This paper was presented at the Association for Education in Journalism and
Mass Communication in San Antonio, Texas August 2005.
If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
rakyat [ at ] eparker.org. For an explanation of the subject line,
send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").
Advanced Learning and Online News: A Test of Cognitive Flexibility Theory
Department of Journalism
The University of Alabama
Tuscaloosa, AL 35487
[log in to unmask]
Kyun Soo Kim
The University of Alabama
Manuscript submitted to the Mass Comm and Society Division, AEJMC 2005
For years online journalists and new media scholars have called for
news organizations to take greater advantage of the Web's unique
qualities – i.e., to make online news more interactive, experiential
and nonlinear (Lasica, 2003; Lanson, 2003; Paul & Fiebich, 2005).
With some notable exceptions, the news industry has turned a deaf
ear, continuing to repurpose content and doing little to innovate new
formats (Lowrey, 2003; Singer, 2003; Randazzo & Greer, 2002). Today
the call for new news forms has been renewed, as institutional
journalism receives an unprecedented challenge from bloggers, who
have taken nonlinearity and associative linking to new levels by
establishing vast, widely dispersed networks of content expertise
(Lasica, 2003; Gillmor, 2003; Rosen, 2005).
Scholarship on Internet news has increasingly focused on new story
forms (Eveland, Cortese, Park, & Dunwoody, 2004; Eveland, Marton, &
Seo, 2004; Eveland & Dunwoody, 2001, 2002; Eveland, Seo, & Marton,
2002; Paul & Fiebich, 2005; Lowrey, 2004). Recently communication
scholars have moved beyond studies of simple effects from story forms
on recall and recognition to more complex studies involving
interaction effects (Southwell & Lee, 2004; Eveland et al., 2002,
2004; Lowrey, 2002; Lee & Tedder, 2003) and impact on advanced
learning, such as the ability to associate concepts (Eveland et al.,
2004; Eveland et al., 2004) and the ability to apply knowledge in
"real world" settings (Lowrey, 2004). The present study extends this
work on advanced knowledge by employing cognitive flexibility theory
(CFT) from the field of education psychology. The theory posits that
when case examples from a knowledge area are interwoven with
conceptual perspectives, cognitive structures should be more flexible
and interconnected, and learning should be more easily applied across
diverse settings. Online news stories with flexible structures should
give readers more thorough and sophisticated understandings of news
issues and therefore a greater ability to successfully apply
knowledge. However, it is not expected that the CFT format will be
entirely beneficial. Prior studies have shown that the increased
complexity of the CFT navigation scheme may burden memory processing,
thus decreasing simple recall and recognition.
Hypermedia and Knowledge Acquisition
Initial studies of online news stories assessed impact of
nonlinearity on knowledge acquisition, an emphasis that reflected
work on cognitive effects of TV editing (e.g., Lang, Zhou, Schwartz,
Bolls, & Potter, 2000) but which was drawn more directly from studies
of hypermedia in education psychology (Graff, 2003; Brunken, Plass, &
Leutner, 2003; Niederhauser, Reynolds, Salmen, & Skomolski, 2000;
Dee-Lucas & Larkin 1995; Dillon & Gabbard, 1998). The study of
hypermedia in education was originally influenced by hypertext
scholars who viewed nonlinear, associatively linked content as
reflective of the way human memory works (Jonassen, 1988; Nelson &
Palumbo, 1992). Eveland and Dunwoody (2001) have labeled this quality
"structural isomorphism." It was thought that hypermedia would
naturally aid learning, but results from experiments showed that this
premise, despite its face validity, was perhaps too simplistic
(Dillon, 1996). In most experiments, nonlinearity correlated with low
memory recall and recognition or there was no correlation
(Niederhauser et al., 2000; Brit, Rouet, & Perfetti, 1996; McKnight,
Dillon, & Richardson, 1990). Researchers began to understand that in
some cases the nonlinear structure caused disorientation
(Niederhauser et al., 2000).
One explanation for disorientation is that hypermedia's unfamiliar
structures and navigation schemes cause cognitive load. According to
cognitive load theory, information overload puts stress on working
memory, which leads to cognitive disorientation. This in turn causes
problems for the development of long-term memory (Sweller, 1988;
Cooper, 1990). A number of factors can increase cognitive load,
including lack of expertise with the content (Gray, 1990), confusing
navigation schemes ("scaffolding," is the term used in the education
literature) (Niederhauser et al., 2000; McKnight, Dillon, &
Richardson, 1990), and erosion of learning efficacy due to loss of
perceived control over the reading experience. Research suggests that
successful learning from hypermedia requires that readers be
carefully oriented to navigation structure prior to use (Chou, Lin, &
Sun, 2000; Curry, Haderlie, Ku, Lawless, Lemon, & Wood, 1999).
Learning is also aided by specific learning goals (Dee-Lucas, 1996).
Results of studies of the impact of perceived control are mixed.
Readers judge hypermedia more favorably if they think they have a
high degree of control over the reading experience (Hannafin &
Sullivan, 1996; Morrison, Ross, & Baldwin, 1992), but study findings
on the relationship between control and learning are inconclusive
(Becker & Dwyer, 1994).
Communication researchers have also assessed cognitive impacts of
nonlinear, hyperlinked media formats. Generally, results have shown
that readers recall and recognize information from traditional linear
formats better than from nonlinear formats. Tewksbury and Althaus
(2000) found that readers of a newspaper print edition remembered
news content better than readers of a hyperlinked Web format, and
they cited a reduction in familiar "importance cues" as a likely
explanation. Eveland and Dunwoody (2001) compared a print version of
a news story with linear and nonlinear Web versions of the same story
and found no significant differences between the Web sites in
subjects' ability to recall story information. In a separate study
they found that readers of the nonlinear format were more likely to
skip information they found less interesting, which led to reduced
learning (Eveland & Dunwoody, 2002). Results from an experiment by
Lowrey (2004) showed that nonlinear formats increased perception of
control over the reading experience but did not aid recognition
memory, and in fact discouraged opinion feedback from readers.
Southwell and Lee (2004) found that giving viewers control over the
order of video clips decreased recognition.
Communication scholars have begun to study effects on advanced forms
of knowledge, such as concept linkage and application of knowledge.
The degree to which concepts are interconnected, or "knowledge
structure density," is particularly important from the perspective
that hypermedia reflect the natural structure of the human mind. If
the structures of hypermedia and the mind have an isomorphic
relationship, then the interconnectedness of concepts in nonlinear,
associatively linked Web sites should encourage development of denser
knowledge structures in readers' minds. Eveland et al. (2004) tested
this impact by having subjects recall concepts from linear and
nonlinear online readings and then rate the strength of relationships
between concepts. They found that the KSD of readers of nonlinear
content was slightly greater than the KSD of readers of linear
content (Eveland et al., 2004), and this relationship was stronger
among frequent Web users (Eveland et al., 2004).
