Behavioral Decisions Insights

Better training does not guarantee meaningful change. Organisations often invest in new content, improved formats and skilled facilitators, yet employees return to workplaces that reward the very behaviours the training was meant to replace. The programme may not be the problem. The surrounding system may be. In his latest article, Nelson Enrique Muñoz Cerda explores why effective organisational learning begins with three questions: What kind of problem are we facing? Technical problems may require new knowledge or skills. Adaptive problems require people to question established assumptions, values and ways of working. Can employees apply what they learn? Supervisor support, opportunities to practise and psychological safety strongly influence whether learning reaches everyday work. What does the organisation actually reward? Training may promote autonomy and experimentation, while performance systems continue rewarding control, speed and error avoidance. When messages and incentives conflict, employees follow the system. The article illustrates this through a leadership programme that received consistently high satisfaction scores but produced little improvement in trust, autonomy or employee turnover. Instead of redesigning the training, the organisation changed its evaluation criteria, involved senior managers and introduced measurable follow-up indicators. After 12 months: 📈 Trust and autonomy scores improved 📉 Annual turnover fell from 18% to 12% ✅ Documented delegated decisions increased from 0.4 to 2.3 per month The lesson is simple: learning does not create change when the workplace makes new behaviour difficult or costly. Before designing another course, diagnose the real problem, examine the conditions for application and align organisational systems with the behaviour you want to see. Read “The True Direction and Meaning of Learning in Organizations” by Nelson Muñoz Cerda.

Most process improvement efforts begin too late. A mistake happens. A deadline is missed. A complaint arrives. The organisation responds with more training, another reminder, tighter oversight or a new control. Even reaching for a behavioural nudge can be premature. Before choosing an intervention, we need to ask a more basic question: What kind of decision are we designing for? Fast, automatic thinking is not always the problem. It helps experienced employees complete familiar tasks efficiently and preserve their attention for situations that require greater judgement. The real design challenge is deciding when to support that automatic response and when to interrupt it. In his latest article, Joe Caccitolo, MBA, PsyD introduces a five-question Decision Diagnostic: What is the exact decision? What does a successful decision look like? What will the user do automatically? Why might that automatic response fail? Should the redesign support or interrupt it? The principle is practical: ✅ Support automatic thinking when a task is frequent, predictable and has a clearly preferred response. Prefill known information, remove irrelevant choices and make the correct action easier. ⚠️ Interrupt automatic thinking when a decision is unusual, difficult to reverse or potentially costly. Use targeted warnings, confirmations or forcing functions to create a deliberate pause. The key word is targeted. Poor controls slow everyone down, including those handling routine cases correctly. Good behavioural design removes friction from the normal path and places it precisely where automatic behaviour creates risk. Once the decision has been diagnosed, choice architecture offers several tools: defaults, feedback, incentives, mapping, error elimination and structured choices. The goal is not to add more controls. It is to design the right amount of friction at the right moment. Before you nudge, diagnose the decision. Read “Before You Nudge, Diagnose the Decision” by Dr. Joseph Caccitolo.

When does something become more valuable not because it changed, but because it became ours? A stray kitten hiding in a garage slowly became part of a family’s routine. Nothing about the kitten objectively changed. But somewhere between feeding him, earning his trust and hearing him purr, he stopped being “a kitten” and became “our kitten.” That shift offers a powerful introduction to two behavioural concepts: 🐾 Psychological ownership - the feeling that something belongs to us, even without formal ownership. 💭 The endowment effect - our tendency to value something more highly once we perceive it as ours. Classic experiments demonstrate this with ordinary objects. Participants randomly given a coffee mug demanded roughly twice as much to give it up as others were willing to pay for the same mug. The object remained identical; only the relationship to it changed. This psychology extends far beyond physical possessions. It can influence why founders resist abandoning features they built, why we defend our own ideas more fiercely, why leaving a job can feel like losing part of our identity, and why an old childhood toy can suddenly seem priceless. Businesses also create opportunities for psychological ownership through free trials, personalisation and hands-on product experiences. Once we begin imagining something as ours, losing access can feel different from never having it. But the effect is not universal. Experience, culture, context, genuine attachment, identity, sunk costs and status quo bias can all influence why we struggle to let go. In her latest article, Haidi Kariem explores these distinctions without reducing every emotional attachment to a single bias. She also leaves us with a useful question: If I didn’t already own this, would I choose it again today? If the answer is yes, we may simply value it. If the answer is no, ownership itself may have quietly become part of the evidence we are using to judge its value. Read “The Stray Kitten That Rewired How I Think About Value.”

