Feeds:
Posts
Comments

Posts Tagged ‘collective behavior’

This is a very cool blog.

This post, a summary of collective behavior research, is a nice follow up to my last few posts.

 I also happened into this visualization of voter behavior from the last presidential election.  It would be fun to extend this into specific issues but it’s pretty clear that consequences shape behavior and how their are different consequences by context (economic status, population density, university density, high mover index, etc. etc.).  What’s hard to figure out from a one year snap shot are the schedules.  How much does this map shift by person/family if we look at the same sample over 20 elections?

Read Full Post »

I received the following from a director of ad operations, Jeremy Jones, in response to yesterday’s post on collective behavior and the other day’s post on vertical media.  His remarks focus on the confusion over social networking advertising (both selling and buying).  I include the full remark here and an annotated remark below.

[Social networks/social sites] struggle with how to sell the concept to advertisers.  No one gets it, not even the sn sites themselves.

I had believed that social networking and behavioral targeting was just a fad.  That it will end up just being another tool in the arsenal once somebody actually figures it out.  I think it won’t really take off until someone develops a standard that everyone can follow.  Until then it’s just a mishmash of technologies and ideas that are just tangents of the same idea.  Being in the industry, we hear about every new concept of trying to aggregate behavior and bucketing those behaviors to try and create useful targeting.  It all revolves around collecting site behavior.  Some of the more interesting ones actually involve trying to apply psychological profiles and social classifications based on the user’s browsing behavior.

Is sn or bt about the consumers or the influencers?  Up to this point, anyone trying to use influencers has typically been exposed.  All those attempts at viral are quickly uncovered as manufactured.  It is true what they say that the internet is the world’s greatest laboratory.  The collective intelligence will typically uncover a insincere effort rather quickly.  SEO has generally been regarded as a good thing.  Even though it has been used by every advertiser to increase their influence.  It has ensured that any content worth knowing can be found.  But what about social networking and behavioral targeting?  The impact of social networking is just beginning to be felt.  So much personal data.  Will the collective eventually see this as a good thing or bad?  It’s influence will be far greater than SEO.  Will they approve of having their personal details used for the purposes of a more efficient marketing machine?  Can they do anything about it if they really wanted to?

One of the current problems with behavioral targeting is that no one has developed a method of success reporting.  What is a success metric for a bt campaign?  Advertisers are still using visits and clicks as success metrics.  Widgets is an interesting method.  It requires adoption by users, and once adopted, do they interact with your widget?  Consume it’s content, pass it along to friends?  Do friends adopt it without prompting?  It certainly opens up a whole new area of measurable behaviors that were never available before.  This gets back to the influencer vs the consumer.  How do you target the influencers?  Heck, how do you identify the influencers.  That, I believe is the tricky part.  The influencers are not a stable group.  They are never the same.  They can be an influencer one day, and an consumer the next.  If you could come up with a method for identifying the influencers for any advertiser then you would really have something.

I think this response underscores the general discomfort with media and advertising right now.  We have the technology to deliver interesting experiences and we collect a lot of data.  We simply don’t have a good enough technology/science of behavior that has left the ivory tower and trickled into business intelligence.

We focus too much on the “biological details”(e.g. who are the influencers, where’s the meme/viral agent/contagion, what is the makeup of the content, onbrand strategies) that we’re losing site of the behavior, in particular the social behavior.  We do not need to know much about the influencers to take advtange of influencer behavior (what are the behaviors, what are the schedules of reinforcement, what are the reinforcers), history (normative behavior, rule sets, discriminant stimulus) and the context of the influence (the environment, the websites, the office, the watercooler, the telephone…).

Here’s my direct response to Jeremy:

[Social networks/social sites] struggle with how to sell the concept to advertisers.  No one gets it, not even the sn sites themselves.

“It” needs definition.   The confusion may stem from unclear values and how to measure achievement of that value.  What do advertisers value?  Transactions?  Clicks? User data? Sales?  “Brand awareness”?  Being hip/cool?  Often advertisers use buying agents (agencies, buying teams, their uncle…), so the values are compounded.  The buying agent may value sticking to budget, being hip, being secure…   The confusion in measurement for media and advertising has partial roots in what the bigger financial market (venture capitalists, board members, stock market/shareholders, banks/lenders) value. Do the financiers value media footprint (eyeballs), transactional value (CPMs), advertising market ownership (spend share), loyalty (repeat spends, repeat users)?   When you look at the two value sets and try to assess whether the social networking sites fulfill any combination/which combination you find that social networks are out of whack with mainstream values.  Social networks drive huge footprints and user loyalty and are good at exposing users to brands/ideas/concepts. Social networks have yet to drive transactions or any offsite activity so most of the values to advertisers with the current advertiser approach cannot be fulfilled.

