Four Reasons Why Paradox Helps Research Get Attention

New research suggests surprise and memorability play key roles

Posted Jul 21, 2017

Think of all the scientific studies you encounter in any given year, and chances are, one or two of them will stay lodged in your memory surprisingly clearly—in some instances, years after you first encountered them. While journalists and PR specialists are hardly strangers to what makes a new item memorable, the usual suspects—proximity, impact, significance, timeliness, prominence, and usefulness—number among the items you're least likely to recall after even a day. In contrast, two other items that make news particularly—well, newsworthy—are controversy and the unusual.

1. This last item, in particular, warrants closer scrutiny, as our brains are hard-wired to process what we expect nearly unconsciously.

2. Conversely, the highly unexpected or even paradoxical story appears to receive extra neural encoding (Kidd & Hayden, 2015) to ensure recall precisely because we find a relationship between, say, cause and effect, runs counter to our assumptions about how things work.

3. In Tulving & Kroll's striking 1995 study, readers were likelier to recall words more clearly the fewer times they had encountered them. So the unexpected-or-paradoxical relationship may snag in most audiences' memories precisely because they're hearing or witnessing or reading of a scarce event.

4. On the other hand, paradox may simply reflect an extreme version of incongruity. When we encounter incongruity, we typically feel driven to seek more information, resolving the incongruity (Loewenstein, 1994).

Published in the May issue of the International Journal of Business Administration, one study sought to clarify the strength of paradoxical relationships in items that transition from academic publications and into the mainstream media by examining the memorability of paradoxes in academic studies. In this study of one hundred undergraduate business students, a single, peer-reviewed study was slightly translated into journalistic prose, describing the correlation between barefoot running and minimal long-term running injuries (Divert, Mornieux & Baur at al., 2005). In addition, subtle changes in wording created three different versions from this single study: paradox-explicit, paradox-implicit, no-paradox. 

All three versions were normed for readability by using Flesch, Flesch-Kincaid, FOGG, Lexile, and Ai's software for analyzing comprehensive syntactic features, including median lengths of clauses to numbers of complex nominals. On Day 1 of the study, readers received random assignment into one of three groups (paradox-explicit, paradox-implicit, no paradox) and an email with one of the three versions of the story, plus a single question—constant across all versions that tested readers' comprehension of the main point of the article. On Days 3-5, readers received an emailed link to the same follow-up question they encountered earlier, without any re-exposure to the initial story.

The results were striking. Immediately after reading the article, 98% of the paradox-explicit readers correctly identified the article's main point. In contrast, only 14% of the paradox-implicit readers grasped the article's meaning immediately after reading. And a mere 4% of the no-paradox group correctly identified the study's main finding—despite its salience to university students at a particularly athletic school, where a significant proportion of students ran or jogged daily, year-round. 

More strikingly, after as many as 5 days between reading the truncated article and their second encounter with the same question about the article's gist, 55% of the paradox-explicit readers still correctly recalled the article's main point. On the other hand, only 7% of the paradox-implicit readers could identify the article's meaning, despite its implications for students' own practice of running. And none of the no-paradox readers correctly grasped the article's meaning. (For full study details, see Douglas, 2017).

Should these findings inform how you shape, report, and publicize your next piece of research? Absolutely. My next blog post will explore the ways in which this data can help your next study break into a crowded, noisy news cycle.