Alex Pentland on idea flow, diversity, and contrarians

Alex Pentland is perhaps the one person that I am most likely to agree with regarding social motivations. In a recent Wired column (Why Startups Should Steal Ideas and Hire Weirdos) he wrote,

We all sail in a stream of ideas, ideas that are the examples and stories of the peers who surround us; exposure to this stream shapes our habits and beliefs. We can resist the flow if we try, and even choose to row to another stream, but most of our behavior is shaped by the ideas we are exposed to. The idea flow within these streams binds us together into a sort of collective intelligence, one comprised of the shared learning of our peers.
The continual exploratory behavior of humans is a quick learning process that is guided by apparent popularity among peers. In contrast, adoption of habits and preferences is a slow process that requires repeated exposure and perceptual validation within a community of peers. Our social world consists of the rush and excitement of new ideas harvested through exploration, and then the quieter and slower process of engaging with peers in order to winnow through those ideas, to determine which should be converted into personal habits and social norms.

The most creative people are found in loose networks of creatives, who are all participating in idea flow of the sort that Pentland is talking about.
In a connected world, the most important decision is who to follow. And one measure to use when evaluating a possible following is idea flow: how many innovative ideas are they generating, and what proportion of those do other creative people find intriguing?
In that same piece, Pentland supports the idea that business has to accept a much higher level of dissensus and diversity. About the first, he advocates finding contrarians who can resist the conservative consensus-building latent in social systems:

When people are behaving independently of their social learning, it’s likely that they have independent information and that they believe in that information enough to fight the effects of social influence. Find as many of these “wise guys” as possible … and learn from them.

And about the power of diversity:

When everyone is going in the same direction, then it’s a good bet that there isn’t enough diversity in your information and idea sources, and you should explore further. A big danger of social learning is groupthink. To avoid groupthink and echo chambers, you have to compare what the social learning suggests with what isolated individuals (who have only external information sources) are doing. If the so-called common sense from social learning is just an overconfident version of what isolated people think, then you’re likely in a groupthink or echo chamber situation. In this case, a surprisingly good strategy is to bet against the common sense.

Consider that a great deal of the energy expended in ‘culture building’ is actually trying to cram a non-diverse, consensus worldview and mindset down the group’s throats, and actively culling out those that don’t ‘fit in’ (see Is ‘cultural fit’ a cop out? In general, yes).
As I have said before, the wisest approach in a time of rapid change is to actively take on more risk, and one technique is to try parallel experiments based on competing premises about the near future:

But it is also important to diversify by considering more than one strategy at a time, because as our environment changes, the old strategies stop working and new strategies take the lead. Therefore, it’s not the strategies that have been most successful that you want to follow; it’s the strategies that will be most successful that you have to find.

Counterintuitively, by accepting dissent, and encouraging contrarian thinking to pursue non-convergent paths toward different futures, the company builds resilience and accelerates learning.

What top performers do, and how to do it

In some recent posts, I’ve explored the notion of increasing talent density in business as a means to stay ahead of the downside of growth — complexity and process encroachment (see What do Amazon and Netflix have in common? and Countering the traps of complexity and growth by creativity and context: the Netflix model). One of the central themes of those discussions is the notion of basing the organization around finding and retaining top performers.

Two questions have been directed my way in discussions about this approach: what do top performers do, and what happens to the less-than-top performers? (It turns out to be important to discuss these in that order.)

Robert Kelly of Carnegie Mellon did a study at Bell Labs in 1985 that sheds light on a key characteristic of top performers. Bell Labs was a top research lab — responsible for critical advances like Unix, C and C++, transitors, the wave nature of matter, and myriad other breakthroughs. However, even though the Labs hired the brightest minds, many did not become stars, and in fact, only a few did.

The network of contacts that a top performer has does not operate like a team on a playing field, which is a sort of a collective. Instead the network of contacts serves as a connective, where the individuals may be totally unaware of the actions or even existence of many of the others.Kelly found that, despite the folklore surrounding higher IQ, or better problem-solving skills, something more fundamental was involved in the performance of stars: they proactively networked in advance of specific needs. And then, later on, when they were engaged in some critical task they could — and did — reach out to other experts with salient skills and knowledge, and where a social connection had been created making it more likely that the contact would in fact make the time and effort to help.

I have often said that in a connected world, the most important decision is who to follow. Perhaps taking that a step farther in light of Kelly’s research is justified. Since we live and work in a connected world, the most important first step is to decide who to follow, and how to gain their respect so that they will follow you back. And maybe someday, cover your back.

This ‘preparatory exploration’ — as MIT’s Alex Pentland calls it — sets the top performers apart from the less stellar, whose networks are less diverse and smaller. In essence, the top performers create a network in which the rate of social learning is simply higher than others, and more likely to involve novel information. This is quite similar to the notion of social capital as developed by Ronald Burt in Structural Holes and Good Ideas:

“Social capital exists where people have an advantage because of their location in a social structure.”

Another way of looking at this differential of performance is that top performers are actually cheating, because they are leveraging their network more than the less connected, who rely more on their own abilities.

This, then, solves the mystery of how top performers do what they do. But what are we to do about all the other people? The answer is obvious: the techniques used by top performers are not a secret, and can be learned. And just as importantly, mentors and managers can intercede to create more diverse and richer social networks by making ‘preparatory exploration’ a part of organizational culture, and a part of every person’s toolkit. Those that are not today’s top performers can get better, and an organization as a whole can improve together.

Note also that this is not ‘teamwork’, which is another term that is bandied around far too much. The network of contacts that a top performer has does not operate like a team on a playing field, which is a sort of a collective. Instead the network of contacts serves as a connective, where the individuals may be totally unaware of the actions or even existence of many of the others. More on this in a later post.

The challenges are significant, because for talent density to grow faster than the company — or its challenges — increasing diversity has to become foundational. Also note that the answer isn’t a sweeping dictate that everyone should be connected to everyone: that leads to groupthink, not high performance. It’s actually essential that personal networks not grow too large, or the laws of diminishing returns start to play. But it is important to see that there is an upward path for the middling players, even in a world where the best connected have a real — and growing — advantage.

There is a way to base a company around unrelenting high performance that avoids a zero sum outcome in which all but the best performers are routinely fired. The key is to increase the rate of social learning, which anthropologists will tell us is largely based on observing and patterning ourselves on what experts do.

With nearly 10M members, Edmodo goes back to school with new version

With nearly 10 million members, Edmodo is on Wednesday rolling out an updated version, with an improved user interface and richer functionality. The changes reflect feedback from teachers, which is critical for the site given educators’ role in driving adoption — and, ultimately, revenue.