Reading and understanding what we read is an essential demand of our society nowadays. Everywhere we go we are meant to understand information set all around us in very different formats and displays with also very different purposes. For example, some daily activities that require reading and understanding are checking a pay sheet, verifying the terms and conditions of a promotion or searching for information about a recent news. As we can see, reading comprehension requires not only understanding single words and sentences but moreover implies constructing and understanding language representations at different levels (phonological, semantic, syntactic, and thematic) (Graesser, 2007). The present dissertation aimed to analyze the effectiveness of different types of elaborative feedback messages to help students learn declarative content from a task involving reading a scientific expository text and answering related questions. As a starting point, in Chapter 1 we introduce the topic by presenting the theoretical framework for the basis of this work. We will start by reviewing the issue of learning by answering questions from texts and the most influential theoretical models involved in this matter. Additionally, we focus on the types of questions and their impact on learning according to previous research. We subsequently address the topic of feedback as a teaching tool were we review the types of formative feedback and previous research outcomes. We also explore text availability and response certitude as variables which impact the outcome of instructional feedback. To finish our first chapter, we include a section on how to correct misconceptions according to the recent literature. We will explore several procedures although we will focus on refutational texts and we will discuss how can feedback help in the refutational process during learning. In Chapter 2 we present the current dissertation, explaining the path followed from one study to the next one and the common procedures among the different studies. We then present our empirical studies in Chapters 3, 4, 5 and 6. The first empirical study of this dissertation (Chapter 3) aims to explore the effectiveness of two types of elaborative feedback vs. corrective feedback when students can decide to search the text for additional information after giving an incorrect answer. Chapter 4 includes a new study that replicates the previous one with a major change: searching was not available at any point of the experimental procedure. For our third study (Chapter 5) we decide to analyze only the effect of elaborated feedback messages and we create this study with three different types of elaborated feedback with the objective of identifying if it is better to direct the information in the feedback message to the right answer, to correct the mistake or using a combination depending on the score. Chapter 6 includes our last experimental study were we analyze the difference between the two most used elaborated feedback messages and we explore how response certitude impacts feedback reading times. Finally, in Chapter 7 we discuss the overall findings from these four empirical studies. Our findings throughout the dissertation outstand the importance of coherence between the students answers and the feedback message given as well as the usefulness depending on the correct or mistaken answer of the student. This coherence effect will help acquiring knowledge tested on a post test. Other variables such as searching and type of question are also discussed. Additionally, we suggest future research directions and we discuss the main practical implications of our findings for education.