A Book Club Guide to
The Psychology of Software Teams
Book Synopsis
To build the future, we need new ways of supporting software teams. This book will give you a secret weapon: the psychology that creates resilience for developers, sustainable practices for software teams, and innovation for organizations. You’ll learn from rigorous empirical evidence gathered from top engineering organizations and thousands of developers around the world, revealing powerful principles software teams can use to guard against failure and drive cultures of collaboration and problem-solving.
Making incredible software doesn’t have to be a death march. This book presents a humane alternative for software teams looking to use the transformative power of behavioral science to understand what drives technology businesses forward.
This book is for the developers and builders of the future. Bringing science and heart together, The Psychology of Software Teams will teach you how to untangle your thinking from pervasive myths about software work and harness the superpowers of psychology to create more joyful, innovative, and thriving environments for software.
- The Brains and Hearts That Build the World
- Breaking Up with Brains-in-Jars
- More Bands and Fewer Rockstars
- The Performance Paradox
- Conflict or Coalitions?
- Becoming an Organization That Wants to Understand Itself
- Fighting Dirty for Good Culture
How to Use This Guide
This guide was made to help any individual or group that wants to dive into the book together and connect its research and ideas to their own lives. You are free to take, share, use and remix this guide and its content in any way that helps you. Each chapter section includes discussion questions, and one or more “try it” prompts you can use to apply a concept to your own experiences and teams, past or present.
You don’t need any particular background to lead or join this discussion, nor do you need to have a certain role. The book (and this guide) are written for anyone who works with, manages, or cares about the people who build software and the psychological experience of technical communities.
However, the book’s ideas can play out differently or feel more relevant depending on scale and the context of your guide use and group. Some things are within a team’s own reach (such as norms, habits, how a manager runs a retro); others require company-level authority, budget, or policy to shift (such as hiring rubrics, promotion criteria, org-wide metrics). Some questions ask you to explicitly split these apart. If you are using the guide for individual reflection, feel free to re-draft items to something more relevant to you, or focus primarily on the reflection and try-it exercises. If you are leading a group where you are acting as a facilitator, you may wish to steer questions more towards the individual or company level depending on your group’s context, and focus on the try-it exercises most relevant to your group's challenges.
Consider assigning one chapter per meeting, or grouping chapters into two or three longer sessions if your club prefers fewer, deeper conversations. Or, use the “deep dive” suggestions to focus on a few thematic sections instead. It's not necessary to work exhaustively through this entire guide. Treat this as a pool of possible questions to discuss!

If your group is short on time, or you’re running this in a workplace format like a lunch-and-learn or a single afternoon workshop, you can use one of these themes for a whole discussion rather than working chapter by chapter. Pick the theme that best matches what your team is currently wrestling with. Themes give you a broader, whole-book conversation compared to a chapter walk-through, and may suit if you can only get people together once.
Key Themes to Explore
Before diving into individual chapters, it can help to name some of the major themes across the book. Here are a few suggestions for the group to discuss:
Software work often gets treated as an isolated, individual, purely cognitive act, as though developers are disembodied, fungible floating brains rather than whole people embedded in teams, histories, and psychological environments. POST argues this framing can actively damage teams, but that it’s sticky because it gets us some short-term outcomes. It is reinforced by common stereotypes about what drives technical work.
Rather than asking just “how do we make developers produce more,” POST reframes a central question for software as “what makes developers resilient, engaged, and able to do good problem-solving over time.” The book explores how these two frames lead to different, sometimes conflicting, answers, and makes the case that lasting technology depends on high-quality problem-solving.
POST repeatedly asks readers to notice when a belief about software teams (the lone genius, believing that technical work operates as a meritocracy, us vs. them stories) is really an untested folk theory, and what it would take to actually test it. Nuances and gaps in our evidence are to be acknowledged honestly, and POST argues that technical communities and scientists need to work together to build understanding past those gaps.
Psychological safety, learning environments, and collaborative groups that can solve conflict aren’t perks, but affordances that determine whether teams succeed or fail, especially under pressure.
