Everyday Sociology Blog

From Cockpit to Concert Hall: Distributed Cognition

Michael Discenza

Columbia University, Class of 2013

mad2200@columbia.edu

Edwin Hutchins, a professor and researcher in the field of cognitive science, has conducted extensive research about distributed cognition. Distributed cognition means that an individual’s understanding of and actions in the world are not merely a product of that one person’s individual decisions or desires, but are influenced by non-human agents.

One of the focuses of Hutchins’ research is the airplane cockpit. In a series of papers about the inner workings of the flight decks of commercial airliners, he explains how much of the cognitive work required for flying an airplane is done by the instrumentation and technology that makes up the flight control consoles. The communication technologies, instrumentation, and industry jargon serve as agents, non-human actors, which store and institutionalize knowledge and past cognitive efforts. The various components of the cockpit free the pilot from a glut of cogitative demands and make piloting a large commercial airliner, which would be an otherwise unwieldy cognitive task, a typical day’s for pilots. The functional system that results from this combination of human and non-human actors is what we call a socio-technical system.

Distributed cognition is a quite straightforward concept, but it requires a certain degree of childlike imagination to internalize, namely a mindset similar to that which is required pass the “how do you put an elephant in the refrigerator” test. I was first exposed to the notion of a socio-technical system in a simple demonstration that Professor David Stark at Columbia University presented to a sociology class that I was taking. Professor Stark walked over to the door of the lecture hall, opened it, and walked away from the door as it closed. He then asked the class what had happened. After a few looks of confusion, the class hesitantly contributed suggestions until one student, in addition to mentioning that the professor had opened the door, included in his explanation that door had closed itself. The point of the demonstration was that the preceding actions were not just the effects of human manipulation of the environment, but that the spring on the door had caused the door to shut itself afterward.

The action of a door closing itself after someone opens it and walk through is one of the most basic examples of distributed cognition across the simplest of networks. The person who opened the door does not have to think about closing it because the architect already decided that the door should close automatically and incorporated the spring technology of the designer of the door and the door’s manufacturer. When we start to recognize non-human agents and give them credit for their influence in our lives, we see can find socio-technical systems everywhere.

The cockpit is an extremely complex socio-technical system, and so is a library—the protocol and tools, card catalogues, databases, and librarians are agents and actors in the system. Any store, home, factory, restaurant, sporting event, or traffic regulation system with stoplights and signage, any system that incorporates even the most simple of technology, is a locus of a socio-technical network across which cognition is distributed.

Photo courtesy of Sarah Sheu

At a recent Electronica concert in a large Mid-town Manhattan venue called Terminal 5, I found myself struck by the salience of the socio-technical system at work. The coordinated efforts of pilots, co-pilots, and air traffic controllers, achieved in the cockpit with the use of complex jargon corresponded to the use of primal cuing with the rhythmic patterns of the music and the lighting, which together induced periodic climaxes of excitement among a crowd of 3000 young adults. The turntable became the flight deck and the DJs and tech crews, pilots not of planes but masses of music fans.

The music groups and DJs at this concert all adhered to similar conventions in their music, crescendos and holds that created tension were followed by a release into up-tempo danceable beats. The corresponding lighting scheme was a mesmerizing lighting pattern that broke into rapidly flashing strobe lights and quickly circling spotlights and lasers. The DJ and support team on stage flashed hand signals from the stage to the tech team in the back of the hall, as the traditional walk-talkie cuing was rendered useless by the deafening roar of the music. Interestingly, as I watched over the shoulders of the tech crew I observed that the lighting patterns were pre-programmed and just required activation from their laptops, lifting a significant cognitive burden from the tech crew.

Although there is no concert-goers handbook that tells people when to put their hands in the air and go crazy, no instructions flashing across the stage signaling the crowd to dance in a certain way, and certainly little cognitive power among the crazed masses, the changing environment induced a crowd reaction that appeared to be carefully choreographed. Everyone’s hands flew into the air at the same time. These gestures were not prompted not by any single person’s will to throw their hands up or by any DJ’s instructions; they were an unconscious response to a complex system employing turntables, lights, tech crews, musical conventions, and previously established signals.

