Regeneron Science Talent Search (STS) 2027: How to Apply
Updated: Aug 28
TL;DR
The Regeneron Science Talent Search is the oldest and most prestigious science and math competition for high school seniors in the United States, and the 2027 application is open right now through 8:00 pm Eastern on Thursday, November 5, 2026. Over 2,600 seniors enter each year, 300 are named scholars and receive $2,000 each, 40 become finalists and receive at least $25,000, and the first place winner receives $250,000. What makes STS different from a science fair is that your research project is only one part of a holistic application that also weighs your recommendations, your transcript, your essays, and how you spend your time. If you want a PhD mentor to help you design and write the independent research project that an STS application is built on, the Lumiere Research Scholar Program is worth a look.
Key takeaways
The 2027 deadline is Thursday, November 5, 2026 at 8:00 pm Eastern Time. Every part of the application, including all three recommendations, must be submitted by then. The Society makes no exceptions.
Over 2,600 students enter annually. In the 2026 cycle, 2,612 students entered, the largest applicant pool since 1967. Of those, 300 became scholars and 40 became finalists.
Total awards are $3.1 million per cycle. First place is $250,000, the largest scientific prize available to a high school student in the United States. Each of the 300 scholars receives $2,000, and so does their school.
You do not need to be a US citizen. Any student of any citizenship who lives in and attends their final year of secondary school in the US or its territories can apply. US citizens studying abroad may also qualify. Non-citizens at American schools abroad cannot.
Only individual research is eligible, and you can only ever enter once. Team projects are disqualified, you cannot split a team project and enter as an individual, and students who have entered a previous STS may not enter again.
What is Regeneron STS?
The Regeneron Science Talent Search is the oldest and most prestigious science and mathematics competition for high school seniors in the US. It was established in 1942 as the Westinghouse Science Talent Search, and it exists to give original high school research a national forum where it is reviewed by a jury of professional scientists.
The thing that separates STS from most other science competitions is that your research project, while it carries the greatest weight, is not the only factor in award decisions. The Society for Science uses a deliberately broad selection process that reads your transcript, your recommendations, your essays, your activities, and your prior research alongside your paper. It is closer to a fellowship or scholarship competition than to a science fair.
STS is run by Society for Science, a non-profit based in Washington, DC that has produced the competition for more than eight decades. Most people know the Society as the publisher of Science News and the organizer of ISEF, the world's largest science fair. Regeneron Pharmaceuticals has been the title sponsor beginning with the 2017 competition, and in 2026 extended that sponsorship through 2036 with a $150 million pledge. Regeneron itself plays no part in selecting winners.
If you are looking at science competitions more broadly, our guides to ISEF and JSHS cover the two closest alternatives, and our roundup of 15 research competitions for high school students maps the wider field.
How the selection process works
Every eligible entry is read by three or more doctoral scientists, mathematicians, or engineers in the relevant discipline. Entries are evaluated in four areas:
Research report and scientific merit
Student contribution to the research
Academic aptitude and achievement
Overall potential as a future leader of the scientific community
Alongside that, a rules committee checks each project for compliance with eligibility and scientific rules, and every entry is run through plagiarism monitoring software.
From that pool, 300 scholars are selected. Those top entries then go to a second judging panel, which selects 40 finalists. The 40 finalists travel to Washington, DC for Finals Week, where they face two kinds of judging: project judging on their research, and panel judging designed to test the depth and breadth of their general scientific knowledge. That second kind is worth internalising early, because it means you are being assessed on your understanding of your whole field, not just your experiment.
A few things the Society states explicitly that it does not consider: category or regional quotas, outside letters of recommendation beyond the required ones, demonstrated interest, and specific score cut-offs. Evaluators are told to weigh your achievements against the resources and lab access you actually had.
How competitive is Regeneron STS?
