How this investigation is built
The starting point is the work already gathered in the project: texts, scientific notes, an inventory of 104 entries and research on cancer, attachment, emotions and the living world. Additional multilingual research was conducted in September 2026, then brought together with this corpus.
The searches used scientific databases, search engines and institutional sources. Accounts and uses were also sought on YouTube, Reddit, X, Facebook, Instagram, V2EX and Bilibili, alongside reading pages, feeds and GitHub resources. The protocol’s twelve channels are recorded in the research log. They offer neither equal richness nor equal access.
Queries and reading mainly used English, French and Japanese, with searches in Chinese and sources in other languages, notably Indonesian and Korean. This linguistic scope is broad without representing every publication worldwide.
Selection prioritises original sources: an article, an organiser’s document, an institutional page or an account by its author. A review helps map a field and locate its studies. A press release establishes what an organisation announces, with particular attention to results not yet published in a scientific journal.
What is counted
The historical inventory mixed products, programmes, events and publications. Entries overlapped, and some brought together several objects. The catalogue retains their relationships and provenance. It does not present the 104 entries as 104 independent health interventions.
The library counts documents. Two analyses of the same cohort remain two publications, with a link to their common origin. An official page, a press report and a video describing the same operation do not become three independent confirmations of effectiveness.
The depth of reading is indicated: relevant full text, partial reading, abstract or document only identified. An inaccessible source is not presented as fully audited. References identified but not read can guide future research; they do not support a clinical conclusion on their own.
How claims are reviewed
Review checks figures and conclusions against the population, protocol and outcome actually studied. It also checks distinctions between correlation and causation, interpretation of non-significant findings and possible cohort overlap.
This work led to corrections of several formulations in the initial corpus. The Pokéwalker’s accuracy had been measured, even though a lasting benefit remained to be established. The brushing study in autistic children measured reported behaviour, not plaque. A study sometimes associated with Pokémon and autism primarily concerned Digimon. These corrections are part of the investigation and its revision log.
AI tools helped search, organise and cross-check sources. Independent passes were used to identify omissions and check the scope of wording. This is neither external clinical validation nor peer review. No external expert interview was conducted during this pass.
The ambition of completeness
My ambition is to bring together the most complete investigation possible into Pokémon and health. This means following favourable uses, negative findings, risks and open questions, including when they shift the starting idea.
The dated comparison examines scientific reviews, books and institutional resources. Several syntheses cover GO in great depth. This guide also seeks to connect Sleep, Smile, objects, prevention initiatives, care and cultural experiences. Breadth of scope and strength of evidence remain two distinct dimensions.
The comparison is not an exhaustive global ranking. Paid content, unindexed material or publications in other languages may complement this work. Methods and sources are made visible so readers can check what is covered and extend the investigation.
Keeping corrections alive
The research date appears at the beginning of the investigation. An update should identify new sources, what they change and corrected wording. Changes to an app, campaign or data policy should be dated separately from earlier scientific findings.
For further reading, the catalogue of objects and initiatives brings together documented cases and the library provides access to sources. The Pokémon and Health page presents the origins of this investigation and the different reading paths.
The phrase I heard in my office has taken me well beyond the bookshelf. It still interests me for what it opens up: understanding the place a familiar world can take in actions, relationships and moments of care.