The Dynamics of Scientific Knowledge - Mapping Scientific Frontiers: The Quest for Knowledge Visualization - page 8

She emphasized that the third category is most suitable to Kuhn's view of
scientific development. Scientific knowledge grows as a result of the invention of
a puzzle-solving device that can be applied to a set of problems producing what
Kuhn has described as “normal science.”
In this topic, we will focus on puzzle-solving examples in this category. For
example, numerous theories have been proposed to explain what caused the
extinction of dinosaurs 65 million years ago; scientists are still debating on this
topic. Similarly, scientists are still skeptical about what causes brain diseases in
sheep, cattle, and human. These topics share some common characteristics:

• interpretations of available evidence are controversial

• conclusive evidence is missing

• the current instruments are limited

Mapping the dynamics of competing paradigms is an integral part of our quest
for mapping scientific frontiers. We will demonstrate some intriguing connections
between Kuhn's view on paradigm shifts and patterns identified from scholarly
publications.
Information scientists are concerned with patterns of scientific communications
and intellectual structures of scientific disciplines. Since the 1970s, information
scientists began to look for signs of competing paradigms in scientific literature,
for example, a rapid change of research focus within a short period of time. In
1974, Henry Small and Belver Griffith were among the first to address issues
concerning identifying and mapping specialties from the structure of scientific
literature by tapping on co-citation patterns as a grouping mechanism (Small and
Griffith 1974 ). In a longitudinal study of collagen research published in 1977,
Small demonstrated how collagen research underwent some rapid changes of its
focus at a macroscopic level (Small 1977 ). He used data from the Science Citation
Indexing (SCI) to group documents together based on how tightly they were co-
cited in subsequently published articles. Groupings of co-cited documents were
considered as a representation of leading specialties, or paradigms. Small used the
multidimensional scaling technique to map highly cited articles each year in clusters
on a two-dimensional plane. An abrupt disappearance of a few key documents in
the leading cluster in 1 year and the rapidly increased number of documents in the
leading cluster in the following year indicate an important type of specialty change -
rapid shift in research focus -, which is an indicator of “revolutionary” changes.
We can draw some useful insights from studies of thematic maps of geographic
information. For example, if people study a geographic map first and read relevant
text later, they can remember more information from the text (Rittschof et al.
1994 ). Traditionally, a geographic map shows two important types of information:
structural and feature information. Structure information helps us to locate indi-
vidual landmarks on the map and determine spatial relations among them. Feature
information refers to detail, shape, size, color, and other visual properties used to
depict particular items on a map. When people study a map, they first construct
a mental image of the map's general spatial framework and add the landmarks
into the image subsequently (Rittschof et al. 1994 ). The mental image integrates