Cognitive Flexibility Theory
The theory of cognitive flexibility from education psychology is
consistent with the idea that hypermedia structure can shape
knowledge structure. Cognitive Flexibility Theory was designed to aid
the instruction and application of complex knowledge areas – i.e.,
knowledge areas that must be learned across different sorts of
real-world cases and from different theoretical perspectives. For
example, a cardiologist must flexibly apply medical knowledge of
heart conditions across a variety of physical conditions. An engineer
must apply knowledge of road construction across a variety of
geographical conditions and local regulations and laws. In such
situations, rote memorization of a technique is not enough. Through
this crisscrossing of the conceptual landscape, advanced learners can
build interconnected and flexible cognitive structures rather than
rigidly prepackaged schema (Spiro, Feltovich, Jacobson, &
Coulson,1991; Spiro & Jehng, 1990). Hypermedia typically aid
CFT-based instruction because of their nonlinearity. They allow
learners to easily and quickly compare different cases from a
knowledge area, to assess cases in light of different conceptual
perspectives, and to revisit these cases and concepts.
Results from tests of the theory show that the CFT format does not
aid rote memory, but it can increase ability to apply knowledge to
complex situations. In an early CFT experiment, Spiro and colleagues
(1991) used the theory to create a hypermedia program to teach theory
about the film Citizen Kane. Video film clips served as the "cases,"
and students navigated from case to case, as well as to different
film theories, comparing and contrasting cases in light of the
theories. Results showed that the CFT format did not aid recall, but
it did help students apply knowledge to problem-solving situations
(Spiro et al., 1991). CFT has proven helpful in aiding understanding
of complex, case-specific problems (Demetriadus & Pombortsis, 2000;
Spiro et al., 1991), and readers say they enjoy reading content in
CFT formats (Siegel, Derry, Kim, Steinkuehler, Street, Canty,
Faasnacht, Hawso, Hmelo, & Spiro, 2000). Consistent with findings
from other research on hypermedia programs, CFT research shows that
design must be carefully tailored for nonlinear formats to have the
desired impact because of a lack of familiarity with navigation cues
(Godshalk, Harvey & Moller, 2004). Without the proper design, readers
may ignore the hyperlinked conceptual perspectives (Siegel et al.,
2000), and they may fail to grasp the "big picture" of the course
content (Rossner-Merrill, Parker, Mamchur, & Chu, 1998). As Spiro et
al. (1991) found, results also show that the CFT hypermedia format
have little or no effect on memory recall and recognition
(Balcytiene, 1999; Demetriadus & Pombortsis, 2000), and it may
actually hinder cognitive processing by adding cognitive load
(Niederhauser, Reynolds, Salmen, & Skolmoski, 2000).
Lowrey (2005) tested the impact of the CFT format on advanced
knowledge by assessing readers' perceived abilities to apply news
knowledge in public discussion of issues and their ability to see the
gray areas of issues. Readers' memory of content and their attitudes
about using the format were also assessed. Consistent with past CFT
research, results showed that readers enjoyed using the CFT site more
than linear and traditional nonlinear versions of the same story.
Also consistent with past findings, there was no significant
relationship between format and recognition memory. However, there
were only weak effects on readers' perceived ability to engage in
public discussion about the topic. Prior knowledge of the story topic
had a significant moderating effect. CFT readers with higher prior
knowledge were less likely to have extreme opinions on the topic,
showing that CFT format communicates complexities and gray areas.
Those with higher prior knowledge also were more likely to scan the
CFT format, but to read the linear story in-depth. In contrast, those
with less prior knowledge were not likely to use the CFT story much
differently than they used the nonlinear and linear stories.
The present study builds on previous findings by testing new
dependent variables and moderating factors, and by including older
non-students as research participants. Primarily, the study seeks to
test the impact of a CFT hypermedia format on advanced knowledge
acquisition and in doing so, to assess the impact of key moderating
variables. The study asks, to what degree does the CFT format affect
the ability to interrelate cases and concepts, and the ability to
"transfer" or apply knowledge to real-world situations? It is assumed
here that the real-world application of journalism knowledge is
public discussion of public issues. The study also assesses
potential moderating effects from knowledge of news story content,
involvement with news story content, and degree of Web use.
Dependent variables are grouped into three categories: (1) advanced
learning (2) memory processing and (3) site use.
Advanced learning involves both the ability to interconnect concepts
and to apply knowledge. Hypermedia theory suggests a relationship
between the structure of the human mind and the structure of
hypermedia (what Eveland and colleagues call "structural
isomorphism"). Nonlinear hypermedia formats, in which concepts are
interconnected via hyperlinks, are said to be similar to the
interrelatedness of knowledge nodes in human memory. Though some
education scholars think this is an oversimplified representation,
the idea has received moderate support in recent empirical research
(Eveland et al., 2004). It should be the case therefore, that
exposure to online content in which there are many hyperlinked
connections among concepts should lead to a greater ability by
readers to make mental connections among concepts. This literature
suggests this should be the case for nonlinear content in general,
and not only for the CFT format, a particular type of nonlinear format.
H1a: Readers of nonlinear Web sites will make more connections among
concepts from the reading than will readers of linear Web sites.
Cognitive flexibility theory has specific applications for advanced
learning. According to CFT, hypermedia formats that crisscross
conceptual frameworks with multiple case examples of a knowledge area
help readers build schematic structures that are flexible enough to
adjust to the variations of real-world settings. It should be the
case, therefore, that a Web site that facilitates cross-referencing
of cases and conceptual perspectives should facilitate application of
knowledge to real world settings. For journalism, the "real world"
application of knowledge is presumed to be public discussion of
content from news stories.
H1b: Readers of Web sites designed for cognitive flexibility will be
more able to engage in public discussion of issues than readers of other sites.
Though it is often predicted that nonlinear formats will aid memory
because associative linking will foster elaboration (Eveland &
Dunwoody, 2002) and increase perceived control over the reading
experience (Hannafin & Sullivan, 1996; Southwell & Lee, 2004), most
studies have found the opposite to be true, as readers of the CFT
format and other nonlinear formats tend to remember less, or no more,
than readers of linear formats (Eveland & Dunwoody, 2002, ; Lowrey,
2004; Niederhauser et al., 2000; Southwell & Lee, 2004). Because
expectations are mixed, and because effect on memory is obviously
important to learning, the impact of format on recognition memory,
recall memory and cognitive load will be tested. Recognition memory
is a check on familiarity. Individuals are provided the information
and tested to see if it is recognized from memory. Recall, which
involves coming up with the information from memory on one's own, has
proven more difficult to achieve. Both of these measures have been
tested in hypermedia studies, and generally, nonlinear formats have
not proven helpful to either (Brit, Rouet, & Perfetti, 1996;
McKnight, Dillon, & Richardson, 1990; Lowrey, 2004, 2005; Eveland et
al., 2002, 2004; Southwell & Lee, 2004). In looking for explanation,
researchers have pointed to the burden of unfamiliar site navigation.