The Society does not publish an official acceptance rate, but it publishes the underlying numbers, so you can calculate it. Using the official 2026 cycle figures of 2,612 entrants:
Stage | Number selected | Share of entrants |
Scholars | 300 | roughly 11.5% |
Finalists | 40 | roughly 1.5% |
Top 10 winners | 10 | roughly 0.4% |
Two things to keep in mind about those numbers. First, entry is open, so this is a selection rate rather than an admissions rate, and the pool is already self-selecting: 2,600 seniors who have finished an independent research project are not a random sample of high school seniors. Second, the 2026 pool of 2,612 was the largest since 1967, so the competition is getting harder, not easier.
For context on what "competitive" means in practice, the 2026 finalists came from just 35 schools across 15 states.
Is Regeneron STS prestigious, and is ISEF more prestigious?
Yes, STS is prestigious. Being named a scholar is a real credential worth putting on an application. Becoming a finalist, or winning a top 10 award, is what moves the needle significantly.
The competition's standing rests partly on its alumni record. Since 1942, STS alumni have gone on to earn thirteen Nobel Prizes, eleven National Medals of Science, and two Fields Medals. Regeneron's own co-founder and Chief Scientific Officer, George Yancopoulos, was a 1976 Science Talent Search winner.
On the ISEF comparison, the honest answer is that they are not really on the same scale, because they are different kinds of competition. ISEF judges you primarily on your project and how you present it, visually and orally, and it is open to high schoolers of all ages. STS is seniors only and holistic. In our assessment, becoming an STS finalist is roughly comparable to becoming an ISEF finalist. That equivalence breaks down at the top: winning at ISEF is arguably more prestigious than winning at STS, though STS pays considerably more.
Who can apply to Regeneron STS, and is it only for US students?
No, STS is not only for US citizens, and this is the single most commonly mis-stated fact about the competition. Here is what the official rules actually require.
On citizenship and residence, you qualify if you are either:
A student of any citizenship who is enrolled in and attending secondary school and maintains full-time residence in the United States, Puerto Rico, Guam, the US Virgin Islands, American Samoa, Wake Island, Midway Island, or the Marianas, or
A US citizen in your final year of secondary school abroad, attending a Department of Defense Dependents School, an accredited overseas American or international school, a foreign school as an exchange student, or a foreign school because your parents are temporarily working abroad. In this case you must provide proof of both citizenship and school accreditation.
Students attending American schools abroad who are not US citizens are not eligible. That is the one common case people get wrong in the other direction.
On school status, you must be in your final year of secondary school, which includes public, private, parochial, charter, and home school. You must not have graduated before the deadline, and you must not have entered any previous Science Talent Search. Homeschooled students are eligible, and need to contact the Society to be added to the school lookup and to submit a course record as a transcript. If you are uncertain about your graduation year at the time you enter, you have to notify the Society by December 15, 2026 either to withdraw or to confirm you will graduate in the 2026/2027 academic year.
On age, you must be over 13 with parental or guardian consent, or 18 or older but under 21, or an emancipated minor over 13.
On research type, only independent, individual research is eligible. Team projects do not qualify, and you cannot split a team project and enter your portion as an individual. If you were part of a team and want to continue alone, you may only submit the work you completed independently. Research conducted alongside adult researchers in an institution is permitted, and is very common, but you must be clear throughout the application about which parts were yours. Continuation projects count, and research completed in any year of high school is eligible, so a project you finished as a sophomore is fine.
You can read the full eligibility section on page 6 of the 2027 Rules Book and cross-check it against the official FAQ.
Paid programs and mentors
You are allowed to do your research in a paid program, and plenty of entrants do. But both you and your mentor must disclose the payment structure, and no research program may guarantee placement in STS. If the Society finds a program making that promise, every entry from that program is disqualified.
You can also be mentored by a relative, or by someone with a professional connection to your family, as long as the relationship is disclosed. The Society's guidance on choosing a program is worth reading directly: a legitimate program publishes the names and credentials of its mentors, publishes transparent pricing, does not promise results, and never asks students for secrecy.
When is the Regeneron STS 2027 deadline?
The 2027 application closes at 8:00 pm Eastern Time on Thursday, November 5, 2026. Here is the full published timeline for the cycle.
