the same visual angle.

The only differences between the conditions, then, were the nature of the large visual transient and the format of display. We suggest that these variables constitute the defining differences between all real-world and on-screen visual interactions, in that they represent both the behaviour of the subject who is viewing and the nature of the object that is being viewed in these scenarios. Thus, the results of this experiment reflect a comparison of the levels of change blindness produced by a single set of visual stimuli in real-world and on-screen viewing, as defined by the nature of the large visual transient and the format of display typical of these conditions.

We present three main findings. First, that change blindness does occur in a museum setting when similar ancient artefacts are viewed briefly one after another in both real-world and on-screen viewing conditions (Table 1 and Figure 3 ).

TABLE 1. Table of results.

FIGURE 3. Levels of change blindness in real-world and on-screen viewing conditions produced by each pair of artefacts and the overall mean. Asterisks denote level of significance ( X 2 test with one degree of freedom. No asterisk = p > 0.05; ∗ = 0.05 > p > 0.02; ∗∗ = 0.02 > p > 0.01; ∗∗∗ = 0.01 > p > 0.001; ∗∗∗∗ = 0.001 > p ).

Second, we found that there was no statistically significant difference between the mean levels of change blindness produced in real-world and on-screen viewing conditions [real-world 42.62%, on-screen 47.35%, X 2 = 1.626, p > 0.05 1 d.f. (Table 1 and Figure 3 )]. The total number of trials per pair of artefacts ranged from 29 to 31 due to a small number of failures by participants to follow the experimental procedure described above (13 failures from 371 trials = 3.5%). The mean level of change blindness produced in the on-screen condition was close to 50%, as intended to allow comparison between the two conditions. One pair of artefacts produced a significantly higher degree of real-world change blindness than on-screen change blindness (Pair 2: real-world 86.7%, on-screen 46.7%, X 2 = 10.800, 0.01 > p > 0.001), while three pairs produced a significantly higher degree of on-screen change blindness than real-world change blindness (Pair 4: real-world 20.0%, on-screen 50.0%, X 2 = 5.934, 0.02 > p > 0.01; Pair 10: real-world 3.5%, on-screen 41.4%, X 2 = 11.997, 0.001 > p ; Pair 12: real-world 70.0%, on-screen 93.3%, X 2 = 5.455, 0.02 > p > 0.01). But in the other eight pairs, and overall, there was no significant difference between the levels of change blindness produced.

Third, following the finding of no significant difference between the levels of change blindness produced in real-world and on-screen conditions, we combined the data from both groups to compare the levels of change blindness produced by each pair of artefacts independently (Figure 4 ). From these results, we consider in particular three pairs of artefacts which produced a level of change blindness greater than 75% (pairs 1, 7, and 12, 79.31–83.33%), and three pairs of artefacts which produced a level of change blindness lower than 15% (pairs 3, 5, and 6, 4.84–12.90%).

FIGURE 4. Levels of change blindness combined from real-world and on-screen viewing conditions produced by each pair of artefacts and the overall mean.

Change Blindness in a Museum Setting

Our first finding, that change blindness does occur in a museum setting when similar ancient artefacts, in this case some more than 2,000 years old, are viewed briefly one after another in both real-world and on-screen viewing conditions, is a significant addition to the body of evidence demonstrating that change blindness can be produced in more naturalistic environments outside of the laboratory.

Change blindness experiments have established that details considered to be important are detected more readily than those that are less important ( Rensink et al., 1997 ; O’Regan et al., 2000 ), even when the changes are of equivalent physical salience ( Kelley et al., 2003 ). These findings suggest that attention plays an important role in prioritising the elements of a visual scene, and in determining what is represented and compared between scenes and what is not ( Simons and Rensink, 2005 ). However, even changes that are made to attended objects can still be missed ( Ballard et al., 1995 ; Simons, 1996 ), which leads to the conclusion that attention is necessary, but not sufficient, for change detection to occur. The other determinants of change detection can be divided between bottom-up, or stimulus-driven, factors, such as visual salience, and top-down, or goal-driven, factors, such as context, gist and motivation ( Borji and Itti, 2013 ). It has been suggested that both bottom-up and top-down factors are enhanced in the visual interactions that take place within museums, due to the exceptional and exemplary nature of the objects being viewed, and the intensity of observation and motivation to form interpretations from what is seen, respectively ( O’Neill and Dufresne-Tassé, 1997 ). However, later discussions have warned against ‘uncritical acceptance of the distinction between utilitarian (ordinary) and the aesthetic (museum) seeing,’ and while it is acknowledged that ‘the notion of the distinction…is, in one form or another, firmly embedded in many account of vision and aesthetic experience,’ in fact, ‘cognitive neuroscience does not supply any facts that could substantiate the sharp divide between the ‘normal’ and the aesthetic perception’ ( Kesner, 2006 ). In the present experiment, we did not compare the levels of change blindness produced within and outside of the museum setting. Instead, we have merely demonstrated that change blindness can be produced among participants viewing artefacts inside a museum. However, in light of this finding and the unresolved questions that surround it, such a comparison would be an appropriate next step.

Change Blindness in Real-World and On-Screen Viewing Conditions

Our second finding was that there was no statistically significant difference between the mean levels of change blindness produced in real-world and on-screen viewing conditions. This means that altering both the format of display from the objects themselves to an on-screen virtual object representation, and the nature of the large visual transient from a camera pan to an eye saccade preceded and followed by VOR, did not significantly affect the rate of change detection in this experiment.

The possible interpretations of this finding are: (1) that neither the format of display nor the nature of the large visual transient had a significant effect on change detection, (2) that the format of display and nature of the large visual transient had equal and opposite effects on change detection, resulting in no combined effect overall, or (3) that the similarities between the two conditions were so great compared to the differences, that any effects produced by either the format of display or the nature of the large visual transient were masked by the intrinsic design of the experiment.

Regarding the first part of the first interpretation, if the format of display had no significant effect on change detection, then this finding provides no support for our hypothesis, which was that the perceptual advantages of binocular stereoscopic vision would produce a greater rate of change detection in the real-world condition compared to the on-screen condition.

Our hypothesis was formulated based on evidence that binocular stereoscopic vision confers an advantage over monocular vision in certain perception performance tasks, including the analysis of complex visual scenes ( Jones and Lee, 1981 ), surface visualisation ( Wickens et al., 1994 ), and the programming of prehensile movements ( Servos et al., 1992 ). We also made reference to the fact that in the real-world condition, binocular stereopsis would provide additional depth cues of binocular disparity and ocular convergence, compared to in the on-screen condition where only monocular depth cues of linear perspective, object occlusion, and motion parallax would be available ( Cutting, 1997 ; Albertazzi et al., 2010 ). According to our first interpretation, this finding runs contrary to the evidence supporting our initial hypothesis. However, due to the equal plausibility of the other interpretations, we cannot reliably contrast our finding with those drawn elsewhere. As previously discussed, evidence of preserved visual performance without stereopsis does exist ( Snyder and Lezotte, 1993 ), and so the overall functional significance of binocular stereopsis must unfortunately remain unclear.

