Why your review is wrong: the psychology of internet commentary

Image source: xkcd

This all started with the phrase, ‘Buying it anyway’. Three little words that appeared as the first comment on our 7/10 Assassin’s Creed 3 review. Several hundred well reasoned words and expert opinion formed from time spent with the game brushed away with one dismissive sentence fragment. This response, and many like it, appear constantly in the world of game reviews. And rarely as politely phrased. Belief and unsubstantiated opinion seem to far outweigh experience when a score doesn’t match expectation. Curious to find out more I spoke to Jamie Madigan, the psychologist behind the psychologyofgames.com.

So what is it that drives internet commenters to decry reviews they don’t like? Often prior to release, where the reviewer(s) are the only people who’ve spent any significant amount of time with the product. “There’s a couple of things going on,” says Jamie Madigan. “The foremost is called ‘confirmation bias’. It’s essentially that our brains are wired to want to seek out information that confirms beliefs or hypotheses we have, and ignore information that goes against them”.

“Our brains are wired to want to seek out
information that confirms beliefs we have”

Essentially, like Mulder, gamers want to believe: in their opinion, in their game, in their choices (hello pre-orders). “People will pay more attention to what confirms their beliefs,” says Madigan, “and will put more weight on that in comparison to other information, and certainly in comparison to information that runs counter to their beliefs”. And not only will people put more emphasis on what goes in their favour they’ll distort other areas to fit their agenda. “If there’s any room for ambiguity, needing to interpret what someone means, they will do so in a way which confirms their beliefs. People will find way to interpret and ignore the information which allows them to keep the opinions which may have been established”.

It’s something hardwired into the brain to the point of affecting our recollection of the past, as well as our perception of the present. “People’s memory will be biased,” explains Madigan. “People forget, or find it more difficult to access, from memory, information that disconfirms their beliefs. There’s been a lot of studies to test memory. Even given incentive people find it more difficult to remember things which upset their existing opinions or hypotheses”.

Perhaps more interestingly, according to Madigan, these highly held opinions aren’t always even formed by the individual arguing them so passionately. “There’s a concept called ‘social proof’, where people will take their cues from a group about what’s expected of them,” he explains.

An example he gives is a series of experiements first conducted in 1951 by Solomon Asch. In these tests a volunteer sat in a group while being shown a series of lines of different lengths and then choosing one that matched the length of a reference line. Without the subject knowing, all the other people in the group were part of the test and told to unanimously select the wrong answer. The result was that the test subject would often go along with the wrong answer to join with the group and avoid standing out. Another example along these lines involves a single person in a room full of actors who have been told to ignore the smoke pouring out from under a door. If none of the stooges react most test subjects sit, often unhappily, in what they think is a burning building but obediently follow everyone else’s lead in ignoring the ‘fire’.

Or, to put it another way, if everyone in a lift faces the wrong way, people will try to fit in:

Interestingly the gaming press, can be partially to blame for creating the social proof that later causes the problem of disputed reviews says Madigan. “I think you can get some early opinion makers and the effect is – if that person is held in high regard [or] enjoys any celebrity status – that effect is even stronger”. So the first big trailer or an early preview can create a mold the audience tumbles into, taking on its shape forever. If the final product doesn’t deliver, or turns out to be better than expectations, it can make little difference to the opinion already formed.

The nebulous subject matter of gaming can exacerbate the issue. “It’s also been shown to be a lot easier to create that effect when the topic is ambiguous,” points out Madigan. “[With] abstract concepts like ‘which of these is prettier’ or ‘which of these would be more fun’, you get that social effect a lot more readily and a lot more strongly”.

Resident Evil 6 ps3 screensAnother thing Madigan points to is something called “social identity theory” Here he cites an experiment where, “professors had two groups of teenage boys and said ‘look at these paintings by two different artists – which ones do you like better?’ It was totally arbitrary and the kids would go, ‘Well I like this one. I like this other one’, and that was all it took in their minds for them to say, ‘Well I’m part of this group who likes this painting, and they are part of the group that likes this other painting’”.

That simple and apparently meaningless grouping was enough to create a bond, highlighted by the boys then playing games. “They had them play this sort of game where they distributed fake money, points. They made it known what group they were in – whether they liked this painting or that painting [and] they found stark favouritism. People would give more points to those in their group because they felt some sort of camaraderie with this person on this completely arbitrary thing”.

“You don’t have to give your name, nothing
can be traced back, you can edit things after
the fact. You’re free of consequences”

So opinion can be group lead but the internet also has its own effects that enable people to voice their opinions. “I think there is a couple of things going on with the internet that exasperates this kind of thing” says Madigan. “One of the things is the anonymity, in psychological speak it’s called deindividuation, it’s de-individualising. If you’re in the context of a forum for debating and arguing then you may pick up on that”.

As he explains, “If you feel anonymous [you then] think ‘it’s okay for me to express belligerent or extreme opinions.’ The fact you don’t have to give your real name, nothing can be traced back to you, you can edit things after the fact. You’re free of consequences. In a way very unlike you are in face to face conversation or other types of communication”.

Fortunately, while most comments suggest otherwise, Madigan thinks, “there’s still something to be said for content experts”. As he points out, “crowdsourcing opinions, or groups to aggregate opinions, to find the true opinion of this game is an appealing idea, but there’s still something to be said for content experts”. As he puts it, “people who know how to evaluate and compare, and communicate experiences have value” and that value is a particular skill set: “Not everybody has that content knowledge and knows about the history of a game or knows about how features in this game are competing with others, to help people make informed decisions in the marketplace”.

So the short of it all is that once people form opinions they don’t like changing them; they actively want to fit in with groups, and say bad things when they think no one knows who they are. That’ll help a little next time someone tells me I’m “not fit to do my job”. I remember you.

Read more about Jamie’s work at www.psychologyofgames.com