There seems to exist a general belief that one needs to learn specifically how to read mathematical texts, that is, a need to develop a special kind of reading ability for such texts. However, this belief does not seem to be based on research results since it does not exist much research that focus on reading comprehension in mathematics.

The main purpose of this dissertation is to examine whether a reader needs special types of knowledge or abilities in order to read mathematical texts. Focus is on students’ reading of different kinds of texts that contain mathematics from introductory university level. The reading of mathematical texts is studied from two different perspectives, on the one hand a cognitive perspective, where reading abilities and content knowledge are studied in relation to reading comprehension, and on the other a metacognitive perspective, where focus is on beliefs and how a reader determines whether a text has been understood or not.

Three empirical studies together with theoretical discussions, partly based on two literature surveys, are included in this dissertation. The literature surveys deal with properties of mathematical texts and reading in relation to problem solving. The empirical studies compare the reading of different types of texts, partly mathematical texts with texts with content from another domain and partly different types of mathematical texts, where focus is on the use of symbols and texts focusing on conceptual or procedural knowledge. Furthermore, students’ beliefs about their own reading comprehension and about texts and reading in general in mathematics are studied, in particular whether these beliefs are connected to reading comprehension.

The results from the studies in this dissertation show that the students seem to use a special type of reading ability for mathematical texts; to focus on symbols in a text. For mathematical texts without symbols, a more general reading ability is used, that is, a type of ability also used for texts with content from another domain. The special type of reading ability used for texts including symbols affects the reading comprehension differently depending on whether the text focuses on conceptual or procedural knowledge. Compared to the use of the more general reading ability, the use of the special reading ability creates a worse reading comprehension.

There seems to exist a need to focus on reading and reading comprehension in mathematics education since results in this dissertation show that courses at the upper secondary level (course E) and at the university level (in algebra and analysis) do not affect the special reading ability. However, the mentioned results show that this focus on reading does not necessarily need to be about learning to read mathematical texts in a special manner but to use an existing, more general, reading ability also for mathematical texts.

Results from the metacognitive perspective show a difference between conscious aspects, such as regarding beliefs and reflections about comprehension, and unconscious aspects, such as the more automatic processes that make a reader understand a text, where also metacognitive processes are active. In particular, beliefs, which have been examined through a questionnaire, do not have a clear and independent effect on reading comprehension.

From the texts used in these studies and the participating students, there seems not do be a general need to view the reading of mathematical texts as a special kind of process that demands special types of reading abilities. Instead, the development of a special type of reading ability among students could be caused by a lack of experiences regarding a need to read different types of mathematical texts where similarities with reading in general can be highlighted and used.