Game accessibility and the seizure problem

disability-symbolsMomentum is growing to eliminate the barriers that make it difficult or impossible for people with disabilities to play video games. An assortment of game accessibility specialists and organizations are advocating with and educating game developers on how to adapt their offerings for gamers with physical, sensory, and/or cognitive impairments many types of disabilities. In addition some of these organizations work with individuals to resolve specific accessibility issues.

Q: How might this trend bring about progress on the seizure hazard that persists in many games?

A: Remains to be seen because game-induced seizures don’t typically receive a lot of advocacy attention amid so many other disabilities.

Some advocacy efforts focus on modifying the user interface to accommodate certain types of disabilities—such as mobility limitations that make a typical game controller too challenging. There are also two well-established game accessibility groups—both founded more than 10 years ago—that address all sorts of disabilities and could therefore actively promote the development of games that are unlikely to trigger seizures.

The leading advocacy players (as it were)

The Able Gamers Foundation is a non-profit, staffed by volunteers, that is supported by donations from individuals and some big names in the industry, including Sony, Harmonix, the Steam storefront, and others. Able Gamers advocates with and provides free consulting to game developers in order to open gaming up to individuals with disabilities. The organization published a set of game accessibility guidelines written by developers and by gamers with disabilities. The guidelines appear under the title Includification (I love that term). Accommodations for mobility impairments make up the largest category in the Includification booklet: remappable keys, compatibility with specialized input devices, and so on. Accommodations for hearing, vision, and cognitive disability are outlined as well. Able Gamers says that flash and flicker guidelines, not currently in the document, will be added in a revision scheduled for later this year.

Able Gamers does a lot of community outreach to gamers with special needs and tours the country demonstrating assistive technology devices that make game play possible for individuals who cannot use conventional controllers or other standard input/output technology. The group reviews games for accessibility features, and it provides online forums where disabled players can connect with each other. Able Gamers also provides grants to individuals who need customized assistive technology for gaming.

The Game Accessibility Special Interest Group of the International Game Designers Association, founded by a member of Stockholm University’s Department of Computer and Systems Sciences, is comprised of game developers and academic and industry experts in accessibility. The group presents at conferences, surveys developers, and organizes accessibility competitions for game design students. The goals of the SIG include advocacy, cross-industry cooperation, and creating a curriculum on accessible design that can be incorporated into existing post-secondary game design courses. The SIG is embarking on an ambitious multi-faceted agenda to move the accessibility agenda forward.

For developers seeking references on how to make games more accessible, the SIG recommends Able Gamers’ Includification and www.gameaccessibilityguidelines.com, a document that received an award from the Federal Communications Commission (FCC). The guidelines detail 100 accessibility features and include a remarkably clear outline of how to reduce the risk of flash- and pattern-induced seizures. The SIG also promotes a list of 14 accessibility features–such as easily readable closed captioning and colorblind-friendly design–that would have the most impact on gamers with disabilities. (Seizure risk reduction didn’t make the top 14 list.)

How photosensitive epilepsy is unlike other disabilities

I exchanged emails with London-based Ian Hamilton, an accessibility specialist and user experience (UX) designer active in the SIG. Hamilton was instrumental in organizing and producing gameaccessibilityguidelines.com. He respects the importance of seizure safeguards. “…even though epilepsy is relatively uncommon compared to things like colourblindness or dyslexia,” he wrote, “it’s one of the only accessibility issues that can result in physical harm, so it’s absolutely critical to get right.”

