Why am I writing this blog post when I have a deluge of deadlines on November 30, including two papers, and a grant due in December, and possible travel to Tanzania in the first week of December? Because I think this is an important, useful lesson for those in teaching/mentoring/academia-ing – the lesson being that students come in the different modalities of thinking and that we as mentors/ teachers truly genuinely need to help them maximize their potential, even if it’s not conventional. I am sharing my experience as someone who is not a verbal thinker.
I am a visual thinker, through and through. Here’s some evidence below from my papers (see supplement at the end) and from my projects. The stuff from my projects, I created NOT for publishing purposes, but I want to look at every single data point visually, and I have done that for every project been part of — which means generating 1000s of household profiles. I do it because it helps me generate hypotheses, and it helps me communicate, and more importantly, it humanizes the data. For a long time, I have STRUGGLED with writing and I still do. That was the whole purpose of starting this blog to improve writing, logic, and communication. I have faced a lot of insults, and I am sure lots of snap judgments on being an immigrant and having grammatical issues. There is a lot of feedback from papers I have written in undergraduate/graduate school, how I need to learn scientific writing, and how this might be hard because “English is my second language.” It is not, but I got that a lot too. Let me tell you, some of the worst grammar Nazis are immigrant Indians trained by the British school systems in India. These people are mortified by my writing and cannot comprehend how someone can only think visually and, with decent analytical skills, cannot write well.
Well, here’s why. I finally googled last night “visual thinker having a hard time writing.” The first website was “The Writing Problems of Visual Thinkers”, written by Gerald Grow, a retired teacher from Florida A&m University (Grow, 1994). I swear this whole article described me to the T. He summarizes visual thinkers’ common problems: excessive use of passive words, weak-narrative, non-linear language, fear of words, and connecting sentences. I can give you an example of each of these from the last decades of my schooling life. Visual thinkers have trouble writing well because they apply spatial thinking to writing, when writing is supposed to be sequential, linear, descriptive…exact opposite of visual thinking (Grow, 1994).
First item on Grow’s list was “Naming imprecise or lacking – The doohickey bollusked up my thingamajig”. This is ME. At home I can only communicate with onomatopoeia, it’s gotten so bad that my partner reminds me to use full sentences and words for the sake of our toddler so she doesnt pick up the way I talk (because trust me, you want a toddler to use words, their default is crying or screaming). This constant struggle for words made me seriously wonder if I had some sort of early onset of dementia because I have a visual memory, but I really really struggle for the words to describe. I have also actively worked towards addressing this writing issue by taking workshops while at Harvard and attending a lot of workshops at Purdue on writing logic and grant writing (at Harvard, we took snippets on Atul Gawande’s work and analyzed line by line how he uses words to evoke imagery, its beautiful and time-consuming process but well worth it). And with a lot of reading because reading is writing, as Dr. Crystal Patil would say. She has informally trained me on how to layout logic in grants. So, this learning to think in words is a continuous learning process for me and it is a struggle and is exhausting – and there are times I do get lazy about it.
Here’s another challenge visual thinkers have — because they can’t describe, they fuse words to evoke words in their heads (from Grow, 1994).
Here’s an example I use: “Dry-through” (Mix of dry-run for slides and drive-through for coffee). I have so many examples where I fuse words that mix Tamil-Spanish-English as well, with a sprinkle of onomatopoeia.
Group writing is also a real struggle for me because organization of the work is so different and everyone writes in their own cadence and tone, and significantly complicated when everyone is trained in different fields. Now back to those writing deadlines with group co-authors.
The second thing that really struck me is a concept that Ms.Grow introduces, called “Semantic memory”. Semantic memory is knowing all the relationship between words and having a network of vocabulary, that could be easily navigated and manipulated. And apparently good writers have this quality. I don’t, I am learning, and thankfully scientific writing is prescriptive. I mentioned in one of my previous blog posts that I always keep good examples of writing to help me when I write either methods, or results, or even discussion. From my reading of various disciplines, here’s my current learned formula that I subscribe to – Write introduction like a sociologist (context + stats), write methods like an epidemiologist (numbers, missing, how this affects results, bias, think trees), write results like an economist (statements, facts, boring, simple), and write discussion like an ethnographer (story, narrative, larger picture, FORESTS).