In the absence of clear evidence either way, it is interesting and perhaps instructive to consider the evolutionary arguments for why we might expect change detection to be enhanced by binocular stereoscopic vision. One could argue that a real-world object, with the potential to act upon its viewer and itself to be acted upon, should be perceived more strongly than an on-screen object, which ultimately remains virtual (although the screen which displays it is itself a real-world object). However, as some of the artefacts used in this experiment demonstrate, the human visual system has been processing two-dimensional representations of three-dimensional objects for thousands of years. The earliest cave paintings discovered date back to 15,000–10,000 BC. Human beings, and especially the human nervous system, have undergone significant changes over hundreds of generations in this time, but we reason that in this period there will have been no drive to either significantly strengthen or weaken the local visual transients formed from the observation of two-dimensional images relative to three-dimensional objects. The subjects of the earliest two-dimensional representations were bison, mammoth, and reindeer, the prey of those who depicted them on the walls of their dwellings. This alone is testament to the fact that the ability to create and understand representations of the surrounding environment and the messages being communicated about them is likely to have conferred a selective advantage over the recent course of our evolution. Indeed, while objects that can act upon us and that we can act upon have remained important for our survival, one can argue that images have come to be just as important to the modern human.

Returning to the second part of the first interpretation of our finding, it is possible that the nature of the large visual transient had no significant effect on change detection. We are not aware of any previous attempts to compare the effect of eye saccades and VOR vs. camera pans on visual performance. And, as above, due to the equal plausibility of the other interpretations, we cannot present this interpretation as a reliable conclusion. Theoretically, one could reasonably argue that activation of neural systems controlling head and eye movements might either enhance or impair the parallel systems involved in change detection. Once again, in the absence of evidence, we might have recourse to consider evolutionary arguments. However, in this situation this seems hardly relevant. Before motion pictures were developed at the turn of the twentieth century, the human visual system would never have been exposed to a change to the contents of the visual field in the absence of head or eye movements – such a thing would simply not have been possible. Consequently, there has been almost no time for natural selection to affect the mechanisms of change detection operating in the context of a camera pan compared to an eye saccade accompanied by VOR eye movement.

The second possible interpretation of our finding is that the format of display and nature of the large visual transient had equal and opposite effects on change detection, resulting in no combined effect overall. Because of the difficulties in drawing conclusions about either variable discussed above, the uncertainty of following this interpretation would be even greater, and as such need not be discussed further.

The third possible interpretation of our main finding was that the similarities between the two conditions were so great compared to the differences, that any effects produced by either the format of display or the nature of the large visual transient were masked by the intrinsic design of the experiment. In any experiment, the pattern of findings will be determined by a balance between both the controls and the variables that constitute the experimental paradigm ( Gozli, 2017 ). Our paradigm included a relatively large number of controls and restrictions: a fixed viewing time, a restricted field of view, proscribed head and eye movements, and a specific set of visual stimuli. It was necessary to institute these limitations to reliably isolate our two experimental variables from a complex naturalistic scenario. However, it is possible that, such was the impact of these controls relative to the difference between the experimental variables, that our two viewing conditions were in effect much more similar than they were different. In this way, it is possible that the similarity in task performance across the two conditions could have masked effects produced by the differences between on-screen and real-world viewing conditions. It is perhaps not possible to determine to what degree any effects may have been masked. However, with this in mind, we can only state that the differences between on-screen and real-world viewing conditions were not large enough to produce a significant difference in participant performance in the context of this experiment.

In summary, then, it is difficult to interpret our finding that there was no statistically significant difference between the mean levels of change blindness produced in real-world and on-screen viewing conditions. The effects of altering the format of display and the nature of the visual transient in this experiment cannot be separated, the possibility of equal and opposite effects cannot be excluded, and the possibility that effects were masked by the overall similarity of the viewing conditions must be considered.

Bottom-Up and Top-Down Factors

The third and final finding of this study came after combining the data across both conditions to compare the levels of change blindness produced by each pair of artefacts independently. We consider in particular three pairs of artefacts which produced a level of change blindness greater than 75% (pairs 1, 7, and 12, 79.31–83.33%), and three pairs of artefacts which produced a level of change blindness lower than 15% (pairs 3, 5, and 6, 4.84–12.90%). Given that the nature of the large visual transients was controlled across the experiment, it follows that these data reflect the fact that local visual transients produced by the changes between the artefacts in pairs 1, 7, and 12 were weaker than those produced by the changes between the artefacts in pairs 3, 5, and 6. These local transients arose from the differences in appearance of the pairs of artefacts.

Taking pairs one and three, both Japanese woodblocks prints, as an example, the artefacts in both pairs are the same size as each other and share the same designs (Figures 5A , 6A ). Pair one, the wave prints, also share very similar colouring (Figure 5A ). The only differences in colouring between this pair are the subtle changes in hue to the border and box containing script. These changes in colour are slight and cover a small proportion of the visible surface of the artefacts. By contrast, pair three, the eagle prints, are more obviously different in colour (Figure 6A ). For instance, the colour of the sky changes from dark blue to light blue between the two prints, and the colour of the boxes containing script changes from pink and red to green and orange. Collectively, these changes represent a more significant colour change and cover a larger proportion of the artefact’s visible surface, compared to the wave prints. It is these local visual transients which account for the lower level of change blindness amongst participants viewing pair three compared to pair one (12.90% vs. 79.31%, respectively).

FIGURE 5. (A) Pair 1: Utagawa Hiroshige, The Sea at Satta in Suruga Province from Thirty Six Views of Mount Fuji . Woodblock prints with bokashi (tonal gradation). 1858-9 AD. 22.4 cm × 34.0 cm. (B) Pair 7: Athenian red-figure lekythoi. Nike flying with phiale (left). Nike flying with thurible (right). 490–480 BC. 32.4 cm (left) and 31.8 cm (right) tall. Images reproduced with permission from Ashmolean Museum, University of Oxford. (C) Pair 12: Isaac Dighton, Silver toilet dressing table service, 2 of 14. 1699–1700 AD. 10.3 cm (left) and 10.5 cm (right) tall. Images reproduced with permission from Ashmolean Museum, University of Oxford.

FIGURE 6. (A) Pair 3: Utagawa Hiroshige, Jûmantsubo Plain at Susaki, near Fukagawa from One Hundred Famous Views of Edo . Woodblock prints with bokashi (tonal gradation). 1856-8 AD. 22.0 cm × 32.8 cm. Images Ashmolean Museum, University of Oxford. (B) Pair 5: Iranian star tiles. Late 13th–14th century AD. 16.0 cm × 6.5 cm, 15.0 cm × 15.0 cm (left), 13.0 cm × 15.0 cm (right). (C) Pair 6: Iranian tiles with interlacing pattern. 13th century AD. Images reproduced with permission from Ashmolean Museum, University of Oxford.

Pair seven, the Athenian lekythoi, produced a high level of change blindness similar to that produced by pair one, the wave prints. The artefacts in this pair are the same size as each other, share the same red-figure colouring, and have near-identical designs, except for the depiction of the object in the figure’s right hand, which changed from a phiale to a thurible (Figure 5B ). As for pair one, this change covers a small area of the visible surface of the artefacts, and so represents a relatively small local visual transient, which translates to a high level of change blindness (83.33%). Similarly, pair 12, the silver flasks, also produced a high level of change blindness (81.67%). The two flasks are practically identical, save only for the uppermost tip which has been displaced atop the first item (Figure 5C ). Again, this change represents a small area of the artefact’s visible surface, and so only produced a small local visual transient.

The artefacts in pair six, the Iranian tiles with interlacing pattern, are the same size as each other and share the same design and colouring (Figure 6C ). However, the first tile has an area of damage to its corner and the second tile carries an extra piece of cement on its front. These changes together account for a large area of the artefacts’ visible surfaces, and as such constitute large local visual transients responsible for a low level of change blindness (11.29%). Pair five, the Iranian star tiles, produced the lowest level of change blindness of all (4.84%), with only three of the 62 participants not noticing any changes between them. These artefacts manifest differences in both the design of their central area, and also in that one of the points on the second tile has been broken off (Figure 6B ). These two changes constitute large local visual transients accounting for a very low level of change blindness.