As he so rightly pointed out, the stakes (that include the potential for serious medical issues and life-long limitations) in the case of photosensitive epilepsy are quite apart from the frustrations of being unable to play because of, say, inadequate captioning. While photosensitive epilepsy absolutely belongs in any set of game accessibility guidelines, it differs in some other significant ways from the other conditions/disabilities that need accommodations:

  • It is particularly important to proactively reduce the seizure risks to all consumers since many people aren’t aware they have the condition and are therefore at risk. They’ve either not yet been triggered or they aren’t aware of seizures they’ve already had. Nobody really knows how many people are affected by photosensitive epilepsy, but awareness is low, diagnostic methods are unreliable, and estimates of how common it is are very conservative.
  • Advocates say many developers are responsive to specific requests for modifications that help those with a particular disability, because many accessibility issues stem from a lack of awareness by product designers of accessible design practices. Although it’s true that accessible design isn’t typically included in the game design curriculum, this argument doesn’t really hold up when it comes to seizure-inducing images. Game publishers have acknowledged for more than 20 years that their games could pose a risk to individuals with photosensitive epilepsy. The legal departments of game publishers should communicate with their own designers and developers about making games less likely to spark seizures. Many popular games continue to contain seizure-inducing images.
Just 340percent of games recommended by GamesBeat passed the seizure safety test.

In my tests, just 40 percent of games recommended in GamesBeat’s 2014 holiday guide were unlikely to trigger seizures.

  • Advocates encourage the involvement of individuals with disabilities in the design and testing of games for the benefit of all parties. For people with visually induced seizures, this isn’t feasible, nor is it necessary. Testing needs to be done without placing people at risk, using existing analysis applications that utilize research-based image specifications. For these consumers there is little chance of developing an ongoing collaborative relationship with the industry.
  • Despite a 2005 consensus paper by the Photosensitivity Task Force of the Epilepsy Foundation of America identifying seizures from visual stimulation as a significant public health problem**, there has been no organized advocacy to reduce visual stimuli in our everyday environment that can trigger seizures. Advocacy efforts (in the USA, anyway) to improve the lives of people with photosensitive epilepsy appear to be practically non-existent. In contrast, the interests of people with vision, hearing, and movement impairments are represented by organizations that proactively take on quality-of-daily-life issues.

Under the heading “Ways to further accessibility in the games industry” Ian published a comprehensive list of steps required for making video games accessible to people with disabilities. In addition to game developers and disabled gamers, the stakeholders involved would include educators in game design, game publishers and distributors, platform manufacturers, government bodies, trade groups, development tool builders, and so on. Among the recommendations: “Include basic access requirements in publisher certification requirements, such as subtitles and avoidance of common epilepsy triggers (both of these examples are required at a publisher level by Ubisoft).” Ian notes in his preface to the list that bringing about this accessibility within the game industry amounts to “wide-scale cultural change.”

Moving forward

I applaud the work of the Games Accessibility SIG, the Able Gamers Foundation, and other groups addressing of accessibility. They are making strides on an important issue while faced with the task of convincing the industry that design changes for accessibility will pay off in improved overall game design and a larger customer base.

But in order to make possible the massive and broad-based cultural change that’s needed, I believe these talented and dedicated advocates need significantly greater resources, buy-in, and recognition. The industry needs to declare publicly that it is committing itself to making gaming more inclusive of people with disabilities. In the face of frequent challenges about the contents of video games and their influence on young people, the accessibility issue offers a win-win public relations opportunity for video games. The Entertainment Software Association is just the organization to proactively announce an industry-wide goal of providing people with disabilities easier access to video game entertainment and learning. The ESA should establish and contribute major resources to a game indistry consortium for promoting and achieving accessibility education, standards, learning, and collaboration, leading to a more inclusive—and larger—customer base.

** “The Photosensitivity Task Force of the Epilepsy Foundation of America believes that a seizure from visual stimulation represents a significant public health problem. No known method can eliminate all risk for a visually induced seizure in a highly susceptible person, but accumulation of knowledge about photosensitivity is now at a level sufficient to develop educational programs and procedures in the United States that substantially will reduce the risk for this type of seizure.

— from Robert S. Fisher et al., “Photic- and Pattern-induced Seizures: A Review for the Epilepsy Foundation of America Working Group.” Epilepsia Volume 46, Issue 9 (September 2005), pages 1426–1441.