What other examples of examples of distributed cognition take place in our everyday lives?

Racial and Ethnic Categories

By Sally Raskoff

Have you ever thought about how your definition of your race and ethnicity differs from the definitions the government and science uses?

The government defines race as white, African American (or Black), American Indian or Alaska Native, and Asian while ethnicity is simply Hispanic or not. Many options are available for what type of Asian, Pacific Islander, or Other race. (See the relevant questions below from Census 2000.)

The Census Bureau defines race as follows:

The racial categories included in the census questionnaire generally reflect a social definition of race recognized in this country, and not an attempt to define race biologically, anthropologically or genetically. People may choose to report more than one race to indicate their racial mixture, such as “American Indian and White.” People who identify their origin as Hispanic, Latino, or Spanish may be of any race. In addition, it is recognized that the categories of the race item include both racial and national origin or socio-cultural groups. You may choose more than one race category.

Scientific definitions of race are best summed up by the American Anthropological Association's (AAA) statement on race. The AAA points out the difficulties in gaining any clear definition of race due to the lack of clarity as to what we are attempting to measure. In short, scientifically, there is no physical or geographic sense of what race is, although it does have cultural distinctions and the definitions vary in different times and places. 

The surveys I’ve done with my students suggest that most of them think about themselves racially and ethnically in terms of their cultural group identities. If they are not in a group considered white, they have a clear sense of being part of an identifiable racial group (black, Asian, Latino, Native American Alaskan Native, or Pacific Islander). No matter their racial identity, most mention important distinctions related to their ethnic groups, e.g., Chicana, Armenian, Mexican American, Chinese, Guatemalan, and Jamaican.

Is it important to notice the disparities between the governmental and scientific definitions? If science says race as a concept is not useful for describing humans, why does the government do it?

It all comes down to the economics and politics of categorizing people. Keeping track of people in different categories, whatever they may be, helps us know more about our society. If we have clearly identified disparities and problems in treating certain groups fairly, studies that track these populations can help us know if our social programs and policies are actually helping or not.

What does it mean that our governmental, scientific, and personal definitions of race and ethnicity vary so much? Are the categories that the government and science use useful at all since they may not match up with those that people use?

If people fill out the census forms as accurately as possible, then the data the government collects can be used as our constitution demands: to allocate political representation and governmental programs appropriately to serve the people. If people aren’t aware what those categories mean or don’t how to accurately fill out the forms, what then is the use of that data?

Good data comes when people understand the concepts and share their definitions with the researchers.

It is fascinating that although science says race is a problematic concept to define, people hold dearly to their racial and ethnic categories of identity. Thomas Theorem comes to mind– if people define situations real, they become real in their consequences.

People grow up having to fill in those forms and thus claim a racial and/or ethnic identity. People participate in some cultural group that may have racial or ethnic distinctions or identities. Thus the concept of race becomes very real. This is especially true historically as society reifies these definitions and establishes policies that either disadvantage or elevate various groups depending on racial identity. Some of the many examples of such policies include Jim Crow laws passed in the South post-slavery, and the racial steering and covenants that were used in housing markets across the country until well into the 1960s.

As social scientists, we prefer clarity of definition rather than ambiguity. What, therefore, do these ambiguous racial definitions teach us about our society?

Doing Research while Watching Sports Center

By Karen Sternheimer

Think about how many people you know whose days are punctuated by checking sports scores, and whose bedtime stories come courtesy of ESPN’s Sports Center. Even something that seems as mundane as tracking sports coverage has sociological meaning.

Suppose you wanted to learn more about this practice. You might use sociological research methods such as in-depth interviews with fans, surveys of television audience members, or ethnography of how news agencies produce sports programming to collect your data.

Another interesting method tends to be overlooked in many research methods texts and courses. Content analysis involves systematically observing the content of a text, including written, visual, and audio texts. Often used in communications research, sociologists also use content analysis to take a deeper look into media we might otherwise take for granted.

