Selected writing

I would describe my writing style as reader-centred information design. For me, good writing is not just about putting good sentences on a page. I think carefully about the reader’s progression through an argument or idea: what they already know, what they need to know, what can wait, and in what order ideas need to be presented so that each builds clearly on the last.

That also means adapting the writing to its context. Different audiences and publications bring different levels of knowledge, expectations and conventions, so I adjust the structure, language, tone and level of detail accordingly. As a behavioural scientist, I naturally think about the reader’s mental journey: the logic they are following, the assumptions they might be making and the cognitive load I am placing on them.

The underlying principle is consistent even when the style changes: design the information around the person who has to make sense of it.

I am often drawn to fields where the terminology is unstable, categories overlap and apparently similar ideas are actually addressing different problems. These pieces show how I synthesise fragmented material, identify underlying distinctions and turn them into frameworks that make a complex landscape easier to navigate.

A literature-based framework mapping different ways behavioural science is being used in relation to AI. The guide distinguishes between competing objects of study, levels of analysis and functions of behavioural science, turning a fragmented emerging literature into a practical map.

A practitioner-facing extension of the field guide above, translating the conceptual map into a tool for people trying to understand where their expertise fits and why different approaches to behavioural science and AI can talk past one another.

A three-dimensional matrix for distinguishing different relationships between behavioural science and AI according to purpose, scale and interaction. Written to help practitioners scope problems more precisely and avoid collapsing very different kinds of work into one category.

Some writing problems are less about building a new map and more about translating an existing body of evidence into something people can actually work with. These examples show how I move from theory and research to practical guidance without losing the important qualifications.

An ESOMAR paper translating behavioural science into a practical method for sharpening research briefs, identifying target behaviours and making insight work more actionable.

The Seven Sins of Memory

Thinking About Behavior

A practical piece on why asking people about past behaviour does not automatically produce reliable accounts, drawing on cognitive psychology to show how memory errors shape qualitative research.

A practitioner guide to the judgement involved in applying behavioural frameworks to real qualitative data, including where categories overlap and where mechanical coding can go wrong.

A public-facing example of turning a confusing evidence base into a clear starting point for decision-making while preserving uncertainty and competing explanations.

Good writing is not only about clarity. It is also about understanding the conventions, expectations and language of the audience you are writing for. These examples show how I adapt structure, tone and level of explanation to different editorial and professional contexts.

A commissioned breed profile written in Gundog Journal’s distinctive house style. Before writing it, I analysed 12 previous breed profiles to identify their recurring structure, expressions, pacing and editorial conventions, then used that model to produce a new specialist feature.

A concise editorial article translating cross-cultural psychology for a broad marketing audience, showing how assumptions about “universal” consumer behaviour are often rooted in Western psychological models.

A longer research-led industry paper examining how culture shapes decision-making and what that means for the application of behavioural economics across markets.

Many of the questions I write about are difficult precisely because the evidence does not support a simple answer. These pieces focus on separating mechanisms, testing assumptions and communicating uncertainty without retreating into vagueness.

An evidence-led myth-busting piece examining genetics, development, environmental influences and measurement uncertainty in a condition often discussed in overly binary terms.

A conceptual clarification piece separating distinct biological mechanisms that are often collapsed into a single category, and showing why the distinctions matter in practice.

An examination of what ingredient lists, manufacturing processes and contamination risks do — and do not — allow owners to infer about food reactions.

A piece on probabilistic reasoning in canine genetics and the limits of prediction, aimed at helping readers distinguish risk from certainty.

Some of my writing is more exploratory: identifying emerging phenomena, testing ideas and developing concepts before the language around them has settled.

An essay exploring what happens when generative AI dramatically reduces the effort required to produce options while increasing the cognitive burden of evaluating, choosing and committing to them.

How LLMs think differently

Artificial Thought

A comparative experiment using the same prompts across multiple language models, examining differences in reasoning style, framing, structure and assumptions and what those differences mean for users.

My broader writing and podcast work on AI through a behavioural lens, focusing on how people think, work, decide and collaborate with AI.

Website copy requires a different kind of writing: understanding what visitors need to know, organising information around their questions, and shaping tone, hierarchy and calls to action so the site feels coherent and easy to navigate.

Website copy for my behavioural strategy practice, designed to explain an abstract consulting offer in concrete terms. The site moves from the core problem through services, working approach and examples, using concise language and progressive disclosure to help prospective clients quickly work out whether the offer is relevant to them.

Copy and information architecture for the international English-language site for the breed. The challenge was to make specialist information accessible to an international audience while supporting different visitor needs, from breed history and standards to breeding guidance, hunting trials and finding a dog.