Video Game of the Year fails seizure safety test

game developers choice logoThe Game Developers Choice Game of the Year Award was given yesterday to Middle-earth: Shadow of Mordor, a title containing scenes that can provoke epileptic seizures.

Many people vulnerable to visually induced seizures are unaware they have this genetic sensitivity, unless they’re aware of prior seizures triggered by visual stimuli. This means that before playing this game, consumers should carefully consider the typically ignored seizure warning in the packaging. The game’s visual sequences violate guidelines for preventing seizures provoked by flashing and/or other provocative visual stimuli.

Game Developers Choice Awards are voted on by an invitation-only group of leading game creators across the video game industry, and are presented at the annual Game Developers Conference, an event that attracts more than 20,000 attendees. Here are some excerpts from the conference website describing the Game Developers Choice Awards:

  • the premier accolades for peer recognition in the digital games industry
  • innovation and excellence in the art of making games
  • recognize and celebrate the creativity, artistry and technical genius of the finest developers and games created in the last year

You get the idea. Consumer safety and inclusiveness, though, are apparently–no surprise–absent from the selection criteria.

Hearthstone: Heroes of Warcraft is the the only nominee for the Developers Choice Best Game Award that passed the seizure safety test. It's a digital version of a trading cards-based strategy game.

Hearthstone: Heroes of Warcraft, the only Game of the Year Award nominee that passed the seizure safety test, is a digital version of a trading cards-based strategy game.

Game of the Year Award

The winning game was chosen from a field of five nominees, only one of which–Hearthstone: Heroes of Warcraft (Blizzard)–tested as seizure-safe.***Please note that while my testing didn’t find unsafe sequences in Hearthstone, a Reddit poster had seizures triggered by the game. I’ve previously posted on the other three nominated games that tested as unsafe:

Bayonetta 2 (Platinum Games/Nintendo)
Destiny (Bungie/Activision)
Alien: Isolation (Creative Assembly/Sega)

At this point I explain how I test games for seizure safety and explain that there are no guarantees your results will match mine.

Testing methodology 

I run downloaded gameplay clips, cinematic clips, and promotional trailers for each game and submitted the sequences to the Harding Flash and Pattern Analyzer. The FPA is widely used by producers and networks in the UK—including by the BBC—to ensure seizure safety of all material on broadcast TV. It examines video sequences for very specific and measurable image qualities that researchers have found can trigger seizures:

  • rapidly alternating light and dark images (flash/flicker)
  • certain stripes and geometric patterns with high contrast
  • large areas of very bright (“saturated”) red
  • problem images take up more than one quarter of the screen

If the first clip I test of a game fails the safety test, I note that and move on to test the next game. I typically test at least 4 or 5 additional clips of a game if no safety violations are found initially. If I have a high level of suspicion due to the game genre and/or overall look of the sequences that there might be unsafe “footage” that I haven’t yet found, I may test a lot more clips.

Disclaimers***

Your results could vary. Games I’ve listed as safe could have seizure-provoking sequences that I was unable to locate. I don’t do this testing while actually playing these video games. Instead I work with video clips available online, some of which are official marketing and gameplay trailers; others are cutscenes and gameplay sessions posted by reviewers or fans. I avoid clips showing games that have been modified with other software.

In addition, the seizure threshold of individuals is affected by a number of factors including illness, hunger, stress, fatigue, length of play, and the player’s menstrual cycle, among others. So a game that seems OK may trigger a seizure under different conditions.


So, what’s new on the seizure thing?

newspaper stackNot much, unfortunately.

There are many reasons for low awareness of the seizures induced by video games, but today I want to focus on a big one: the lack of current research.

Without new research or a concerted outreach effort by advocacy organizations, it’s tough to keep the issue alive in the media. New research means new findings to announce. The press isn’t likely to cover a subject that’s producing no news.

