This paper reports on a study that is concerned with the cognitive
aspects of reading in a hypertext environment. The study focuses
on text based electronic documents. A cognitive model for
hypertext document reading proposed in an earlier work is here
developed and validated with the use of think aloud protocols.
Navigational strategies that readers employ in hypertext reading
and hypertext links' selection are also under study. The results
indicate that 100% of the task related data correspond to the
components of the cognitive model, allowing us to conclude that
the proposed model sufficiently describes the cognitive processes
involved in hypertext reading. In addition, three navigational
strategies are revealed: linear, mixed, and mixed review. The
quantitative data show no significant differences between
different reading goals and comprehension.