According to the limited capacity model (Lang, 2000), the mind can
only process so much information before working memory becomes
stressed. This phenomenon, called cognitive load, leads to
disorientation, which in turn causes problems for the development of
long-term memory (Sweller, 1988). Based on past studies then, it is
expected that the CFT format will increase cognitive load and hinder
recall and recognition memory – that is, cognitive processing or
H2: Readers of Web sites with a CFT format will find memory
processing more difficult than readers of other sites.
Prior research shows that nonlinear sites are used differently from
linear sites. Readers of nonlinear stories tend to scan content
rather than read in-depth (Eveland & Dunwoody, 2002), and this is
particularly true for readers who have a greater degree of prior
knowledge about the story content (Lowrey, 2005). The impact on time
spent on reading has not been thoroughly explored in communication
research, but it should follow that readers would tend to spend less
time with nonlinear stories if they are more likely to read these
stories less thoroughly. It should be the case therefore, that users
of CFT stories, which are designed to be read in a flexible, even
adventurous manner, will spend less time with the stories and will
be more likely to read them selectively – i.e., scan them rather than
read in-depth. In other words, they will read the stories less thoroughly.
H3: Readers of Web sites designed for cognitive flexibility will read
less thoroughly than readers of other sites.
Finally, prior research on nonlinear sites has shown the importance
of moderating variables. Degree of prior knowledge of a content area
has been most often studied, though with mixed results. Studies have
shown that prior knowledge can aid learning from hypermedia (Gray,
1990), and can increase the benefits of control over the reading
experience (Shyu & Brown, 1995). However, other studies have shown
that those with lower prior knowledge may actually benefit more from
hypermedia (Balcytiene, 1999), and that prior knowledge increases
likelihood of scanning nonlinear formats (Lowrey, 2005).
The degree to which readers are motivated by or feel involved with
the reading experience has also been studied. Generally, increased
involvement and motivation lead to increased learning from hypermedia
(Jonassen & Grabinger, 1993), and assignment of specific tasks prior
to reading also benefit learning (Curry et al., 1999; Dee-Lucas,
1996), though this finding is not unanimous (Foltz, 1996).
Degree of experience using the Web is another frequently studied
variable. Generally, research shows that increased experience with
hypermedia improves learning from nonlinear formats, though results
are somewhat mixed (Eveland et al., 2004; McKnight, Dillon, &
Richardson, 1990; Rouet, 1990). Similarly, research shows that when
readers are given clear and specific guidance for use of a hypermedia
format, learning improves (Dee-Lucas & Larkin, 1995; Jacobson,
Maouri, Mishra, & Kolar, 1995).
In sum, research shows that prior knowledge, level of involvement
and prior experience with hypermedia can shape the impact of format
on learning and use, but results from these studies are mixed.
Therefore, a research question is posed rather than a hypothesis.
RQ: What impact do degree of prior knowledge, degree of prior
involvement and degree of prior hypermedia use have on the
relationship between story format and advanced learning, memory
processing and use?
Research participants include 106 undergraduates and 78
non-undergraduates over the age of 23, most of whom were recruited
from the local community. Students were undergraduate communication
majors and were recruited from mass communication classes with an
incentive of extra credit. Ages ranged from 18 to 25, with 88 percent
being 21 or under. Older participants were recruited through
advertisements in the community newspaper and flyers posted on the
researcher's campus and in the larger community. Older participants
were offered an incentive of $25 for full participation. Ages ranged
from 24 to 70, with half being over the age of 40. Experiments with
students took place in summer and fall 2003, and experiments with
older participants took place in fall 2004. In each of the
experimental sessions, written consent was obtained, and participants
were randomly assigned to three groups, one for each of three
Design and Procedure
For the three conditions of format type, three Web-based versions of
a news report on the issue of cloning were created. The first report
was created with a structure informed by cognitive flexibility
theory, the second was created with a general nonlinear structure,
and the third was created with a linear structure. Reports contained
identical photos and graphics, and text was close to identical, with
some adjustments necessary for format changes. Because prior research
on hypermedia has demonstrated the importance of orienting users to
navigation (Dee-Lucas & Larkin, 1995; Jacobson, Maouri, Mishra, &
Kolar, 1995), the home pages of all three sites contained limited
guidance on site use. No verbal guidance was offered, as this would
have diminished external validity.
The CFT news report contained five short news stories, each 200 to
250 words in length. Brief "perspective" paragraphs, between 30 to
100 words each, were linkable from each of the news stories. Readers
first encountered a home page with a headline across the top of the
page, and down the left side below the headline were five story
"teases" and accompanying photos that linked to the five news stories
(Figure 1). When a tease or accompanying photo was clicked, the full
story loaded in a column running down the center of the page. A brief
line of instruction was given across the top of the home page, which
read "Choose the brief stories on the left and then read different
perspectives on each story." In the space in the center of the page
where stories appeared, another cue was provided: "Click on the
stories on the left and read them here." Each story served as one of
the multiple cases for the learning experience. Jump menus appeared
at the top and bottom of each full story with the label "Choose a
perspective", allowing readers to link to the four "perspectives" on
each story. When linked, a few sentences offering interpretation of
the story from a particular perspective appeared in a column running
down the right side of the page. A line of instruction appeared in
this right-hand column, which read "Perspectives on (name of story
clicked) appear here." By hyperlinking among stories and
perspectives, readers are able to crisscross the conceptual landscape
of the news report.
A nonlinear site that did not have the CFT format was also assessed.
This format was tested because it is similar to nonlinear story
formats commonly used for magazine-style "Web specials" in which a
story is segmented into components and interconnected through
hyperlinks and a navigation bar (Figure 2). Also, previous studies of
nonlinear stories have adopted similar formats (Eveland & Dunwoody,
2001; Lowrey, 2004).With this format, readers could begin reading at
any of the six story segments, which were written to stand alone, and
could read the segments in any order. In addition, links embedded in
the story text allowed users to link across pages within the site.
Because all sites needed to have the same amount of content, no links
to external Web sites were offered.
The linear format of the report consisted of four pages linked
together so readers could move forward and backward but could not
skip around in the page order. The only links appeared in the form of
backward and forward arrows at the bottom of each page, and there
were no embedded links. The format reflected the format of a lengthy
"repurposed" news story, used routinely in many Web sites. Photos and
graphics were indented into the text at points where graphic content
reflected story content (Figure 3).