Milestone | Date |
Application opens | June 1, 2026 |
Customer support deadline | November 4, 2026, 8:00 pm ET |
Application and recommendation deadline | November 5, 2026, 8:00 pm ET |
Top 300 scholars announced | January 7, 2027 |
Top 40 finalists announced | January 21, 2027 |
Finals Week, Washington DC | March 11 to 17, 2027 |
Top 10 winners announced | March 16, 2027 |
The recommendation deadline is the same as your deadline. All three recommendations are due at 8:00 pm ET on November 5, not before it and not after. If a recommender misses it, your application is incomplete. Request your recommendations the day you open your application.
The customer support deadline is 24 hours earlier. If you report a technical problem by 8:00 pm ET on November 4, the Society guarantees it will be resolved before the deadline. Report it later and there is no guarantee. Put both dates in your phone.
No portion of the application is accepted after the deadline for any reason. The Society also does not consider any materials or updates sent after submission.
On terminology: STS has no "semifinalist" round, though people search for one. The 300 announced in January are scholars, and the 40 announced two weeks later are finalists.
What do you win at Regeneron STS?
Total awards are $3.1 million per cycle.
Award | Amount |
First place | $250,000 |
Second place | $175,000 |
Third place | $150,000 |
Fourth place | $100,000 |
Fifth place | $90,000 |
Sixth place | $80,000 |
Seventh place | $70,000 |
Eighth place | $60,000 |
Ninth place | $50,000 |
Tenth place | $40,000 |
Remaining 30 finalists | $25,000 each |
Each of the 300 scholars | $2,000 |
Each scholar's school | $2,000 |
Beyond the cash:
All 40 finalists get an all-expenses-paid trip to Washington, DC for Finals Week, which includes visits to historic sites, meetings with prominent scientists and national leaders, a public exhibition of their research, and a black-tie gala where the top 10 are announced.
The 300 scholars receive an invitation to apply to Regeneron's summer internship program.
Every eligible entrant, including the roughly 2,300 who are not named scholars, receives a t-shirt, laptop stickers, and a free one-year digital subscription to Science News.
Scholars' schools get $2,000 to support science, math, or engineering education. If you are homeschooled, that money goes to a public school district in your state or to a science non-profit of your choosing.
Awards under $20,000 are paid in the year they are awarded. For awards of $20,000 or more, payments are made annually either to you for educational purposes or directly to your college, you must be an undergraduate or graduate student in good standing to receive them, payments must begin within six years, and they must conclude within eight years of finishing high school.
There is no fee to enter. The 2027 rules book and official FAQ list no application or participation fee at any stage.
What does the Regeneron STS application consist of?
The application resembles a college application to a highly selective STEM program. It runs across 13 numbered tasks in the online portal, and you can work in stages and keep editing up to the deadline.
The 2027 rules book enumerates Tasks 1 through 12; the portal refers to Tasks 1 through 13. The thirteenth is most likely the Ethics Statement and submission step.
Task | What it covers |
1 | Eligibility |
2 | About you (demographics) |
3 | School information |
4 | Recommendation requests |
5 | Disclosures of all support received |
6 | Science research description |
7 | Rules Wizard (determines your paperwork) |
8 | Research report upload |
9 | Previous research |
10 | Essay questions |
11 | Activities, interests and awards (up to seven activities) |
12 | Test scores (optional) |
Tasks 1 through 4 are mostly demographic and administrative. The Society's own advice is to knock them out in an hour, which gets you about 20% of the way through, and to request your recommendations at the same time.
Can you use AI in your application for STS?
Task 5 requires you to disclose all support of any kind received across your research and application: paid support, AI use, and any pre-existing relationships. This is not a formality. Entrants who fail to disclose involvement in a paid program face disqualification, and the rules make the point twice that you should err on the side of sharing more rather than less. Conflicts you have to declare include a parent working in the same department as your mentor, even if they barely know each other.
The research report
This is where most of your time goes. The report is evidence of research ability, scientific originality, and creative thinking, and it should read like a journal article explaining your experiments and conclusions. It is not identical to a paper you would submit to a journal, and the formatting requirements are specific:
20 pages maximum, no minimum. Anything past page 20 is not read.
