Contributing Factors

Adolescent sensitivity to peer influence

Adolescence is a time of transition and uncertainty as those of this age strive for greater independence and the development of a personal identity. During any period of uncertainty and change related anxiety, a natural reaction is to look to others one regards as similar as a guide to approved behavior and comfort. The unique sensitivity of adolescence to peer influence is evident in laboratory tasks comparing the likelihood of conformity in participants of different ages (Costanzo & Shaw, 1966) and more intensely among adolescents who are socially anxious (Zhang, Deng, Yu, Zhao & Liu, 2016).

The anxiety and proneness to conformity typical of adolescence makes them more sensitive to their online experiences and more easily influenced by these experiences. Of course, educators who are targeted by our materials work with learners of this age and these educators understand that factors that influence their students out of the classroom are not forgotten by adolescents as they come to class.

Self imposed and other encouraged biases

Student access to online resources has resulted in some educators taking on the challenge of preparing students to evaluate the credibility of the content they discover. While needed, I would argue that this is not enough. The problem is not solely an issue of flawed content. Bias is part of the human condition and our biases influence the information we choose to attend to and our biases influence how we process this information. We tend to fit experiences and information to our existing models of the world. This tendency is studied as confirmation bias (Nickerson, 1998). Understanding that we are prone to this tendency is an important part of encouraging more careful selection and processing of information and experiences so explaining how this works to students should be an educational goal.

The way we encounter information through social media can support existing or challenge existing biases. Some services such as Twitter and Facebook give users control over those who author more of the content a user sees. You see mostly what "friends" write and what they share. The feed resulting from these self-selected influencers is not necessarily neutral and the reliance on this feed as a primary source of information can create a skewed view of the world through a type of self-imposed information bias.

As was mentioned previously, certain services can also impose an information bias on our online experiences (filter bubble). In an effort to offer search results that will satisfy what the service thinks we want to know (Google) or what we want to see in our news feed (Facebook) could be supported by the content selection biases services use to feed our interests. An optimistic view would argue that this bias is an unintended consequence of these services trying to be helpful and using what the services have learned about us to improve our efficiency. Again, understanding how this works would seem important so that individuals recognize potential problems in a service trying to be helpful by telling us what we want to hear. I think of the issue as similar to the bias of watching certain television news channels. It is important I am aware as I watch a given channel that other perspectives exist and that what I am viewing may not be include an effort to present these other perspectives.

Emotional content

If you want to attract attention on social media, not all content you share is created equally. Content that might be described as emotional is more likely to engage the attention of others (Berger & Milkman, 2012). Controversy, fear, and anger create an emotional reaction. Both individuals and online services wanting to encourage higher levels of engagement can do so by offering such content. This reaction is one way to meet the goals of getting and sustaining attention. This tendency creates a spiraling effect. When people are riled up they consume and they also share more content.

There is a second finding related to emotional content that could be important. One characteristic of following and friending is that many users tend toward developing a collection of those with similar interests and attitudes. There is an interesting consequence of the discussions that occur within groups of individuals with mixed versus similar beliefs and attitudes. Researchers investigate this phenomena by first measuring the degree to which participants endorse a specific attitude and then forming groups of individuals with similar or different preexisting attitudes for a period of discussion. An interesting thing happens in a group holding a similar attitudinal position. When the attitude scale scale administered before the discussion is given again, members of such a group positions on the attitude scale tend to become more extreme. This phenomena is described as polarization (Isenberg, 1986).

Consider how these two factors might interact online. Groups of individuals left to their own selection of online acquaintances tend toward others with similar interests and attitudes. How do you get attention within a group? You take positions that generate an emotional response. It makes some sense that interaction combining these two characteristics would move the commonly held positions by members of a group in a more extreme direction.

Skinner and his rats should have warned us

B.F. Skinner is famous for his research and subsequent laws of operant conditioning. Much of his research, which you likely studied in introductory psychology courses, was conducted with rats but is equally useful in understanding certain human behaviors. Simply put when a behavior is following by a positive consequence, we tend to engage in that behavior more frequently (positive reinforcement). When the occurrence of a consequence is unpredictable, we tend to accelerate the behavior that generates the eventual positive consequence (random schedule of reinforcement).

It can be argued that social media sites include multiple components that can be interpreted as positive reinforcers. Most individuals value users/friends and find an increase in this number to be motivating. Most of us value likes, shares, and positive comments related to content we produce. We typically don't know what posts will produce these reactions and check our accounts frequently in hopes of seeing new likes and shares (random schedule). Some social media services document "streaks" - the number of consecutive days a certain thing has happened (friends have shared a post with each other). A streak is also perceived as a positive consequence (when it happens) and a concern when streaks end. The combination can be quite motivating and I would guess such consequences are of greater concern to those most tuned in to how they are perceived by others.

Some would call these techniques a form of "gamification", but all rely on long understood principles of conditioning. Were these techniques designed to create the dependency some seem to exhibit for participation on some social media sites? This is hard to say. Some might explain that these strategies were intended as potentially useful features and people seem to like using them. Unintended or not, these features increase the commitment of individuals to the services including such services. (McNamee,2019; Zuboff, 2019)

Conclusion, recommendations, and references