Chapter-by-Chapter
Discussion Questions
The Brains and Hearts That Build the World
This opening chapter makes the introductory case that software is built by whole humans, with histories, emotions, and social needs, and that technical people do vital work in environments that often fail to listen to and understand them.
- Watch: The Psychology of Technologists. Cat’s posit::conf(2025) keynote (1:02:43) is an overarching exploration of a developer science research agenda in one talk. Watch to learn about our mental models for the environments around us, hear a live version of some of the stories in the book, and hear background details about some of the studies in the book.
- Read: Hicks, Lee, & Ramsey (2024), Developer Thriving, a short-form, practical IEEE Software paper on the four sociocognitive factors of the LABS model for resilient teams. Read to learn about the methodology details of the study and use the specific measures in the study.

If your group has both managers and developers, the “visible and valued” question can surface disagreement between the two. While this might feel scary, it could also be tremendously valuable for your group. Spend a few minutes thinking about how you as a facilitator plan to respond if a tension emerges. It might be helpful to establish expectations (e.g., mutual listening before response). Directing attention to shared goals can go a long way toward helping people feel like different viewpoints are tolerable, and can help your group find resolution.
- What assumptions about developers do you walk in with? Where did you learn them?
- Hicks’ research has found a large discrepancy between managers’ agreement that technical work is visible and valued, and developers’ agreement with the same question. Have you seen this in your own lives?
- Team vs. company.A team can informally acknowledge the full humanity of developers by practices like making room for a bad day, modifying how a standup is run, making it easy for someone to hand off a task, or small moments of support. How do companies encode (or fail to encode) developers’ “full humanity”? For instance, are developers treated like more than “Brains-in-Jars” in PTO policy, performance reviews, and HR systems?
- Team vs. company.Where does Brains-in-Jars treatment show up most in your own experience: in how your immediate team behaves, or in policies that come from above your team?
Think of one moment at work where your emotional state visibly shaped your technical output (a good day or a bad one). What would it have looked like for your team to account for that? Make an if-then plan (“if I have a bad day, then I will…”) to take care of yourself on a bad day.
Think of a moment where you’ve seen a technical person or team successfully make their work more “visible and valued.” What barriers did they face in doing so? Is there any wisdom they would pass along to others? If you are that person, what advice would you give your younger self? To make it tangible, write down one concrete piece of advice, or even a whole letter, to your younger self.
Breaking Up with Brains-in-Jars
In Chapter 2 we dig further into the “Brains-in-Jars” metaphor and the myths it produces across the three “Traps” that describe the negative cycles running through tech communities.
- Watch: Where we’re going wrong with developer productivity. Cat at LeadDev London 2023 (25:50). Here you can see an early draft version of some of the frameworks in this book, including naming productivity myths at the heart of the Brains-in-Jars mindset.
- Read: Lee, Hicks, & Foster-Marks (2024), “My code is shit”: Negative automatic thoughts and outcomes of a behavioral experiment for code review anxiety. This field study explores Chilly Climates as they show up in the internal narrative of developers facing technical tasks.

The Chilly Climate topic can bring up experiences of exclusion tied to identity. Think carefully about whether your group is the right space to surface these, and be prepared to offer supportive facilitation.
- POST describes “Brittle Productivity,” “Lone Genius,” and “Chilly Climates” as three different traps driving the Brains-in-Jars experience for software teams. Did one of these resonate most with you?
- Who benefits from the belief that great developers are simply born, not made or supported?
- Hicks compares brittle productivity to black ice: invisible precisely because the surface looks fine. What’s a moment you can recall where a team’s success masked mounting stress fractures underneath? What made those fractures hard to see at the time?
- The Murphy-Hill et al. research found that developers who read as different from the “default” stereotype of a developer (by age, gender, or race) get systematically more pushback when they try to make technical contributions. That means something as ordinary as a code review comment can be a site where chilly climate gets reinforced. Have you seen a pattern where certain people’s contributions get more scrutiny or pushback than others, even when the work itself was comparable? What made that pattern easy to miss, or hard to name, while it was happening?