Title page, abstract, and bibliography do not count toward the 20 pages. Appendices do count.
Font must appear at least as large as Times New Roman 11pt. Captions can be smaller if legible.
1.5 line spacing, 1 inch margins on all sides, no multiple columns.
Page numbers bottom right, starting after the abstract.
PDF only, 4MB or smaller, named LASTNAME.FIRSTNAME.ZIPCODE using your home zip.
Every single image, graph, table, and chart must be cited individually, directly under or next to the graphic. A reference list for graphics is not accepted. This includes images you made yourself, and it includes naming the software you used. APA style is preferred. Failure to cite an image can get you disqualified.
No links anywhere except in the bibliography. Evaluators are not permitted to click links.
Research proposals, unfinished investigations, literature reviews, and essays are all ineligible. You need data and results. You may submit published work, but the Society recommends against it unless you are the sole or first author, because co-authorship makes your contribution hard to assess. If you have published as part of a group, acknowledge the paper and submit your own version highlighting what you actually did.
One small but genuinely useful permission: if your field conventionally writes in the first person plural, you may use "I" instead of "we," which helps clarify what you did independently.
If you are earlier in the process, our guides on what makes a good research question and publishing your research in high school are the right places to start.
Recommendations
You need three types, and none of them is a letter.
Educator Recommendation (up to 2). A high school teacher who has taught you in a classroom in any year, preferably in science, engineering, or math. If you are homeschooled and your parent is your only educator, request a second from a non-relative.
Project Recommendation (up to 2, at least 1 required). The person closest to your day-to-day research, which is often a graduate student rather than the head of the lab. If no adult guided you, have a parent or teacher complete it saying so.
High School Report. Your counselor uploads your official transcript as a PDF. Make clear to them that this is a transcript request and not a letter request. Current semester grades are not needed.
The Society does not accept PDF letters of recommendation. Recommenders answer specific questions inside the online system, with word limits generally in the 75 to 250 word range. Project Recommenders get asked pointed questions: how you got the idea for the project, what the size and composition of the lab was, whether other high school students in the group are also entering STS, and which specific components of the work (research question, procedural design, data collection, data analysis, drawing conclusions) they can credit to you. Recommenders are also barred from using AI, and nobody other than the recommender may draft or edit their responses.
Essay questions and other components
You can preview the current questions yourself in the 2027 Application Questions PDF linked from the official FAQ. Expect questions about you, your goals, and your project, with defined word counts.
The Society's guidance on the essays is worth taking at face value: you do not need to max out every word count, but you should give full detail in the sections that ask you to explain your process. If you are stuck, try recording yourself answering out loud.
On previous research, you list conferences or journals where your work has appeared. If this is your first project, that is fine, and it is true of many successful entrants each year. On activities, you may list up to seven, and the framing is how you spend your time rather than a formal extracurricular list. On test scores, submission is optional and choosing not to submit is not held against you. You may enter SAT, ACT, AP, and IB scores.
Paperwork and approvals
Task 7 is the Rules Wizard, and it is the step most applicants underestimate. It walks you through questions that determine exactly what paperwork your project requires.
You almost certainly need pre-approval if you did any of the following: asked humans to take any action at all, including filling out a survey or testing your invention; used human tissue or cell lines from any source, including a public catalog; or worked with any data relating to humans, whether you collected it, received it from a mentor, or pulled it from a public source.
The critical point about Institutional Review Board approval is timing. A properly constituted IRB must approve your research plan before you begin recruiting or interacting with human participants. Approval after the fact does not work. If you change your research plan, you have to repeat the approval process before resuming data collection.
If your project was at a federally regulated institution such as a university or medical center, use that institution's IRB. If it was at school or at home, a school or district-level IRB is usually acceptable, and any high school can form one. It needs a minimum of three members: an educator not involved in the project, a school administrator, and one qualified person capable of evaluating physical or psychological risk, such as a physician, psychologist, registered nurse, or licensed social worker. No IRB member may be related to you, and a teacher who oversaw your project cannot sit on the board reviewing it. An improperly constituted IRB invalidates the approval and results in disqualification.
