Archive for August, 2009

The Social Psychology of Hulu

Social psychologists have for decades recognized that cognitive dissonance, the uncomfortable sensation of holding two conflicting ideas in mind, leads to some strange behavior. For example, Leon Festinger once infiltrated a religious cult on the brink of an end-of-the-world prophecy. Upon seeing that there deeply-held belief of salvation before a great flood failed to materialize on a specific date, the followers were faced with a choice: recognize the failure and abandon their convictions or rationalize the mistake and become more devoted.

As has happened countless times throughout history, most of the members chose the latter path. The reason, according to Festinger, is that they had publicly and privately expressed their beliefs, a process that has powerful effects on thoughts and behavior. Out of a desire to act consistently, the members concluded that the world was going to end, but their belief spared the planet of suffering.

While these findings are fascinating, it may seem to be an extreme case, and thus not applicable to people in general. However, other experiments have shown that cognitive dissonance compels all people to behave in unexpected ways. People are far more likely to show affinity towards products or services when they are first induced to state a preference for it. While this effect is greatest when individuals state it publicly, even a seemingly ‘private’ choice, wherein a preference is written but unshared, has a measurable effect on how one will perceive a product in the future.

I mention all of this because I saw a new advertising feature on Hulu.com that harnesses the effects of cognitive dissonance and a desire for consistency. Hulu, as you probably know, is a website that presents television shows, movies, and other professionally-produced videos under legal license from studios. Generally, it is supported through short advertisements (presented during standard broadcast commercial breaks, but much shorter than what one would see on television). Typically, the experience is fundamentally the same as watching TV.

Yet recently, the site has used a different model. Before some videos, users are given the option to select which company they would prefer sponsor the video. For most users, this probably seems like an insignificant choice. I highly doubt that people prefer which commercial they view, and as such little conscious thought likely goes into the choice. On a subconscious level, however, this is very important.
For one, users are probably likely to automatically select a brand that they already have some affinity for. So selecting that brand will reinforce existing ideas. More importantly, however, are the implications of cognitive dissonance and a desire for consistency. Because the viewer chose to see that particular advertisement, subconscious processes will force them to feel that they did so for a reason. Specifically, because they felt a connection with that particular brand.

Ad options

Which shall I choose? (click for full size)

If I chose to watch a BlackBerry advertisement, I would most likely respond more positively to BlackBerry than to Applebee’s a short time after viewing the video. With enough repitition, this process would create a long-term connection between myself and the brand. More to the point, however, I would be more receptive to BlackBerry after choosing the ad than if the exact same ad were presented without a choice.

In other words, the consistency bias will really have an effect not simply because I chose BlackBerry over Applebee’s, but because I simply made a choice. As soon as I have done that, the subconscious machinery of my mind will fool me into feeling like I made a selection I cared about, when in reality, I probably just clicked a link.

With that in mind, I wonder about the effects of a different form of advertising the site uses. Occassionally, Hulu will give me an option to watch a full-length trailer for a film before watching a video, or viewing thirty-second commercials during the video. Is this, as Advertising Age suggests, simply a nice bit of choice given to the consumer?

Just making a choice (click for full size)

Just making a choice (click for full size)

I doubt it. The very act of making a selection activates specific, powerful forces of social psychology. Watching TV, I never have to think about how I want to watch an ad for All About Steve. But if the site forces me to choose, my brain will have to decide why I chose how I did. And that process leaves an indelible mark on my perception of the RomCom.

I Always Have More Questions

Over at the Groundswell blog, Josh Bernoff has an interesting update. According to the latest numbers, the number of people engaging in social media as joiners or spectators (ie, members of Facebook, blog readers, etc.) has been growing steadily. However, creator and collector activity (writing blogs, participating in digg, etc.) has grown slowly, and critical activity has actually declined.

Josh writes:

If you believe in the future that everybody will be creating or organizing content, we disagree — it’s a matter of temperament, not technology.

I have tried to say the same thing repeatedly in the past. In fact, this is the crux of my central disagreement with Clay Shirky’s Here Comes Everybody; the fact of the matter is that only some people are coming. Bernoff and Eszter Hargittai have both recognized that different demographic groups have different levels of participation, and these differences will have significant effects on the structure (and thus output) of the social web.