What looks like a personality clash may actually reflect different ways of using knowledge. In many teams, the same tensions appear: One person wants more evidence before moving forward. Another wants to stop discussing and start implementing. Someone introduces new possibilities. And someone questions whether the team is solving the right problem at all. These differences are often interpreted as resistance or incompatibility. In his latest article, Nelson Enrique Muñoz Cerda offers another perspective through an exploratory model based on two dimensions: ↔️ Conservation or transformation - Do we apply existing knowledge or challenge and reconfigure it? ↕️ Problem or action orientation - Do we prioritise understanding or execution? Together, they produce 4 orientations: 🔍 Conservative-Analytical - Protects rigour and established boundaries. ⚙️ Conservative-Executor - Turns proven knowledge into results. 💡 Transformative-Executor - Experiments and applies knowledge in new ways. ❓ Transformative-Analytical - Questions assumptions and reframes problems. None is inherently better than the others. Each contributes something valuableand each creates risks when it operates without the balance of the rest. A team that only analyses may never act. A team that only executes may move quickly in the wrong direction. A team that only preserves existing knowledge may struggle to adapt, while one that constantly transforms may lack stability. The most useful question, therefore, is not “Who is right?” It is “When does the team need each perspective?” The profiles are not fixed personality types or validated diagnostic categories. They are flexible orientations designed to help teams replace judgement with curiosity and turn cognitive differences into more productive collaboration. Read “Four Ways of Using Knowledge in a Team: A Model for Understanding How People Learn and Act in Collective Work” by Nelson Muñoz Cerda.

What happens when self-esteem becomes something we earn, spend and lose in public? Every social platform runs on attention. But that attention rarely remains external. A like becomes evidence of approval. A follower count becomes a scoreboard. A carefully curated feed becomes a mirror, one that only shows certain angles. Welcome to the ego economy: a system in which our sense of worth can be updated in real time by strangers, algorithms and visible engagement metrics. The instinct behind it isn’t new. Social comparison theory suggests that we often evaluate ourselves by comparing our lives with those of others, particularly when there is no objective standard available. Social media didn’t create that instinct. It industrialised it. Today, we can continuously compare our ordinary moments with the most polished parts of other people’s lives. The platform doesn’t need to tell us that we are falling behind. It only needs to keep showing us people who appear to be doing better. But the research tells a more nuanced story than “social media is bad for self-esteem.” Upward comparison can make someone else’s success feel like proof of our own inadequacy. Yet the same content can sometimes inspire us by suggesting, “Someone like me can achieve that too.” Similarly, active use may help build connection and online support, while passive scrolling may invite more comparison. However, the overall effects are often smaller and less consistent than popular narratives suggest. Neither type of use is automatically safe or harmful. What matters is the interaction between the content we see, the comparisons it encourages and the way platforms quantify approval. So, what can we do? We may not be able to eliminate comparison, but we can become more conscious of how we participate in it. We can curate feeds that don’t constantly trigger feelings of inadequacy, question what visible metrics actually represent and practise feeling glad for someone else’s success without automatically turning it into a judgement about ourselves. Platforms are not neutral. Visible likes, follower counts and algorithmically selected content transform social interaction into a continuously updated score. But we are not entirely passive either. In her article, Nomrota S. examines what the research really says about social media and self-esteem without reducing a complicated relationship to a simple verdict. Because self-esteem was never meant to be a public ledger. And a number generated by an algorithm was never a reliable measure of anyone’s worth. Read “The Ego Economy: The Impact of Social Media on Self-Esteem.”

Is it a hat or a boa constrictor digesting an elephant? The famous drawing from The Little Prince captures something fundamental about human psychology: we don’t experience reality exactly as it is. We interpret it through the lens of our backgrounds, biology, language, motivations and past experiences. Psychologists call this a perceptual set, our tendency to interpret new information according to familiar patterns. It helps us make sense of the world efficiently, but it can also limit what we are able to see. The problem begins when we mistake our interpretation for objective reality. This is known as naïve realism: the belief that we see the world as it truly is and that reasonable people should naturally reach the same conclusions. It can strengthen confirmation bias, create a false sense of consensus and contribute to polarization when others see things differently. Our perceptions don’t only affect how we understand a situation. They also influence how we behave within it. If we expect rejection, for example, we may become more distant or withdrawn. Other people may then respond less warmly, reinforcing the belief that we were going to be rejected. What began as an assumption becomes a self-fulfilling prophecy. So, how can we interrupt this belief–behaviour feedback loop? In her latest article, Daria A. introduces the APR technique: 🔎 Awareness: Recognise the belief or assumption shaping your interpretation. ⏸️ Pause: Create space between that initial interpretation and your response. 🔄 Reframe: Consider alternative explanations that leave room for a different outcome. Reframing isn’t about denying reality or convincing ourselves that everything will go perfectly. It is about replacing an overly certain prediction with a more open and useful interpretation. When we question the reality we have constructed, we create space to behave differently and give different outcomes a chance to emerge. After all, what looks unmistakably like a hat may still conceal a boa constrictor digesting an elephant.