Essentially, content analysis involves counting the occurrence of specific phenomena that researchers are interested in learning more about. One of the first tenets of good content analysis is to establish specific guidelines to make sure researchers are clear and consistent about what they are counting. More than just watching TV and writing an overall opinion, content analysis requires clear definitions of a sample and the procedure the researchers will use to analyze their findings.

Something as familiar as Sports Center can be systematically analyzed, with clear quantitative and qualitative results. Sociologists Michael Messner and Cheryl Cooky recently completed a content analysis of Sports Center and the sports coverage of the three Los Angeles network news affiliates. Their report, Gender in Televised Sports: News and Highlight Shows, 1989-2009, features the results of their content analysis over five separate periods: one in 1989, 1993, 1999, 2004, and 2009.

As their title suggests, the researchers wanted to learn about how much sports coverage is devoted to male and female athletes, as well as to better understand the context of the coverage about sports. They measured six weeks of local news coverage, selecting three two-week periods at various times of the year. In addition, they also recorded three weeks of Sports Center during similar time periods as the local news. The researchers timed the amount of coverage men’s sports and women’s sports received, including the ticker at the bottom of the screen.

If you are a regular viewer of sports coverage, you won’t be surprised to learn that men’s sports coverage dramatically overshadows women’s coverage. You might even conclude that this finding is so obvious that it isn’t necessary to conduct a systematic study.

But you might be shocked to learn that local coverage of women’s sports has dramatically declined over the past twenty years (and in the last ten especially), as the graph below illustrates. This is something we can only learn through repeated content analysis studies.

After a rise in local news coverage between 1989 and 1999, women’s sports coverage declined dramatically from 8.7% of all sports coverage to 1.6%. Likewise, Sports Center’s coverage (only included in the last three time periods of the study) declined as well, albeit not as dramatically because women’s sports coverage was low to begin with.

Why the decline? We might even predict that coverage would increase rather than decrease; as the researchers point out, girls' and women's participation in sports has increased dramatically over the past twenty years. In 1989, 1.8 million high school girls participated in sports, compared with 3.1 million in 2009. Women are more likely to play collegiate-level sports, and women’s opportunities to play professional sports—most notably in the WNBA—expanded too. So it seems counterintuitive that coverage would decline.

While Messner & Cooky note that fans now have numerous ways besides television to get sports updates (including fan websites and smart phone apps, for example), they conclude that to understand why women’s sports coverage declined it would be necessary to study “the assumptions and values [that] guide the decisions of producers, editors and TV sports commentators on what sports stories are important to cover, and how to cover them.”

The researchers hypothesize that the structure of televised sports coverage might help us understand the lack of women’s sports coverage. As Messner & Cooky point out, the network’s promise to deliver an audience of young men to advertisers provides a financial incentive to not only keep coverage of women’s sports to a minimum, but also to portray women in specific ways:

A foundational assumption of those who create programming for men on programs like SportsCenter seems to be that men want to think of women as sexual objects of desire, or perhaps as mothers, but not as powerful, competitive athletes.

This long-term study found another interesting change: despite the reduced coverage of women’s sports, when women are part of sports coverage the tone was more respectful in 2009 than it had been in the past. Most of the time women appear in sports coverage as male athletes’ girlfriends, wives, or mothers, and yet some coverage now focuses more on women’s athletic prowess.

By contrast, the authors observed that in the past, sports coverage frequently ridiculed women. For instance, a 1999 story on women who bungee jumped nude and a 2004 story about a “weightlifting granny” made women appear more like comic relief than serious athletes. These stories haven’t completely gone away, but they were less common in 2009. The authors conclude that involving more women in the sports news production process might produce more favorable coverage of women’s sports, but concede that there’s no guarantee. For example, female reporters might be hired more for their looks than for their sports background.

As you can see, content analysis can take an everyday activity like watching sports coverage and analyze its deeper meanings. What other kinds of sociological information might content analysis give us?

