David Kolb (1984) argued that learning is not a passive process of absorbing information — it is a dynamic cycle of transformation, triggered by real experience. Every other reflection model in this page is, at its core, an elaboration of Kolb's central insight. Understanding Kolb helps you understand why reflection works — not just how to do it.

Graham Gibbs (1988) built on Kolb's cycle to create a more detailed, structured framework specifically designed to support deep reflection on significant experiences. Its defining feature is Stage 2 — Feelings — which explicitly acknowledges the emotional dimension of experience. This makes Gibbs particularly powerful for emotionally charged clinical events: a patient death, a near-miss, a complaint, or a consultation that left you feeling unsettled.

Gibbs' model is cyclical — the Action Plan from one reflection feeds into the next Concrete Experience, and learning deepens with each complete cycle.

Terry Borton (1970) created the simplest and most accessible model of reflection — three questions drawn from Gestalt therapy. Its power lies precisely in its simplicity: almost anyone can start using it immediately, even without training in reflective practice. It is an excellent entry point for trainees who are new to log entries, and it remains useful at all stages of training for quick, regular reflection.

The key insight is that "So What?" is where almost all the real reflection happens. The questions are simple; the depth you bring to them is what makes the difference.

Christopher Johns (1994) designed this model specifically for healthcare professionals — originally for nursing, but widely used in medical education. Unlike Borton or Gibbs, Johns uses a comprehensive set of structured questions to guide the practitioner through five distinct domains of reflection. Its strength is thoroughness: it explicitly considers the perspectives of the patient, the team, and the wider healthcare system — not just the practitioner's own experience.

Johns can feel prescriptive when first encountered, and some trainees find the question structure produces a "box-ticking" approach rather than genuine reflection. Used thoughtfully, however — especially for ethically complex or team-dynamic cases — it produces some of the richest reflective entries possible.

John Driscoll (1994, updated 2007) took Borton's three simple questions and added structured trigger questions under each stage — specifically designed for healthcare practitioners reflecting on clinical experience. The result is a model that is almost as simple to remember as Borton's, but far more guided in practice. If Borton is the skeleton, Driscoll is the skeleton with muscles.

This model is particularly well suited to GP trainees who find Borton's model too open-ended but find Gibbs' 6 stages slightly overwhelming. It sits neatly in the middle ground.

Sue Atkins and Kathy Murphy (1993) developed their model following a comprehensive review of the reflective practice literature. Their key insight was that most models treat reflection as something you do after any event — but the most powerful reflections are triggered by a specific emotion: discomfort. A feeling of uncertainty, unease, guilt, surprise, or conflict is the body's way of signalling that this event contains important learning.

This model is particularly valuable for GP trainees because it helps them identify which cases actually deserve a deep reflective entry — not every consultation generates genuine discomfort, and those that do are the richest sources of learning.

PUNs and DENs is the only reflection method on this page invented by a GP, for GPs. Dr Richard Eve, a GP in Somerset, developed it in the mid-1990s as a practical, patient-centred way of identifying learning needs directly from consultations. It was subsequently recognised by the RCGP as a valid method for learning needs assessment and has remained embedded in UK GP training and appraisal ever since.

Unlike the other models on this page, PUNs and DENs is not primarily a reflective writing framework. It is a way of identifying what to reflect on — a filter for selecting the right cases and learning needs from the stream of everyday general practice. Used well, it makes your log entries more relevant, more patient-centred, and more genuinely useful.

This is an upgraded, deepened version of the Borton/Rolfe framework, specifically adapted for GP training use. The key addition is Stage 2 — "Why?" — which forces analysis of your thinking process rather than just the events. Most trainees skip straight from "what happened" to "what will I do differently" without ever asking why they acted the way they did. This model puts the "why" at the centre where it belongs.

One of the most impressive types of reflection a GP trainee can write is an analysis of their own clinical reasoning — specifically, the biases that may have influenced their thinking. This model structures that reflection. It demonstrates advanced clinical insight and shows awareness of safe practice — both highly valued by educational supervisors and ARCP panels.