The issue of cloning was chosen as the topic of the story because it
is a complex subject with various competing frames of reference. The
issue has been debated at the political level, the religious and
ethical level, and at the scientific level. The issue of cloning also
offers a variety of case examples. For the present study, two cases
involved cloning for the purpose of medical research, one case
involved cloning attempts to help infertile couples, one case
involved cloning to produce a racing mule, and one case involved a
religious sect's attempt to achieve immortality by cloning human
babies. Readers were able to examine each story in light of four
different angles, or perspectives: a politics/legislation
perspective, a science perspective, an ethics perspective and a media
portrayal perspective (which pertains to the implications of public
Text for each perspective provided interpretation or implications for
each cloning case. For example, for the case involving cloning for
infertile couples, the clickable "Ethics" perspective offered
observations from two expert sources: "Leon Kass, chairman of the
President's Council on Bioethics, says infertile parents should not
recreate lost children. 'It's a sign of our growing despotism over
the next generation,' Kass said. 'Cloning introduces the possibility
of parents making choices for their children far more fundamental
than whether to give them piano lessons or straighten their
teeth.' But professor of philosophy Gregory Pence says people have
always had self-serving reasons for having children 'whether to
ensure there's someone to care for them in their old age or to relive
their own youth vicariously.'" For the same case, the "Politics"
perspective offered the observation that "it has been politically
difficult to ban cloning that produces children while allowing
cloning for medical research. Supporters of reproductive cloning have
found this political confusion helpful and have tried to entangle the
two even more."
The experiment was administered to students in a department computer
lab. After random assignment to the three experimental conditions,
each subject was seated at a computer terminal and instructed to fill
out a pretest survey containing 22 questions. Questions addressed
demographics, use and expertise with the Web, and perceived expertise
and involvement in the issue of cloning. Subjects were then asked to
open the browser windows on their computer screens and read the
story. Subjects were given 20 minutes to read the story but were told
they could stop reading at any time. Time spent reading was measured
and recorded by research assistants. Four minutes into the reading,
subjects were asked to fill out a brief three-question survey
measuring cognitive load, and they were then told to resume reading.
At the end of the 20-minute reading time subjects were told to close
the browser and were given a 32-item post-test questionnaire. The
questionnaire measured the dependent variables Self-efficacy,
Elaboration, Format Understanding, Recognition Memory, Recall Memory,
Cognitive Load (for a second time) and Selectivity of Reading.
Older participants were then given a questionnaire that measured
knowledge structure density (KSD was not measured in the earlier
student experiments). In this questionnaire, participants were asked
to list concepts recalled from their reading along the top of a grid,
and then to list these same concepts down the left side of the grid.
They were then asked to rate the strength of connectedness between
concepts by marking a number between 1 and 7 in the cells at which
concepts met, with 1 being very weakly connected and 7 being very
Finally, both students and older participants were given a fourth
questionnaire, in which they were asked to read two brief cloning
cases, to provide opinions on these cases and to write reasons for
their opinions. Prior to writing reasons, participants were told they
would share their reasons with the person next to them. After writing
reasons, brief discussions were held among participants. Participants
were told about these discussions ahead of time so that as much as
possible, the reasons participants provided would represent
application of knowledge in a real-world setting. That is, reasons
would be produced with the knowledge that they would be used in
actual public discussion.
Measures of variables.
Advanced Learning measures
In order to test impact of conceptual interconnectedness, as
suggested in the first hypothesis, the ability of readers to
elaborate on the readings was assessed. Elaboration takes place when
new information is interconnected with related concepts in the
reader's mind. Elaboration therefore serves as a measure of concept
connectedness. Hyperlinked concepts and perspectives in the CFT
format should encourage elaboration, and as readers consider
relationships and linkages, learning should increase (Eveland &
Dunwoody, 2002). Elaboration was measured through three statements:
"I thought about how what I read related to other things I know," "I
thought about what actions should be taken by policy-makers based on
what I read," and "I found myself making connections between the news
report and what I've read and heard about elsewhere. (Strongly
Disagree = 1, Strongly Agree = 7). These measures had an alpha of .72
and were summed to form the variable elaboration (M = 15.12, SD =
4.08). These measures were used with success by Eveland & Dunwoody (2002).
The variable Knowledge Structure Density, also a representation of
conceptual interconnectedness, was measured by giving respondents a
matrix. Participants were asked to recall concepts from the reading
and to list these concepts across the top of the matrix, and then to
list the same concepts down the left side of the matrix. Participants
then rated the degree to which concepts from the readings were
related (0 = not related, 5 = very closely related). A formula
adopted from education psychology (Wasserman & Faust, 1994), and also
employed by Eveland, Marton, & Seo, (2004), was used to calculate the
degree to which concepts (or "nodes," to use the language of social
network theory) are interconnected. The formula divides the sum of
the participants' ratings of relational proximity by n(n-1)/2, where
n is the number of concepts participants recalled. The result is a
measure of the density of the participants' knowledge structure, or
the degree to which ideas and concepts are associated. Number of
concepts recalled from the reading was also used as a second measure
of memory recall, but only for the older non-student participants, as
the undergraduates were not given this questionnaire.
As posited in Hypothesis 1b, another component of advanced learning
relates directly to cognitive flexibility theory – the ability to
transfer, or apply knowledge of complex content to real-world
settings. As discussed earlier, it is assumed that public discussion
of issues is the real-world application of journalism knowledge.
Presumably increased ability to discuss issues corresponds with
perceived ability to discuss issues. Therefore Self-efficacy of
ability to apply journalism knowledge in public discussion was
measured by four items, each on a 7-point scale: "I feel I could
understand a debate by policy makers about human cloning," "I feel I
could carry on an intelligent conversation about the topic of human
cloning," "I feel I could make an effective argument to defend my
opinions about human cloning," and "I feel I could successfully
answer questions about the story I just read." Previous research on
self-efficacy suggests this focus on measuring "feeling of
confidence" (Bong & Hocevar, 2002). These four items were tested for
reliability (Cronbach's alpha = .78) and were summed, with a range of
4 to 28 (M = 18.79, SD = 4.67).
The variable reasons for discussion measured readers' preparedness
for applying news knowledge in public discussion, as it was predicted
that the CFT format would encourage ability to apply knowledge in
real-world settings. As mentioned, research participants were asked
to give answers to two opinion questions, they were told to write
between zero and five reasons for each opinion, and they were told
before writing that they would be discussing these reasons with
others. The variable was measured by summing the number of reasons
for the two questions (M = 4.66, SD = 1.88).
Memory processing measures
The recognition measure was a sum of correct answers (each correct
answer = 1) for 14 multiple choice and true-false questions about the
cloning story. The final measure ranged from 3 to 14, M = 9.36 and
s.d. = 2.47. Memory recall was measured through five
fill-in-the-blank questions. Correct answers (scored as 1) were
summed, and the measure ranged from 1 to 5, with M = 1.60, s.d. =
1.30. Cognitive load was measured by three 7-point items: "How
difficult was it to follow what this news report was about?"