The characteristics of changes that produced the most easily detected local visual transients include a large visible area of change and high contrast changes in colour. Both of these characteristics, area and contrast, can be directly related to the retina, where light from the visual field is transduced by photoreceptor cells, and contrast is enhanced by lateral inhibition of neurons in the layers between the photoreceptors and retinal ganglion cells. Because these characteristics are amongst the first to be encoded by the visual system, they are possible candidates for bottom-up influences on the prioritisation of what is represented and compared during the process of conscious change perception. In line with this, it has been shown that highly salient objects, where salience includes colour, intensity, and orientation ( Koch and Ullman, 1985 ; Itti and Koch, 2000 ), attract visual fixations earlier than less salient objects ( Underwood and Foulsham, 2006 ; Underwood et al., 2006 ), and it is well-established that the larger a surface is within the visual field the more likely it is to be fixated ( Peschel and Orquin, 2013 ).

However, it is clear that areas undergoing change can be fixated within a change blindness paradigm without the change itself being perceived ( O’Regan et al., 2000 ). It has also been shown that bottom-up factors can at times be overridden by top-down cognitive influences, such as the consistency of an object within the gist of a scene ( Underwood and Foulsham, 2006 ; Stirk and Underwood, 2007 ), and the specific task the viewer is asked to perform when observing a stimulus ( Underwood et al., 2006 ). In this experiment, there are likely to have been many top-down influences derived from the artefacts themselves, such as the prior knowledge that ancient pottery is more likely to exhibit differences in terms of damage, while prints may be more likely to exhibit colour differences. However, the two groups of artefacts which produced the lowest and highest levels of change blindness, respectively, both exhibited differences of colour, design, and damage. This suggests that top-down influences concerning types of changes had a minimal effect on the level of change blindness produced by each artefacts. For this reason, we suggest that bottom-up factors were relatively spared from being over-ridden by top-down effects, and were therefore able to exert their own influence on the processes of representation and comparison, and ultimately change blindness. In this way, our findings support a role for bottom-up factors including a large visible area of change and high contrast colour change in determining which elements in a visual scene are represented and compared in the process of conscious change perception, in both real-world and on-screen viewing conditions.

Limitations

The methods used in this study carry their own limitations. We will discuss them in relation to the two main comparisons performed in this experiment. Namely, the comparison of real-world and on-screen viewing conditions, and the comparison of the 12 pairs of artefacts. Regarding the former, first, by comparing the performance of two different groups of participants in real-world and on-screen conditions, we introduced the potential for selection bias. We saw no practicable alternative to this, as a change cannot be shown to the same participant more than once in a change blindness experiment. To mitigate this bias, we recruited over 30 participants that we randomly allocated to each group, which resulted in near-identical demographics being represented in both.

Second, while it was important to control the conditions in which the artefacts were observed, this was at the expense of the naturalism of the viewing experience. The viewing distance and placements of the objects were similar to what would be found in a natural museum environment, but the brief periods of observation and the removal of peripheral vision using modified goggles were both unnatural. However, the conditions were the same for participants in both groups. Third, by recording changes which participants described incorrectly in the same way as changes that were not described at all, we set a relatively high threshold for change detection to be achieved. Our methodology did not distinguish between the experience of completely missing a change and the experience of sensing that a change had occurred but not being able to describe that change correctly. It is also possible that the head movement of the real-world observer provides an extra attentional cue to the on-screen observer by centering on the change.

Regarding the comparison between the 12 pairs of artefacts, first, it is possible that the performance of participants changed over the course of the experiment as they advanced through the 12 sets of observations. It is both conceivable that their performance may have improved due to a learning effect, or conversely have worsened due to fatigue. We expect that because each observation was only brief (less than 3 s), and the number of observations was relatively few, neither of these effects are likely to have impacted significantly on the levels of change blindness recorded over the course of the experiment. Each set of 12 trials took less than 10 min to perform. Although the order in which the artefacts were viewed was not varied between participants (which could have mitigated any such effects), the levels of change blindness produced from pair one to pair 12 bear no relation to either an increasing or decreasing trend. Finally, the collection of artefacts used as visual stimuli did not contain a control pair, in that there was no pair of artefacts that were truly identical to each other. If such a pair had produced a change blindness level of 100% it would have strengthened the confidence with which we can draw conclusions from our data.

Change blindness is a testable phenomenon of visual perception that can be used to investigate the nature of visual perception in different conditions. It has been produced in naturalistic scenarios outside of the laboratory before using everyday objects, but until now it has not been produced in a setting such a museum, where visual perception may be enhanced. We have for the first time demonstrated that change blindness can be produced inside a museum, using ancient museum artefacts as visual stimuli, under both real-world and on-screen viewing conditions. We anticipate further experiments will be required to fully investigate the notion of altered visual perception inside museums.

While in society, on-screen interactions are increasingly coming to replace real-world ones, there is a relative lack of experimental comparisons between visual perceptual performance in real-world and on-screen conditions. We have for the first time directly compared the levels of change blindness produced by a single set of visual stimuli viewed in both on-screen and real-world conditions, and found that there was no statistically significant difference between the levels of change blindness produced in the two conditions. This does not appear to support our original hypothesis that change detection would be enhanced in real-world conditions relative to on-screen due to the perceptual advantages of binocular stereoscopic vision. We discuss the difficulty of interpreting this finding and caution against generalising the result of this experiment too readily.

In light of this finding, we combined the data from both viewing conditions to identify groups of artefacts that were independently associated with high and low levels of change blindness, and found that change detection rates were influenced mainly by bottom-up factors, including the visible area and contrast of changes, more than top-down factors. In this way, our findings support a role for bottom-up factors in determining which elements in a visual scene are represented and compared in the process of conscious change perception, in both real-world and on-screen viewing conditions. Finally we discuss the intrinsic limitations of this experiment which must be considered alongside its results. We hope, nevertheless, that our attempt to add to the understanding of visual perception within museums, the phenomenon of change blindness, perceptual performance in real-world and on-screen conditions, and the role bottom-up and top-down factors in change detection will motivate further research into these increasingly relevant questions.

Author Contributions

JA, CK, JH, GH, and CAA designed the research, revised and improved the manuscript. JA and CAA analysed the data and prepared the figures. JA, CK, JH, and CAA discussed the results, advised on interpretation and contributed to the final draft of the manuscript. All authors contributed to and had approved the final manuscript.

JH is funded by the Andrew Mellon Foundation and CAA is supported by the Dementias and Neurodegenerative Diseases Research Network (DENDRON), the NIHR and UCB.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

We would like to thank all of our participants, whose enthusiasm and encouragement have been a great support to us throughout our work on this project. We are also very grateful to the staff of the Ashmolean Museum of Art and Archaeology for their generosity and guidance in making this unique experiment possible. This manuscript is dedicated to the memory of Professor GH.

Albertazzi, L., Van Tonder, G. J., and Vishwanath, D. (eds). (2010). Perception Beyond Inference: The Information Content of Visual Processes . Cambridge, MA: MIT Press.

Google Scholar

Ballard, D. H., Hayhoe, M. M., and Pelz, J. B. 1995). Memory representations in natural tasks. J. Cogn. Neurosci. 7, 66–80. doi: 10.1162/jocn.1995.7.1.66

PubMed Abstract | CrossRef Full Text | Google Scholar

Binder, M. D., Hirokawa, N., and Windhorst, U. (eds). (2009). “Gaze Shift,” in Encyclopedia of Neuroscience , (Berlin: Springer), 1676.

