Friday Flux 2: Reading Better, Thinking Better

FRIDAY FLUX

Knowledge over noise. Evidence over ego.

Learn how to read scientific sources
Build real research‑literacy skills
Understand what the evidence actually says

Introduction — Why Reading Research Needs a System

Every week in Friday Flux, we explore how to think more scientifically about self-improvement – not just what to change, but how to learn from evidence without drowning in it.

If you’ve ever opened a research paper and felt your eyes glaze over halfway through the introduction section, you’re not alone. Maybe you don’t even know where to find a research paper (we’ll talk about this next week). Academic writing is dense by design: it’s optimized for precision, not readability. But that density makes it hard for students, creators, and practitioners to extract the insights that actually matter.

That’s why this week is all about building a repeatable system for reading research – one that helps you move from “I don’t get this” to “I can explain this to someone else.”

Enter: S. Keshav’s Three-Pass Method, adapted for real-world learners. It’s not speed-reading. It’s strategic comprehension – a way to layer your understanding so you see the skeleton of a paper before you dive into its organs.

And this week, we’re putting it to work on three habit-science sources that pair perfectly with the next chapters of Atomic Habits.

This Week’s Sources

To ground the Three-Pass Method in something concrete, I’m working through three papers that each illuminate a different angle of habit formation:

1. Lally et al. (2010) – How Are Habits Formed?

The foundational real-world study that modeled daily automaticity over 84 days. This is the source of the “66-day median” – and the much more important insight that habit formation varies dramatically across individuals and degree of complexity of the habit.

“Automaticity increases following an asymptotic curve…”

(Lally et al., 2010)

2. Carden & Wood (2018) – Habit Formation and Change

A sweeping review of how habits form, how they resist change, and how environmental cues shape behavior more than willpower ever could.

“Context cues automatically bring habitual responses to mind…”

(Carden & Wood, 2018)

3. Lally & Gardner (2012) – Making Health Habitual

A practical, clinical perspective on habit formation: how to help people build stable, context-triggered behaviors in everyday life.

“Repeat a chosen behaviour in the same context, until it becomes automatic and effortless.”

(Lally & Gardner, 2012)

Together, these papers give us a spectrum: mechanism → environment → application.

The Next Four Chapters of Atomic Habits

This week’s reading in Atomic Habits covers the “Cues” portion of the book.

These chapters map surprisingly well onto the research above – though in simplified form. Friday Flux introduces self-help and pop-science claims and puts them next to peer-reviewed evidence and seeing where they converge, diverge, or oversimplify.

The Three-Pass Method is how we bridge the two.

How the Three‑Pass Method Works

To make this system easier to apply, I built a visual worksheet that breaks down Keshav’s original framework in a step-by-step process. Each pass builds on the last – from quick orientation to deep reconstruction – so you can move from skimming for relevance to thinking like the author.

It’s designed for anyone who wants to read smarter, not necessarily faster.

Three-Pass Method Worksheet

Pass 1 - The "Should I Read This?" Pass

Goal: Get the big picture in 5-10 minutes.

1. Skim the structure

Title:
Authors & Affiliations:
Year:
Journal:

2. Read only these sections

Read these sections, but don’t get caught up in the sticky details:

Abstract → What is the paper about?
Introduction → What problem are they trying to solve?
Section headings → What’s the roadmap?
Conclusion → What did they claim?

3. Answer these questions

You should be able to answer what Keshav refers to as the five C’s after completing your first pass:

Category → What type of article is this?
Context → What relevant papers and theories are referenced?
Correctness → Do the assumptions seem valid?
Contributions → What is the main contribution to science?
Clarity → How well is the paper written?

4. Should I do a second pass?

Pass 2 - The "Understand the Methods" Pass

Goal: Understand the logic, not the math. ~1 hour.

1. Read more carefully

Focus on:

Figures and tables
Methods overview
Key arguments
Definitions of terms
Limitations section

2. Fill in these prompts

What problem is the paper solving?
What approach/method did they use?
What assumptions are they making?
What are the key results? (Bullet points)
Do the results logically follow from the methods?
- ☐ Yes
- ☐ No

- ☐ Unsure
What questions do I still have?

3. Third pass?

If the paper does not directly answer your research question, or if you still don’t understand the article, there are three ways to proceed as specified by Keshav:

If it truly seems irrelevant, set it aside.
Return to the paper later, and consult background information as necessary.
Chug along and move onto the third pass.

Pass 3 - The "Rebuild It Yourself" Pass

Goal: Be able to reconstruct the paper from memory. 1–4 hours.