This model uses the ICE framework — Ideas, Concerns, Expectations — which is already central to GP consultation skills, and adapts it as a reflection tool. It is particularly well-suited to GP trainees because it is familiar, GP-specific, and directly connects reflective learning to consultation quality improvement. It is excellent for entries that arise from missed connections with patients.

This model is specifically designed to maximise the value of your log entries for ARCP evidence. It structures your reflection around the RCGP Professional Capabilities, ensuring that every entry you write is directly and explicitly evidencing your progression. It is particularly useful when you know you have a gap in a specific capability area and want to address it deliberately.

This model is designed for consultations where your emotional response was significant — where you felt pressured, frustrated, anxious, distressed, or out of your depth. These emotions are not weaknesses to hide; they are data. When analysed honestly, they reveal the most important insights into how you think and behave under pressure. This is one of the most impressive categories of reflection a trainee can produce.

This model is specifically designed for safety-critical learning — near misses, significant events (SEAs), prescribing errors, or cases where patient harm nearly occurred or did occur. It uses a systems-thinking approach to understand not just what happened, but why the system allowed it to happen. This is directly aligned with how the RCGP and GMC expect you to approach SEAs in your ePortfolio.

One of the most consistent pieces of advice from experienced GP trainees is this: don't let perfect be the enemy of good. Waiting for time to write a comprehensive Gibbs entry on every case is a recipe for backlog, avoidance, and panic before ARCP. The Micro-Reflection model is for the end of a busy clinic when you have 2 minutes, an interesting case, and a risk of forgetting it entirely by tomorrow.

ERA is one of the simplest and most sustainable models for day-to-day reflective writing. Three stages, each with clear purpose. It is particularly well-suited for routine clinical encounters that don't require the depth of Gibbs but still deserve more than a one-liner. Use it when you have 10–15 minutes and want to write something meaningful.

ERA+ adds a fourth stage to the basic ERA cycle: Review. This is written after you have tried your action plan — typically days or weeks later. It answers: did the change work? The Review stage is what transforms a log entry from a record of intention into evidence of genuine learning and practice change — exactly what ARCP assessors and the RCGP want to see.

Originally developed for improvement cycles, APDR adapts very naturally to reflective log entries — particularly when reflecting on quality improvement activities, practice-wide changes, or a series of similar clinical problems. It is the most structured model for showing that your reflection has translated into tested, evaluated change rather than just stated intention.

The STARR model (Situation, Task, Action, Result, Reflection) is adapted from the standard STAR interview technique — with a crucial addition: the Reflection stage that makes it a genuine learning tool rather than just a narrative. It is particularly effective for "showpiece" log entries that need to demonstrate a specific capability with clarity and impact — leadership, handling complaints, teaching, or managing complex ethical situations.

The 3R model is one of the simplest frameworks available — three words, three stages, immediately usable after any clinical encounter. It is based on academic guidance for reflective writing and works for quick log entries while still producing meaningful, assessable reflection. Think of it as a Borton variant with tighter language.

Medical education research consistently shows that reflective tasks underuse emotional content — even though emotions powerfully influence clinical behaviour. This model adds a dedicated Emotion stage between Experience and Reflection, making the emotional dimension explicit rather than optional. It is particularly useful for reflections on burnout risk, difficult consultations, value conflicts, or any situation where your emotional state noticeably shaped your actions.

Origin: Educational psychology. Jenny Moon, a UK academic, proposed that reflective writing is not simply present or absent — it exists on a spectrum of depth. Most importantly, she defined what each level looks like in actual writing, making it a practical self-assessment tool rather than just a theory.

Why it matters for GP trainees: Moon's levels map almost exactly onto the Bradford VTS ISCE criteria (Not Acceptable → Acceptable → Excellent). Using Moon's descriptions alongside ISCE helps trainees understand what deeper reflection actually looks like on paper — not just what it means in theory.

