Study methods (8)
- Active Recall#
- Deliberately retrieving information from memory, rather than re-reading or highlighting.Active recall is the practice of testing yourself on material instead of passively reviewing it. Research (Karpicke & Roediger 2008) consistently shows that active recall produces 50-200% better long-term retention than re-reading. Methods include: self-quizzing, practice problems, teaching the material aloud (Feynman technique), and fill-in-the-blank flashcards. Every BackRow quiz generator implements active recall by design.
- Related: Spaced Repetition, Feynman Technique, Retrieval Practice
- Cornell Notes#
- A note-taking format with a summary column, note column, and bottom-of-page synthesis.Cornell Notes is a structured note format invented at Cornell University in the 1950s. The page is divided into: left column (cues/questions), right column (lecture notes), and a bottom summary. Promotes active review because the cue column becomes a self-quiz tool. BackRow's AI Notes format auto-generates notes in a Cornell-compatible structure when requested.
- Feynman Technique#
- Learning by explaining a concept in simple terms, as if teaching a beginner.The Feynman technique, named after physicist Richard Feynman, is a study method: (1) choose a concept, (2) explain it in plain language as if to a 12-year-old, (3) identify gaps in your explanation, (4) revise and simplify. It forces you to surface the concepts you actually understand vs. ones you only think you do. BackRow's AI Tutor implements a Feynman mode where you explain a topic aloud and it identifies unclear spots.
- Related: Active Recall, study-method
- Forgetting Curve#
- Ebbinghaus's finding that memory retention decays exponentially without reinforcement.The forgetting curve describes how memory of new information decays exponentially over time without review — 50% lost within the first hour, ~75% within a day. The curve flattens with each round of spaced review. This is why last-minute cramming fails and spaced repetition works: each review resets the decay curve.
- Interleaving#
- Alternating between multiple topics during study, rather than blocking one at a time.Interleaving is the practice of mixing topics during a study session instead of massing practice on one topic. Though it feels harder in the moment, research (Rohrer 2012) shows it produces 40-80% better retention and transfer. Example: alternating 5 calculus problems, 5 statistics problems, 5 linear-algebra problems instead of doing 15 calculus problems in a row.
- Pomodoro Technique#
- A time-management method: work 25 minutes, break 5 minutes, repeat.The Pomodoro Technique, developed by Francesco Cirillo in the late 1980s, alternates focused work sprints (25 minutes) with short breaks (5 minutes), with a longer break every 4 cycles. Named after the tomato-shaped timer Cirillo used. Effective for managing attention fatigue and preventing study-session burnout.
- Retrieval Practice#
- The act of pulling information from memory (e.g., via flashcards or quizzes) to strengthen learning.Retrieval practice is a cognitive-science term for any activity that requires you to actively pull information from memory. It strengthens memory traces and improves long-term retention far more than re-reading. Quizzing, flashcards, and "blank sheet" reviews are all forms of retrieval practice. This is the single most effective study method in the cognitive-science literature.
- Related: Active Recall, Spaced Repetition
- Spaced Repetition#
- Reviewing information at increasing intervals to maximize long-term retention.Spaced repetition is a learning technique where information is reviewed at increasing intervals — 1 day, 3 days, 7 days, 14 days, etc. — based on how well you recall it. Pioneered by Hermann Ebbinghaus (1885) and formalized as the SM-2 algorithm by Piotr Woźniak (1987). Apps like Anki, Quizlet, and BackRow use spaced repetition to surface cards just before you would otherwise forget them, cutting review time by ~60% vs massed practice.
- Related: Active Recall, Forgetting Curve, flashcard
AI & tooling (9)
- AI Detection#
- Software that attempts to identify whether text was generated by an AI.AI-detection tools (Turnitin AI, GPTZero, Copyleaks) analyze text for statistical patterns typical of LLM output: low perplexity, uniform sentence structure, and common AI phrase patterns. Accuracy is imperfect — false positive rates of 10-30% are common, penalizing students whose genuine writing resembles AI patterns. BackRow's Essay Humanizer rewrites AI-assisted drafts to read in the student's own voice, vocabulary, and style while preserving meaning, for use on assignments where AI is permitted.
- AI Hallucination#
- When an AI generates content that sounds plausible but is factually wrong or fabricated.An AI hallucination is when a language model generates output that sounds confident and coherent but is factually incorrect, fabricated, or nonsensical. Common in math (wrong intermediate steps), citations (fake papers), and dates. Mitigated by: RAG (grounding in real documents), verification against structured databases, using symbolic computation for math, and hybrid systems. BackRow mitigates hallucination by routing math queries to real symbolic solvers, not LLM generation.
- API Key#
- A secret token that authenticates programmatic requests to an API.An API key is a secret token an application uses to authenticate requests to an API. BackRow keys are shaped bk_live_<prefix>_<secret>, are scope-gated (each key grants only the permissions you choose), and are created in the developer console at backrow.ai/developers on Pro and Max plans. The secret is hashed before storage and sent on each request as an Authorization: Bearer header.