1. Rebuild the argument

Try to restate the paper’s logic without looking:

If I had to explain this paper to someone else, how would I do it?
What is the core mechanism or insight?

2. Challenge every assumption

Which assumptions are weak?
What alternative explanations exist?
What would break the authors’ argument?

3. Compare your reconstruction to the real paper

Where does my version differ from theirs?
Do those differences reveal hidden weaknesses?

4. Identify missing pieces

Missing citations?
Unaddressed limitations?
Unjustified methodological choices?

5. Generate future work

What follow‑up experiments or analyses would I propose?
How could I extend or apply this work?

Example: The Three‑Pass Method Applied to Lally et al. (2010)

A real demonstration using one of this week’s papers.

Pass 1 — Big Picture (5–10 minutes)

1. Skim the structure

Title: How are habits formed: Modelling habit formation in the real world
Authors & Affiliations: Phillippa Lally, et al., University College London
Year: 2010
Journal: European Journal of Social Psychology

2. Read only these sections

Read these sections, but don’t get caught up in the sticky details:

Abstract → What is the paper about?
- Measuring daily automaticity using SRHI over 12 weeks to evaluate the habit formation process.
Introduction → What problem are they trying to solve?
- Trying to fill a gap in literature to explore the process of habit formation on an individual level by modeling the relationship between habit behavior repetition and automaticity in a real-world setting.
Section headings → What’s the roadmap?
- Abstract
- Introduction/The Present Research
- Method
  - Participants
  - Procedures
  - Measures
  - Analysis
- Results
  - Does an Asymptotic Curve Model the Relationship Between Repetition and Automaticity?
  - How Many Repetitions are Needed to Reach a Plateau of Automaticity?
  - Do Time to Asymptote or Asymptote Value Vary by Behaviour Type?
  - Does Missing an Opportunity Compromise Habit Development?
  - A Validity Check
- Discussion
Conclusion → What did they claim?
- No external rewards are necessary for habit development, but internal rewards may help.
- A nonlinear, asymptotic model that reaches an eventual plateau is the best fit for the measured daily automaticity.
- The time to reach the plateau varies widely.
- Continued support through interventions may be necessary to form habits for some individuals.
- A missed day of performing a habit does not contribute to your overall success, but there is likely a threshold beyond which you are less likely to succeed.

3. Answer these questions

You should be able to answer what Keshav refers to as the five C’s after completing your first pass:

Category → What type of article is this?

This is an experimental (empirical) study.

Context → What relevant papers and theories are referenced?

SRHI questionnaire for automaticity measurements
Cited automaticity studies: Verplanken, 2006 and Wood & Neal, 2007
Hull’s work showing an asymptotic fit between repetition and habit strength
Situations are better cues than time: McDaniel & Einsteint, 2000
Methods of quantifying habit strength
- Response frequency measure (RFM)
- Multiplying ratings of frequency by consistency of context
- Self-report habit index (SRHI)
Identity may not be fundamental to habit formation: Verplanken, Myrbakk, & Rudi, 2005
Complex behaviors have a lower automaticity plateau than simple behaviors: Verplanken, 2006
types of automaticity and goal-directed automaticity for complex habits: Wood & Neal, 2007

Correctness → Do the assumptions seem valid?

While I can’t say that their assumptions are valid because I don’t know enough about the statistical methods they used yet, they are very honest about the limitations of the study. This builds credibility and makes me believe the conclusions drawn may be valid. There is no circular logic or obvious falsification in their conclusions.

Contributions → What is the main contribution to science?

This provides empirical data for statistical analysis to reveal trends in habit formation. Importantly, it measures habit formation through automaticity on an individual and daily basis in a real-world setting.

Clarity → How well is the paper written?

The paper was well-written but methods heavy. However, because they were introducing a new statistical framework for evaluating habit formation, this was a good choice for building credibility.

4. Should I do a second pass?

Is it interesting?

Yes, this article is well-written, heavily cited, and presents novel information.

Do I understand the subject background?

Yes, I’m familiar with the SRHI questionaire, automaticity, and know what an asymptote is.

Is it relevant to my research question?

Yes, it applies to my research question for this week: What role do cues play in habit formation?

Does it have invalid assumptions?

No, there are no obvious invalid assumptions and limitations are well-presented.

Verdict: Yes, I should move on to do a second pass.

Pass 2 — Methods & Logic (1 hour)

1. Read more carefully

Focus on:

Figures and tables: I looked at the figures and tables. For the figures I checked the titles, axes, caption, error bars, legend, and general trend of the data. For the tables, I read all the column headers, making sure I understood the measurement and the units associated with it.