Look at your team through the lens of all three traps (Brittle Productivity, Lone Genius, and Chilly Climate), then pick whichever one feels most salient right now. Name one concrete, observable sign of it (for example, a sprint that “succeeded” but left everyone wrecked; a story where one person is framed as the only source of breakthroughs; a cue about whose identity is really welcomed in the room). Then identify one small step you could take this week that would help break the thinking trap.
Look at a tech community you’re part of (an open-source project, a meetup, a Discord or forum, a professional network) through the same three traps. Pick whichever trap feels most salient. Brittle Productivity might appear in unpaid maintainers burning out while everyone praises the project’s “amazing momentum.” Lone Genius might show up in who gets thanked in the release notes versus who did the coordination and community work. Chilly Climate might be present in a community’s inside jokes, its default assumptions about who’s “actually technical,” or who stops showing up to events. Name one concrete sign of the trap, and then write down one thing you personally could do this month to interrupt it.
About the Evidence

This section is a bit different from the rest of Chapter 2. Instead of a research finding to apply, Hicks turns the lens on her own methodology, laying out exactly how she weighed and selected the evidence behind the whole book. For some groups, this can be worth its own deep dive because it is a chance to model what good “evidence citizenship” looks like, and because the same critical eye can be turned on any claim your own workplace relies on.
- Read: Hicks (2026, May 21), Software research must become more reusable in the Fight for the Human newsletter. An essay from Cat on why community-based work coupled with methodological rigor can create new insights for the technical people that software research is about.
- POST describes a knowledge gap: software research often fails to draw on rigorous theory about how our minds work, while psychology studies human behavior broadly but rarely studies developers specifically. Have you ever encountered the gap? What do you wish psychologists would study with developer populations? What do you wish software research would bring in from the behavioral sciences?
- This section describes four families of methods and some of their trade-offs: surveys (self-report and memory bias), lab experiments (artificiality), observational studies (hidden confounds), and large-scale trace data like software metrics (variables that weren’t designed to answer human questions in the first place). Which of these does your own technical community or organization lean on most heavily when it makes decisions about people? What blind spot does that create?
- Hicks flags the “WEIRD” sample problem: a long history in research of generalizing about all humans from participants who are Western, Educated, Industrialized, Rich, and Democratic. Has your team or company ever imported a “best practice” or intervention from research or a well-known book that may never have been tested on a workforce that looks like yours? What groups and experiences are particularly underrepresented in evidence about software teams?
- POST lays out three criteria in the book for evaluating evidence: (1) understand how the evidence was gathered, not just the headline finding; (2) require an argument to be supported by multiple studies and multiple kinds of evidence, not just one; (3) treat findings as something to dialogue with and compare against your own experience, not something to passively accept. Pick a “best practice” claim you’ve heard repeated at work (a productivity habit, a hiring criterion, a team ritual) and run it through these three questions. What do you actually know about it, and what are you assuming?
- The section closes with, “rigor is a form of care.” What does it mean to treat methodological rigor (sample size, replication, construct validity) as an act of care for the people a claim is about, rather than as a dry academic virtue? Does that reframe change how impatient you feel with “boring” evidentiary questions?
Pick one claim about “what works” for developers or software teams that gets repeated in your technical context without much scrutiny. Try to trace where it actually came from. Would it survive the three evidence criteria?
More Bands and Fewer Rockstars
This chapter shifts the unit of analysis from the individual “rockstar” to a collaborative unit, the “band.” In this chapter, we trace the origins of a few myths about individual differences in programming ability, review the impact of metacognitive strategies, and look at the attribution errors that push our minds to explain software in terms of individuals.
- Watch: The Innovation of Cumulative Cultures and Developer Problem-Solving. Cat at Virtual DDD (1:29:53). A dive on social learning as the alternative to the rockstar frame and how we miss the broader ecosystem of human problem-solving processes when we perform "cognitive cherrypicking."
- Read: Hicks & Hevesi (2024), A cumulative culture theory for developer problem-solving. A framework paper that describes developer problem-solving through the lens of social learning cultures, tying together cultural psychology, cognitive science, and practical experience from developer community design.
