Scientific writing 3.0: A reader and writer's guide - Jean-Luc Lebrun, Justin Lebrun 2021

The halo effect and confirmation bias
Strategic writing
The reading toolkit

Unfortunately, we must begin this paragraph with a somewhat depressing but sobering truth. Pull out your manuscript, and gaze upon it lovingly. Have you not spent countless hours writing, rewriting and re-rewriting this document? You know that this finely tuned text into which you’ve poured your blood, sweat and tears will have judgment passed upon it by the reviewer.

They’ll at least spend a solid hour or two combing through it before coming to a decision, correct?

…correct?

As trainers who have taught scientific writing skills for more than two decades, we’ve had thousands of scientists from all fields come through our workshops. Inevitably, many have had experience reviewing papers. Course after course, we have asked this group of reviewers a simple question: “How long does it take you, on average, to make up your mind about whether or not a paper is worth publishing?” There was of course a range of answers. Some reviewers spent large amounts of time reading a paper before coming to any conclusion about its value. However, those reviewers were the exception to the rule. The majority spent an average of only 15 minutes.¹⁷

If you find yourself thinking that it is impossible to review a paper in such a short amount of time, you would be correct. In fact, 15 minutes is insufficient to read through the whole manuscript. Logically, this would imply that reviewers are deciding whether or not your paper is publishable based on incomplete information. To understand why this occurs, we must foray into the field of psychology.

Reviewers take the responsibility of vetting new contributions to science seriously. But they are also human, and so like any of us, they suffer from evolutionary cognitive biases. The first of these, the halo effect, helps explain why they confidently decide whether or not a paper is publishable after just 15 minutes of reading.

Although the term “halo effect” is relatively recent, having only been coined and scientifically proven in 1920¹⁸, the effect itself has been intuitively exploited since the dawn of civilization. Have you ever noticed how in ancient times benevolent deities were portrayed as beautiful, and evil spirits as repulsive or scary? Or in a modern context, have you noticed that the main characters of your favourite TV shows tend to be attractive individuals? Neither one of these scenarios are coincidental: they demonstrate the halo effect at play. We associate positive attributes such as intelligence, creativity and charisma with attractiveness, and negative characteristics with unattractiveness¹⁹. Objectively, such extrapolations from appearance to character are completely baseless; yet, humans consistently prove that such extrapolations exist and are made subconsciously.

The halo effect does not relate simply to people and their appearance. On a larger scale, it dictates that if our initial impression of someone or something is positive, that impression bleeds into other attributes of the targeted person or object²⁰.

What does this have to do with scientific writing and reviewers? Whereas we make conclusions about someone’s appearance instantaneously, getting a first impression of a manuscript takes a little longer (let’s say … 15 minutes on average. Sounds familiar?) Once the impression has been made, the halo effect dictates that even without having read the rest of the manuscript, the reviewer will attribute to it positive or negative qualities. If the reviewer is impressed by the clarity of the abstract and the significance presented in the conclusion, he or she may subconsciously assume that the methodology is robust and the introduction is sufficiently comprehensive (although again, there is no clear reason to believe that this is objectively true). The reviewer has already decided to “like” the manuscript overall, even without comprehensive reading.

Once that preference has set in, the second cognitive bias comes into play: confirmation bias. Having initially felt the paper was worthy of publication, the reviewer will continue reading it focusing on the positive elements of the paper and minimizing its faults. On the other hand, if the reviewer felt the paper was not ready for publication, he or she would then comb through it with the aim of pointing out every little mistake or lack of methodological detail.

Taken together, the halo effect and confirmation bias act as a powerful synergistic mechanism that guides the reviewer’s positive or negative response to your paper. Knowing that this process only takes 15 minutes on average, you must make certain that the parts the reviewer reads first, the critical parts of the paper, are well written. What are these critical parts?

The title, abstract, structure, introduction (if the reviewer is a non-expert), conclusion, and the visuals (particularly the diagrams in the methodology section and the figures that closely support your contribution). Given how essential these sections are in setting the reviewer’s expectations, we have given each one its own chapter in this book.

The halo effect is neither friend nor foe. It is simply a psychological reality that your reader will face. Will you let it affect you detrimentally, or write in such a way as to put it to good use?