It’s not as if conclusive studies don’t exist on seizures triggered by video games. Many clinical studies on video games as a cause of photosensitive seizures have been published, beginning in the early 1980s. The findings and methods have been refined over time, but the results have been fairly consistent and haven’t stirred up controversy. In fact, photosensitive epilepsy has always been considered a niche disorder and an out-of-the-way area of research. What’s left to prove?

Today’s studies involving video games are all about how games influence behavior and attitudes. Social scientists are investigating questions such as how much video games affect cognitive performance, memory, physical response time, behavior, and attitudes about violence. The results of ongoing studies like this—unlike the old news about seizures triggered by video games–frequently appear in the news media, keep the issues alive, and are topics of public debate.

The last time significant research findings were published and the advocacy community acted upon this public health problem was in 2005, when consensus papers on photosensitive seizures were published and the Epilepsy Foundation acted by issuing new guidelines for families on preventing them. Very little has occurred or has been published since then to keep the issue before the American public. (One exception: in 2008 the Epilepsy Foundation publicly responded to an Internet attack that placed seizure-inducing images on its website.)

Here’s an example of how things work in the news business.

Last month a pretty balanced piece titled “Video games can be good for you, studies say” appeared in the Chicago Tribune. The reporter, John Keilman, begins the piece by putting video game research into historical context.

Researchers have done thousands of studies on gaming since the 1980s, often with unmistakably negative results. Some associated video games with an increased risk of epileptic seizures, while others cautioned that the games could provoke dangerously elevated heart rates. Many studies also linked violent games to aggression and anti-social behavior.

Then the article turns to Christopher Ferguson, a psychology professor at Florida’s Stetson University who’s found positive contributions of video games:

Ferguson has done dozens of studies on the subject and has consistently found that violent video games do not contribute to societal aggression. One recent project actually concluded that some children who play violent games are less likely than others to act like bullies. [That’s pretty newsworthy!–JS] Ferguson said early research into any new technology is often flawed. Studies that aim to find negative effects get funded and promoted, while those with more benign findings are unpublished and forgotten, he said. When a new generation of scholars more familiar with the technology comes along, different results often appear — and that’s what is happening with gaming, he said. “We’re just not seeing the kind of data emerge that would support the techno-panic that was common in earlier years,” he said.

Although Keilman balanced Ferguson’s remarks with quotes from another researcher, who’s quite skeptical about recent positive findings, there is no further mention in the article of studies about the seizure problem. Readers can easily assume that—as with early studies linking violent games to violent behavior—all the research showing video games can trigger seizures stems from the “negative attitude by technophobes” in the early days of gaming. If one follows Prof. Ferguson’s line of thinking, one could expect that newer studies—if there are any–on seizures and games can be expected to reverse the earlier studies.

I contacted Keilman to point out that the seizure problem hasn’t gone away. He responded that he was aware of this and was interested in writing about it sometime.  He also mentioned continuing research on the seizure issue. I think he’d be interested in writing about the subject if there was some new study that’s newsworthy, but there simply isn’t.

In the absence of new research, or a Pokémon-type episode, it’s hard for journalists to write about a topic that just isn’t news.

 


Wearable seizure detector as consumer product

photosensitive epilepsy, smart computing

The Embrace watch doesn’t look like a medical device, but its sensor technology has been used in research published in leading neurology journals. Photo credit: Empatica

Later this year parents will be able to monitor a child’s seizure activity by using wearable technology in the form of a sleek, stylish watch with sensors underneath. The $199 Embrace watch, which will debut in October, could eventually offer a way for parents to learn whether the video games their child is playing are triggering seizures.

The Embrace won’t be able to do a thorough job of that yet—primarily because the device is currently most accurate for detecting tonic-clonic (grand mal, convulsive) seizures. Many seizures aren’t that type; they’re complex or absence, so the device would need to reliably pick up all seizure types. In addition, the Embrace hasn’t been tested specifically for picking up seizures triggered by visual stimuli. In its initial release, though, Embrace will be able to alert parents to tonic-clonic seizures while their child is asleep or in another room.