(Extremely easy = 1, Extremely difficult = 7), "I felt lost reading
this news report" (SD = 1, SA = 7) and "It was clear how all the
information in the news report fit together" (SA = 1, SD = 7). The
three questions were administered twice – once five minutes into the
reading period and once immediately after the reading period. Similar
measures have been used in previous studies (Gellevij, van der Meij,
de Jong & Pieters 2002; Eveland & Dunwoody, 2001). Results were
summed (Cronbach's alpha = .77) and ranged from 6 to 33, M = 13.89,
s.d. = 6.88. A higher score indicated greater cognitive load.
Reported selectivity of reading, a measure of "thoroughness of
reading," as stated in Hypothesis 3, was measured with two 7-point
items: "I skimmed through the news report" (Strongly Disagree = 1,
Strongly Agree = 7) and "I only read sections of the news report that
looked interesting or important" (SD = 1, SA = 7). These measures
were used reliably by Eveland and Dunwoody (2002). Results were
summed (Cronbach's alpha = .78) and the final variable ranged from 2
to 14, M = 4.57, SD = 3.11, with a higher score indicating more
selectivity, or scanning.
Time of reading was measured by noting the time to the nearest minute
that participants finished reading. All participants started reading
at the same time, this time was noted, and participants were asked to
signal the researcher upon finishing.
Independent variable measures
The independent variable format type had three values, with 1=
linear format, 2 = nonlinear format and 3 = CFT format. Prior
knowledge had two values, with 1 = low prior knowledge of the cloning
issue and 2 = high prior knowledge. Prior knowledge questions were
asked on the pretest: "How knowledgeable are you about the issue of
human cloning?" and "How extensively have you studied the issue of
human cloning?" These two 7-point items were summed (Cronbach's alpha
= .83) and dichotomized at the median. The Web use variable was also
dichotomized, with 1 = low Web use and 2 = high Web use. Participants
were asked on the pretest how many days out of the past 30 days they
had used the Web, and how many hours they used the Web on weekdays
and on weekends. Hours were multiplied by days, and the sample was
split at the median score. Finally, Involvement in the cloning issue
was assessed through nine measures, each on a 7-point scale. Measures
were adopted from Zaichowsky's (1986) Personal Involvement Inventory.
The nine measures were summed and then split along the median to
produce a dichotomized measure, where 1 = low involvement and 2 =
Multivariate Analysis of Variance was performed in order to test the
relationships of story format type and interaction variables on the
three sets of dependent measures. First, the main effect of story
format was assessed. According to all four algorithms commonly used
in MANOVA (Wilks', Hotelling's, Pillai's, Roy's), format type had no
significant main effect on the advanced learning variables and no
significant main effect on the memory processing variables. Therefore
Hypotheses 1 and 2 were not supported in this overall analysis.
However, Hypotheses 3 received support as there was an overall main
effect on the use variables, primarily due to the impact on time of
use [F (2, 178) = 5.9, p < .01]
Univariate F tests were conducted to assess individual measures of
advanced knowledge, memory processing and use (Table 1). Hypothesis
1a again received no support, as the advanced knowledge variables,
when tested as a group, revealed no differences by format, and none
of the individual measures registered differences. Therefore these
results provide no support for the notion that interconnectedness of
concepts in online content translate to increased association of
concepts within readers' minds. Hypothesis 1b also received no
support, thus undermining the claims of cognitive flexibility theory.
The crisscrossing of cases and perspectives facilitated by the format
has no main effect on ability to connect concepts or to apply content
in public discussion.
Results of univariate tests of the memory processing variables
indicate that readers of linear stories are significantly more likely
to recall concepts from the reading than are readers of the nonlinear
formats [F = 4.0, p<.05], thus lending limited support to Hypothesis
2. There was no significant difference between the nonlinear and the
CFT formats in number of concepts recalled. As previously mentioned,
this measure was only administered to older participants, as it was
part of the knowledge structure density assessment. No other
individual measures of memory processing showed main effect
differences among the formats.
As mentioned, Hypothesis 3 was supported, primarily through the
measure of time on task. Readers spent significantly more time on the
linear and nonlinear sites than on the cognitive flexibility story.
Interestingly, however, readers did not report scanning the CFT story
more than the other two formats. This suggests many readers of the
CFT format did not know they had missed pages, despite the fact that
they were given brief instructions on the home page, which stated how
to link to stories and "perspectives" briefs, and which indicated
where these stories and briefs would load on the page. This problem
with the CFT format has been found before by Siegel et al. (2000) in
their tests of the format for educational purposes. As Tewksbury and
Althaus (2000) surmised, lack of familiarity with cues in a format
may lead readers to misread content, or at least read it less
thoroughly. Eveland & Dunwoody (2002) also found that readers tended
to ignore certain pages in the nonlinear format, but in the present
study readers of the traditional nonlinear format spent the same
amount of time reading as readers of the linear format.
After main effects were assessed, the impact of three interaction
variables were tested: Amount of time participants typically spend on
the Web, degree of prior knowledge about the issue of cloning, and
level of involvement in the issue of cloning (Significant results
shown in Table 2).
First, the impact of time spent on the Web was assessed. If lack of
visual cues disoriented readers of the CFT format, it seems
reasonable to suggest that more experienced Web users would be less
likely to miss pages because they would have developed a more
sophisticated understanding of online navigation. However a MANOVA
testing possible interaction effects of Web time on the relationship
between format and advanced knowledge showed no overall difference on
this group of variables, and no univariate effects.
However, there was a slight overall interaction effect of Web time on
memory processing variables according to the Roy's Largest Root
algorithm [F = 2.40, p<.10], primarily due to a significant
relationship with the memory recognition variable [F = 3.2, p<.05].
Findings show that readers with more Web experience recognize more
concepts from reading the CFT story than readers with less Web
experience. At the same time, readers with more Web experience
recognize fewer concepts from reading the linear story than readers
with less Web experience. This finding is in line with expectations
from cognitive flexibility theory and with prior research on
nonlinear formats. Greater Web use should lead to greater familiarity
with online formats, which should in turn allow these more expert Web
users to explore content in the unfamiliar CFT format more
successfully. The literature suggests that cognitive load should
explain this finding, as less expert Web users would be struggling to
process unfamiliar navigation cues in addition to story content.
However, results show no significant differences in the cognitive
load measure between high and low expert Web users. Again, this seems
to point to the possibility that readers were not aware they were
missing content. Less expert Web users missed pages, and therefore
remembered less, but it may be that they did not report stress during
reading because many were unaware of the site's greater complexity.
Prior knowledge of cloning was also assessed as an interaction
variable. Prior knowledge had no significant impact on the
relationship between format and the advanced knowledge variables, or
on the relationship between format and the memory processing
variables. Prior knowledge did interact with the impact of format on
use variables [F (5,175) = 3.74, p<.01]. Readers with high prior
knowledge spent more time on the CFT story than readers with low
knowledge, whereas readers with high knowledge spent slightly less
time with the other two formats than readers with low knowledge.