Don’t forget to add the set of rapidly changing distribution mechanisms (the APIs, social networking platforms, ad implements, publishing toolsets, scripting languages, AJAX vs FLASH vs Silverlight, the tracking systems…)

Thus confusion.

Resolution?  Experience.  That’s it.  Everyone will gain experience and we’ll integrate social networking advertising and not even notice because their will be some other “new thing” chewing up attention on blogs and at water coolers.

I had believed that social networking and behavioral targeting was just a fad.

A common response to an unknown stimulus.  Until you experience new methods enough, everything we don’t know about is a fad.

That it will end up just being another tool in the arsenal once somebody actually figures it out.  I think it won’t really take off until someone develops a standard that everyone can follow.  Until then it’s just a mishmash of technologies and ideas that are just tangents of the same idea.  Being in the industry, we hear about every new concept of trying to aggregate behavior and bucketing those behaviors to try and create useful targeting.  It all revolves around collecting site behavior.

Right, see my note above.

Some of the more interesting ones actually involve trying to apply psychological profiles and social classifications based on the user’s browsing behavior.

Pyschological profiles?  ugh.  It’s behavior that matters.  The psychology, in typical usuage, is a useless abstraction.  As a thought experiment, profile yourself.  What’s you profile?  Another experiement: Go to your amazon.com page – based on the products you’ve purchased and they recommend, what does it really say about you and can you do much to target yourself that you couldn’t do by matching like categorized products that get ok user reviews?

Behavior classifications based on schedules of reinforcement,with historical and contextual classification is the way to go.  Read up on schedules on reinforcement to understand the concept and how it might be applied to online advertising.  Basically, a publisher needs to understand rates of reinforcement and what schedules a user behaves under.  You can have the perfect product and a great message but if you miss a users schedule, they won’t buy.  Schedules can be:

  • buying schedules
  • reading schedules
  • relationship schedules
  • job schedules
  • cultural schedules
  • learning schedules

One topical example that might help understanding. Think about this years elections and why the candidates might be having more success with their platforms vs. in previous years.  The voters schedules are inline with some politicians message.  The context of economy, health care, unpopular war, social upheaval, housing issues are both on social schedules/cycles and personal cycles.   Certain politician presentation styles and stance on these issues fit better in the current schedules than others.

Your own schedules impact how you vote and how others might influence your vote.

I suppose you could say “timing is everything.”

Is sn or bt about the consumers or the influencers?  Up to this point, anyone trying to use influencers has typically been exposed.  All those attempts at viral are quickly uncovered as manufactured.  It is true what they say that the internet is the world’s greatest laboratory.  The collective intelligence will typically uncover a insincere effort rather quickly.

Is there a difference between consumers and influencers?  and does the exposition of that difference matter at all?

Authenticity and credibility are very much tied to values.  A manufactured attempt a buzz is usually the result of the manufacturing body not understanding the values of the consumers. Often agencies, marketers and publishers get the optics (the outside packaging) of a product or service but totally miss the utility. (hint: ipods are much more than music on the go with a nice digital store)

SEO has generally been regarded as a good thing.  Even though it has been used by every advertiser to increase their influence.  It has ensured that any content worth knowing can be found.

I suppose.  A sidenote: SEO was also a product of the search companies themselves to improve their own product.  The better content can be found the more useful the search engine.

But what about social networking and behavioral targeting?  The impact of social networking is just beginning to be felt.  So much personal data.  Will the collective eventually see this as a good thing or bad?

It’s neither good nor bad and will never be labeled as such.  It’s all about value.  You give all your professional data to LinkedIn and Monster because they help you get a job, you do not always share this information with your land lord because he might use it against you.  You give your friends names to Facebook because it helps you communicate.  You give google all of your search queries because it finds the info you need, but you’d never share your search query list with your girlfriend – you’d have far too much explaining to do.

It’s influence will be far greater than SEO.  Will they approve of having their personal details used for the purposes of a more efficient marketing machine?  Can they do anything about it if they really wanted to?

