Category Archives: writing

Writing Process Blog Tour – Elizabeth Heller, PhD


My friend James Herman invited me to join the writing process blog tour. James and I have been friends for almost two decades, and have recently become scientific collaborators in our shared field of neuroscience. As a scientist, most of my current writing aims to disseminate my work through the publication of primary research articles. To this end, I was fortunate to receive an education that provided me with a strong foundation in concise prose. My sixth grade teacher taught how to construct an outline, and from this outline to craft an essay. The technique appealed to my natural sense of organization, and I have relied on it to this day. Throughout high school and college, I honed my writing and learned to include elaborate and symbolic language, and to formulate a thesis. I was fortunate to have guidance from my parents as well. My father, a chemist, is an excellent writer and poet, and he edited my essays and prose from an early age. In the ninth grade I wrote a research paper on tuberculosis for my introductory biology course. My father provided his medical dictionary to decipher the original research papers I used as source material. So impressed was my dad with the final product that he shared it with a colleague, the director of a tuberculosis research center, and encouraged him to let me join the lab. The director agreed (I worked on Staph, not TB) and my research career was launched. Today I am a postdoctoral fellow studying the molecular biology of drug addiction and depression, hoping to land an academic faculty position in the next year.

What am I working on:

I am currently crafting what is known as a Research Statement. This is a 2-3 page document that introduces my research background and future goals to a panel that will evaluate me as a candidate for an academic tenure-track research position. The market for such positions is tight and competitive, and the career statement intends to prove that I am a productive scientist, capable of generating data and securing funding. Furthermore, it should be clear that I have developed a unique scientific perspective and have crafted a plan that differs from that of my current and previous mentors.

It is a daunting task.

The Research Statement consists of an introductory paragraph that highlights the research accomplishments of the postdoctoral fellowship, and alludes to the main focus of the candidate’s future research. Next is a concise description of previous work, both in graduate and undergraduate research, as well as work with collaborators. This shows a history of research productivity, a trajectory of scientific interest, as well as the ability to seek the expertise of other researchers. If possible, it is helpful to include representative data figures of recent work, since the panel reviewing the application consists of scientists that typically digest material through graphical representation (a thousand words…). These sections are peppered with allusions to awards and fellowships won, presentations at scientific meetings, and citations. For most applications it is important to include a section on teaching and mentorship, as these are critical aspects of the research career and often teaching is required for an academic post. At least one third to half of the statement is devoted to a detailed description of the plan for future research. Academic research is costly, and hiring committees have a commitment to select candidates that will be able to secure independent funding. The research statement is thus an opportunity to present a research plan that can be imagined as several successful grant applications.

So far, I have generated a 1-page research statement that I have submitted for some soft interviews, and a few applications to meetings. The sections on past and current research, as well as teaching, mentorship, and collaboration are relatively complete, but I am struggling with crafting a plan for future research. As I am still collecting data for my current project, and feel that I am just now confident in my expertise in this work, I have had a hard time thinking creatively about a new direction. I have two project ideas, but they are not comprehensive or unique enough to qualify as a research ‘identify.’ So where do new ideas come from? As with writing, reading helps. Diving into the current literature uncovers holes in our existing knowledge that often lead to interesting questions. I am working on reading and discussing ideas with my colleagues, to find inspiration.

How does my work differ from others of its genre:

Scientific writing is in general more concise than other forms of writing. This is due to word restrictions set by publishers, the use of figures instead of text to convey information, and the fact that scientists must consume a terrifying lot of literature to stay current. I tend to find concise writing appealing and natural, a gift I am grateful for in a profession for which I don’t often feel I am a natural. The Research Statement is also a concise piece of prose and, as with research articles, each sentence has to pack in a considerable amount of meaning. One mentor advised me to imagine that it will be read by a search committee member on her mobile device, while walking from one seminar to another at a meeting. Concise writing requires greater clarity, as it lacks the opportunity to state one idea in several ways. Scientists are aided by writing primarily for other scientists in their field, and can thus rely on familiar composition and sentence structure to convey meaning. I find that it is the precision and formulaic nature of scientific prose that enables complex ideas to be stated with minimal language. This thrust towards similarity between my writing and that of others is what allows it to be read and digested while multitasking, i.e. most scientists fit in their reading while doing 8 other things.