Borji, A., and Itti, L. (2013). State-of-the-art in visual attention modeling. IEEE Trans. Pattern Anal. Mach. Intell. 35, 185–207. doi: 10.1109/TPAMI.2012.89

Cutting, J. E. (1997). How the eye measures reality and virtual reality. Behav. Res. Methods Instrum. Comput. 29, 27–36. doi: 10.3758/BF03200563

CrossRef Full Text | Google Scholar

Fielder, A. R., and Moseley, M. J. (1996). Does stereopsis matter in humans? Eye 10, 233–238. doi: 10.1038/eye.1996.51

Gozli, D. G. (2017). Behaviour versus performance: the veiled commitment of experimental psychology. Theory Psychol. 27, 741–758. doi: 10.1177/0959354317728130

Grimes, J. (1996). “On the failure to detect changes in scenes across sccades,” in Perception:Vancouver Studies in Cognitive Science , ed. K. Akins (Oxford: Oxford University Press), 89–110.

Grossman, G. E., Leigh, R. J., Abel, L. A., Lanska, D. J., and Thurston, S. E. (1988). Frequency and velocity of rotational head perturbations during locomotion. Exp. Brain Res. 70, 470–476. doi: 10.1007/BF00247595

Itti, L., and Koch, C. (2000). A saliency-based search mechanism for overt and covert shifts of visual attention. Vis. Res. 40, 1489–1506. doi: 10.1016/S0042-6989(99)00163-7

Jones, R. K., and Lee, D. N. (1981). Why two eyes are better than one: the two views of binocular vision. J. Exp. Psychol. Hum. Percept. Perform. 7, 30–40. doi: 10.1037/0096-1523.7.1.30

Kelley, T. A., Chun, M. M., and Chua, K.-P. (2003). Effects of scene inversion on change detection of targets matched for visual salience. J. Vis. 3, 1–5. doi: 10.1167/3.1.1

Kesner, L. (2006). The role of cognitive competence in the art museum experience. Mus. Manage. Curator. 21, 4–19. doi: 10.1080/09647770600302101

Koch, C., and Ullman, S. (1985). Shifts in selective visual attention: towards the underlying neural circuitry. Hum. Neurobiol. 4, 219–227.

PubMed Abstract | Google Scholar

Levin, D. T., and Simons, D. J. (1997). Failure to detect changes to attended objects in motion pictures. Psychon. Bull. Rev. 4, 501–506. doi: 10.3758/BF03214339

Levin, D. T., Simons, D. J., Angelone, B. L., and Chabris, C. F. (2002). Memory for centrally attended changing objects in an incidental real-world change detection paradigm. Br. J. Psychol. 93, 289–302. doi: 10.1348/000712602760146224

Levy, E. K. (2011). An artistic exploration of inattention blindness. Front. Hum. Neurosci. 5:174. doi: 10.3389/fnhum.2011.00174

McConkie, G. W., and Currie, C. B. (1996). Visual stability across saccades while viewing complex pictures. J. Exp. Psychol. Hum. Percept. Perform. 22, 563–581. doi: 10.1037/0096-1523.22.3.563

O’Neill, M.-C., and Dufresne-Tassé, C. (1997). Looking in everyday life/gazing in Museums. Mus. Manage. Curator. 16, 131–142. doi: 10.1080/09647779708565838

O’Regan, J. K., Deubel, H., Clark, J. J., and Rensink, R. A. (2000). Picture changes during blinks: looking without seeing and seeing without looking. Vis. Cogn. 7, 191–211. doi: 10.1080/135062800394766

O’Regan, J. K., Rensink, R. A., and Clark, J. J. (1999). Change-blindness as a result of “mudsplashes”. Nature 398:34. doi: 10.1038/17953

Pelisson, D., and Guillaume, A. (2009). “Eye-head coordination,” in Encyclopedia of Neuroscience , eds M. D. Binder, N. Hirokawa, and U. Windhorst (Berlin: Springer), 1545–1548. doi: 10.1007/978-3-540-29678-2_3257

Peschel, A. O., and Orquin, J. L. (2013). A review of the findings and theories on surface size effects on visual attention. Front. Psychol. 4:902. doi: 10.3389/fpsyg.2013.00902

Rensink, R. A., O’Regan, J. K., and Clark, J. J. (1997). To see or not to see: the need for attention to perceive changes in scenes. Psychol. Sci. 8, 368–373. doi: 10.1111/j.1467-9280.1997.tb00427.x

Servos, P., Goodale, M. A., and Jakobson, L. S. (1992). The role of binocular vision in prehension: a kinematic analysis. Vis. Res. 32, 1513–1521. doi: 10.1016/0042-6989(92)90207-Y

Simons, D. J. (1996). In sight, out of mind: when object representations fail. Psychol. Sci. 7, 301–305. doi: 10.1111/j.1467-9280.1996.tb00378.x

Simons, D. J., and Chabris, C. F. (1999). Gorillas in our midst: sustained inattentional blindness for dynamic events. Perception 28, 1059–1074. doi: 10.1068/p281059

Simons, D. J., and Levin, D. T. (1997). Change blindness. Trends Cogn. Sci. 1, 261–267. doi: 10.1016/S1364-6613(97)01080-2

Simons, D. J., and Levin, D. T. (1998). Failure to detect changes to people during a real-world interaction. Psychon. Bull. Rev. 5, 644–649. doi: 10.3758/BF03208840

Simons, D. J., and Rensink, R. A. (2005). Change blindness: past, present, and future. Trends Cogn. Sci. 9, 16–20. doi: 10.1016/j.tics.2004.11.006

Snyder, Q. C., and Lezotte, D. C. (1993). Prospective assessment of stereoscopic visual status and USAF pilot training attrition. Aviat. Space Environ. Med. 64, 14–19.

Stirk, J. A., and Underwood, G. (2007). Low-level visual saliency does not predict change detection in natural scenes. J. Vis. 7, 3.1–10. doi: 10.1167/7.10.3

The Nielsen Total Audience Report - Q1 2016 (2016). The Total Audience Report: Q1 2016 . Available at: http://www.nielsen.com/us/en/insights/reports/2016/the-total-audience-report-q1-2016.html

Underwood, G., and Foulsham, T. (2006). Visual saliency and semantic incongruency influence eye movements when inspecting pictures. Q. J. Exp. Psychol. 59, 1931–1949. doi: 10.1080/17470210500416342

Underwood, G., Foulsham, T., van Loon, E., Humphreys, L., and Bloyce, J. (2006). Eye movements during scene inspection: a test of the saliency map hypothesis. Eur. J. Cogn. Psychol. 18, 321–342. doi: 10.1080/09541440500236661

Varakin, D. A., Levin, D. T., and Collins, K. M. (2007). Comparison and representation failures both cause real-world change blindness. Perception 36, 737–749. doi: 10.1068/P5572

Wickens, C. D., Merwin, D. H., and Lin, E. L. (1994). Implications of graphics enhancements for the visualization of scientific data: dimensional integrality, stereopsis, motion, and mesh. Hum. Fact. 36, 44–61. doi: 10.1177/001872089403600103

Keywords : change blindness, vision, perception, real-world, on-screen, binocular, museum, artefact

Citation: Attwood JE, Kennard C, Harris J, Humphreys G and Antoniades CA (2018) A Comparison of Change Blindness in Real-World and On-Screen Viewing of Museum Artefacts. Front. Psychol. 9:151. doi: 10.3389/fpsyg.2018.00151

Received: 26 July 2017; Accepted: 29 January 2018; Published: 16 February 2018.

Reviewed by:

Copyright © 2018 Attwood, Kennard, Harris, Humphreys and Antoniades. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Chrystalina A. Antoniades, [email protected]

† Deceased

Change Blindness: 10 Examples, Definition & Types

Change blindness is what happens when a person fails to recognize the change in an object or scene when that change coincides with a brief visual disruption (Simons & Levin, 1997).