Interracial Marriage among Newlyweds in the U.S.

By Janis Prince Inniss

Last week, I received an envelope in the mail that was clearly an invitation. I recognized the return address as that of a couple that my husband and I have been friends with for almost ten years—let’s call them the Smiths. The Smiths host several parties annually for which we receive written invitations. Still, this looked more formal than an invite for a summer gathering. The envelope was as thick as a wedding invitation, but the Smiths have been married for many years. What could it be? A ”major” birthday?

Inside was a wedding invitation for their son’s nuptials. You don’t know the Smiths, but if I told you the race of their son, would you be able to guess the race of his soon-to-be bride? How about if you had information about whom most newlyweds marry?

Who do people marry? Much has been written about romance and the challenges of finding suitable dating partners, but once people find a mate and decide to marry, who do they choose? Let’s focus on the race/ethnicity of the newly married who wed someone outside of their race/ethnicity.

By looking at the chart below, you will see the percentage of newlyweds who married someone of a different race/ethnicity than their own in 2008. Notice that that the percentage of people “marrying out” (marrying someone of a different race) varies across racial/ethnic groups.

The group with the most out- marriages—Asians—did so at a rate of almost one third (30.8 percent). Whites had the lowest out- marriage rate of the groups, with fewer than one in ten whites (8.9 percent) married to someone of a different race than their own.

Does any of this surprise you? Did you expect any of these numbers to be higher? Lower? More similar across groups? Why would one racial/ethnic group “marry out” at a rate that is particularly different from another? For example, why do Asians and Hispanics “marry out” so much more than blacks and whites--especially whites? Or to flip the question around, why are out-marriage rates for blacks and whites so low? Do you think that some groups have cultural attitudes that shape their attitudes towards intermarriage? What role, if any, do you think the numbers of available people within one’s racial/ethnic group play in any of this?

 

Source: Marrying Out: One-in-Seven New U.S. Marriages is Interracial or Interethnic (Pew Research Center)

Who are newlyweds marrying when they do marry out their racial/ethnic group? As indicated in the series of pie charts below, the answer depends on the group. For minority groups though, the majority of intermarriages do not occur with other minorities but with whites. Of newlywed Asians, 75.1 percent married whites, of Hispanics 80.5 percent, and of blacks 57.5 percent. So who do whites marry when they marry outside of their race/ethnicity? Almost half (48.8 percent) of all newlywed whites married Hispanics.

Note: Other includes American Indian, two or more races and “some other” race categories.

Data reflect marriages to someone of a different race/ethnicity in the previous 12 months.

Source: Marrying Out: One-in-Seven New U.S. Marriages is Interracial or Interethnic (Pew Research Center)

What patterns, if any, can we detect in looking at the spouses of men and women in new interracial marriages? The charts below provide such data. One of the more striking differences between males and females who recently “out-married” is among whites: Of those who intermarried in 2008, far more white men married Asian women than white women married Asian men (26.9 percent compared to 9.4 percent), while white women were far more likely to marry black men than white men were to marry black women (20.1 percent compared to 6.9 percent). Another noteworthy difference is that of Hispanics who married someone of a different race/ethnicity, the proportion of Hispanic women who married black men was much higher than Hispanic men who married Black women (13.2 percent compared to 4.5 percent).

What about the “desirability” of certain groups as spouses? (I presume that marriage is some indication of someone’s desirability—at least desirability as a marriage partner.) The lack of desirability of black women and Asian men as spouses for those who intermarried in 2008 is worth noting. White, Hispanic, and Asian men in mixed marriages married women of every other racial category more than they did black women. Similarly, white, black, and Hispanic women who entered interracial marriages in 2008 did so with men from other racial/ethnic groups ahead of Asian men. Both white women and white men however, were desired as partners by blacks, Hispanics and Asians in interracial marriages in 2008. As the data in the pie charts illustrates, the majority of minorities in intermarriages—both male and female—married whites (ranging from 57.2 percent to 83.3 percent).