- Hicks compares software ecosystems to shared infrastructure like water or electricity: vast networks of other people’s solved problems that we all draw on. How often do you actually think of your own code as part of a shared commons, versus as your own individual output? Does that framing change anything about how you’d want to work?
- One early reader called this “the chapter that nukes the 10x developer myth from orbit.” In contrast to early claims, Hicks’ research has found that software work is highly variable, and that much of that variation is driven by organizational and team factors, not just individual factors. What are some places you've noticed your own speed or quality of work impacted by these team or organizational factors? Are there times you have seen how much work gets done become more or less predictable?
- What’s the difference between a team that has a rockstar and a team that acts like a band? Have you experienced both?
- The concept of “free revealing” describes people openly sharing novel solutions back to a community, producing benefits for the whole system. Does your technical community or organization actually reward that behavior — sharing code, documentation, or hard-won knowledge publicly or across teams — or does its incentive structure only recognize privately attributed, individually credited work? What would change if free revealing were something people got promoted for?
- Team vs. company.At the team level, the 10x myth might show up in who gets handed the hard problems, or who’s quietly excused from “boring” work like documentation or reviews. At the company level, it’s often baked into hiring bars and promotion. Which version is easier for your group to notice? Which is harder to change without going above your team?
- Team vs. company.A team can build its own informal habits for spreading credit around (imagine rotating who presents in demos or intentionally acknowledging less visible roles). A company builds the formal machinery that decides who actually gets rewarded for collaborative work (imagine promotion packets, performance ratings, bonus structures, org charts that flatten a band into a single named lead). Which of these is your group already doing well, and which would require an organizational change?
Think of a recent technical win on your team. Trace how many people’s contributions actually fed into it, even indirectly. Who tends to get the credit, and does that match the trace?
Identify a hiring or performance review practice you’ve encountered that assumes individual, fixed ability. How might it look different if it assumed ability is shaped by environment and support?
Pick a recent piece of code or a solution you’re proud of and trace your thinking’s lineage. How much of it was genuinely built from scratch, versus adapted from documentation, Stack Overflow, a coworker’s earlier PR, or an open-source library? What does that trace tell you about how “individual” your work actually was?
The Performance Paradox
Chapter 4 tackles how understanding our own motivational system, and how we respond to the social pressures around performance and disclosure, shapes performance. A quick frame for the group: expectancy-value theory says we can think about our motivation as riding on two separate ingredients: your expectancy (how confident you are that you’ll succeed) and the value you place on the work itself (how much you care about or enjoy it) or on the outcome (what it will bring to your life, and why that matters).
- Watch: Building for the new developer. Cat at LeadDev London 2024 (28:57) presenting empirical work on developer identity threat in the AI era. Watch for a behind-the-scenes take on the study featured in this chapter.
- Read: Hicks, Lee, & Foster-Marks (2025), The New Developer: AI Skill Threat, Identity Change & Developer Thriving in the Transition to AI-Assisted Software Development. How developers’ belief structures shape their psychological resilience during rapid technological change; read for the empirical deep dive, including methodological details and measures you can use.
- “Being a high performer was an identity that was important to Shaun.” At what point does an identity built around achievement markers (grades, titles, promotions) start working against the thing that made you want to achieve in the first place? Has anyone in the group noticed this happening in their own career, or in a hobby that turned into something they had to be good at?
- POST shares the goal orientation research that has found different outcomes for learners depending on whether they tend to use Mastery-Approach, Mastery-Avoidance, Performance-Approach, and Performance-Avoidance strategies. Which quadrant do you feel like you fall into more readily? Which feels more important to your company or community? Have you ever shifted your strategies?
- Hicks points out that leaders under time pressure often habitually reach for person praise about outcomes and innate talent (“you’re such a great developer”) rather than process praise that notices effort, strategy, or improvement. Think of the last piece of praise you gave or received at work. Was it about who someone is, or what someone did? Would it have landed differently if it had been the other type of praise?