- Related: REST API, MCP (Model Context Protocol)
- LLM (Large Language Model)#
- An AI model trained on large amounts of text to generate and understand human language.A Large Language Model (LLM) is a neural network trained on billions of words of text, learning statistical patterns that let it generate coherent human-like language. LLMs power AI chatbots, code completion tools, translation services, and study assistants like BackRow's AI Tutor. BackRow routes student queries across multiple LLMs intelligently — task-tuned models for study chats, research, and general queries, picked automatically per request.
- MCP (Model Context Protocol)#
- An open standard that lets AI agents securely use an app's tools and data.The Model Context Protocol (MCP) is an open standard for connecting AI agents to external tools and data sources through a uniform interface. Instead of bespoke integrations, an MCP-capable agent can discover and call an app's capabilities directly. BackRow runs an MCP server on Pro and Max plans, so an agent — with the user's API key — can record, transcribe, generate study materials, and read knowledge-state on the user's behalf, making study and meeting workflows automatable.
- Related: API Key, REST API, LLM (Large Language Model)
- RAG (Retrieval-Augmented Generation)#
- A method where an AI retrieves relevant documents before generating an answer, improving accuracy.Retrieval-Augmented Generation (RAG) is a technique where an LLM first retrieves relevant documents from a database or document store, then uses those documents as context for its answer. This drastically reduces hallucination because the model grounds its response in real sources. BackRow uses RAG to answer questions from your lecture recordings and uploaded materials — answers are sourced from YOUR content, not the model's training data.
- REST API#
- An HTTP interface that lets external software call an app's features programmatically.A REST API exposes an application's features over HTTP so other software can call them programmatically. BackRow's REST API (Pro and Max plans) covers transcription, notes and AI transformations, flashcards with SM-2 scheduling, knowledge-state, the AI tutor, and server-side engine tools — authenticated with scoped bk_live keys and rate-limited at 60 requests/minute. It is what lets developers build automations and integrations on top of BackRow.
- Related: API Key, MCP (Model Context Protocol)
- Speaker Isolation#
- Identifying and labeling who is speaking in an audio recording with multiple voices.Speaker isolation (also called speaker diarization) is the process of automatically identifying different speakers in an audio recording and labeling their utterances. Critical for lecture recording where you want to distinguish the professor from student questions. BackRow separates speakers on every plan — the free tier keeps two voices apart so a question from the room never reads as part of the lecture — and paid plans LABEL them, so your notes separate "Professor:" from "Student Q:" by name, for up to 4, 8, 20 or 32 voices depending on plan.
- Transcription#
- The process of converting spoken audio into written text.Transcription converts spoken audio to text. Modern AI transcription services use speech recognition models to produce text with punctuation, speaker labels, and timestamps. BackRow uses industry-leading AI transcription with automatic fallback for reliability, achieving 95%+ accuracy on standard classroom audio. Premium tier uses an upgraded engine for highest accuracy.
Academic frameworks (10)
- APA Citation#
- The American Psychological Association's format for academic citations, used in social sciences.APA (American Psychological Association) citation format is the standard in psychology, education, and social sciences. 7th edition (2020) requires author-date in-text citations and a References list. Example: (Smith, 2023) in-text; Smith, J. (2023). Title. Publisher. in References. BackRow's Citation Generator formats APA 7 correctly for books, articles, websites, and more.
- Bluebook Citation#
- The citation format for American legal writing, used in law school and courts.Bluebook (The Bluebook: A Uniform System of Citation) is the standard citation format for U.S. legal writing. 21st edition (2020). Used in law reviews, court filings, and law-school work. Complex — handles cases, statutes, regulations, law reviews, secondary sources, all with specific formatting. BackRow's Citation Generator supports Bluebook 21 for all source types.
- Chicago Citation#
- The Chicago Manual of Style's citation format, used in history, humanities, and some social sciences.Chicago citation format has two styles: Notes-Bibliography (footnotes + bibliography) for humanities, and Author-Date (parenthetical) for sciences. 17th edition (2017). Widely used in history and publishing. BackRow's Citation Generator supports both Chicago styles.
- IRAC Method#
- Law-school framework: Issue, Rule, Analysis, Conclusion — used to structure case briefs.IRAC is the standard framework law students use to brief cases and structure essay answers on law school exams: Issue (what legal question is presented?), Rule (what law applies?), Analysis (apply the rule to the facts), Conclusion (answer the issue). BackRow's Case Brief Generator outputs briefs in IRAC format automatically.
- MCAT#
- Medical College Admission Test — a standardized exam required for US medical school admission.The MCAT (Medical College Admission Test) is the standardized exam required for admission to U.S. and Canadian medical schools. 7.5 hours, 230 questions covering Biology, Chemistry, Physics, Biochemistry, Psychology, and Sociology. Widely considered one of the hardest standardized tests. BackRow generates MCAT-style practice questions from any Kaplan/Princeton/UWorld material students upload.
