Laboratory Experiments

Every day, our behavior is directly influenced by a number of factors, some of which we may not even be consciously aware of. How things are designed is one of those factors.

Psychology researchers call it fluency while web developers call it usability, but they’re both basically talking about the same thing — how well something is designed can directly impact how much people use it. And not only the degree to which they use it, but also the amount of self-disclosure a person makes while using it.

Online researchers have repeatedly referred to the disinhibition effect of online behavior — people tend to disclose more about themselves or their personal details online than they do in similar face-to-face interactions. But why do people — especially teens — over-disclose on websites like Facebook or Twitter? What might be a contributing factor to encouraging people to disclose more than they ordinarily would?

Adam Alter from New York University and Daniel Oppenheimer (2009) from Princeton University set to answer this question in a recently published research study. Through the use of three laboratory experiments and one real-world experience with a live website, the researchers discovered that fluency (or usability) directly impacts people’s self-disclosure amounts.

Researchers manipulated fluency in the first three experiments by making the font harder to read on a number of exercises. In the first experiment, 33 undergraduates were asked to fill out a test that measured how strongly people “claim 18 virtuous but implausible attributes (e.g., “No matter who I’m talking to, I’m always a good listener”) and deny 15 common human frailties (e.g., “I like to gossip at times”). Lower scores on the scale indicate a willingness to disclose potentially self-incriminating flaws.” The study found that subjects tended to choose a greater percentage of socially desirable, non-disclosing responses when the test was printed in the difficult-to-read font. This experiment was replicated with another test just to make sure it wasn’t the test itself that was leading to the findings.

In the second experiment, subjects need to add a letter to an incomplete word in order to form a real word, the first one that popped into their head. According to the researchers, “eight of the [incomplete words] could be completed to form words associated with risk (e.g., “ris_” could be completed to form “risk” or “rise”), 5 could form words associated with self-presentational concerns (e.g., “_iked” could be made to form “liked” or “hiked”), and the remainder [could form] words associated with neither concept (e.g., “_og” could be made to form “dog”). Sixty-seven adults completed the study. Words show in the difficult-to-read font were more likely to be risk-related works.

In the third experiment, subjects were asked to complete a self-disclosure questionnaire in which they rated how comfortable they would be discussing their views on 30 self-relevant issues. The researchers found that “participants who found the font harder to read expressed greater discomfort and exhibited a diminished willingness to disclose their opinions on the 30 topics. However, this relationship between fluency and self-disclosure was mediated by the experience of negative emotions, [suggesting] that at least part of the reason why people prefer not to disclose self-relevant information when they experience disfluency is that disfluency enhances discomfort.”

The fourth and final experiment involved an existing confession-based website called grouphug.us that changed its design from gray text on a black background (harder to read and therefore less fluent) to black text on a white background (easier to read and therefore more fluent). Researchers had volunteers analyze the responses made on this website before and after the design change. They found that responses after the design change tended to disclose more embarrassing information.

The researchers summarized their results by noting that people self-disclosed more in high-fluency conditions — that is, when text was easier to read. When text was harder to read, subjects were more likely to hide their flaws and to think more about risk and concern. In the website experiment, people disclosed more revealing information about themselves when the website was easier to read.

The easier it is for us humans to process information means the more likely it is that we’ll engage in more of the behavior encouraging us to disclose. Alternatively, if we make it harder for people to read a website or fill out a form, a person is less likely to self-disclose.

This finding has many real-world implications, since self-disclosure is a big part of the psychotherapy relationship. Health (and mental health) professionals have many opportunities to induce greater fluency. The researchers suggest one example — by using simple words rather than long-winded alternatives, professionals can create more fluency (and more self-disclosure) from their patients. “Evidence from the negotiation literature similarly suggests that mutual disclosure benefits both negotiating parties, enabling them to identify their otherwise opaque mutual interests,” the researchers also noted.

Reference:

Alter, A.L. & Oppenheimer, D.M. (2009). Suppressing Secrecy Through Metacognitive Ease: Cognitive Fluency Encourages Self-Disclosure. Psychological Science. DOI: 10.1111/j.1467-9280.2009.02461.x.