- Team vs. company.What’s one concrete habit a manager or teammate could build to protect mastery-oriented goals specifically during crunch time? How might mastery-oriented language land differently depending on whether it comes from a teammate, or a leader?
Take a piece of praise you’ve recently given or received that emphasized talent or outcome (“you’re a natural at this,” “great job hitting the deadline”). Rewrite it to emphasize effort, strategy, or improvement instead.

The next Try it is more personal than a typical discussion prompt. Depending on your group’s comfort, consider inviting the group to think about it privately, then asking only whether it felt useful, not what came up.
Bring to mind a recent moment when your technical work did something negative you didn’t expect, like created a bug in production, put you through a review that went badly, or locked you into a design you now regret. Instead of moving past it, sit with it for a minute. Try to recreate the feeling you had when you realized that negative thing was happening. Now, shift your attention to how you're thinking about yourself, and try saying these three sentences about that moment. This might feel a little cheesy at first, but notice if it has an impact as you try on these thoughts:
Think back to a version of yourself before your current skills became tied to evaluation, your own equivalent of Shaun’s “motivated kid” who loved the puzzle before it became a performance. Pick one piece of technical work this week with nothing riding on it (a side project, an unfamiliar language, a bug with no deadline attached). Deliberately set a mastery goal for it instead of a performance goal: define success as “what did I learn” or “how did my strategy improve” rather than “did this look impressive” or “did I get it right on the first try.”
Conflict or Coalitions?
This chapter looks at common ways that disagreement and tension play out on technical teams, what separates destructive conflict from productive coalition-building, and how understanding group identities and group psychology can help us steer our groups toward better.
- Read: Hicks (2024), Psychological affordances can provide a missing explanatory layer for why interventions to improve developer experience take hold or fail. A framework paper that dives into examples of how understanding the psychology in software environments can predict whether interventions succeed or fail. A good source to find language and examples for telling other people about the power of good environments for technical teams!

If your group has a history of working together, this chapter’s questions about outsiders and coalitions can potentially raise old tensions. On the first pass, consider framing examples around past teams or third parties rather than the current room. Consider setting a group intention to find the superordinate goal (a shared objective that even two different, opposing sides can agree matters to them).
- Can you recall experiencing a technical disagreement (a design debate, an architecture fight, a code review dispute) that felt like it divided into us vs. them? What tipped it that direction?
- “Superordinate goals help us break through in-group biases as we both realize that cooperative problem-solving is truly necessary, and help us see the humanity in the folks solving next to us.” In a group conflict you can remember, did the two sides have clear “superordinate goals” that were shared and bigger than the group dispute? Are there superordinate goals you can identify in retrospect?
- What does it take for a team to disagree well and to treat group-norm-conflict as information rather than threat? POST proposes that people inside of groups (like "loyal dissenters") can play key roles in helping to model productive conflict and coalition-building. Have you ever seen a person do this? Have you been that person yourself?
- This chapter reviews research that suggests dissent lands best not when it’s watered down, but when the dissenter’s loyalty to the group is unambiguous. Can you think of a technical disagreement that went well specifically because everyone understood the person was “on the team’s side”? What signaled that?
Think of someone on a past team you were in coalition with, even if you didn’t always agree with them. What made that relationship work? How did it change your thinking? If you’re still in contact, make a plan to reach out to that person and thank them.
Think of a disagreement with a group that you’re experiencing or seeing right now. Whether or not you agree, take two minutes to experiment with reframing the conflict as coming from loyal dissent, putting it into the context of a push for better coming from someone committed to the team’s success. Try writing a few sentences from the dissenter’s point of view.
Becoming an Organization That Wants to Understand Itself
Chapter 6 makes the case for organizations building self-understanding and self-correcting processes through developing an evidence strategy. This requires actually wanting to know what’s true about our teams, not just wanting to confirm what we already believe. It also requires creating a measurement plan that technical people can trust, give feedback on, and believe in.
- Watch: Measuring Cycle Time with Dr. Cat Hicks (44:53). A deep dive conversation about the complexities of measurement, where Cat describes the challenges and translating her empirical work on cycle time into advice for leaders.
