All articles
Case study

I Found My Dissertation Topic in an Afternoon After Six Weeks of Despair

PA
PromptDock AIVerified creator — vouched by Prompt Dock
Jun 22, 2026 · 7 min read
promptdock.ai/blog

The hardest part of a dissertation is not the research. It's finding the question. I'm a third-year PhD student in organizational behavior, and by the end of my sixth week of reading I had reached a very specific kind of rock bottom: convinced that everything interesting about work motivation had been studied by someone more qualified than me, published somewhere I'll never reach, and probably replicated twice for good measure. I called my advisor in a tone she later described, charitably, as 'pre-spiral.'

Her advice should have come in year one: 'Stop reading for what we know. Start reading for what we don't.' It's obvious in hindsight and impossible to actually do when you're drowning. You can't read for absence — your brain keeps grabbing onto the presence of things, the studies that exist, the findings that are there. What I needed was something that would systematically ask 'what's missing here?' of a body of work I'd already collected.

Turning a 40-page review into a gap map

I took the most comprehensive thing I'd found — a 40-page meta-analysis on self-determination theory in workplace settings — and ran it through Gemini 3 Pro Preview with the Find the Real Gaps in a Literature Review prompt, field set to 'organizational behavior' and research goal set to 'identify a viable dissertation topic.' I used Gemini deliberately: a 40-page review is a long-context job, and Gemini chews through long documents and returns a clean comparison table without losing the thread halfway down the page. The prompt asks for a GAP MAP table, which turned out to be exactly the right shape — five categories of gap, side by side, each rated for tractability.

The 'Understudied Population' row named something I'd half-noticed for weeks and never managed to articulate: nearly every study in the meta-analysis used full-time, permanent employees at Western firms. Contract workers, gig workers, and non-Western contexts were barely present, and where they showed up they were a footnote about generalizability. The Evidence-in-Text column pointed me to the exact sentences in the review where the authors mentioned, almost in passing, that their samples skewed this way. The tractability was rated High — standard validated survey scales would work, and the population is reachable through gig-work platforms. That 'High' is what turned a vague unease into a project I could actually start on Monday rather than someday.

What got me was that I had read that meta-analysis. Twice. The sentences the Evidence-in-Text column quoted back at me were sentences my own eyes had passed over, because I was reading for findings — for the satisfying 'they discovered X' moments — and these were limitations, the part of a paper your brain treats as throat-clearing. The gap wasn't hidden. It was sitting in plain sight in a section I'd been trained, by years of reading for results, to skim. The prompt's only real magic was reading the part I'd taught myself to ignore.

From a row in a table to a real question

A gap is not yet a question. The TOP THREE DIRECTIONS section did the bridging. It sketched a study: does self-determination theory predict motivation differently for contingent workers, who lack the job security the theory's evidence base assumes? I brought exactly that to my advisor, who has a finely calibrated detector for 'student is grasping' versus 'student has something.' She said, 'That's a real question. Write the proposal.' I have heard her say the first half of that sentence many times. The second half was new.

Here's the honest part: the prompt did not find my topic. It found the shape of what hadn't been done, in a body of work I'd already read but couldn't see clearly because I was reading for the wrong thing. After six weeks of cataloguing what existed, having something map the holes was the difference between spiraling and starting. I'm now nine months into a dissertation on motivational dynamics among gig workers, and the GAP MAP for that original review is still pinned above my desk, a little smug.

If you're staring at a literature review certain that every good question is taken, you're reading it the way I was. Grab the Find the Real Gaps in a Literature Review prompt on Prompt Dock, feed it the densest review you've got, and let it show you the absences. Then run my Create a Reproducible Search Strategy for a Systematic Review prompt to actually go confirm the gap is real before you bet two years on it.

The prompt behind this post
Premium
Find the Real Gaps in a Literature Review

Paste a literature review, meta-analysis, or research summary and get a categorized map of unanswered questions, understudied populations, conflicting findings, and methodological holes — each rated by how tractable it is. The starting point for a dissertation topic or grant proposal.

View promptGemini 3 Pro Preview
Keep reading
Our Support Bot Confidently Made Up Enterprise Pricing — One System Prompt Fixed It

The agent wasn't hallucinating wildly, it was extrapolating plausibly. That's worse. Here's the exact system prompt that taught it to know its own limits.

Elite Prompting: Why Better Prompts Beat Random AI Instructions
I Designed Branded Wrapping Paper for My Candle Business for $28, Not $400

Packaging is part of my product, but a custom print run starts around $400. Print-on-demand plus one seamless-tile prompt got me there for the price of lunch.

Related prompts
Start in two minutes

Find a verified prompt for the job.