Economic valuation methods are tools used to value goods and services in situations where markets fail to provide an adequate measure of value. The methods are based on the trade-offs that choosing one option over another entails. The assumptions underlying the econometric analysis of these choices treat humans as perfectly informed utility maximizers. While many economists are aware of the fact that these assumptions are overly simplistic and that humans are not always rational, the assumptions are still considered good enough and treated as the standard approach, resulting in the majority of valuations studies being conducted under these assumptions. However, a growing body of research is now explicitly incorporating the decision process into the analysis of choices, showing that relaxing the underlying assumptions regarding the decision process affect the value estimates. The four papers in this thesis fall into two categories: one paper follows the standard approach, implicitly assuming perfect rationality. The three other papers explicitly consider the decision process in the analysis of the choices. All the four papers are based on different stated choice experiments (SCE), a specific type of stated preference economic valuation method. Paper i, representing a standard approach to environmental valuation, uses an online sce to value riparian buffer strips (RBS) in Denmark. The findings indicate that the value of this agri-environmental measure not only lies in its ability to improve the condition of aquatic environment, but that it provides valuable co-benefits as well. In papers ii and iii, the decision process is explored with the use of eye-tracking. The findings in paper ii indicate that the way in which respondents acquire information in a SCE indicate their level of certainty in choice. Paper iii introduced the gaze-contingent eye-tracking paradigm to SCE, which enables the disentanglement of choice and information non-attendance, yielding insights into the operationalization of attribute non-attendance, a type of decision strategy. The findings indicate that model performance can be improved by using visual non-attendance captured by the eye tracker at the alternative level, and that controlling for information non-attendance removes choice non-attendance for some attributes. Paper iv analyses the effects of presenting information in an online SCE using narrated videos as opposed to text and illustrations. The aim is to provide guidelines on how information can best be provided in online SCEs. The results are mixed, with choices from the narrated video treatment being more accurate, however, this might be due to selective attrition, i.e. that certain respondents choose not to complete the survey. Overall, this thesis contributes to the understanding of the decision process in SCE, exploring how the decision process can be used to explain choices, as well as, how information presentation can impact the decision process and choice. These findings will hopefully provide stepping stones for future analyses of the decision process, which, in turn, will pave the way for more realistic behavioral models in SCE, thus improving the validity and reliability of SCEs.