- Read: Flournoy, Lee, Hicks, & Wu (2025), No silver bullets: Why understanding software cycle time is messy, not magic also published in Empirical Software Engineering. Read for our evidence practice in action, including an analysis that questions the predictive value of a software metric widely perceived to be predictive. If you are tasked with handling software metrics data yourself, you can find and build on our statistical analysis code available at the paper’s repo.

This chapter asks questions that can veer into critique of specific leaders or decisions “when has leadership ignored inconvenient data,” for instance, could call up real memories. If this is a concern in a workplace group, consider asking participants to frame the discussion around patterns rather than people (“we’ve seen this happen when…”) and around past examples rather than current tensions.
The Evidence Readiness Levels shows nine different levels of evidence to consider when moving interventions from “lab” to “real world ready.” Where have you seen software team evidence used to drive decisions around you, and which level does it match?
- What would it take for your organization to actually want to understand itself, rather than defend its existing narratives?
- Have you seen a case where leadership asked for data, got an inconvenient answer, and ignored it? What happened?
- The chapter suggests that unpacking a fuzzy question like “does this practice work” often reveals you actually care about something more specific, like “will this practice improve our sense of autonomy and speed up delivery.” Try that unpacking on a belief your own team holds. What was the fuzzy version, and what did the specific version turn out to be? Did it surprise you?
- This chapter offers the “evidence audit” as an option for people who don’t have the power to run real interventions at work: just cataloguing what you currently know, and how certain you actually feel about it. If you ran an evidence audit on one belief driving decisions on your team right now, what do you think you’d find? Would there be solid evidence, a reasonable guess, or something closer to folklore?
- Team vs. company.A team can often start small on gathering evidence without waiting for permission, like acting on what its own retros surface or running an informal pulse check. But some questions (why is turnover high across the whole engineering org? is a metric that leadership relies on actually measuring the right thing?) need company-level investment: dedicated research capacity, budget, and access to cross-team data. Which kind of evidence question is in front of your team right now, and does it need a team-level answer or a company-level one?
Pick one causal claim currently floating around your technical community or workplace (“X practice makes us more productive,” “Y tool improves quality”). Write out the claim. Then, see if you can answer these questions for it: what specific effect is this, over what timeframe? What evidence could you realistically get to assess it? What would a counterfactual comparison look like? Is there an important confound (a third variable that changes the relationship) that might change whether you get this effect?
If you could run one evidence project inside your own team or org, what question would you actually want answered?
Fighting Dirty for Good Culture
The closing chapter is a call to action: protecting good culture takes deliberate, sometimes uncomfortable effort. But POST argues that this effort is worth it, because it brings more meaning and agency to our lives.
- Watch: Helping developers thrive. Cat at the Data Science Hangout (57:51). An informal community conversation on advocacy, psychological safety, and being strategically visible.
- Read: Lee & Hicks (2024), Understanding and effectively mitigating code review anxiety in Empirical Software Engineering. Read for a full breakdown of the Code Review Anxiety intervention, including intervention design details.

When talking about culture change, people often express frustration about workplaces that feel unmovable. Venting and validating frustration may be a valued use of time for your group, but the closing prompt suggests pushing for action to avoid discussions getting too mired in pessimism. If the discussion tilts into venting, consider gently redirecting toward the one small thing each person is actually willing to try.
- What does “fighting dirty for good culture” mean to you after reading this book? Does it feel aggressive, necessary, both?
- POST names four components of Developer Thriving measured in Hicks' research that together support resilient teams: Learning, Agency, Belonging, and Self-efficacy (the LABS model). Which of the four feels most under threat on your team right now? What would “fighting dirty” for that specific component look like? Would you consider yourself a culture champion for one of these components in particular?
- How can we, in this group right now, support each other in believing in our own power to create positive change?
- Team vs. company.“Fighting dirty for good culture” looks different depending on your scope of authority. At the team level, it might mean pushing back on a bad norm in the room where it happens. At the company level, it might mean building a coalition, going to leadership with evidence, or waiting for the right moment to raise something that’s above your team’s control. Which fight is actually yours to have right now, and which one would require recruiting allies outside your team first?