Stay tuned to this space as we answer some of the questions raised in this piece and continue to learn more about intermarriage in the U.S. And regarding young Mr. Smith, like 84.5 percent of people in his racial/ethnic group, he is marrying within his race.

“Good” Meetings

By Sally Raskoff

I recently blogged about frustrating meetings. I usually prefer to focus on the positive, so I am following up with a blog about “good” meetings. To me, a non-frustrating meeting would be one in which the meeting leaders address the reasons for calling the meeting in an effective and efficient manner.

What makes an efficient meeting possible? How can sociology help us better understand what makes meetings effective?

1. To hold a meeting is to gather people together outside of their typical realm of activity. The meeting should have some relevant focus and purpose whether it is discussing issues or making decisions.
2. Effectiveness has to do with implementing decisions. Actions need to come from the decisions made, and the discussion should have some impact on what goes on outside the meeting.
3. Efficiency is different from effectiveness. It means having clear objectives and getting things done with as little “noise” as possible. We are efficient if we deal with what is before us rather than distracting the task at hand with other issues.
4. The social context is important. If people know their social roles, they are more likely to act accordingly to do as others would expect. If everyone shares the meaning of the meeting itself and knows the symbols that are used within, communication is easier. If all key players are present, that can support efficiency since all those with something to say or do are involved in the process.
5. Social norms exist in this microcosm. Clarity about the rules of the meeting-- in terms of behavior and process-- are crucial to having a good meeting. If everyone knows how to behave, those social norms ensure that people will bring appropriate and expected actions into the room.

Many meetings even come with pre-existing formal rules. Groups often use Robert’s Rules of Order in meetings to provide the basic guidelines for what happens and when. The rules include how to make a motion (suggest a decision or policy be made), how to discuss it (after the motion is made and seconded), and when to vote on it (after discussion). These rules also specify not to discuss other issues while discussing a motion – this effectively limits discussions to one issue at a time, which enhances efficiency and potential effectiveness.

Many sociologists have discussed how organizations create rules and order, most notably Max Weber and Emile Durkheim. Organizations form bureaucracies to create a set of protocols for how the organization will operate. Durkheim discussed how the division of labor provides the role definitions so that people can know what they are to do and who has the authority to make decisions or change the rules. Weber wrote extensively on how bureaucratic organizations have hierarchical structures of offices and positions in which individuals have clear career paths. These provide a conduit for the distribution and use of power and authority within the tasks of the organization. Thus, if the boss calls a meeting, the workers show up and pay attention. If a co-worker calls a meeting, perhaps fewer people will show up or will take the process seriously.

Formal and informal networks operate within a bureaucratic structure and are also linked to power. The formal network, tied to the hierarchy, has clear connections with power and authority. The informal networks that develop over time due to proximity or personal ties can supplement or detract from the official uses of power within the organization. Thus if a co-worker who calls a meeting is connected to many other workers through the informal networks and if those other workers have a lot of respect for the person calling the meeting, that meeting may draw more people and participation than the one called by the boss!

While Weber’s ideal type of bureaucracy did point out the (potential) for efficiency and rationality of organizations, it is not appropriate to assume that bureaucratic structures are always effective. A lack of focus can get in the way of implementing the decisions made.

A recent meeting I participated in illustrates this well. We gathered together to deal with issues that are partly defined and controlled by a union-management contract, state law, and many other bureaucratic structures. We focused on specific tasks, using the process to discuss the information we had gathered. Those with the formal authority expressed their opinions, as did those with less formal authority. Decisions were made based on those discussions. At the end of the meeting, many felt the process was somewhat efficient and we had made progress on the tasks at hand.

However, we will truly know how effective this meeting was only in the future. Will any of those decisions be implemented or affect the day-to-day operations of the organization? In the days after the meeting, many informal conversations occurred where people speculated on those decisions and whether or not they will be implemented. Thus a meeting that feels “good” while it is taking place may or may not continue with that qualifier if the outcome of the meeting is for naught.

What other aspects of “good” meetings can you identify? What sociological concepts can help illuminate those aspects?