Similar pre-approval regimes apply to vertebrate animal research (IACUC approval, plus significant restrictions on what you may do at all) and to potentially hazardous biological agents. Projects above BSL-2 are prohibited outright, and culturing biological agents at home has been prohibited since 2020, including apparently harmless things like baker's yeast fermentation and mold growth.
The US Department of Health and Human Services Office for Human Research Protections hosts the training the Society recommends completing before you plan any human participant study.
What are the AI rules for Regeneron STS 2027?
The 2027 rules book includes a full acceptable-use chart for generative AI, and the restrictions are stricter than most applicants assume. This is worth reading carefully, because violations are grounds for disqualification.
Never acceptable:
Using AI to write your application responses. You are personally responsible for drafting all of them.
Using AI to draft the research report, or any section of it, including the abstract.
Asking AI to add points to your paper, or to produce your conclusions or future steps.
Asking AI for a starter bibliography, or to generate citations. The rules note that AI hallucinates references, and discovery of a fake reference results in disqualification.
Reproducing an AI summary of a source in your literature review without reading the source.
Using AI to diagnose or address medical issues, including mental health.
Acceptable, with conditions:
Using AI to summarise an article before you read it, as a starting point. No citation needed.
Using a chatbot to generate and develop ideas. May require citation. Keep a log of your prompts in your research notebook.
Sharpening the language of an abstract you wrote yourself, but only for grammar and syntax, without modifying or adding to the substance. Must be credited.
Writing initial code, with explicit citation of which portions were AI generated plus a prompt log.
Identifying appropriate statistical tests or software tools, with a prompt log. You must interpret the data yourself.
Producing a flowchart or image, clearly marked as AI generated and cited, and reverse-searched to make sure it has not incorporated copyrighted material.
Fixing the formatting of a bibliography you built yourself. You must verify every citation.
Using AI tools within your actual research project is permitted, and should be disclosed and cited. The rule is about who writes the words and who does the thinking. You will also attest in the Ethics Statement that you did not use AI tools to draft the paper or your responses, and every submission is run through plagiarism screening that authenticates essays, research reports, and recommendations.
Sample winning projects: what the 2026 top 10 actually researched
The most useful thing you can do before you enter is look at what winning work actually looks like. Note how varied these are in field, and how many are computational rather than lab-based.
Place | Category | Project |
1st | Mathematics | Wrote a program to enumerate every possible "noble polyhedron," concluding there are 146 beyond the two known infinite families |
2nd | Neuroscience | Trained AI models on retinal images from a public dataset to screen for autism and ADHD, then built a prototype tool |
3rd | Animal sciences | Tested whether a marine clam with naturally occurring blood cancer can serve as a cheaper, more ethical model for human leukemia drugs |
4th | Mathematics | Extended a 1997 result to describe a whole system of quantum particles under a magnetic field using Temperley-Lieb diagrams |
5th | Computational biology | Compressed protein structural information into numerical strings so structures can be compared across massive databases quickly |
6th | Medicine and health | Built a rotating device to simulate microgravity, then showed red light sped up tissue regeneration in injured flatworms by 95.2% |
7th | Medicine and health | Developed a cellular model of juvenile idiopathic arthritis, a disease she was diagnosed with in third grade |
8th | Neuroscience | Identified two FDA-approved lung-scarring drugs that reduced cell damage linked to the APOE4 Alzheimer's risk variant |
9th | Mathematics | Investigated a new "unrestricted finite factorization" property and how it relates to other kinds of factorization |
10th | Physics | Designed a three-network machine learning system to handle measurement error in particle collision data without losing efficiency |
Two patterns worth noticing. Four of the top 10 were pure mathematics or physics projects requiring no lab access at all, which matters if you do not have a university lab nearby. And several started from a personal or local starting point rather than a grand one, including a project built on a disease the student lives with.