Most seizure detection devices (except EEG) rely on motion sensors that transmit an alert to a caregiver about a convulsive seizure. The Embrace wristband works primarily by detecting subtle changes in the flow of electrical charges on the skin. These changes in skin conductance are reliable indicators of changes in deep-seated areas of the brain associated with seizures, the hippocampus and amygdala. In times of cognitive, physical, or emotional arousal—and during seizures, these parts of the brain are activated. In some instances, changes on the skin can even be detected prior to the onset of a seizure.

As I’ve noted elsewhere, although EEG is the gold standard used by doctors to diagnose seizures, it’s limited in what it can pick up from structures deep inside brain. EEG uses scalp electrodes to detect electrical signals that first must penetrate the outer brain layers (cortex) and the skull. Because electrodes are affixed to the other side of these layers of tissue and bone, not all seizure signals can be detected at the surface.

The Embrace sensor technology was initially developed by the MIT Media Lab to help people on the autism spectrum to identify and communicate their stress level. The MIT team subsequently discovered that the sensors could detect not only stress/arousal levels but also seizures, including some seizures not picked up on an EEG. A start-up was formed, Empatica, to bring the wearables to market with the help of Indiegogo crowdfunding. I’m intrigued by the company’s roots in technology to aid people with autism; the autism community is probably at highest risk for photosensitive epilepsy and would therefore significantly benefit by being able to identify visually triggered seizure activity.

Sensors on the underside track skin conductance and body movement. Credit: Empatica

Sensors on the underside track skin conductance and body movement. Photo credit: Empatica

If your child is fatigued and kind of “out of it” at the end of a gaming session, imagine being able to find out whether hidden seizures are occurring during certain games. Wouldn’t it be great to be able to correlate seizure activity with specific video games? I’d love to see this application of the Embrace watch come about.

A video about Embrace is available on Empatica’s Indiegogo page.


2 of these 4 Xbox games don’t cause seizures!

Halo: The Master Chief Collection was the only game of the recommended Xbox titles that poses a seizure risk.

First-person shooter game Halo: The Master Chief Collection was one of the recommended Xbox titles that poses a seizure risk.

Half of the Xbox-exclusive games selected by GamesBeat for its 2014 holiday gaming guide appear to be free of seizure-provoking visuals. But since there were just 4 Xbox games in the buying guide, we’re talking about exactly 2 safe games. I don’t know how representative this sample is of the universe of Xbox-exclusive titles.

About these games and test results

Disney Fantasia: Music Evolved isn’t a game in the conventional sense; it’s a rhythm game played by the motion of the player’s arms, enabled by the motion-sensing Kinect controller. There are no battle scenes that might set off visually risky explosions, or races with crashes and fast-moving scenery. Instead, when players “conduct” to the beat of music selections, all sorts of colored patterns are created and set in motion on the screen. It’s visually interesting yet not overstimulating. There’s a lot of safe black space separating the smaller colorful elements.

GamesBeat Xbox chart

Sunset Overdrive, released by Microsoft Studios, is a third-person shooter game, a relatively safe genre seizure-wise. Because these games tell a story from the perspective of an observer of the action, scenes are shown from a wide angle. This means that, unlike first-person shooters, any bright, explosive flashes and rapidly swirling debris don’t dominate the field of view. The larger the screen area of patterns flashing images, the higher the likelihood is that those visual effects might provoke a seizure.

The same Halo image at the top of this post is shown here in the upper left of a screen capture from the seizure safety testing application. That image occurs at the start of a stretch of flashing red that goes well beyond safety guidelines for photosensitive seizure prevention.

The Halo image at the start of this post is shown here in the upper left of a screen from the seizure safety testing application. The image occurs in a stretch of gunfire with flash levels exceeding seizure safety guidelines, as shown by the red line on the graph.

So it’s not surprising that first-person shooter Halo: The Master Chief Collection contains seizure-provoking images. Its predecessor Halo 4, which I tested 2 years ago along with other FPS games, failed the flash and pattern analyzer safety test, too.