Social networking traffic is different than search traffic.  It’s not greater or lesser in any quantifiable sense, certainly not yet.

Yes, you can do something about your data.  Always.

One of the current problems with behavioral targeting is that no one has developed a method of success reporting.  What is a success metric for a bt campaign?  Advertisers are still using visits and clicks as success metrics.  Widgets is an interesting method.  It requires adoption by users, and once adopted, do they interact with your widget?  Consume it’s content, pass it along to friends?  Do friends adopt it without prompting?  It certainly opens up a whole new area of measurable behaviors that were never available before.

See my earlier notes on the confusion of values.

This gets back to the influencer vs the consumer.  How do you target the influencers?  Heck, how do you identify the influencers.  That, I believe is the tricky part.  The influencers are not a stable group.  They are never the same.  They can be an influencer one day, and an consumer the next.  If you could come up with a method for identifying the influencers for any advertiser then you would really have something.

My focus is on coming up with a way to track, analyze and respond to schedules.  If I know your schedules, I don’t need to know you.  Tracking schedules doesn’t require demographic studies or surveys or other intrusive methods.  I can use the behaviors you exhibit.  The web is excellent for this.

My point is that I can spend time classifying people into various categories by gender, “personality”, social status in an effort to predict behavior OR I can just observe the behavior and schedules and attend directly to that.

~R

add to del.icio.us :: Add to Blinkslist :: add to furl :: Digg it :: add to ma.gnolia :: Stumble It! :: add to simpy :: seed the vine :: :: :: TailRank :: post to facebook

Read Full Post »

There’s a remarkable feature on Edge.org.  I point it out because it’s robust dialogue about collective behavior.  In particular, the discussion is decidedly not casual agentish, mind-body dualist nor monocasual.  Dr. Couzin is refreshing!  His approach ties very well to analysis of media (collective behavior!)  (Read his other stuff like this essay, too!)

how can a colony decide between two food sources, one of which is slightly closer than the other? Do they have to measure this? Do they have to perform these computations?

We now know that this is not the case. Chris Langton and other researchers have also investigated these properties, whereby individuals just by virtue of the fact that one food source is closer, even if they are searching more or less at random, have a higher probability of returning to the nest more quickly. Which means they lay more chemical trail, which the other ants tend to follow. You have this competition between these sources. You have an interaction between positive feedback, which is the amplification of information—that’s the trail-laying behavior—and then you have negative feedback because of course if you just have positive feedback, there is no regulation, there is no homeostasis, you can’t create these accurate decisions.

There’s a negative feedback, which in this case is the decay of the pheromone, or the limited number of ants within the colony that you can recruit, and this delicate balance of positive and negative feedback allows the colony to collectively decide which source is closest and exploit that source, even though none of these individuals themselves have that knowledge.

Great exposition on the web of contingencies and the feedback loops capable of reinforcing complex behavior that we typically claim is “free choice” like or conscious decision making.

One of the big challenges that still remains, and one that we’re beginning to address—I’m not saying it’s a new question; I’m not saying people haven’t addressed it—is the level at which selection is acting within populations. The view of individual level selection, and selection at the level of genes, of course, holds. But if you consider, say, a school of pelagic fish—these large schools that make these dramatic maneuvers—the individuals are unrelated to each other. They drift around as pelagic larvae, so when their schools are comprised as adults, they’re completely unrelated.And yet the individuals’ functioning is entirely within the context of these schools; you can see the integration of the behavior when they are attacked by predators, you can see why in the ’40s people thought there must be thought transference, must be telekinesis, because of these remarkable maneuvers. We now know that these maneuvers are created by the relatively local interactions among the individuals. But if you take an individual, say, a herring, from the school and isolate it, it will die of stress.It is a bit like taking cells from your body—when you take them outside the body, they are unable to function. Of course it is not as closely integrated as a body, it is not as closely integrated as an ant colony. But there is this high level of integration among unrelated individuals. And in terms of how the genes are going to propagate, genes that allow individuals to function collectively as a group are going to be extremely important. So one has to then begin to think about the level at which selection is actually functionally acting. There is nothing new in terms of the genetics here, but it just in terms of how you begin to understand how the collective behaviors emerge and evolve within these types of systems.

Yes! Selection by consequences. Selection happens at many levels – genetically, epigenetically, and behaviorally.  This is a clearer and more accurate description of collective behavior than some previous discussion on edge.org.