And a final thing that really struck me from Mr.Grow’s article is that apparently, humans think in 4000 words for minute, and that visual thinkers need to be encouraged to communication through variety of media as opposed to just writing.
So yeah, this is why I love figures, tables, icons (thank you ANH and the free noun project), I like listening, and I am ok with words. And I know in education/academia, people push written prose as the primary way to communicate science and that’s fine, I do take classes to get better and its continuous journey for me – but I do hope it changes in the future. Because, one of the big lessons learned from this pandemic on vaccine hesitancy is that prose doesn’t stand a chance to fb memes, or stupid podcasts.
References and supplement stuff:
Grow, G. (1994). The writing problems of visual thinkers. Visual language, 28(2), 134. Link: https://longleaf.net/wp/wp-content/uploads/2015/02/WriteVisual.pdf
P.s 1. I wrote this in 25 mins because I have been thinking of this for a long time and as Dr.Crystal Patil says, “Reading is writing”. As always, visual thinkers or verbal thinkers are not monoliths; there are lots of flavors. The editing took another 1 hour.
P.s.2. As always, feedback welcome!
P.s.3. I am every so thankful for Dr.Gwentyh Lee for first teaching me to write loops to make 5000 graphs in Stata, which is what lead me to make those plots for each study.
P.s.4 Recently, I taught myself how to write loops in R for VISUALIZATION (thank you stack overflow!). Here’s a sample script for box plots that labels individual data points (plot is not for individual profiles but could be modified). Enjoy and think visually!
uniq_indi = unique(dataset$var)
for (i in uniq_indi){
print(ggplot(subset(dataset, var == i ), aes(x=catx,y=y label = person, color=secondcat))+
geom_boxplot(width=.5)+
# jittered text with geom_text helps with labelling all the datapoints
geom_text(check_overlap = F,
position=position_dodge2(width=0.15, padding=.05), fontface = “bold”, size = 4)+
theme_minimal() + theme(legend.position=”none”) +
ylab(“i”) +
xlab(“category x”) +
theme(axis.text=element_text(size=18),
axis.title=element_text(size=25,face=”bold”)) +
theme(axis.title.y = element_text(margin = margin(t = 10, r = 20, b = 20, l = 10))) +
theme(axis.title.x = element_text(margin = margin(t = 10, r = 20, b = 20, l = 10))) +
ylab(i)) +
ggtitle(i)
ggsave(paste0(i,”_ group.png”), width = 14, height = 15)
}
Supplement stuff — Below are four examples of visuals/profiles/report cards I have created. I have more but deadlines; please email if you’re interested in seeing them!
1) In the example below, I visualized breastfeeding patterns for each individual child across 6 months from-MAL-ED study in Bangladesh and Pakistan. We collected twice weekly data for ~200 kids from eight countries. This paper studies how different bf patterns can affect metrics as measured by demographic health surveys. This whole paper was motivated by the visuals that were created internally to understand the trends. Here’s the article, and you can check out other 6 countries patterns as well: https://onlinelibrary.wiley.com/doi/abs/10.1111/mcn.12352
Breastfeeding trajectory plot of 50 children from Loreto, PEL. Each number/row on the y-axis indicates the pattern of feeding for a child with age in days on x axis. Blue represents exclusive breastfeeding (EBF); orange represents predominant feeding (Predominant BF); yellow represents partial breastfeeding with liquids only (Part BF:liq); brown represents partial breastfeeding with solids (Part BF: sol) and red represents no breastfeeding (No BF). ‘|’ in the sequence indicates when the visit was made. The preceding visit feeding is assumed in the days in between for illustrative purposes. For example, child 25 starts with exclusive bf, shifts to predominant bf ~day 40, shifts back to exclusive at day 60, which stops at ~day 90. Thus, the total gain of EBF days in the first episode of EBF is 40 days and in the second episode, gain is 30 days.
2) Market food availability data vary by season and agroecology, so yes, in some study contexts, you cannot have static indicators that measure market food diversity or market access. Article here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7745109/
Proportion reporting market food availability (as indicated by two key informants in each village) by agro-ecological zone: (a) lowland, (b) midland, and (c) highland agroecology zones in Ethiopia
3) Dietary profiles of DECIDE participant #66 (mainly for looking purposes):
4) Farm and livestock profiles from each of the 2000+ households from the ATONU study – (create loops in Stata and in R, see sample R script above for examples).
\\\\