- MLA Citation#
- The Modern Language Association's format for academic citations, used in humanities.MLA (Modern Language Association) citation format is standard in English, literature, and humanities. 9th edition (2021) uses parenthetical author-page citations and a Works Cited list. Example: (Smith 23) in-text; Smith, John. Title. Publisher, 2023. in Works Cited. BackRow's Citation Generator supports MLA 9.
- NPV (Net Present Value)#
- The present-day value of a series of future cash flows, discounted to account for time value of money.NPV (Net Present Value) is the sum of a project's future cash flows, each discounted back to present value at a chosen discount rate. NPV > 0 typically means the project adds value; NPV < 0 destroys value. Standard tool in corporate finance, investment analysis, and MBA coursework. BackRow's NPV Calculator handles arbitrary cash flow schedules.
- SWOT Analysis#
- Strategic planning framework: Strengths, Weaknesses, Opportunities, Threats.SWOT analysis is a strategic-planning framework used in business and case study analysis. Identifies internal Strengths and Weaknesses, and external Opportunities and Threats. BackRow's SWOT Generator analyzes a company or business case and produces a four-quadrant SWOT automatically.
- USMLE Step 1#
- First of three exams required for US medical licensure, focused on basic medical sciences.USMLE Step 1 is the first of three United States Medical Licensing Examination steps, taken after the second year of medical school. Covers anatomy, biochemistry, microbiology, pathology, pharmacology, physiology, and behavioral science. Became pass/fail in 2022. BackRow generates Step-1-style vignette questions from First Aid or class notes.
- WACC#
- Weighted Average Cost of Capital — a company's blended cost of debt and equity financing.WACC (Weighted Average Cost of Capital) is a company's cost of capital, weighted by the proportion of debt and equity in its capital structure. Formula: WACC = (E/V × Re) + (D/V × Rd × (1−Tc)). Used as the discount rate for DCF valuations. Standard in MBA finance coursework. BackRow's WACC Calculator handles the full formula including tax-adjusted cost of debt.
BackRow specific (5)
- AI Engine (BackRow)#
- A subject-specific collection of tools in BackRow — e.g., Math Engine, Chemistry Engine.In BackRow, an "engine" is a subject-specific collection of 7-24 specialized tools. BackRow has seven engines: Math (23 tools), Computer Science (19), Chemistry (14), Physics (15), Business (17), Language (24), and Creative (7). Each engine uses purpose-built tools (symbolic computation for math, real syntax parsers for CS, actual periodic-table data for chem) rather than relying on LLM generation.
- Related: LLM (Large Language Model)
- AI Tutor (BackRow)#
- BackRow's interactive AI chat, tuned for academic explanations.BackRow's AI Tutor is an interactive chat interface powered by frontier AI models with deep-reasoning mode for complex problems. Unlike general-purpose chatbots, it's tuned for academic tasks: step-by-step explanations, worked problem walkthroughs, Feynman-style teaching mode, and citation-backed research. Part of the Scholar plan and above.
- Related: LLM (Large Language Model), Feynman Technique
- Credits (BackRow)#
- BackRow's usage system — each AI query costs credits that refresh monthly. Free tier resets on the 1st with no rollover; paid plans roll unused credits forward.Credits are BackRow's usage-metering system. Free users get 300 credits per month (no rollover, resets on the 1st of each UTC month). Paid plans get a larger monthly allowance (Starter: 600, Scholar: 1,500, Pro: 3,000, Max: 8,500) and unused credits roll over month-to-month. Most quick actions cost 5–10 credits; standard generations (flashcards, quizzes, mind maps) cost 10–25 credits; premium features like the essay humanizer (50), cinematic video (300–3,000), and recording with transcription (100 credits/hour) cost more.
- Learn Hub (BackRow)#
- BackRow's mastery-tracking feature that identifies knowledge gaps and suggests review.The Learn Hub is BackRow's progress-tracking interface. It analyzes quiz scores, flashcard accuracy, and time-spent per topic to identify weak areas and recommend targeted review. Maps closely to the cognitive-science concept of "desirable difficulty" — surfaces topics you're close to forgetting, spaces out topics you know well. Unique to BackRow among study apps.
- Related: Spaced Repetition, Active Recall
- Rowly (BackRow)#
- BackRow's AI Tutor mascot and default conversational persona.Rowly is BackRow's AI Tutor mascot — a friendly, chill conversational persona that handles the study-chat experience. Technically, Rowly is a prompt template running on top of BackRow's AI engine. Stylistically, Rowly doesn't use "as a large language model" preamble, doesn't over-hedge, and treats users like adults. Named as a playful twist on "back row" plus the "-ly" suffix common in AI assistant names.
- Related: AI Tutor (BackRow)
Want to use these concepts?
BackRow implements every technique in this glossary (spaced repetition, active recall, Feynman technique, and more) automatically.
Try BackRow free →