It doesn’t stop there though.  Dr. Couzin reminds us yet again observed behavior emerges from simple, often unexpected, contingencies.

Another example that we’ve been investigating arehuge swarms of Mormon crickets. If you look at these swarms, all of the individuals are marching in the same direction, and it looks like cooperative behavior. Perhaps they have come to a collective decision to move from one place to another. We investigated this collective decision, and what really makes this system work in the case of the Mormon cricket is cannibalism.

You think of these as vegetarian insects—they’re crop pests—but each individual tries to eat the other individuals when they run short of protein or salt, and they’re very deprived of these in the natural environment. As soon as they become short of these essential nutrients, they start trying to bite the other individuals, and they have evolved to have really big aggressive jaws and armor plating over themselves, but the one area you can’t defend is the rear end of the individual—it has to defecate, there has to be a hole there—and so they tend to specifically bite the rear end of individuals. It is the sight of others approaching and this biting behavior that causes individuals to move away from those coming towards them. This need to eat other individuals means you are attracted to individuals moving away from you, and so this simple algorithm essentially means the whole swarm starts moving as a collective.

I mean, really. Think about this… how different is human behavior? (I know, I know, it’s more complex, but…).   Consider the elections, consider online behavior, and consider office politics.  We move towards what we move away from and then you get this behavior that appears collective.  Perhaps in our rush to be anti-brand, unique, a cut above, a stand out, we all come together????

In the human sphere it takes us a great deal of effort and complex cognitive abilities to decide what to do if we are going to decide in a collective. What we can now show is that animal groups, with really simple cognitive powers, can actually perform these types of computations. What we now want to understand is under what conditions these types of cognitive capacities work. How can these animal groups take information from multiple sources; how do they filter out noise and yet amplify weak signals?

Right, we should get some data to back that up!  We are on to it…

Second idea that is interesting is one reason I’m interest in NKS Summer School – simple rules applied over and over and sometimes compounded can generate very complicated behavior.  Sometimes the behavior is so complicated we’re inclined to model it with complex rules.  I want to explore this in depth and showcase it visually.

Dr. Couzin makes a great case for why the study of behavior is key to understanding.

The locust is one of the best-studied organisms for physiology and neurobiology. It is really an amazing model system for looking at these principles. Yet the last time people looked at the swarming behavior of locusts in the lab was in 1954. So there is this dearth of information, and no matter how much you look at a locust, and how much you know about the biology of a locust, you cannot predict what will happen when you start to put these organisms in swarms. …. Issues like the dimensionality of the problem are important; issues such as certain details of the interactions are not important if you want to understand the general principles of how it changes—the phase transition is very important for us in the case of locusts because, as everyone knows, locusts are always around. But then suddenly there is this transition from one state to another almost liquid state—so its driven liquid-type state—where the swarms can become enormous.

He, too, arrives at a similar curiousity in the application of these abstractions to the study of media.

Individuals can have a certain opinion on certain topics, and we can allow individuals to interact across a social network, and of course the social network’s topology is partly defined by what opinions you have. You tend to interact more often with people who have similar opinions to yourself because you are more likely to meet them in your sphere of life. But interacting with people can change your opinions, which can then change your social network, which can change your opinions, so again we have this recursive feedback. And so we are using it to explore these types of properties?I am sure that these types of principles also would apply to understanding dynamics on the Web.

I know there’s some excellent work by Duncan Watts on how individuals buy on-line, or how they judge information that they have on-line, how your judgment of something is dependant on what previous people have said about it. What they used was an on-line music store, where you either have information about what previous people have thought about a song, or you have no information and you just have to rank the song without that previous buyer. This strongly changes people’s behavior because of course when you see what other people have been doing, you can have this autocatalysis, this positive feedback. You can tend to buy into that because you have seen other people do it.

Yes, like so many of my posts, we’ll leave with a simple take away:

We are constantly looking for areas where we can create a more data-driven science behind the spread of these normative behaviors.

And maybe this constant looking is also a spread of a normative behavior?

~R

p.s.
For info on Duncan Watts weave your way to his Kevin Bacon paper and his take on online music/social trends.  You may want to hit up his wikipedia page for more links.

And Yahoo! research, where he works, has some kickass concepts, publications, apps.

add to del.icio.us :: Add to Blinkslist :: add to furl :: Digg it :: add to ma.gnolia :: Stumble It! :: add to simpy :: seed the vine :: :: :: TailRank :: post to facebook

Read Full Post »