Seemingly then, the CFT format encourages those with greater
familiarity with an issue to explore cases and perspectives more
persistently. At the same time, the CFT format does not entice those
with low prior knowledge to explore the content. Low knowledge CFT
readers spent much less time with the content than low or high
knowledge readers of the linear and nonlinear stories.
Finally, level of involvement with the cloning issue was assessed as
an interaction variable. Involvement did not interact with the effect
of format on memory processing measures, but it did interact with the
effect of format on advanced learning variables [F = 2.7, p<.05].
Univariate tests show that high involvement readers of the CFT format
scored higher on the elaboration measure than low involvement
readers. They also scored higher than low or high involvement readers
of the other two formats. Apparently the CFT format leads readers to
connect concepts in the reading with existing concepts in the mind.
This is the only evidence in the study that the CFT format
There is little support for the supposition that the CFT format,
designed to interlink perspectives and cases, directly increases
ability (or perceived ability) to apply knowledge. And there is
little support that the CFT format directly strengthens
interconnections among concepts in readers' minds. This study does
reinforce previous findings that nonlinear formats in general do not
aid recall or recognition, and there are some notable interaction effects.
There is a substantial difference across formats in time spent on the
reading. However, it is surprising that there is no difference in
cognitive load among formats, and that there is no change in
selective reading (degree to which a story was scanned),. A possible
explanation is that readers missed pages in the CFT readings and were
simply unaware that they had done so, thus causing little cognitive
stress. Some previous CFT research has produced similar findings
(Siegel et al., 2000).
The study suggests the importance of interaction effects in
hypermedia research, a point made by a number of hypermedia
researchers (Southwell & Lee, 2004; Lee & Tedder, 2003). Readers with
more Web experience exhibited better recall with CFT formats, readers
with more prior knowledge spent more time with CFT formats, and
readers with higher issue involvement exhibited greater elaboration
with CFT formats. Conversely, the CFT format was detrimental to those
with less knowledge, less involvement and with less Web expertise.
These results suggest the CFT format should not be used to introduce
new issues to readers but could be helpful if used as a supplement
for those who are already familiar with an issue and are ready for
deeper, more complex treatments of issues. Findings also suggest the
CFT format may aid learning of complex issues that that have been
publicly prominent, frequently covered or highly resonant, as
readers' conceptual structures for these issues are likely to be more
developed. For example, a CFT Web site that explores U.S. involvement
in Iraq and the Middle East – a complex issue with high public
salience – may be more beneficial than a site on cloning. Cognitive
flexibility theory has been envisioned as a way to support advanced
learning rather than to introduce a knowledge area (Spiro & Jehng,
1990), and findings here reinforce this vision. Clearly, however,
another way should be found to orient infrequent Web users to the CFT
format, as the format in this study obstructs learning for those with
fewer technological resources.
The impact of issue involvement reinforces previous findings in the
education literature that motivation to learn is an important
predictor of learning outcomes. Findings here suggest level of
interest in the topic strengthens the impact of the CFT format on the
ability to connect concepts from the reading with previously held
concepts, which is a key step toward advanced learning.
These interaction effects are important, but most notable is the weak
direct impact from hypermedia format change. For the most part
readers of the linear format, CFT format and the traditional
nonlinear format were able to recognize and recall story content
equally well. There was no impact on perceived ability to discuss the
issue with others, nor on the strength of relationships among
concepts from the readings. Perhaps most surprisingly, there were no
differences on degree to which stories were scanned or skipped over,
nor on level of cognitive stress. These last two findings contradict
the Eveland et al. studies (2002, 2004) on general nonlinear formats,
though the finding for the CFT format is new. There was consistency
with previous hypermedia news studies that found nonlinear formats
did not aid memory processing (Eveland et al., 2002, 2004; Lowrey,
2004) and that CFT formats in particular did not aid memory
processing (Lowrey, 2005; Niederhauser, Reynolds, Salmen, & Skomolski, 2000).
Why the null results? It may be that overall, readers are getting
used to nonlinear formats, and that therefore attributes of the
reader are taking precedence. Though effects of reader attributes
were not hypothesized, readers' level of prior knowledge and level of
involvement in the issue had strong significant main effects on both
advanced knowledge and memory processing variables.
It may be that a more in-depth and qualitative measurement such as
think-aloud protocols would have detected differences in conceptual
interconnectedness and ability to apply knowledge. Likewise, a
contextual analysis of actual discussion with others may be a more
direct and sensitive assessment of ability to apply journalism
knowledge than the measures used in the present study.
It is also likely that the lack of impact from the formats lies
partly in the lack of control that journalists have over their
readers. Theoretical perspectives in education psychology that
predict direct links between hypermedia and the workings of the mind
are not always supported in empirical studies, but often when they
are, an intervening factor is the guidance or orientation given to
readers. The present study provided several brief lines of
instruction in the CFT and nonlinear formats, but the experiment
administrator offered no verbal guidance, as this would have violated
the test's external validity. In the real world, teachers have the
authority and proximity necessary to encourage crisscrossing of
perspectives and cases, but journalists have neither. Journalists are
both distant from and powerless over the habits and motivations of
readers, and as research on uses and gratifications and media
dependency suggests, readers' individual motivations have a
substantial impact on media effects.
The present study suggests journalism researchers should be wary of
uncritically adopting theory from education psychology. The
differences in professional goals and the structural differences in
message sending relationships are substantial. Cognitive effects from
the CFT format may change in the transfer from an educational
environment to a journalistic environment. For example, linear news
stories may affect readers differently than linear instructional text
because the journalist is more likely to emphasize story-telling so
as to engage readers. Also, readers are more likely to leisurely scan
an online newspaper than an assigned online reading for a class. In
education experiments, the expectation is that subjects will read
carefully because they will be tested. Of course, news audiences are
not tested, and urging subjects to read thoroughly or to anticipate a
test would diminish external validity. Further study of attention on
message may be warranted in future studies. Finally, there is no
recognized "canon" of knowledge and theoretical perspectives in
journalism as there is in an academic discipline, and this makes it
difficult to choose perspectives for "crisscrossing" with cases, as
required in the cognitive flexibility format.
Perhaps the chief limitation of this study is the use of only one
news issue. It may well be that, as mentioned earlier, a salient, but
complex issue such as terrorism or our involvement in the Middle East
would lead to different effects. A strength of this experiment is the
inclusion of older non-students in the sample – something that is not
often done in experimental research. As mentioned, findings did not
differ substantially by age or by education level. Future studies
might test the CFT format on different types of issues to see if
effects are shaped by the degree to which an issue has been
previously covered. It would also be helpful to vary the amount of
guidance given to participants, and to perhaps vary types of
motivation. For example, formats may have a different impact for
those searching for specific information vs. those who are casually
browsing. This study suggests future research that explores the
impact of these and other interaction variables may be particularly
helpful in explaining the impact of online media on cognitive processes.