Change blindness occurs when we don’t perceive the motion that happens when a stimulus moves from one location to another in a visual scene (McConkie & Loschky, 2006). It often happens as a result of selective attention .

Change blindness happens regularly, for example, in magicians’ tricks, where the magician will use a distraction in one hand while changing cards in the other hand, which prevents observes from observing the switch.

Change Blindness Definition and Overview

Simons and Fracnoneri (2000) define change blindness as: “a surprising inability to detect large changes to scenes from one view to the next.”

Of note in this definition is that change blindness can occur even when the stimulus that has changed is quite large. This can happen with objects presented in still images, motion pictures, and even in real-world situations involving interpersonal interactions.

In a representative study, Simons and Levin (1998) conducted an ingenious investigation of change blindness in a real-world situation.

An accomplice of the study, carrying a map, asked directions from a stranger near a university campus. While the stranger was giving directions, two other accomplices carrying a large door walked between them.

In that moment, the first experimenter was replaced with another experimenter that looked similar, but was in fact dressed differently and was in fact, a different person: “yet, despite clear differences in clothing, appearance, and voice, only 7 of the 15 pedestrians reported noticing the change of experimenters” (p. 646).

Types of Change Blindness

There are two types of change detection: via a sensory transient signal and via inference.

• Sensory transient signal change: Thisoccurs immediately as a result of one object immediately replacing another. The change is noticed as a result of visual detection produced by a transient signal. This signal happens automatically and brings the change into conscious awareness (Reichardt, 1961).
• Inference : Inferential change detection occurs after a gap in the presentation of the two stimuli and involves a more cognitive process. For example, it can take some time to notice the change in hairstyle of a coworker. The gap from one stimulus presentation (coworker on Friday) to the second presentation (coworker on Monday) produces change blindness. This type of change detection occurs as a result of comparing the current image with information stored previously in memory.

James (1950/1891) made note of this distinction between the two types of change detection over 100 years ago:

“With such direct perceptions of difference as this [sensory transient signal], we must not confound those entirely unlike cases in which we infer that two things must differ because we know enough about each of them taken by itself to warrant our classing them under distinct heads. It often happens, when the interval is long between two experiences, that our judgments are guided, not so much by a positive image or copy of the earlier one, as by our recollections of certain facts about it.” (p. 496-497, words in square brackets added)

Since those earliest days of discussion, research has found change blindness can occur during a brief flash on a computer screen (Rensink, O’Regan, & Clark, 1997; Simons, 1996), in movies (Levin & Simons, 1997), or in the blink of an eye (O’Regan, Deubel, Clark, & Rensink, 2000).

Change Blindness Examples

• In Movie Scenes: There are lots of examples in the movies of an object being misplaced or even absent from one scene to the next. Although the director did not intend those errors to be included in the final cut, sometimes they get through anyway. Audiences rarely take notice.
• In Rescue Operations: The Coast Guard sometimes has to attempt a rescue in turbulent seas. Trying to lower oneself to rescue a drowning victim requires a continuous monitoring of the victim’s location, who is being thrown about by large swells and might actually get submerged for several seconds at a time. In this kind of emergency, there is no room for change blindness.
• In Checking Programming Code: Programming languages can be quite complex and developing software can involve hundreds and thousands of lines of code, worked on continuously by a dedicated team of programmers. Being able to check those lines through multiple iterations is a painstaking task that requires a keen eye for detail. Change blindness can make that process ever more time-consuming and extremely inefficient.
• In Older Drivers: Older drivers are more likely to miss changes in their visual field compared to younger drivers. This was demonstrated in research using photos of intersections that altered images of pedestrians, vehicles, signs, or traffic control devices.
• When Spies Switch Drinks: In a lot of spy movies there is always a scene in which two enemies pretend to be having a friendly drink together. As one of them turns their back, the other puts poison in their glass. If the one getting the poisoned drink experiences change blindness, then its game over for them.  
• In Eyewitness Testimony: Although most people are confident in their memory, research on eyewitness testimony says otherwise. Research has demonstrated that change blindness can occur when the perpetrator of a burglary is switched half-way through a video reenactment of a crime. Over half of viewers did not notice the change at all.   
• Air Traffic Controllers: If an air traffic controller experiences change blindness, it could result in disaster. The screen they monitor contains a large number of images that are in constant motion. Failing to recognize a change on the screen is a mistake that can have deadly consequences.
• Playing Video Games: So many games involve rapidly changing images and scenes. Players have to keep their eyes focused on the screen at all times. Even a momentary lapse of attention can lead to change blindness, and then…game over.  
• In Combat Situations: Operating modern military equipment involves personnel being heavily loaded with visual processing tasks, situation assessments, voice communications, and the tactile manipulation of numerous displays. Research has demonstrated that change blindness can easily occur in these highly stressful situations and lead to substantial negative consequences.
• When Conducting Research: Naturalistic observation is a type of research that involves watching and recording behavior. That could be the behavior of people or animals. The researcher has to have a trained eye for detail and needs to be able to notice very minute changes in behavior. Having a bout of change blindness can mean missing key moments and jeopardizing the study’s validity.
• Monitoring Airplane Cockpit Instrument Displays: If you have ever seen the inside of an airplane’s cockpit, you will be amazed at the dizzying number of instrument displays. A pilot has to monitor the information presented by all of those instruments and be acutely aware of any significant changes. Even a brief moment of change blindness may lead to an emergency.

Applications of Change Blindness 

1. in the movies.

As Simons and Levin (1997) point out, directors and editors exert considerable effort to eliminate flaws in the continuity of scenes.

Unfortunately, because scenes are often shot out of sequence, perhaps even on different days, there are many occasions in which the objects in one scene have been moved slightly or might even be absent in the next.

Fortunately, nine times out of ten, those errors will not be noticed by the audience due to change blindness (Dmytryk, 1984).

Film makers are especially knowledgeable of how people process visual scenes. They often use that knowledge to create a sense of cohesion and continuity to a scene.

Some even employ different techniques to facilitate that perception by cleverly inserting elements that will cause the audience to blink at just the right moment. In this way, they capitalize on change blindness to construct a desired perception.

2. In Eyewitness Testimony

Our understanding of change blindness has also been applied to the accuracy of eyewitness testimony (Loftus, 2019). Research has already demonstrated that eyewitness memory is easily altered and manipulated ( Loftus & Palmer, 1974).  

Change blindness can also be an issue. For instance, Davies and Hine (2007) produced a short video depicting a burglary. However, half-way through the 2-minute video, the burglar was switched.

Burglar 1 was 170 cm tall, slightly built with an oval-shaped face. Burglar 2 was 188 cm tall, heavier, and had an oval-shaped face. Both burglars wore dark clothing, but according to the researchers, the style and detail differed considerably.

Afterwards, the research participants were given a questionnaire which asked “Did you notice anything change about the burglar throughout the film?”

The results showed that “only 39% of participants noticed the identity change” (p. 431). This percentage is similar to that obtained by Davis et al. (2008). Similar results were also obtained by Levin and Simons, (1997) and Levin et al. (2002).

While 95% of participants in Nelson et al. (2011) experienced change blindness while watching a simulated crime video, they also found that increased severity of the crime led to fewer errors identifying the perpetrator.

The overarching message from this line of research is that change blindness can lead to inaccuracy identifying criminals, create errors in eyewitness testimony, and may ultimately lead to wrongful convictions.  

Change blindness refers to not noticing a change from one visual stimuli to the next. Some of the earliest discussions of change blindness occurred in the film industry. Film makers noted that most viewers failed to detect errors in the continuity from one scene to the next.

Although seemingly unimportant at first, researchers have since connected change blindness to a variety of consequential situations.