From Manager to Detective: Why Recurring Errors Point to Design Problems, Not People Problems An employee makes a mistake. The instinct is often to ask: Who made the error? A better question may be: What made the error predictable? When the same mistake keeps appearing across competent employees, different locations, or over time, it may not be a people problem at all. It may be a decision-environment problem. Behavioral science gives managers a different lens: → Is the right choice obvious at the moment of action? → Are defaults pushing people toward the wrong outcome? → Does the workflow create unnecessary cognitive load? → Are warnings appearing too late—or being ignored because there are too many? → Are incentives unintentionally rewarding speed over accuracy? Recurring errors are data. Instead of immediately adding training, reminders, or accountability measures, investigate the system that repeatedly produces the behavior. Good managers correct mistakes. Great managers investigate why the mistake was easy to make. The next time an error keeps recurring, put on the detective hat. What does the decision environment reveal? A new article by Joe Caccitolo, MBA, PsyD “From Manager to Detective: Why Recurring Errors Point to Design Problems, Not People Problems.”

🤖 What happens when AI stops being just a tool and starts influencing how we think? As AI becomes an everyday workplace companion, a new set of psychological questions is emerging. How much should we trust what AI tells us? What happens when we begin turning to it for reassurance, judgement, or even major decisions? And where is the line between productive AI use and psychological dependence? In a new article, Susan A. R. explores these questions through an interview with Dr. Adjoa Smalls-Mantey, an emergency psychiatrist and physician-scientist. One of the most important points is that “AI psychosis” is not a formal psychiatric diagnosis. It is an emerging concept being discussed in research around situations where AI interactions may contribute to or reinforce delusional thinking in vulnerable individuals. The broader workplace question, however, goes beyond psychosis. AI can be incredibly useful. But if employees gradually begin to: 🔹 Trust AI's judgement over their own 🔹 Accept its answers without verification 🔹 Struggle to explain the reasoning behind their work 🔹 Use AI as a substitute for human connection or professional expertise 🔹 Feel uncomfortable acknowledging how heavily they rely on it …we may need to start thinking differently about what healthy AI use looks like. The goal isn't to fear AI. It is to develop the psychological literacy needed to work alongside it responsibly. As AI becomes embedded in our workplaces, perhaps the next form of digital literacy isn't only knowing how to use AI. It is knowing when not to trust it. What do you think organizations should be doing now to encourage healthy relationships with AI?

🧠 What if your strongest memories are not always your most accurate ones? We often think of memory as a recording of the past; a mental archive of what really happened. Every time we recall an experience, our brain may be creating a new version of it. ✍️ Memory is not a fixed file we simply retrieve. It is reconstructed based on our current emotions, experiences, expectations, and the information we encounter afterward. A single misleading question can change what someone remembers. A conversation with another person can unintentionally reshape our own recollection. And confidence does not always mean accuracy. These insights have important implications - from everyday interactions to one of the most critical areas where memory matters: eyewitness testimony. The key lesson? Our minds are not passive recorders of reality. They are active meaning-makers, constantly updating our understanding of the world. 🌍 In her article, Haidi Kariem explores the science behind memory reconsolidation, the misinformation effect, and why the memories we trust most may not always be the memories that happened exactly as we remember them. 📝 Sometimes, what we remember is not the original version, it is the latest edit. Read the full article: Yesterday, Re-written: The Science Behind Memory 👇

AI simulations are becoming a powerful tool for understanding people; but are we asking them to do something they are not yet ready for? Synthetic users can help teams test ideas faster, explore scenarios, and identify patterns before investing significant resources. But human decisions are rarely made in ideal conditions. People do not simply respond to information. They make choices while dealing with emotions, habits, uncertainty, social pressure, stress, and countless contextual factors that are difficult to model. This raises an important question: Are AI simulations helping us understand human behaviour — or are they sometimes showing us how an easier-to-persuade version of humans might respond? The future of behavioral testing will likely not be about choosing between AI and human research. It will be about understanding where each approach is strongest. AI can help us explore possibilities. Humans help us validate reality. The challenge is knowing the difference. A new article by Beatrice Widmark explores the "synthetic persuasion paradox" and why organizations should treat simulation results as hypotheses to test, not decisions already made. Read the full article: The Limits of AI Simulation: What AI Still Misses About Human Decision-Making 👇