Why do I write what I do:

The main mode of advancement in a research career is through the successful publication of original research articles. Typically I am engaged in some form of communication of my data, either through abstracts to be presented at scientific meetings or original articles to be submitted for publication. The publication process is lengthy, and I currently am in a phase of revision, in which I gather additional data to satisfy the requirements of scientists that have reviewed my recent submission and asked for additional data to determine if the manuscript is appropriate for publication. While I gather this data, the manuscript is on the back burner. Once this paper is published I will be “on the market,” that is, I will seek and apply for an academic position as a tenure-track research faculty at either a medical center or a university. Hence my current focus on the Research Statement, described above.

As a scientist I am also required to write grant and award applications, in order to secure funding to continue my work. The structure of these differs from that of a research article, in that a grant application describes unpublished, preliminary data, with a focus on the intention for the work, rather than its successful completion. As a student and trainee, I have had only minimal experience with this type of writing, but as a faculty member my main focus will shift abruptly from designing experiments and gathering data, to writing grant applications.

How does my writing process work:

My writing process begins with outlining. For a research paper, after I have generated the figures, I outline the various sections (using the very technique I learned in sixth grade) and then fill in the outline with prose. I will spend the majority of the writing process developing and detailing the outline, as the prose will follow easily from this. A manuscript that describes four years of work in 2500 words requires the crafting of exquisitely precise sentences. I tend to craft sentences meditatively, either while jogging, showering, or working at the bench. I find that process delightful, and I will chew over a sentence for weeks before putting words to paper. Once I have a draft, I ask a few scientific peers for comments and edits. In addition to the co-authors, I ask for the edits of a colleague that is less familiar with the research. This is helpful, as it is often hard for me to assess clarity on work that I am so intimately familiar with. Next I share the manuscript with my mentor and other senior authors, and after their revisions I submit to a journal for publication.

I am fortunate to be a native English speaker and a naturally gifted writer that was educated with great emphasis on prose. This skill has been acknowledged by several of my scientific mentors and has aided me throughout my career. In fact, the reviewers of my current manuscript did not ask for any textual revisions. A welcome gift.

How does my writing process not work:

Procrastination. Writing is hard, publishing is harder, and I find that I am reticent to start the task. I will even delay making the figures, because I am so fearful of entering the review/revision/rejection cycle. Once I do begin, the process is quite pleasurable, as designing figures and writing prose is a complete departure from my typical lab grind, and relies on a different set of instincts and skills. I have procrastinated lately on completing my Research Statement, and have turned instead to looking for inspiration to mold my scientific identity and plan. I hope that as I continue my career, and have more success in publishing, my confidence will grow and I will not be as fearful of starting the writing process. In the mean time, I am grateful for this Writing Process Blog Tour. It has greased the prose wheels for me, and inspired me to keep at it.

Writing Process Blog Tour – jph


I was asked by my mother, Joanna Clapps Herman, a writer of fiction, a teacher or writing, and one of my main writing influences to contribute to this Writing Process Blog Tour. She and my late father William Herman (also a fiction writer and a professor of English) were always writing, our apartment was filled with it. From an early age, I conceived of writing as an important job, a task to be taken seriously, but also a source of joy, a labor of love, and at times a Sisyphean process. Naturally, writing became part of my life as well, though in a very different form than it had occupied in my parents’. I’m a neuroscientist, a post-doctoral fellow at the moment, and thus my writing is focused on the efficient communication of my research findings. It was very surprising to me to realize near the end of my PhD that I was pretty terrible at scientific writing. Having spent a great deal of time so close to my parents’ efforts, and having also spent a great deal of time with them editing my own papers for school, I had come to the misguided conclusion that I was a marginally above average writer, and thus should be able to turn my modest skill into at least acceptable scientific writing. Now I have come to realize that it’s truly a different genre entirely, and the expectation of translating general writing skill into science writing skill was misguided. Instead, I am trying to approach science writing as a new skill I am trying to acquire essentially from scratch. As such, at present, I am trying my best to follow the process favored by my lab head, Dr. Richard Krauzlis. What I present below is a window into my current efforts along those lines.

What am I working on:

Presently, I am working on a draft of a manuscript that I plan to submit for publication in a scientific journal. Along with another post-doctoral fellow and the head of my lab Dr. Richard Krauzlis – the paper’s eventual co-authors – I conducted a series of experiments in human “psychophysics”. This field is concerned with quantifying various aspects of human perception. For us, this involved asking several people to sit in a dark room staring at a computer monitor for several hours (cumulatively, not continuously) while we presented flickering patches of color that sometimes became more intensely colorful by degrees and asking them to report whether (under systematically varied conditions) they thought they had detected such an increase in intensity. Thus, we were (more specifically) studying color perception.