For instance, an eyewitness to a crime can fail to correctly distinguish between two suspects due to change blindness. Similar looking suspects are sometimes not even noticed as being two different people.

Older drivers are more likely to experience change blindness, which can mean not noticing pedestrians abruptly appearing in the street or a change in traffic signals.

Change blindness also has significant implications for professionals that operate sophisticated machinery such as airplanes or military equipment. The failure to recognize changes in instrument displays of essential data can have profoundly negative consequences.

Caird, J. K., Edwards, C. J., Creaser, J. I., & Horrey, W. J. (2005). Older driver failures of attention at intersections: using change blindness methods to assess turn decision accuracy. Human Factors , 47 (2), 235-249.

Davies, G., & Hine, S. (2007). Change blindness and eyewitness testimony. The Journal of Psychology , 141 (4), 423-434.

Davis, D., Loftus, E. F., Vanous, S., & Cucciare, M. (2008). ‘Unconscious transference’ can be an instance of ‘change blindness’. Applied Cognitive Psychology, 22 , 605–623.

DiVita, J., Obermayer, R., Nugent, W., & Linville, J. M. (2004). Verification of the change blindness phenomenon while managing critical events on a combat information display. Human Factors , 46 (2), 205-218.

Dmytryk, E. (1984). On film editing: An Introduction to the art of film construction. Focal Press.

Hochberg, J. (1986). Representation of motion and space in video and cinematic displays. NASA STI/Recon Technical Report A , 1 , 22_1-22_64.

James, W. (1950/1891). The principles of psychology (Vol. 1). New York: Dover.

Kuleshov, L. (1987/1920). Selected Works: Fifty Years in Films (D. Agrachev & N.

Belenkaya, Trans.). Moscow: Raduga Publishers.

Levin, D. T., Simons, D. J., Angelone, B., & Chabris, C. F. (2002). Memory for centrally attended changing objects in an incidental real-world change detection paradigm. British Journal of Psychology, 93 , 289–302.

Loftus, E. F., & Palmer, J. C. (1974). Reconstruction of automobile destruction: An example of the interaction between language and memory. Journal of Verbal Learning and Verbal Behavior , 13 (5), 585-589.

Loftus, E. F. (2019). Eyewitness testimony. Applied Cognitive Psychology , 33 (4), 498-503.

McConkie, G., & Loschky, L. (2006). Change blindness, Psychology of. In L. Nadel (Ed.), Encyclopedia of Cognitive Science . Vol. 1 pp. 491-495. London, UK: Nature Publishing Group.

Nelson, K. J., Laney, C., Fowler, N. B., Knowles, E. D., Davis, D., & Loftus, E. F. (2011). Change blindness can cause mistaken eyewitness identification. Legal and criminological psychology , 16 (1), 62-74.

Kevin O’Regan, J., Deubel, H., Clark, J. J., & Rensink, R. A. (2000). Picture changes during blinks: Looking without seeing and seeing without looking. Visual cognition , 7 (1-3), 191-211.

Reichardt, W. (1961). Autocorrelation, a principle for evaluation of sensory information by the central nervous system. In Symposium on Principles of Sensory Communication 1959 (pp. 303-317). MIT press.

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.

Rensink, R. A. (2002). Change detection. Annual Review of Psychology, 53, 245–27.

Simons, D. J. (1996). In sight, out of mind: When object representations fail. Psychological Science , 7 (5), 301-305.

Simons, D. J., Franconeri, S. L., & Reimer, R. L. (2000). Change Blindness in the Absence of a Visual Disruption.  Perception ,  29 (10), 1143–1154.  https://doi.org/10.1068/p3104

Simons, D. J., & Levin, D. T. (1997). Change blindness. Trends in Cognitive Sciences , 1 (7), 261-267.

Simons, D. J., & Levin, D. T. (1998). Failure to detect changes to people during a real-world interaction. Psychonomic Bulletin & Review , 5 , 644-649.

Dave Cornell (PhD)

Dr. Cornell has worked in education for more than 20 years. His work has involved designing teacher certification for Trinity College in London and in-service training for state governments in the United States. He has trained kindergarten teachers in 8 countries and helped businessmen and women open baby centers and kindergartens in 3 countries.

• Dave Cornell (PhD) #molongui-disabled-link 25 Positive Punishment Examples
• Dave Cornell (PhD) #molongui-disabled-link 25 Dissociation Examples (Psychology)
• Dave Cornell (PhD) #molongui-disabled-link 15 Zone of Proximal Development Examples
• Dave Cornell (PhD) #molongui-disabled-link Perception Checking: 15 Examples and Definition

Chris Drew (PhD)

This article was peer-reviewed and edited by Chris Drew (PhD). The review process on Helpful Professor involves having a PhD level expert fact check, edit, and contribute to articles. Reviewers ensure all content reflects expert academic consensus and is backed up with reference to academic studies. Dr. Drew has published over 20 academic articles in scholarly journals. He is the former editor of the Journal of Learning Development in Higher Education and holds a PhD in Education from ACU.

• Chris Drew (PhD) #molongui-disabled-link 25 Positive Punishment Examples
• Chris Drew (PhD) #molongui-disabled-link 25 Dissociation Examples (Psychology)
• Chris Drew (PhD) #molongui-disabled-link 15 Zone of Proximal Development Examples
• Chris Drew (PhD) #molongui-disabled-link Perception Checking: 15 Examples and Definition

Leave a Comment Cancel Reply

Your email address will not be published. Required fields are marked *

Change Blindness (Definition + Examples)

Have you ever seen a movie with continuity errors? It's astounding how many we fail to notice. Interestingly, about 58% of people don't spot these errors due to a phenomenon known as change blindness. This intriguing facet of perception reveals how much our brains can sometimes miss, even when the information is right before our eyes!

What Is Change Blindness?

Change Blindness is a cognitive glitch wherein people fail to detect visual environmental changes. They might seem unmistakably evident if you're actively looking for these changes. Yet, if you’re preoccupied, or something else distracts you, these changes might completely elude your observation, no matter how glaring.

Examples of Change Blindness In Movies

Creating a movie is meticulous, but continuity errors occasionally make it to the final cut. In fact, according to a schema that details common cinematic mistakes, about 32% of all movie errors are related to change blindness.

For instance:

• In the movie New Moon, Jacob’s tattoo relocates! Sometimes, it's higher up, lower down.
• Blade Runner has its share of the evident stunt double and the contrasting weather conditions between scenes.
• And who could forget Ace Ventura: When Nature Calls, where a packed chessboard suddenly becomes empty and then refills mysteriously?

Inattentional Blindness vs. Change Blindness

In 1999, psychologists Chris Chabris and Dan Simons conducted what is now known as “The Invisible Gorilla Experiment.” Participants watched a video of people passing basketballs back and forth in the experiment. They were instructed to count how many times the basketballs were passed.

During the video, a person in a gorilla suit walked through the circle. But a shocking number of participants didn’t notice the gorilla’s presence.

This experiment led to the creation of the term “ inattentional blindness. ” This term is often confused with change blindness. Let’s go over the difference between these two phenomena.

Inattentional blindness is failing to recognize visual objects when focused on something else. The participants were so focused on the people passing the basketballs that they failed to see the gorilla.

If the gorilla had always been in the environment, but participants failed to see the gorilla change fur colors or put on a shirt, then change blindness would have occurred.

Change blindness is the failure to notice changes to visual objects as they happen. During the change, you have recognized all the visual objects in your environment. You don’t see when those objects have shifted, transformed, or changed entirely.

A sharp focus on something in your environment causes inattentional blindness. A lack of focus or other factors may cause change blindness.

What Causes Change Blindness?