Write down one sentence you could say in your next team meeting that puts one of this book’s ideas into practice.
Take two minutes to write down a concrete value that matters to you. It can be broad, but should be personal to you. For instance, you might consider family, friendships, craft, honesty, care for other people, learning as a lifelong value, or intellectual fulfillment. Don’t necessarily try to connect it to your work; just write about why the value matters to you and one specific way it has shown up in your life. Then think about your technical work. How might you show up for and demonstrate this value in your specific role, workplace, or life in technical communities?
The Culture Cycle

This section is a bit different from the rest of Chapter 7 as it offers a working group framework that maps culture across four interlocking levels. For some groups, this can be worth its own deep dive because it is a chance to move from discussion to concrete planning, and because the same framework can be applied to any change your group is trying to make.
This section draws on the “culture cycle” framework from Hamedani, Markus, Hetey, and Eberhardt’s (2024) American Psychologist article, “We Built This Culture (So We Can Change It).” The model maps culture across four interlocking levels: ideas (shared narratives and assumptions about what’s good, right, and effective), institutions (for example, an organization’s formal laws, policies, and practices), interactions (the everyday exchanges among people, groups, and tools), and individuals (people’s own beliefs, identities, and behaviors). The four levels continuously shape one another, and change is most durable when it happens at multiple levels at once, rather than being left to a single “culture initiative.”
A chapter 7 table proposes contrasting ways that Brains-in-Jars and Thriving organizations take up beliefs and norms about ideas, institutions, interactions and individuals. Do any of these resonate in particular with this group?
- Map your own team or org onto the four levels: What’s an idea about “good developers” or “good code” that seems to circulate unquestioned? What institutional policy operationalizes it (hiring rubrics, promotion criteria, review templates)? What interactions reinforce it day to day (standups, code review comments, Slack norms)? How do individuals internalize it, in their own self-talk or self-doubt?
- The article argues culture change sticks best when the four levels are in alignment, and that misalignment (“culture eats policy for lunch”) is why well-intentioned policies often fail. Can you think of a policy at your workplace that didn’t “take” because it wasn’t backed up at the interactions or individuals level? What would alignment have looked like?
- The framework distinguishes top-down change (initiated by people with power or authority) from bottom-up change (initiated by people with less power, often because they’re the ones most affected by the status quo). Where have you seen each type play out in a software organization, and which tends to be trusted more?
- One principle from the paper: culture change is easier when it leverages existing core values, and harder when it challenges deep-seated defaults and biases. What’s a “default” in software culture (about who counts as technical, whose feedback carries weight, what “10x” looks like) that is hard to dislodge?
- The framework argues that culture change often involves power struggles and identity threat, and that resistance or backlash is a normal part of the process rather than proof that the change has failed. Has your team ever mistaken pushback for failure and abandoned a change too early? What would it have looked like to expect and plan for that resistance instead?
- Timing and readiness matter, and change attempted in a moment of crisis or exhaustion can land very differently than the same change attempted when a team feels some stability. Have you seen a “fight for good culture” that failed mainly because of bad timing, not a bad idea?
Pick one small change you’d want to see in how your team works. Using the four levels, sketch one concrete strategy you could try at each: an idea to name and question, an institutional practice to adjust, an interaction to change, and something for yourself as an individual to try differently.
Big Picture Discussion Questions

This is the closing conversation, and it invites more personal reflection than earlier questions did, especially the two questions about naming an experience the book helped you understand and identifying one idea to act on. You might consider giving people permission to answer briefly or to pass. You might also consider ending with a round-robin where each person briefly names one thing they’re taking from the book with no discussion. That structure gives everyone a way to be heard but manages time and energy at the end of a session.
- Which chapter changed your mind about something, and why?
- This is an unusual “software book” in how much it focuses on human psychology, our experiences, and concepts like self-compassion and belonging. Did that combination change how you read the book compared to a typical management or productivity book?
- If you could hand this book to one person at your workplace, who would it be, and what would you want them to take from it?