A successful neuroscientist will write several papers per year describing experiments they have completed. These papers go into journals that are ranked by “impact factor”, a numerical value which attempts to quantify the average importance of the papers appearing in its pages, which generally comes down to counting how often other cite these works.

How does my work differ from others of its genre:

In practice, there is little room for distinction in terms of style of presentation in scientific writing. Almost all scientific papers have the same structure, they consist of (1) an Abstract – a brief summary of the work encapsulating the hypothesis, a whiff of background, and the work’s main findings , (2) an Introduction – more expansive background, a contextualized presentation of the hypothesis (or hypotheses) being tested in the work, (3) a Methods section – a description of the procedures followed, equipment used, the nature of the test subjects, and any other relevant information regarding how the experiments were conducted, (4) a Results section – the actual data resulting from the experiments, (5) a Discussion section – a description of the relationship of the results to previous findings, and conclusions regarding the success of the experiments, as well as any plans for future follow-up studies. Some journals (particularly the ones with high impact factor) limit the number of words in these sections, which can be quite a challenge if your work is complex, which it nearly always is, forcing one to either artificially gloss over explanatory points that deserve more attention, or compress things in strange ways. Furthermore, it is expected that a certain dry style will be adhered to. This is serious business after all, and jokes have no place here!

However, one aspect that I try to focus on a great deal is the visual presentation of my data. I feel that too often, visual presentation in scientific papers is quite bland. I think this is a particularly grave error because the writing must be so technical and provides little opportunity to truly captivate the reader. In contrast, aesthetically appealing presentation of data is a chance to tantalize one’s audience. A striking experimental result can easily be missed with poor presentation, and humdrum results can be made more appealing by presenting them in a way that is pleasing to the eye.

Why do I write what I do:

There are a few core reasons to write scientific papers: (1) you have to – the old chestnuts hold: “publishing is the lifeblood of the scientific community”, “publish or perish”. If you don’t publish at least one paper a year, your fellow scientists will wonder what you have been doing if not working hard at writing up your results! Essentially every scientist lives in fear of become obscure or irrelevant because it will hurt their chances at getting grants, and without a grant, it is impossible to work because it is how they pay for every aspect of their extremely expensive work (graduate students salaries and tuition, technicians salaries, materials, equipment, et cetera). (2) You don’t want to “get scooped”. If you don’t write the experimental results up and publish a paper, somebody else eventually will do the same experiment and write it up first, and then your paper is irrelevant and will have to go to a lower impact journal and then we end up back at the first point. (3) You’re excited about the result and you want to tell the world! This is the most fun, and sadly the least common reason to write papers, but it does happen, and it keeps one invigorated about being a member of the scientific community.

How does my writing process work:

Here’s the current process I’m using with the work I mentioned above.

Once all the experiments were completed, I spent several weeks to months analyzing the resulting data and generating a series of graphical representations (figures) of the trends present. This was an iterative process: initially I would present a graph I considered representative of some interesting aspect of the data to my co-authors for discussion. The outcome of that discussion could be (1) that we all liked the figure (graphic) and felt it displayed an important and relevant aspect of the data that needed to be included in the manuscript, (2) we realized it was garbage and should be discarded, (3) we disagreed and further more subtle analyses needed to be run in order to disambiguate between (1) & (2). After we had arrived at what we thought were the most important points to be communicated about our findings, I generated a series of final figures to be included in the paper.

From the final figures, I generated a series of one to two main points that I was communicating with each figure. The idea that Rich had given me was that these would serve as topic sentences for the results section.

Once I had my topic sentences, I went over them with Rich and we tweaked them a bit, shuffling, rewriting and perfecting them. I then expanded these into the full results section, which may sound somewhat robotic, but this was actually a nuanced moment because I needed to truly guide my reader through the data, easing them in and carrying them from one point to the next. Telling them a cohesive story that ended with a complete picture of what we had found.