There are a few reasons why we might experience change blindness, including:

• Distractions

Expectations

Manipulation, distraction.

Would the change blindness experiment have also worked if the door hadn’t passed before the pedestrian and the actor? No! Without the door, the pedestrian would have remained focused on the actor, and it would have been easy to notice the change. But a distraction takes our attention away from the visual object that changes; therefore, we don’t see the change happening.

Take the scene in Ace Ventura. We can take in the scene with the chess board, Jim Carrey, and Vincent Cadby. Several shots include the chess board filled with pieces.

A lot goes on as the camera cuts back and forth, but we’re in the same room. When we get to the scene without the chess pieces, our minds have already started to “fill in the blanks.” We don’t have to visually process every piece of the set the cameras are showing - that would be exhausting. We expect there to be chess pieces on the board. Why wouldn’t there be? Plus, we have to hear how Ace will solve the case!

So our mind takes “shortcuts” and fills in the blanks. Unfortunately, this often means filling in the blanks where changes have occurred.

Change blindness doesn’t always happen by accident. Movie magic (or magic tricks) is made more exciting by change blindness. When manipulation goes wrong, we see blatant changes between actors and their stunt double. When the scene is manipulated just right, we notice no difference.

A Brief History of Discovering Change Blindness

This phenomenon came into the limelight in 1995 when researchers explored why movie continuity errors bypass viewers. The introduction of a flicker or an intervening shot, like Bella's scenes in New Moon, enhanced the occurrence of change blindness.

Furthermore, a 1998 study titled “Failure to detect changes to people during a real-world interaction” showcased that change blindness isn't just restricted to screens. Remarkably, 50% of the participants didn't detect the actor switch during an in-person interaction!

Who Discovered Change Blindness?

In 1995, researchers studied how continuity errors could go completely unnoticed by viewers. They found that these errors were more likely to go unnoticed if there was a flicker in the screen between changes. Viewers might have noticed Jacob’s tattoo moving up and down his arm, but they probably didn’t because it was interrupted by shots of Bella or other shots in the movie.

Changes in Pedestrians

One of the most famous studies on change blindness took place shortly after the study on continuity errors. DJ Simons and Daniel Levin published this study in 1998 called “Failure to Detect Changes to People During a Real-World Interaction.” 

This study became famous because it showed that change blindness doesn’t just occur in a 2D space. We experience change blindness “out in the real world,” too.

In the study, an actor initiated a conversation with a pedestrian on the street. They began to ask for directions. As the pedestrian gave the directions, two more actors walked in between the pedestrian and the first actor. During this time, the first actor was switched out with another actor but pretended to be the first actor.

Half of the pedestrians didn’t notice the change!

Change Blindness Is Normal.

We are all tricked by change blindness in our everyday lives. Maybe you fail to notice someone’s haircut or that they’ve changed into a new outfit. Maybe that magic trick does look like magic. Change blindness isn’t bad, but it’s something to be aware of. Our minds don’t always catch everything that happens in front of us.

Related posts:

• Inattentional Blindness (Definition + Examples)
• The Invisible Gorilla (Inattentional Blindness)
• The Psychology of Long Distance Relationships
• Attention (Psychology Theories)
• Operant Conditioning (Examples + Research)

Reference this article:

About The Author

Perception:

The 5 Senses

Mirror Neurons

Bottom Up Processing

Top Down Processing

PracticalPie.com is a participant in the Amazon Associates Program. As an Amazon Associate we earn from qualifying purchases.

Follow Us On:

Youtube Facebook Instagram X/Twitter

Psychology Resources

Developmental

Personality

Relationships

Psychologists

Serial Killers

Psychology Tests

Personality Quiz

Memory Test

Depression test

Type A/B Personality Test

© PracticalPsychology. All rights reserved

Privacy Policy | Terms of Use

Cognitive Bias: Change blindness

• 13 April 2022
• Behavioral finance , Cognitive Bias

Look at the video below.

Did you see the trick? If not, watch the video again more carefully!

This video is a famous example of change blindness.

Change blindness affects our daily decisions

It refers to the fact that we are sometimes so focused on a specific thought or movement than we ignore the rest. People’s poor ability to detect changes has been argued to reflect fundamental limitations of human attention. This phenomenon has been studied since the XIXth by researchers like William James.

Cognitive psychologists expanded the study of change blindness into decision-making. In a study by Johansson, Sikstörm and Olsson, participants were shown ten paris of faces and asked to choose which face was more attractive.

At the end of the experience, researchers changed discreetly the chosen faces for others and asked the participants to validate their choice. Only 26% of subjects noticed the mismatch between their choice of face and the different face they were shown instead!

We are so focused on being consistent with our past choices and decisions that we do not even notice if their output changes!

How to limit the impacts of the change blindness bias?

Research showed that change blindness can be counteracted by a number of methods like attention guidance, signal gradation, direct comparison or increased concentration.

Time to test!

Johansson, P.; Hall, L.; Sikström, S.; Olsson, A. (2005).  “Failure to detect mismatches between intention and outcome in a simple decision task” .  Science .  310  (5745): 116–119.  Bibcode : 2005Sci…310..116J .  doi : 10.1126/science.1111709 .  PMID   16210542 .  S2CID   16249982 .

• Subscriber Services
• For Authors
• Publications
• Archaeology
• Art & Architecture
• Bilingual dictionaries
• Classical studies
• Encyclopedias
• English Dictionaries and Thesauri
• Language reference
• Linguistics
• Media studies
• Medicine and health
• Names studies
• Performing arts
• Science and technology
• Social sciences
• Society and culture
• Overview Pages
• Subject Reference
• English Dictionaries
• Bilingual Dictionaries

Recently viewed (0)

• Save Search

Related Content

Related overviews.

inattentional blindness

motion-induced blindness

See all related overviews in Oxford Reference »

More Like This

Show all results sharing these subjects:

change blindness

Quick reference.

Failure to notice a change in the visual field occurring during a saccade or when vision is otherwise interrupted. Compare inattentional blindness, motion-induced blindness.

From:   change blindness   in  A Dictionary of Psychology »

Subjects: Science and technology — Psychology

Related content in Oxford Reference

Reference entries, change blindness n..

View all related items in Oxford Reference »

Search for: 'change blindness' in Oxford Reference »

• Oxford University Press

PRINTED FROM OXFORD REFERENCE (www.oxfordreference.com). (c) Copyright Oxford University Press, 2023. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single entry from a reference work in OR for personal use (for details see Privacy Policy and Legal Notice ).

date: 03 December 2023

• Cookie Policy
• Privacy Policy
• Legal Notice
• Accessibility
• [66.249.64.20|162.248.224.4]
• 162.248.224.4

Character limit 500 /500

Skip navigation

• Log in to UX Certification

World Leaders in Research-Based User Experience

Change blindness in ux: definition.

Summary:  Significant changes in a web page can remain unnoticed when they lack strong cues, due to the limitations of human attention.

By Raluca Budiu

• Raluca Budiu

on 2018-09-23 September 23, 2018

• Psychology and UX Psychology and UX
• Share this article:

In usability testing, we often observe users who overlook a change on the screen that the designers thought would be obvious and highly noticeable. As always in usability, if you worked on a design, then you know what to look for, where to look for it, when it will appear, and what it means . So yes, it’s obvious to you and you wouldn’t miss the appearance of an important message or data object when you review your own design. But users often do. Why? Because of change blindness, which is a million-year old fact of human (and protohuman) perception, and not likely to go away any time soon.

So what is change blindness? In Alfred Hitchcock’s Psycho , one of the most famous movies of all time, detective Arbogast looks at Norman Bates’ house projected on a dark cloudless sky. The camera moves back to the detective’s face, and follows him as he starts walking toward the house. The scene is still dark, but the sky has suddenly become full of clouds.