Not so surprising, either, that the racing game Forza Horizon 2 violated seizure safety guidelines. As I’ve previously found, the quick cuts, fast-moving scenery, and dramatic collisions featured in the racing genre result in games that exceed or approach seizure safety guidelines. The original version of the game, which I tested nearly a year ago, contained a lot of scenes that flirted with safety limits, but I didn’t locate actual safety violations.

Summing up GamesBeat’s 2014 holiday gift guide

This completes my safety assessment of games chosen for GamesBeat’s 2014 holiday gift guide. In recent posts I’ve reported on the seizure safety of the guide’s recommended multiplatform games and games exclusive to Nintendo, Sony, and PCs. After testing all 43 games in the GamesBeat 2014 holiday guide, it appears that 26 titles–60 percent–contain image sequences capable of provoking seizures.

Just 40 percent of games recommended in the GamesBeat holiday guide passed the seizure safety test.

These are supposed to be the best games–how seizure-safe are the worst games? Or the not-especially-noteworthy? If I had unlimited time and testing resources, we could find out. Consumers deserve to see reviewers rate games for seizure safety, not just for graphics quality, speed/performance, modes of play, characters’ personalities, levels of difficulty, modes of play, frame rate, music, and so on.

Disclaimers

Your results could vary. Games I’ve listed as safe could have seizure-provoking sequences that I was unable to locate. I don’t do this testing while actually playing these video games. Instead I work with video clips available online, some of which are official marketing and gameplay trailers and cutscenes; others are gameplay sessions posted by reviewers or fans. I do not test fans’ gameplay from games that were acknowledged modified with other software.

In addition, the seizure threshold of individuals is affected by a number of factors including illness, hunger, stress, fatigue, length of play, and menstrual cycle, among others. So a game that seems OK may subsequently trigger a seizure under different conditions.

Testing methodology 

I run downloaded gameplay clips, cinematic clips, and promotional trailers for each game and submit the sequences to the Harding Flash and Pattern Analyzer. The FPA is widely used by producers and networks in the UK—including by the BBC—to ensure seizure safety of all material on broadcast TV. It examines video sequences for very specific and measurable image qualities that researchers have found can trigger seizures:

  • rapidly alternating light and dark images (flash/flicker)
  • certain stripes and geometric patterns with high contrast
  • large areas of very bright (“saturated”) red
  • problem images take up more than one fourth of the screen

If the first clip I test of a game fails the safety test, I note that and move on to test the next game. I typically test at least 4 or 5 additional clips of a game if no safety violations are found initially. If I have a high level of suspicion due to the game genre and/or overall look of the sequences that there might be unsafe “footage” that I haven’t yet found, I may test a lot more clips.

 


Seizures from 7 of 11 best Nintendo games

A 1994 case history in an Irish medical journal documented seizures in a 13-year-old boy that were triggered by a Sonic the Hedgehog video game. Current versions of the game continue to include seizure-inducing sequences.

A case history in a 1994 medical journal documented seizures in a 13-year-old boy that were triggered by a Sonic the Hedgehog video game. The Sonic games just tested continue to include seizure-inducing sequences.

Been thinking about buying any of the 11 titles that GamesBeat thinks are the season’s best Nintendo-only games? Testing shows that just four seem to be free of seizure-provoking visuals.

Among the Nintendo games that don’t comply with seizure safety guidelines are several that one would expect to be seizure-inducing. Two brand-new animé-style games, Persona Q and Hyrule Warriors, predictably failed the seizure safety test because animé typically does. Unsafe graphics in Sonic the Hedgehog and Super Smash Bros., two of the most well-established game franchises, are consistent with seizure-provoking games in their respective product families.

Cooking Mama 5, Bayonette 2, Sonic Boom, Super Smash Bros., Theatrhythm Final Fantasy, Hyrule Warriors, Persona Q

GamesBeat picked these Nintendo-only titles for its early holiday gaming guide this year.

