Our findings demonstrate that ease of acquisition and nutritional potential are the most important determinants of the knowledge and consumption of wild functional and medicinal foods by local experts from the community of Caeté-Açu. Ease of acquisition is important not only in terms of specific differences in plant importance but also in terms of explaining the abandonment of a whole set of wild foods and medicines , as in many regions deforestation is distancing people from natural resource areas.
Studies using ecological indicators of availability to test their relationships with species’ importance have not indicated ease of acquisition as an important predictor [20, 21]. Our study strengthens the idea that accessing people`s perceptions of availability instead of ecological data may provide evidence of certain patterns not highlighted by the latter technique, which may happen because incorporating perceptions of local people allows for a wider number of plants in the analyses, given that wild species occurring outside of forest areas may also be included, as well as species with different life forms. Studies including ecological data commonly focus on specific life forms – more often on woody plants [20, 21]. Furthermore, high values of ecological abundance or frequency do not necessarily ensure that the plants are easy to acquire since they can occur in areas difficult to access. At the same time, low values of ecological abundance or frequency may not indicate that the species is difficult to acquire, as it may occur precisely on people`s common pathways.
However, such arguments do not intend to infer that local perception is a better indicator of species availability/ease of acquisition. Under some circumstances, ecological indexes are related to plant use while local perception of availability is not. In a case study on the drivers of fuelwood use, Hora et al.  encountered this situation and inferred that most of their interviewees were older people who were no longer accessing forest areas and acquired fuelwood from younger community members. Therefore, their perceptions of availability could be strongly related to a past landscape, which could weaken the relationship between perceived availability and fuelwood use. Therefore, we need to perform more studies to understand which portions of availability each technique (ecological indicators vs local perception) is specifically capturing.
Nutritional value has proven to be an important driver of wild food plant importance in other contexts [10, 12]. The fact that nutritional value remained in the final model, while medicinal value did not, shows that, although local experts recognize the important of medicinal attributes in wild food plants, nutritional aspects are more relevant for designating culturally important plants in the food-medicine continuum than medicinal properties.
Our results are consistent with the social-ecological theory of maximation. One of the theory`s assumptions is that the most important natural resources to local people are those that afford the maximum return, considering the balance among different explanatory variables that affect resource use . This maximum return is the best possible achievement from the combination of different variables driving the use of a natural resource . Additionally, the most important drivers as evidenced by our research are also among those deeply studied in the context of optimal foraging theory  – in which nutritional value is analogous to energy intake and ease of acquisition is analogous to energy spent during foraging. Therefore, a balance between the costs of acquiring a plant and the nutritional benefits of consuming it seems to influence cultural relevance in the intersection between food and medicine.
Contrary to other studies, attributes such as taste did not explain plants` importance. The higher importance of nutritional and availability issues compared to taste may be explained by the sociocultural background of the community. Most of the elders (including those from our sample) are former miners or descendants of former miners. In the region, miners worked for extended periods, forcing them to spend much time in natural areas, which were often far from settlements. Wild plants therefore constituted an important source of food in the mining context with a better return for miners implicit in the consumption of nutritionally rich, widely available species. This process probably influenced the importance of plant species, which was transmitted culturally to the miners’ descendants. For this reason, nutritional value and ease of acquisition remain important drivers of plant importance even now when these inhabitants do not rely much on wild foods as an important source of daily energy intake.
However, we need to conduct further studies including nonspecialists (particularly the younger members of the community) to provide evidence for the main drivers of plants’ importance in the intersection between food and medicine. It is possible that the culturally mediated effects of nutritional value and ease of acquisition lose strength among younger dwellers, who are mostly unfamiliar with mining activity and never consumed wild plants as an important source of energy intake. Among people who consume wild plants as a supplementary source of food, taste may be more important, considering that their energy intake is probably being fulfilled by other (nonwild) foods. However, this claim needs to be tested using proper research designs.
Several other variables that were not included in our research may be important drivers of plants’ importance in this food-medicine system. Moreover, qualitative aspects that we did not access can affect the use of food items. People have emotional links (both positive and negative) to food related to childhood memories, family togetherness, sharing care through foods, amusing cooking times or mealtimes, anxiety or sadness . These memories may act as bridges between past individual experiences and present life, affecting the way an individual approaches food . For some individuals from rural communities, eating wild foods can elicit positive memories related to early childhood or can cause them to feel more connected to nature (Cruz et al., 2014; Hora et al., 2020). In such cases, certain food/medicine items may be culturally important because people can experience strong emotional ties to them. Such aspects may not be properly captured by our research design, which is definitely a limitation of our study.
This study has some other limitations. The investigation would have been a stronger tool for finding drivers of plant use if the conventional approach (of directly asking people their reasons for preferring/consuming certain items) had been combined with our statistical model. Therefore, we strongly recommend this combination of approaches for future studies on this subject. Additionally, our sample of ethnospecies was small, which makes our model with two independent variables not as reliable as it would have been if it had included a high number of plants. Finally, other options to classify plants by the attributes would have been preferable over ranking. We obtained a high number of ties (i.e., people stating that several species occupied the same position in terms of taste, for example). When ties are common, scoring exercises  are much better options than rankings.