Background It is only recently that written sources of local knowledge on plants are not being ignored by scholars as not belonging to “traditional” knowledge. Ethnobotanical texts, however, if they at all focus on knowledge from written sources, hardly ever pay any attention to the actual processes of interaction with written texts and illustrations. During our research, we examined people’s interactions with texts, illustrations and herbarium specimens of plants they collect or are familiar with. We focused on a small community of Shiri people in the mountainous village and in the lowland settlements in the Republic of Daghestan, Russia. In the paper we address the following questions: how do Shiri people interact with illustrations, written text, and herbaria specimens? How is this interaction influenced by the practice of plant collection? What are the methodological implications of the ways people interact with illustrations, texts, and herbaria specimens? Methods Our research was based on long-term ethnographic fieldwork: co-designing of a booklet showing edible plants people collect in Shiri, semi-structured interviews and video-recordings, and observing interactions between people and text/illustrations/voucher specimens.Results We identified three kinds of interactions between individuals and text/illustrations: “text-wayfaring” – predominantly a bodily interaction between an individual and illustrations and text; “fact/spelling checking” – predominantly discursive and information focused; “between wayfaring and fact-checking” – the mix of the two. Using the idea of textual poaching, as well as the knowledge-making approach, we show that the mode of interaction with text/illustrations influences what is acquired, and how. This process influences readers’ LEK. The mere presence of an information in the text available to people does not imply that they will acquire it, make use of it, and change their LEK. Photographs and pressed specimens of locally known plants are often not (or only partly) recognized by the interlocutors. Video-recording is essential for analysing the above mentioned interactions.Conclusions In ethnobotanical research, it is important to pay more attention to people’s interaction with their sources of knowledge, including text and illustrations. The discursive part of LEK is more easily influenced by written sources. The practice of plant collection is not as easily influenced. Ethnobotanists function in a particular context and are embedded in discourses oriented towards conservation of bio-cultural diversity that value heritage as such, so it is important to be aware of one’s positionality. A methodology that relies on showing pressed specimens or photographs to interlocutors may be a very misleading way of collecting ethnobotanical data.