You can watch each winner explain their work on the official 2026 winners page. If you are still looking for a direction, our list of 15 research project ideas for high school students is a reasonable starting point.
There are 21 categories, and your choice barely matters
You select one category when you enter, from Animal Sciences, Behavioral Sciences, Biochemistry, Bioengineering, Cellular and Molecular Biology, Chemistry, Computational Biology and Bioinformatics, Computational Humanities, Computer Science, Earth and Planetary Science, Engineering, Environmental Science, Genomics, Materials Science, Mathematics, Medicine and Health, Neuroscience, Physics, Plant Sciences, Social Sciences, and Space Science.
Your category determines only which specialists read your entry first. Scholars and finalists are selected without regard to category, and winners are not distributed proportionally across categories. Evaluators can pull in readers from other categories or reassign your project if it fits better elsewhere. So pick the closest fit and stop worrying about it.
Case study: what a mentor learned guiding an STS applicant
We spoke with Matthew Olonoff, who mentored Anna Jang through her STS entry. Anna was part of Lumiere's research fellowship, where she studied hearing loss in elementary school children and its impact on musical ability. Matthew was pursuing a PhD in bio-psychology at Northwestern at the time and worked with Anna over several months. This interview dates from an earlier cycle, so treat the timelines as illustrative rather than as current-cycle guidance.
How long did the project take? Six months from start to finish, and it would have been much better with another two. That included settling on a research question, preparing IRB materials, running the study, and writing.
What was the main challenge? The IRB, because they were at the mercy of other people's schedules.
How important is mentorship? Choosing the right mentor was the most important step. Anna's study involved generating self-report survey questions and submitting human subjects IRB material. Matthew had done that process twice himself, so he knew what an IRB looks for.
Advice for students? Be prepared to do almost double the work, because you are generating IRB documents on top of running a study and writing an article. Building on an existing study makes the process considerably easier.
Advice for mentors? Be honest and realistic about the timeframe. Matthew planned one to two months to develop the project, two months for IRB prep, and two months for data collection and writing. Even if a student attaches themselves to an existing study to collect or analyse data, the existing IRB needs to know, and the mentor has to take responsibility for finalising documents.
Biggest single piece of advice? If you are committed to STS, it is vastly easier if you are already somewhat along with your research idea before you seek out mentorship. That reduces time pressure and lets you choose a mentor whose interests actually match yours.
(Parts of this interview have been paraphrased and edited for brevity.)
How to win Regeneron STS in 2027
STS rewards a coherent application where the parts reinforce each other. Five things we would prioritise:
Get a mentor, and get the right one. STS asks for scientific rigour and genuine reflection in roughly equal measure. A mentor who is an expert in your specific area adds far more than a generalist, and as Anna's case shows, prior IRB experience can be the difference between finishing and not.
Plan for eight months, not four. Underestimating the length of this application is the most common mistake we see. The report alone is substantial, and the paperwork is a project of its own. Note also that some students complete their research in as little as six weeks, so a short timeline is not disqualifying, but a rushed application is.
Do not treat the non-research components as filler. Recommendations, essays, and your activities list all feed the judging criteria. Set aside at least two months for them. Your recommenders need that time regardless.
Be curious beyond your project. Finals Week includes panel judging on the breadth of your general scientific knowledge, which you cannot cram for. If your research is on carcinogens, know something about the history of carcinogen research and cancer treatment. This is a long-term habit rather than a preparation task.
Remember it is not a science fair. If you approach STS like ISEF, you will over-index on the project and under-prepare the essays, the disclosures, and the recommender selection. Approach it like a holistic application and you will service every part of the rubric.
Common reasons entries fail to qualify
The rules book devotes an entire appendix to this, and the recurring themes are worth knowing before you start rather than after:
Paperwork problems, especially IRB approval obtained after research began, or an improperly constituted IRB.
Uncited graphics, including your own.
Fabricated or hallucinated references.
Team research submitted as individual work, or a split team project.
Undisclosed paid program participation, mentorship payments, or family relationships.
Submitting a literature review, a proposal, or a project without results.