With a draft of the results completed, I then turned to the introduction section. Since I now knew everything that I wanted to present to the reader, I could decide how much background information and framing was necessary to include in the introduction. The idea is that if I had written the introduction first, without knowing precisely which points I was going to make in the results and how I was going to link them together, it would be difficult to predict exactly what I should include in the introduction; perhaps I’d include too much on one part of the background or too little on another. This way, I could avoid including extraneous discussions of things not germane to the results I was presenting, and also write the introduction in such a way to make it lead into the (it is to be hoped) successfully crafted narrative I created in the results.

After the introduction draft, it was on to the discussion. A similar principle applies to the ordering of results first then introduction. Now that I have framed the story with the intro, and told it effectively in the results, I can bring it on home in the discussion section. I don’t need to discuss every paper that has ever mentioned the topic of color, I only need to mention those that bear on the conclusions I want to make based on the results I’ve presented. If I want to conclude that human color perception is relatively poor when looking at purples and relatively good when looking at green hues, I should discuss any papers that have found the same or the opposite and why our results might agree or conflict. I might also discuss any papers that mention disparities in perception depending on hue as well, but I certainly don’t need to discuss papers concerned with color perception under varying conditions of stress, as this is not relevant to my own work. The general principle at work in all of this so far is to tell an effective story and not be exhaustively presenting everything about the work, but rather to present only the details necessary to getting one’s point across effectively.

Finally, I get to write perhaps the easiest section of the paper, a reward for all my hard work on the other sections, I’ve left the dessert for last. Again, I’m only going to include information on relevant details. Does it matter to my color perception study that the temperature of the room was 72 degrees? No! But it does matter that I used a particular type and brand of computer-display, and how I chose the colors and the temporal order of stimulus presentation definitely matters. All this goes in.

Now that I have a working draft, I submit it to my co-authors, and we then go through (if we are lucky) very few revisions amongst the three of us before we feel we are ready to submit the article to a journal. We choose a journal that we think is appropriate for the paper in terms of (1) audience, (2) impact factor relative to the importance of the result, and (3) probability of the paper being accepted. There’s no formal calculation here, just heuristics.

Once the paper is submitted, one of several editors that work for the journal will receive the manuscript and choose reviewers for it. These will be people who probably have some degree of expertise in the field covered in the paper. Those reviewers will read the manuscript, make comments, and send them back to the editor with a recommendation regarding publication. This is something like: “they need to edit it a lot more before it’s suitable”, or “the methodology is totally flawed”, or “this is brilliant work that the world must see!” Or rather, they’re nothing like that, but instead include some combination of points that live in the space defined by comments like that at its outer edges. This can happen several times before the journal agrees to publish the paper, or the paper can be rejected, and one needs to go back to the drawing board. I’ve had both experiences, and I’ve also had a paper accepted after just some minor revising. Anything is possible, really.

How does my writing process not work:

There are two main factors that impede effective communication of my science: (1) once I’m at the stage of writing the results, I’m so myopically close to the work that it’s extraordinarily difficult to get to a mindset other than that of the total believer in the results. However, I’m writing the paper for an audience that may think it’s total baloney, or simply not be very familiar with the methods, or not really agree that the question being asked is all that interesting, or any of a host of other possibilities. This makes it a struggle to know how the work should be written to make it most appealing. I know how to write it so I understand it, but “me” is the person I least need to pitch it to. (2) Knowing how to tell a good science story is a big challenge for me. On some level, scientists get into what they do because its technical and clean and clear, but telling a good science story is sorta about spinning it just right. Keeping in the details that keep it true to what it really is, but excluding the points that might give a reader the wrong impression. Making everything link up just so. In my mind, its all connected by the order I did it, I did the second experiment because there was that weird thing in the data for the first experiment and we realized we had to control for it by, blah, blah, blah. But you can’t say that! You need to make it seem like you knew all along that you were going to do experiment 1 and experiment 2.

I’m confident that if I keep doing this, and keep focusing on it, eventually I’ll be a good science story teller, but for now, it’s a struggle.

On Reading The Unreadable

Phaistos Disc

This is the Phaistos Disc. It is the first piece of printed writing and nobody can read it. It is loosely dated to 1900BC (the Greek Government will not allow it to be subjected to thermoluminescent testing which would clear up its time of origin). It is clearly printed in the sense that each character was stamped onto this piece of clay, not etched or written as all other pieces of writing from this period. Many attempts have been made to decipher it, but the prevailing feeling is that these have been unsuccessful. I find mysteries like this beautiful and exciting because they area great challenge and reminder of how much our knowledge relies on a kind of collective remembrance of things. Without such distributed knowledge, where would we be?