Continue reading ‘I Always Have More Questions’

The Parallels are Uncanny

The headline of my last post compared Twitter to MySpace. The message I was trying to send was that Twitter was a fast-growth company with a lot of media hype that it couldn’t possibly back up. The fact that neither company will be able to live up to the challenge of Facebook was also implicit in the headline. I have since realized that there is a much better analogy here:

 

equation

Continue reading ‘The Parallels are Uncanny’

Twitter = MySpace, or How I Learned to Stop Worrying and Love to Gloat

It seems like Facebook may finally be poised to snuff out Twitter, something I predicted back in December. Ten days ago, the ‘Book announced that it was purchasing FriendFeed, presumably in an effort to finally open up the Publisher feature to the public. To reiterate what I said then:

…my whole take on Twitter is that it’s probably an excellent feature, but that’s pretty much all that can be said about it. If Facebook goes about implementing a similar feature properly (which I suspect they will), there probably isn’t much of a future for the little-microblogger-that-could.

I’m surprised that Facebook is only now getting around to doing this, but it’s probably better to get it right than to do it quickly, especially considering how puny Twitter’s userbase is (relatively).

I’ve read a lot of comments online wondering why people are trying to compare Facebook to Twitter. This argument assumes that since the two services have different goals, both are sustainable for different purposes–a foolish idea. I say this because, at the end of the day, a vast majority of users (ie, those that don’t TechCrunch, Valleywag, etc.) want to use as few services as possible. For most people, having to learn and use both Facebook and Twitter is a pain, especially when there is an opportunity to use only one.

In other words, Twitter is a good feature. But a good feature does not a successful company make.

I am not a Visual Thinker

The title of this post should be obvious to anyone who reads this blog. I think in words and ideas. Never in images.

This has recently come into sharp focus from me as I have been reading Envisioning Information by Edward R. Tufte. Meant as a guide for graphic designers and visual thinkers in developing charts and graphs to display information, the book is relatively short on words and long on illustrations. For me, reading it has been an exercise in tenacity; very often, the author will make a concise assertion and present an image as evidence.

I continually find myself asking ‘why?’ For those of us who don’t think in pictures, or at least have difficulty accepting visualizations, such evidence is bewildering. As an example, I was just reading a section wherein Tufte suggests that (counter-intuitively) sans-serif fonts are harder to read than those with serifs.  As proof, he shows side-by-side images of equal text, with no other explanation.

He obviously knows more about the topic than I, but I have trouble accepting his explanation. If he had provided evidence from a study that concluded the same thing, I would no doubt feel differently. But how do we know that this is the case? Perhaps one is simply more pleasing to him.

Continue reading ‘I am not a Visual Thinker’

Finally, some empirical data

One of the reasons I started this blog was to continue exploring the effects of social media on political campaigns. For well over a year now, I’ve been saying that I thought that Democrats had a big advantage over Republicans in this area, and that this advantage was not due to strategy differences between the parties. Rather, these differences are largely because of the breakdowns of who participates in these media.

Until now, I had to rely on indirect methods to demonstrate this. Democrats were younger, more likely to be in college, etc., and thus more likely to use social media tools. But I’ve finally come across some data that finally reinforce these ideas. Check out this page. [With thanks to Josh Bernoff at Forrester Research].

Bernoff writes:

Right away you can see there is a persistent tendency for Democrats to participate more fully in social technologies. Looking at the index (all adults = 100), you can see that Democrats are at least 10% more likely to do just about anything involving social technologies. The Republicans are the opposite — they’re a lot LESS likely to participate (like Nixon’s “silent majority”). They’re 22% less likely to be a social network (Joiners) and 21% less likely to be uploading video or blogging (Creators). These are not extreme differences, but there are definite tendencies here, likely correlated to the fact that Republicans tend to be older than Democrats on the average. Notice that the Republicans are near par for Spectators, though — they’re watching, even if they’re not as active participants. The Independents are somewhere in the middle, approaching the average in Joiner and Critic activity.

Interesting, no? But I wonder what these technographics will look like in a few years as social technology becomes an even greater part of our lives. Will Republicans finally catch up? I don’t think so, because I believe that they have done too much to alienate the creative class, which overwhelmingly supports Democrats.