A report that pushes past 20 pages, or shrinks font and margins to fit.
Incomplete paperwork remains one of the most common disqualification routes, and it is entirely avoidable.
Frequently asked questions
When is the Regeneron STS 2027 application due? The deadline is Thursday, November 5, 2026 at 8:00 pm Eastern Time. All three recommendations are due at exactly the same time. The customer support deadline is 24 hours earlier, on November 4 at 8:00 pm ET.
How many people apply to Regeneron STS each year? Over 2,600 high school seniors enter annually. The 2026 cycle drew 2,612 entrants, the largest applicant pool since 1967. From that pool, 300 became scholars and 40 became finalists.
What is the Regeneron STS acceptance rate? The Society does not publish one, but based on official 2026 figures, roughly 11.5% of entrants became scholars, roughly 1.5% became finalists, and roughly 0.4% reached the top 10. Because entry is open to any eligible senior, these are selection rates rather than admissions rates.
When do Regeneron STS results come out? For the 2027 cycle, the top 300 scholars are announced on January 7, 2027, and the top 40 finalists on January 21, 2027. Finals Week runs March 11 to 17, 2027, with the top 10 winners announced on March 16.
Is Regeneron STS only for US students? No. Any student of any citizenship who lives in and attends their final year of secondary school in the US or its territories can apply. US citizens attending school abroad may also qualify under specific conditions. However, non-citizens attending American schools abroad are not eligible.
How long should the abstract be, and does it count toward the page limit? Your abstract goes on the second page of the report, and it does not count toward the 20-page limit, nor do the title page or bibliography. Appendices do count. The rules do not set a fixed abstract length, so follow the convention in your discipline.
How many images can the research report include, and do they need citations? There is no cap on images, but every figure, table, graph, chart, photograph, and logo must be cited in full directly under or beside it. That includes graphics you created yourself, where you must state that you made it, name any software you used, and give the year. A reference list for graphics is not accepted, and failure to cite an image can result in disqualification.
Do I need to submit test scores? No. Test scores are optional and choosing not to submit them will not be held against you. If you do submit, you can enter SAT, ACT, AP, and IB scores.
Can I enter Regeneron STS twice? No. Students who have entered any previous Science Talent Search may not enter again, and you may submit only one project per cycle. STS is also seniors only, so for most students there is exactly one chance.
Do I need a research program or a fancy lab to be competitive? No. The Society states plainly that it does not believe a student must participate in a research program to succeed, and evaluators are instructed to weigh your achievements against the lab access and resources you actually had. Four of the 2026 top 10 projects were mathematics or physics work requiring no lab at all. That said, mentorship genuinely helps, particularly with research design and IRB paperwork. If you want structured support from a PhD mentor while keeping the work unambiguously your own, you can apply to the Lumiere Research Scholar Program here.
Build the research project your STS application depends on
An STS application is only as strong as the independent research underneath it, and that research takes months. If you want a PhD mentor to help you shape a research question, design a study, navigate IRB paperwork, and write a paper that stands up to review by three doctoral scientists, apply to the Lumiere Research Scholar Program. Last year over 2,100 students applied for roughly 500 spots.
If cost is a barrier, the Lumiere Research Inclusion Foundation offers full financial aid to students who qualify. You can also review our admission results here.
Sources: Regeneron STS 2027 Rules and Entry Instructions, Society for Science application requirements, official STS FAQ, and the 2026 STS winners page. Verified August 2026 for the 2027 application cycle.
Stephen Turban is the co-founder of Lumiere Education and a graduate of Harvard College, where he earned an A.B. in Statistics before beginning a PhD at Harvard Business School. A former McKinsey & Company consultant and former researcher, his research has been covered by the Harvard Business Review, The Economist, BBC News, Freakonomics Radio, and the Washington Post. In his articles for Forbes, he writes on what students and professionals can learn from high school research, high school passion projects, critical thinking, and mentorship. At Lumiere, he leads a selective research program where students work one-on-one with a research mentor to develop an independent research paper. Connect with him on LinkedIn.