Pokémon video games vs Pokémon video game trailers

Of the three Pokémon games, one isn’t really a game so much as a set of digital drawing and painting tutorials. Interesting that the other two Pokémon games are safe, too. I instantly associate Pokémon with the so-called Pokémon incident in 1997. That event was a Pokémon cartoon broadcast, when about 600 Japanese children simultaneously experienced seizure symptoms severe enough that they were taken to emergency rooms.

Maybe because of that history, I thought, the developers of these games have taken extra care with image safety? To test this theory I put a few other Pokémon games through an image safety assessment tool—the others passed, too.

Hold on, though. Although the Pokémon games themselves still use the same not-very-animated, not-very-2014, unlikely-to-provoke-seizures style of animation of much older games, a promotional video for these same games violates seizure safety guidelines big time. How can that be? Here’s how: a just-released trailer for Pokémon Alpha Sapphire and Omega Ruby is done in a more traditional animé style. So technically the Pokémon games pass the seizure safety test–but beware of seizures from the videos that market the games.

A new animé-style  trailer uses much bolder graphics than the Pokémon game it promotes.

The new animé-style trailer for Pokémon Alpha Sapphire and Omega Ruby uses much bolder graphics …

...than the game it promotes.

…than the Pokémon games it’s promoting. While the game is safe, the eye-catching trailer fails the seizure safety test.

Unaware of the seizure hazard from animé, Sports website Sportskeeda.com commented:

“To celebrate the upcoming release, the Pokemon Company has launched an anime style trailer for the two games…It’s pretty cool of Pokemon to release an anime styled trailer, given how popular the series and manga is. And even though the anime and game are very different from each other, fans of the series are still going to appreciate the extra effort.”

When The Pokémon Company in 2012 used this same marketing strategy with an animé promotional trailer to “celebrate” the release of non-animé Pokémon Black 2 and Pokémon White 2, Youtube viewers of the combined trailer lamented that the (seizure-inducing) animé in the promotional clip looked a lot better than the humdrum (visually safe) appearance of the games’ graphics.

These visually jazzed-up trailers would seem to raise customers’ expectations that in future releases The Pokémon Company intends to incorporate animé style into cutscenes and perhaps gameplay.

Testing Methodology

To test the games I submit downloaded gameplay clips and promotional trailers to the Harding Flash and Pattern Analyzer. The FPA is widely used by producers and networks in the UK—including by the BBC—to ensure seizure safety of all material on broadcast TV. It examines video sequences for very specific and measurable image qualities that researchers have found can trigger seizures:

  • rapidly alternating light and dark images (flash/flicker)
  • certain stripes and geometric patterns with high contrast
  • large areas of very bright (“saturated”) red
  • problem images take up more than one fourth of the screen

Sequences submitted for analysis are video clips available online, some of which are official marketing and gameplay trailers; others are gameplay sessions or cutscenes posted by reviewers or fans. I do not test clips uploaded by fans who have modified the game software.

Action game Bayonetta consists of long fight scenes filled with motion and bright colors.

Action game Bayonetta 2 consists of long fight scenes filled with motion and bright colors.

I typically test at least 4 or 5 clips of a game if no safety violations are found in the first clip. If I have a high level of suspicion due to the game genre and/or overall “look” that there might be unsafe images that I haven’t yet found, I may test a lot more clips. Before finding a clip of Bayonetta 2 that failed the test, I tested about a dozen clips. Its large, fast-moving, brightly colored designs led me to believe there had to be seizure-inducing sequences.

Disclaimer

Your results could vary. Games I’ve listed as safe could have seizure-provoking sequences that I was unable to locate. Furthermore, the seizure threshold of individuals is affected by a number of factors including illness, hunger, stress, fatigue, alcohol, and the player’s menstrual cycle, among others. So a game that has caused no problems over an extended period of use could still trigger a seizure under different conditions.