- The book argues that psychological safety and learning culture are “infrastructure.” If your organization or technical community treated them that way (budgeted for them, staffed for them, treated building them as productive work), what would change first?
- What’s one part of your experience this book helped you name and understand?
Further Reading & Resources
The book
Hicks, C. (2026). The psychology of software teams. Routledge. https://doi.org/10.1201/9781003589112
Research
Flournoy, J. C., Lee, C. S., Hicks, C. M., & Wu, M. (2025). No silver bullets: Why understanding software cycle time is messy, not magic. Empirical Software Engineering, 30, Article 174. https://doi.org/10.1007/s10664-025-10735-w. Also available on arXiv (arXiv:2503.05040): https://arxiv.org/abs/2503.05040
Hamedani, M. G., Markus, H. R., Hetey, R. C., & Eberhardt, J. L. (2024). We built this culture (so we can change it): Seven principles for intentional culture change. American Psychologist, 79(3), 384–402. https://doi.org/10.1037/amp0001209
Hicks, C. M. (2024). Psychological affordances can provide a missing explanatory layer for why interventions to improve developer experience take hold or fail [Preprint]. PsyArXiv. https://osf.io/preprints/psyarxiv/qz43x
Hicks, C. M., & Hevesi, A. (2024, November 21). A cumulative culture theory for developer problem-solving [Preprint]. PsyArXiv. https://doi.org/10.31234/osf.io/tfjyw
Hicks, C. M., Lee, C. S., & Foster-Marks, K. (2025, March 15). AI skill threat: How the structure of developers’ beliefs about software development ability impacts their psychological resilience during rapid technology shift [Preprint]. PsyArXiv. https://doi.org/10.31234/osf.io/2gej5_v2
Hicks, C. M., Lee, C. S., & Ramsey, M. (2024). Developer thriving: Four sociocognitive factors that create resilient productivity on software teams. IEEE Software. https://ieeexplore.ieee.org/abstract/document/10491133
Lee, C. S., & Hicks, C. M. (2024). Understanding and effectively mitigating code review anxiety. Empirical Software Engineering, 29(6), Article 161. https://doi.org/10.1007/s10664-024-10550-9
Lee, C. S., Hicks, C. M., & Foster-Marks, K. (2024). “My code is shit”: Negative automatic thoughts and outcomes of a behavioral experiment for code review anxiety [Preprint]. PsyArXiv. https://osf.io/preprints/psyarxiv/hz3et_v1
Find Cat at…
- Fight for the Human: https://www.fightforthehuman.com/
- Change, Technically: https://www.changetechnically.fyi/
- Personal site: https://www.drcathicks.com/research
- Catharsis: https://www.catharsisinsight.com/
Invite Cat to your book club
Cat is glad to visit book clubs and technical communities that are using this guide for a short virtual Q&A. If your group would like her to join a session, reach out through drcathicks.com.
Research measures and additional open-science materials are available through the Research and Catharsis links above.
To facilitate broad use (including inside of your organization), this Book Club Guide to The Psychology of Software Teams by Cat Hicks, PhD is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0). You are free to share, adapt, and remix this guide for any purpose, including commercially, provided you give appropriate credit and note what changes you made.
Please cite as: Hicks, C. (2026). A Book Club Guide to The Psychology of Software Teams. drcathicks.com/#book. Based on The Psychology of Software Teams by Cat Hicks, PhD (Routledge, 2026, ISBN 978-1-032-96338-9).
Other citation formats
Hicks, Cat. A Book Club Guide to The Psychology of Software Teams. 2026, drcathicks.com/#book.
Hicks, Cat. 2026. A Book Club Guide to The Psychology of Software Teams. https://drcathicks.com/#book.
@misc{hicks2026guide,
author = {Hicks, Cat},
title = {A Book Club Guide to The Psychology of Software Teams},
year = {2026},
url = {https://drcathicks.com/#book}
}
However, the book The Psychology of Software Teams and its cover art are separately copyrighted; the CC BY 4.0 license applies to this guide, not to the book.