Table 1: MANOVA univariate results: Impact of story format on
Sum of reasons for opinion
Knowledge structure density**
Number of concepts recalled from matrix**
Time spent reading
* SD = standard deviation; F value = strength of effect. Measure of
dependent variables: high score means higher value of the
measurement. For example, the measure of Self-efficacy ranges 4-28,
higher score means more self-efficacy. The data meet the criteria for
MANOVA. The determinant value of Bartlett's Sphericity test is small,
with a significance of .001, which indicates the dependent values are
sufficiently correlated. Homogeneity has not been violated, as
results of a Box's M test are not significant.
** This measure was administered only to older non-student participants.
Table 2: Significant MANOVA univariate results: Impact of
interaction between story format and prior knowledge, story format
and Web use, and story format and Involvement on dependent variables.
Level of prior knowledge
Time spent reading
Level of Web use
Level of involvement in cloning issue
*F value = strength of the effect. The data meet the criteria for
MANOVA. The determinant value of Bartlett's Sphericity test is small,
with a significance of .000, which indicates the dependent values are
sufficiently correlated. Homogeneity has not been violated, as
results of a Box's M test are not significant.
Altheide, D. L. & Snow, R.P. (1979). Media logic. Beverly Hills: Sage.
Anzur, T. (March 2001). Broadcast news: The Web's beached whale.
Online Journalism Review. Retrieved May 14, 2004, from
Balcytiene, A. (1999). Exploring individual processes of knowledge
construction with hypertext, Instructional Science, 27, 303-328.
Baylor, A.L. (2001). Perceived disorientation and incidental learning
in a Web-based environment: Internal and external factors. Journal of
Educational Multimedia & Hypermedia, 10, 227-251.
Becker, D. A., & Dwyer, M. M. (1994). Using hypermedia to provide
learner control. Journal of Educational Multimedia and Hypermedia, 3
Bong, M. & Hocevar, D. (2002). Measuring self-efficacy:
Multitrait-multimethod comparison of scaling procedures. Applied
Measurement in Education, 15, 143-171.
Britt, M. A., Rouet, J., & Perfetti, C. A. (1996). Using Hypertext to
Study and Reason about Historical Evidence. In J. Rouet, J.J.
Levonen, A. Dillon & R.J. Spiro (Eds.), Hypertext and Cognition, (pp.
43-72) Mahwah, N.J.: Lawrence Erlbaum Associates, Inc.
Brunken, R., Plass, J. L., & Leutner, D. (2003). Direct Measurement
of Cognitive Load in Multimedia Learning. Educational Psychologist, 38, 53-61
Chan-Olmsted, S.M. & Park, J.S. (2000). From on-air to online world:
Examining the content and structures of broadcast TV stations' Web
sites. Journalism and Mass Communication Quarterly 77, 321-339.
Charney, D. (1994). The impact of hypertext on processes of reading
and writing. In S.J. Hilligoss and C.L. Selfe (Eds.), Literacy and
computers: The complications of teaching and learning with
technology, (pp. 238-263). New York: Modern Language Association.
Cappella, J. A., & Jamieson, K. H. (1997). Spiral of cynicism. New
York: Oxford University Press.
Chou, Chien; Lin, Hua; Sun, Chuen-Tsai, International Journal of
Instructional Media, 2000, Vol. 27 Issue 2, p165, 18p,
Cooper, G. (1990). Cognitive load theory as an aid for instructional
design. Australian Journal of Educational Technology, 6, <
http://cleo.murdoch.edu.au/gen/aset/ajet/ajet6/su90p108.html> (24 May 2003).
Curry, J., Haderlie, S., Ku, T., Lawless, K. A., Lemon, M., & Wood,
R. (1999). Specified Learning Goals and Their Effect on Learners'
Representations of a Hypertext Reading environment. International
Journal of Instructional Media, 26, 43-52.
Dee-Lucas, D. (1996). Effects of overview structure on study
strategies and text representations for instructional hypertext. In
J.-F. Rouet, J. J. Levonen, A. Dillon & R. J. Spiro (Eds.), Hypertext
and Cognition (pp.73–107). Mahwah, NJ: Lawrence Erlbaum.
Dee-Lucas, D. & Larkin, J. H. (1995). Learning from Electronic Texts:
Effects of Interactive Overviews for Information Access. Cognition
and Instruction, 13, 431-468.
Demetriadis, S. & Pombortsis, A. (2000). Novice student learning in a
case-based hypermedia environment: A quantitative study. Paper
presented at the International Conference of the Learning Science,
Ann Arbor, Mich.
Dillon, A. (1996). "Myths, Misconceptions, and an Alternative
Perspective on Information Usage and the Electronic Medium" in
Hypertext and Cognition, eds. Jean-Francois Rouet, Jarmo J. Levonen,
Andrew Dillon and Rand J. Spiro (Mahwah, N.J.: Lawrence Erlbaum
Associates, Inc., 1996), 25-42.
Dillon, A. & Gabbard, R. (1998). Hypermedia as an Educational
Technology: A Review of the Quantitative Research Literature on
Learner Comprehension and Style," Review of Educational Research, 68, 322-349.
Domke, D. & Shah, D.V. & Wackman, D.B. (1998). "Moral referendums":
Values, news media, and the process of candidate choice, Political
Communication, 15, 301-321.
Eveland, Cortese, Park, & Dunwoody, (2004). How web site organization
influences free recall, factual knowledge, and knowledge structure
density. Human Communication Research, 30, 208-233.
Eveland, Marton, & Seo, (2004). Moving beyond "just the facts": The
influence of online news on the content and structure of public
affairs knowledge. Communication Research, 31, 82-108.
Eveland, Seo, & Marton, (2002). Learning from the news in campaign
2000: An experimental comparison of TV news, newspapers and online
news. Media Psychology, 4, 352-378.
Eveland, W.P. Jr. & Dunwoody, S. (2002). An investigation of
elaboration and selective scanning as mediators of learning from the
Web versus print, Journal of Broadcasting & Electronic Media, 46, 34-53.
Eveland, W.P. Jr. & Dunwoody, S. (2001). User control and structural
isomorphism or disorientation and cognitive load? Learning from Web
versus print, Communication Research, 28, 48-78.
Eveland, W.P. Jr. & Dunwoody, S. (2000). Examining information
processing on the World Wide Web using think aloud protocols, Media
Psychology, 2, 219-244.
Fallows, J. (1997). Breaking the news. New York: Vintage.
Fishman, M. (1980). Manufacturing the news. Austin: University of Texas Press.