Best new PC-only games are safe–pretty much

ArcheAge, World of Warcraft: Warlords of Draenor, Civilization: Beyond Earth, Gauntlet, Neverending NightmaresBetter news than usual. For people susceptible to photosensitive seizures, the five best exclusive-to-PC games for the holiday shopping season (per GamesBeat) appear safe. Well, mostly.

There’s one very important caveat: While the gameplay itself seems to stay within visual safety guidelines, a cinematic clip from World of Warcraft: Warlords of Draenor failed image safety tests. You can apparently play World of Warcraft safely, but you need to beware of the risk of seizures from the cinematic game openers and other in-game cutscenes and cinematic material that move the story along. I didn’t give the game a passing rating because it’s likely that players will see the problem video sequences.

Going forward, when testing games for seizure safety, I will routinely test cutscenes in addition to gameplay clips. Unfortunately it’s quite possible that in preparing prior posts I’ve been unaware of some safety failures because I wasn’t systematically looking to test these sequences. I plan to revisit the test results I’ve recently posted and will be updating those posts with any new findings.

Four of the five games are either role-playing (RPG) or strategy games, which tend to be shown from a wide angle that permits scenes showing entire battle fields. These “big picture” scenes are less likely than close-up shots to have areas of flash, saturated color, and patterns that take up a major portion of the screen. Unless these visual effects cover more than one quarter of the screen they do not typically pose a seizure risk.

The flash problem

In the problem sequence in Warlords of Draenor excerpted below, a bright orange screen alternates with a darker screen. The switch between dark and bright happens several times within a fraction of a second. The rapid sequence creates a flash, and because it occurs faster than 3 times per second, it exceeds the flash rate safety limit for photosensitive epilepsy.

Alternating images in the video sequence create a flash effect.

The bright orange frames alternating with a black and white image in this sequence from World of Warcraft: Warlords of Draenor creates a flash effect. This sequence of screens, captured by the Harding Flash and Pattern Analyzer, represents roughly a third of a second of video.

For people with photosensitive epilepsy, flashing red (or shades of red) creates an even higher seizure risk. When the video sequence (at bottom of the analysis tool screen capture below) is assessed for seizure safety, an unsafe level of flash is shown in the green line on the graph. The level of red flash (red line on the graph) is even further beyond the safety limit.

Disclaimers

Your results could vary. Games I’ve listed as safe could have seizure-provoking sequences that I was unable to locate. I don’t do this testing while actually playing these video games. Instead I work with video clips available online, some of which are official marketing and gameplay trailers; others are gameplay sessions posted by reviewers or fans. I do not test fans’ gameplay from games that were acknowledged modified with other software.

In addition, the seizure threshold of individuals is affected by a number of factors including illness, hunger, stress, fatigue, length of play, and the player’s menstrual cycle, among others. So a game that seems OK may trigger a seizure under different conditions.

Testing methodology 

I run downloaded gameplay clips, cinematic clips, and promotional trailers for each game and submitted the sequences to the Harding Flash and Pattern Analyzer. The FPA is widely used by producers and networks in the UK—including by the BBC—to ensure seizure safety of all material on broadcast TV. It examines video sequences for very specific and measurable image qualities that researchers have found can trigger seizures:

  • rapidly alternating light and dark images (flash/flicker)
  • certain stripes and geometric patterns with high contrast
  • large areas of very bright (“saturated”) red
  • problem images take up more than one fourth of the screen

If the first clip I test of a game fails the safety test, I note that and move on to test the next game. I typically test at least 4 or 5 additional clips of a game if no safety violations are found initially. If I have a high level of suspicion due to the game genre and/or overall look of the sequences that there might be unsafe “footage” that I haven’t yet found, I may test a lot more clips.

In recent posts I’ve shown the results of testing editors’ picks for best-of-Sony and best-of-multiplatform games for this year/holiday season. Both groups included lots of seizure-inducing titles. In coming posts I’ll look at best of Nintendo and Microsoft games.

 

 

 


Follow

Get every new post delivered to your Inbox.

Join 44 other followers