And while social technology becomes more relevant in setting our national agenda, this will become an even greater problem for the GOP. Even if conservatives can catch up, their supporters will still have far less experience (and, presumably, skill) using these tools.

I’ve said all of this before, but it’s nice to see some direct data.

Side note: the especially high indices for Al Gore among social media participants makes me wonder what would have happened if those tools were available in 2000. With such a razor-thin margin that year, do you think that those differences would have swayed the election? I do.

In my own experience…

Over at Slate, there’s an interesting little debate happening in response to an article on bullying. The article was written by Alan Kadzen (former president of the American Psychological Association) and John Musser (professor of child psychology at Yale), and is intended as a guide on how to respond to bullying. It’s a worthwhile read, but I’m interested in what is happening in The Fray, Slate’s in-house forum.

While it isn’t really surprising (considering that the authors literally and hilariously asked for it), a majority of the responses have taken on a distinctly anti-intellectual tone. “These professionals don’t know anything”, posters seem to say, “this isn’t how it happened in my life. I’m gonna knock them out of their ivory tower and take them down to the real world”.

These users might be entirely correct, but their reasoning isn’t valid. It seems so obvious to me, but it’s worth stating explicitly: actual, large-scale research is way more valid that subjective experience. The Fray’s users seem to not understand that.

I mention all this because I think it relates to the issues of social media and their effects on society that I write about here. Specifically, it stands as evidence against techno-utopianism, the notion of a marketplace of ideas, and memetics. In varying ways, each of these theories is based on an analysis of information that is divorced from the users–the processors and distributors of that information.

Techno-utopianism assumes that, in response to the free creation and distribution of information, technology will remove the artificial barriers that are holding people back. In the marketplace of ideas, there is the assumption that ideas compete with one another directly, and that the best ideas will surface to the top naturally. Memetics treats abstract information as if it were genetic material, mutating, adapting, and spreading based on ‘fitness’.

All of these approaches are missing an appropriate consideration for the users themselves. Humans do not simply process information like machines, reproducing ideas with some degree of fidelity. We process information according to an incredibly complex web of cognitive and subconscious properties, and that has a very significant effect on the transmission of information.

In the Slate example, I can’t make any valid judgments regarding the posters backgrounds, but it seems likely that there aren’t many people with backgrounds in the psychology of bullying. Yet they are disagreeing with the psychologists on that very issue, based on personal experience. I don’t mean this to be a rant against these users; it is entirely natural (and very human) to consult personal experience against presented evidence. The problem is that it isn’t valid, and is subject to a wide variety of biases and errors.

But it is part of human behavior. We need to recognize that if we are to develop good models of information.

Head-End DVR

Josh Bernoff from Forrester research has written about a new ‘remote DVR’ service proposed by Cablevision, a New York cable-TV provider. The basic premise of the service is that television programs would be stored at the company’s central servers, rather than on home DVR units. Retrieving recorded programs would be more like selecting video on demand content than today’s DVRs.

As Bernoff explains, the system would be vastly more efficient than current designs. Rather than having every person who wants to record popular programs (he mentions Lost) redundantly use disk space to make hundreds of thousands of copies, a single file could supply the program to all interested users.

This is a fascinating concept, not only because it eliminates available disk space as a limitation to the DVR consumer. More importantly, it would effectively remove the time element from broadcasting. Assuming that at least one customer on the system requested a recording of any given program at any time, Cablevision’s system would become a massive archive of every moment of every broadcast from the moment it was launched. Consumers could go back in time and watch anything ever broadcast.

Continue reading ‘Head-End DVR’

Two More Charts

These come from data gathered from a random sample on Yelp.com of neighborhoods in Chicago. Sorry, don’t have time to get a real post in because I’ve been trying to do a full analysis, which will be up soon. For now, what do you think? See a trend here? (hint: look at the trendlines [see what I did there?]).

Y-axis is number of responses, X-axis is number of stars rated.

localservices

Number of Responses by Rating for Local Services in sample of Chicago neighborhoods (Yelp.com)

healthmedical

Number of Responses by Rating for Health and Medical Services in sample of Chicago neighborhoods (Yelp.com)

Completed Chart

metacriticlegit

Y-axis represents total count; X-axis represents user ratings.

N (ratings)=3,064 ;N (Movies in sample)=100

Next Page »