Origin: Adult education and critical pedagogy. Stephen Brookfield, a professor of adult learning in the USA, argued that truly critical reflection is impossible from a single perspective. We are blind to our own assumptions. Seeing our practice through multiple "lenses" is the only way to reveal what a single autobiographical lens will always miss.

Why it matters for GP trainees: Most log entries are written entirely through Lens 1 (the doctor's own view). Lenses 2 and 3 are almost always absent — yet the RCGP explicitly values patient-centred practice and team awareness. Brookfield gives a structured prompt for the multi-perspective analysis that the ISCE "Critical Analysis" stage requires. It is also particularly powerful for trainers, IMGs, and any entry involving team dynamics.

💡 Using Brookfield in Practice: One-Paragraph-Per-Lens

For a complex case, write one paragraph per lens. For quick use: ask which lenses your current entry is missing — then add one sentence each for the missing ones.

Origin: Organisational psychology and systems thinking. Developed by Harvard professor Chris Argyris in the 1970s, popularised by Peter Senge in The Fifth Discipline (1994). The model has been applied to medical education — a 2022 Canadian Medical Education Journal paper specifically references it in the context of clinical reasoning and cognitive bias in trainees.

Why it matters for GP trainees: The Ladder of Inference is the theoretical foundation behind cognitive bias in clinical reasoning. It shows precisely how anchoring bias, premature closure, and availability bias actually happen — not as abstract errors, but as identifiable mental steps. Using it in reflection means "climbing down" from your action to examine the reasoning steps you didn't question.

Origin: Organisational learning and action science. Chris Argyris and Donald Schön (who also gave us Reflection-in/on-Action) described three levels of learning. The BMJ has referenced double-loop learning in the context of NHS quality improvement and clinical learning from errors (Heathcote, BMJ 2004).

Why it matters for GP trainees: Most log entries achieve single-loop learning at best. Double-loop learning is what ARCP assessors mean when they say a reflection "shows genuine insight." It is the difference between "I'll check the BNF next time" and "I need to understand why I felt I didn't need to check it this time."

Origin: Psychology and group dynamics. Joseph Luft and Harrington Ingham developed this model at UCLA in 1955 — the name "Johari" combines their first names. It has been used in healthcare professional development for decades, and HEE has a dedicated Johari Window resource document for clinical educators.

Why it matters for GP trainees: The Johari Window explains why external feedback from WPBAs (MSF, PSQ, CbD, COT/MiniCEX) is so valuable — and how to use it more purposefully. Most importantly, it contextualises the discomfort of critical feedback: your Blind Spot is not a character flaw; it is simply information that others have about you that you do not yet have about yourself.

Johari Reflection Questions for Log Entries

Origin: Coaching psychology and executive coaching. Developed by Sir John Whitmore, GROW is the most widely used coaching framework in the world. It is used in business, education, sport, and healthcare professional development. It is the implicit structure behind most effective supervisor-trainee conversations.

How to use it in GP reflection: GROW is not a complete reflection model — it is a structure for the action planning stage of any entry. It transforms "I will improve my safety-netting" into a specific, testable, time-bound commitment. Use it as the final section of any Borton "Now What?", ISCE "E", or Gibbs "Action Plan."

Origin: Positive psychology and organisational development. David Cooperrider and Diana Whitney developed Appreciative Inquiry (AI) at Case Western Reserve University. It deliberately inverts the deficit-focused logic of most improvement frameworks — rather than starting from "what's wrong?", it starts from "what's working well — and how do we build more of it?"

Why it matters for GP trainees: Almost all log entries are deficit-focused. AI is the right model for the cases that went well — which are just as worth learning from as cases that went badly. A portfolio that documents only difficulty gives assessors an incomplete picture. Using AI to analyse a successful consultation reveals what you actually did well, which is the first step to replicating it intentionally.