Whether the change in the sky’s texture was intentional (perhaps Hitchock’s subtle warning for what was to come) or a continuity error, chances are that most viewers won’t notice it. Motion pictures often have small inconsistencies like this from one scene to the next — changes in the background, in the actors clothing, makeup, or positions, but these get ignored when they are made during cuts between scenes.

This phenomenon is called change blindness and goes beyond movies — it applies to how people perceive scenes in general, whether projected on a screen or in real life. Change blindness is very robust: even when people are warned that a change may happen, changes in a scene can go undetected.

Definition : Change blindness refers to people’s tendency to ignore changes in a scene when they occur in a region that is far away from their focus of attention.

In psychology, change blindness is perhaps best illustrated by a series of experiments performed in mid 1990s. In these experiments, participants were shown a picture of a complex scene for about half a second; then the display went blank for a fraction of a second, and finally the same picture was shown again. However, the second time the scene was shown, some detail was modified — for example, an object changed size, color, or an element was added or removed. Participants in the experiment were supposed to identify the changes in the two images.  In some of these studies, people cycled through the two versions many times, until they were able to detect the change. For many it took 20 or 40 alternations to be able to find it.

The flickering quality of the display (one scene, followed by a blank screen, followed by what seems to be the same scene) is an essential component of change blindness. In movies, the scene cut plays the role of a flicker; and, in interacting with a computer, the flicker is caused by the loading of a new page (or UI element) once the user has pressed a button. But it turns out that the flicker doesn’t have to be literal — scene modifications that happen when the user is blinking or during a saccade (when she’s moving her eyes to another region of the screen) are also highly susceptible to change blindness.

Magicians take advantage of humans’ propensity to change blindness as their eyes move: they use an attention-attracting device to cause eye movements to that spot, and, at the same time, engage in unobserved actions that are essential to their tricks.

Why Does Change Blindness Happen?

The fundamental reason for which change blindness occurs is a limitation in our attention capacities. Any complex scene has a multitude of details, and it’s hard and inefficient for people to attend to all of them. What we normally do is take shortcuts.

Change is and has always been important for humans — it can convey essential information about our environment. Most changes in nature are mediated by movement (with few exceptions such as chameleons, a physical object can’t change instantly with no movement being involved), and movement is easily detected by human peripheral vision. Once the human eyes detect movement in the periphery, they look at the source of movement — central vision kicks in and provides supplemental, detailed information.  This behavior is a remnant of our life in the wild, where any move could signal a predator waiting to hunt us.

Change blindness occurs when movement as a cue for change is weak or completely absent . When the screen flickers, the movement is shunted out — we see two static versions of the scene. The only way to find out what changed is by inspecting all the corresponding elements in the before and after versions and comparing them. This task is very difficult — not only is it tedious to inspect the myriad of little details in the scene, but our memory of the prior version of the scene is likely to be quite poor. In fact, chances are that we paid no attention to many of the elements of that scene anyhow.

But change blindness also happens during an eye movement. In other words, if two (possibly movement-cued changes) compete with each other — like in the magic show, one change usually wins and attracts the eye, but that eye movement blocks the detection of the second change. This cause is particularly important in interface design.

Change Blindness in Interface Design

In normal interactions with a UI, change blindness often occurs when a new element (such as a different image in a product-image carousel , or the contents of a dropdown menu) appears on the screen as a result of user action and other areas of the screen also change.  The locus of the change is expected to be the visible design element that responds directly to the user action and the user moves the eyes in that direction — yet in fact, the change is spread across multiple regions of the screen.

For example, when users tap on the hamburger menu in Aldiko’s Android app, they expect that the changes on the screen be related to that action — namely, that the new elements will be localized in the area enclosing the menu contents. Their eyes will stay around that area and they will be unlikely to notice that the action-overflow button in the top right corner of the screen has been replaced with a search icon.

Elsewhere, we discussed how the search box should not be replaced by a search icon on the desktop; on a mobile screen, however, the pattern is more usable than on the desktop, as our research shows that the magnifier tool is fairly discoverable even when the search box is absent. However, if the search box is not visible by default, when users click on the search icon, the text field should appear next to it (rather than farther apart on the screen) to ensure that people will not miss it.

Semipersistent navigation bars or floating buttons that appear towards the top of the page are also in danger of staying undetected, because often the scrolling of the page masks them. For example, semipersistent navigation bars appear at the top of the screen when the user stars scrolling up. The hope is that people will notice these bars and select one of the options inside them instead of swiping their way up to the very top of the page. Unfortunately, the movement of the page can easily block the movement caused by the apparition of the bar, especially when the color of the bar blends in with the color of the page, like in the New York Times example below.

(Our peripheral vision is responsible for picking out movement and also shadows. When the “shadow” profile of a page changes, for example because a big block of contrasting color has appeared in one corner, it is easier for us to detect its appearance than when the same block subtly blends in with the rest of the page and does not significantly alter the shadow profile of the page.)

There are many other design aspects that can be affected by change blindness — error messages or other notifications, results that appear too quickly , or changing navigation bars may also remain ignored due to this phenomenon, as discussed on our companion article on the same topic .

How to Prevent Change Blindness in Interface Design

To avoid change blindness, analyze your design for any competing changes that may happen at the same time and that may divert attention from each other. Here are some techniques for doing it:

• Make one change at a time . In the Aldiko example above, search could be placed in the top right corner and be visible always.
• Group all elements that will change simultaneously in the same region of the screen, to make sure that the motion will draw attention to all of them. For example, another easy fix for the Aldiko design is to move search inside the menu. (Note however that hiding search under a menu will seriously impact its discoverability and may only be acceptable on browse-heavy sites.)
• Use animation to signal change , but avoid having too many competing animations on the screen to prevent a dilution of attention.
• Dim the areas of the screen that do not change, in order to attract attention to changes.
• If you are adding floating elements to the page as the user scrolls, display them next to the user’s focus of attention (for instance, towards the bottom of the page for Back to Top buttons ) and use colors that contrast with the rest of the page.

Ronald Resnick. 2002. Change Detection. Annual Review of Psychology , 53, p/245-277.

Ronald Rensnick. 2005. Change Blindness. In In McGraw-Hill Yearbook of Science & Technology . pp. 44-46.

• Share: 

About the Author

Raluca Budiu is Director of Research at Nielsen Norman Group, where she consults for clients from a variety of industries and presents tutorials on mobile usability, designing interfaces for multiple devices, quantitative usability methods, cognitive psychology for designers, and principles of human-computer interaction. She also serves as editor for the articles published on NNgroup.com. Raluca coauthored the NN/g reports on tablet usability, mobile usability, iPad usability, and the usability of children's websites, as well as the book Mobile Usability . She holds a Ph.D. from Carnegie Mellon University.

Subscribe to our Alertbox E-Mail Newsletter:

The latest articles about interface usability, website design, and UX research from the Nielsen Norman Group.

• Scroll-Triggered Text Animations Delay Users
• Microinteractions in User Experience
• Working Memory and External Memory
• Designing for Kids: Cognitive Considerations
• Timing Guidelines for Exposing Hidden Content

UX Conference Training Courses

• The Human Mind and Usability
• Persuasive and Emotional Design

Subscribe to the weekly newsletter to get notified about future articles.

Watch Change Blindness in User Interfaces , 3 minute video with Maria Rosala :

You must have javascript and cookies enabled in order to display videos.

ELIZA Effect: Why We Fall in Love With AI

3 minute video

Primacy Effect: Put Important Things First

UX Animations

4 minute video

Animations Are Distracting!

2 minute video