Methods overview: The methods section was dense, but well-organized, and it revealed some limitations and introduced new analysis techniques I had not seen before. This includes Mitscherlich’s law of diminishing returns, which allowed researchers to model the automaticity asymptote curve over time.

Key arguments: I looked in the abstract, results, and discussion sections to find key claims.

Definitions of terms:

asymptote: a straight line associated with a curve, where the distance from the line and the slope of the curve approach the line as the curve goes to infinity (Merriam Webster).

SRHI: A self-report questionnaire containing 12 items, measured on a response scale with 5 or more response categories measuring the repetition history, automaticity, and expressing identity associated with a habitual behavior (Verplanken & Orbell, 2003).

automaticity: a behavior characterized by: “efficiency, lack of awareness, unintentionality, and uncontrollability” (Tally et al., 2010).

idiographic approach: relating to or dealing with something concrete, individual, or unique (Merriam Webster)

intrinsic: belonging to the essential nature or constitution of a thing (Merriam Webster)

extrinsic: not forming part of or belonging to a thing (Merriam Webster)

salient: standing out conspicuously (Merriam Webster)

Likert scale: a rating system used in questionnaires that measures an individual’s attitude, opinion, or perceptions on a degree of agreement scale (Britannica).

quartile: any of the three values that divide the items of a frequency distribution into four classes with each containing one fourth of the total population (Merriam Webster).

ceiling effects: a statistical error when the scale exceeds the range of the questionnaire, so most participants score near or beyond the maximum (Research Prospect).

Limitations section: There was an extensive limitations section including small sample size, missed days in completing the SRHI, self-report bias, and potential ceiling effects (Tally et al., 2010).

2. Fill in these prompts

What problem is the paper solving?

The paper is filling the gap in literature for real-world analysis of habit formation on an individual basis, using automaticity to quantify the results.

What approach/method did they use?

They used the SRHI to measure automaticity with a 5 option Likert scale. The data was then modelled using the law of diminishing returns.

What assumptions are they making?

They are assuming that the SRHI is an accurate way to measure automaticity, that identity is not necessary for habit formation, the asymptotic fit is better than other nonlinear regression methods, and that this model applies to the general population (not just college students).

What are the key results? (Bullet points)
The asymptotic curve model for measuring the relationship between repetition and automaticity was a better fit than the linear model, producing a good fit for 39 of the 82 participants that produced sufficient logged data.
The number of repetitions to reach a plateau of automaticity ranged from 18 to 254 days with a median time of 66 days to reach 95% of the asymptote.
Three behavior types were chosen: eating, drinking, and exercise behaviours; There was no significant difference in time to reach the 95% of the asymptote.
- Percent compliance was significantly higher for the water drinking group than the eating and exercise groups.
Missing an opportunity to complete the habit behavior does not result in long-term negative effects, but it did result in a non-significantly lower SRHI score the following day.
A validity check was completed by comparing the number of repetitions of the individual’s chosen behavior by the curve parameters used to model the asymptotic fit → passed the test.
Do the results logically follow from the methods?
- ✓ Yes
- ☐ No

- ☐ Unsure
What questions do I still have?

3. Third pass?

While this paper contains a lot of relevant information, it does not directly answer my research question. However, the background information and references make it very useful in the research this week.

Verdict: No third pass right now, but it will be set aside for later when the topic is more relevant to the specific research question.

Teaser for Next Week’s Friday Flux

Next week, we’re zooming out from how to read scientific papers to something even more foundational: how to find them in the first place.

Most people rely on whatever floats across their social feeds or whatever Google Scholar spits out on page one — but that’s not how researchers do it. There’s an entire ecosystem of databases, search strategies, filters, and citation trails that make the difference between “I found a paper” and “I found the right paper.”

So next Friday, we’ll cover:

where scientists actually search for research
how to tell if a source is credible before you even open it
how to use citation networks to find the “parent” and “child” papers
how to avoid paywalls (legally)
how to build a personal research pipeline you can reuse every week

If this week was about reading smarter, next week is about searching smarter — and together, they’ll form the backbone of your scientific literacy toolkit.

Closing Thoughts

Reading research doesn’t have to feel like wading through wet cement. With a system like the Three‑Pass Method, you can move from confusion to clarity in a way that’s repeatable, sustainable, and actually enjoyable. This week was about building that foundation — learning how to approach a paper so you can extract the signal without getting lost in the noise.

Thanks for reading this week’s Flux. Keep thinking in systems, keep reading with intention, and I’ll see you next Friday — where we’ll take the next step and learn how to actually find the scientific articles worth reading.

Check Out My Recent Posts

Friday Flux

Catch up on the Latest Friday Flux

Friday Flux