Flanagin, A., & Metzger, M. (2000). Perceptions of Internet
information credibility. Journalism and Mass Communication Quarterly,
Foltz, P.W. (1996). Comprehension, Coherence and Strategies in
Hypertext and Linear text. In Rouet, J.F., Levonen, J.J., Dillon,
A.P. and Spiro, R.J. (Eds.) Hypertext and Cognition. Hillsdale,
Lawrence Erlbaum Associates.
Fredin, E. (1997) Rethinking the news story for the Internet:
Hyperstory prototypes and a model of the user, Journalism and Mass
Communication Monographs, 163.
Fredin, E. & David, P. (1998) Browsing and the hypermedia interaction
cycle: A model of self-efficacy and goal dynamics, Journalism and
Mass Communication Quarterly, 75, 35-54.
Garrand, T. (2001). Writing for multimedia and the Web. Oxford: Focal Press.
Gellevij, M., Van der Meij, H., De Jong, T. & Pieters, J. (2002).
Multimodal versus unimodal instruction in a complex learning
environment. The Journal of Experimental Education, 70, 215-239.
Gillmor, D. (2003, Fall). Moving Toward Participatory Journalism.
Nieman Reports, 79-80.
Godshalk, V.M., Harvey, D.M., & Moller, L. (2004). The Role of
Learning Tasks on Attitude Change Using Cognitive Flexibility
Hypertext Systems, Journal of the Learning Sciences, 13, 507-526.
Gray, S.H., "Using Protocol Analysis and Drawing to Study Mental
Model Construction During Hypertext Navigation," International
Journal of Human-Computer Interaction 2, no. 4 (1990): 359-377.
Graziano, C. & McGrath, K. (1986). Measuring the concept of
credibility. Journalism Quarterly, 63, 451-462.
Hannafin, R. D., & Sullivan, H. J. (1996). Preferences and learner
control over amount of instruction. Journal of Educational
Psychology, 88, 162-173.
Hendrickson, L.J. & Tankard, J.W. (1997). Expanding the news frame:
the systems theory perspective. Journalism & Mass Communication
Educator, 52, 39-46.
Jacobson, M.J., Maouri, C., Mishra, P., & Kolar, C. (1995). Learning
with hypertext learning environments: Theory, design, and research.
Journal of Educational Multimedia and Hypermedia 4, 321-364.
Jacobson, M.J., & Spiro, R.J. (1995). Hypertext learning
environments, cognitive flexibility, and the transfer of complex
knowledge: An empirical investigation. Journal of Educational
Computing Research, 12, 301-333.
Jaffe, M. J. (1995). Media interactivity, cognitive flexibility and
self-efficacy. Doctoral Dissertation, University of Michigan.
Johnson, T. & Kaye, B. (2002). Webelievability: A path model
examining how convenience and reliance predict online credibility.
Journalism and Mass Communication Quarterly, 79: 619-642.
Johnson, T. & Kaye, B. (2000). Using is believing: The influence of
reliance on the credibility of online political information among
politically interested Internet users. Journalism and Mass
Communication Quarterly, 77, 865-879.
Jonassen, D.H. (1988). "Designing Structured Hypertext and
Structuring Access to Hypertext," Educational Technolog, 28, 13-16.
Lang, A., Zhou, S., Schwartz, N., Bolls, P. D., & Potter, R. F.
(2000). The effects of edits on arousal, attention, and memory for
television message: When an edit is an edit can an edit be too
much? Journal of Broadcasting and Electronic Media, 44, 94-109.
Lang, A. (1995). The limited capacity model of mediated message
processing, Journal of Communication, 50, 46-70.
Lanson, J. (2000, August 24). It's Time to Make Web an Equal Partner.
Online Journalism Review. Retrieved May 14, 2004 from
Lasica, J.D. (2003, August 7). What is Participatory Journalism?
Online Journalism Review. Retrieved May 14, 2004 from
Lasica, J.D. (2003, Fall). Blogs and Journalism Need Each Other.
Nieman Reports, 70-74.
Lee, M.J. & Tedder, M.C. (2003). The Effects of Three Different
Computer Texts on Readers' Recall, Computers in Human Behavior, 19(6), 767-783.
Lowrey, W. (2005). The Web News Story and Cognitive Flexibility. In
X. Li (Ed.), Internet Newspapers: Making of a Mainstream Medium.
Mahwah, N.J., Lawrence Erlbaum Associates.
Lowrey, W. (2004). Linear vs. nonlinear Web stories: An assessment of
reader perceptions, knowledge acquisition and reader feedback.
Newspaper Research Journal 25.
Lowrey, W. (2003). Explaining decision-making about online news at
small newspapers. Newspaper Research Journal, 24, 83-90.
McAdams, M. (December 2000). Lesson 2: Why News is Old News.
NewsLink. Retrieved Nov. 15, 2003 from http://newslink.org/mmcol2.html
McKnight, C., Dillon, A., & Richardson, J. (1990). A Comparison of
Linear and Hypertext Formats in Information Retrieval. In R. McAleese
& C. Green (Eds), Hypertext, the State of Art, (pp. 10-19), Oxford,
England: Intellect Books.
McQuail, D. (2000). Mass communication theory. London: Sage.
Morrison, G. R., Ross, S. M., & Baldwin, W. (1992). Learner control
of context and instructional support in learning elementary school
mathematics. Educational Technology Research and Development, 40, 5-13.
Nelson W. A., & Palumbo, D. B. (1992) Learning, instruction, and
hypermedia. Journal of Multimedia and Hypermedia, 1, 287-299.
Niederhauser, D. S., Reynolds, R. E., Salmen, D.J. & Skomolski, P.
(2000). The influence of cognitive load on learning hypertext.
Journal of Educational Computing Research, 23, 237-255.
Outing, S. (2000, October). Are days of innovation waning for online
news?" Editor and Publisher, 11.
Outing, S. (2003, July 31). Blogs have a place on news Web sites:
Reporters and readers can contribute. Editor & Publisher. Retrieved
May 14, 2004, from http://www.editorandpublisher.com
Paul, N. & Fiebich, C. (2005). The Elements of digital storytelling.
Retrieved May 14, 2004, from http://www.inms.umn.edu/Elements/index.php
The Pew Research Center for the People and the Press. (2000, December
3). Internet election news audience seeks convenience, familiar
names. Retrieved May 14, 2004 from
Randazzo, R. & Greer, J. (2002). Context in print and online
environmental articles. Presented to the annual convention of the
Association for Education in Journalism and Mass Communication. Miami, FL.
Rich, C. (1999). Creating online media: A guide to researching,
writing and design on the Internet. New York: McGraw-Hill.
Rosen, J. (2005). Blogging vs. Journalists is Over. PressThink: Ghost
of Democracy in the Media Machine. Retrieved March 1, 2005 from
Rossner-Merrill, V., Parker, D., Mamchur, C. and Chu, S. (1998).
Using constructivist instructional design featured in two online
courses: notes from the field, Journal of Educational Media
International, 35, 282-288.
Shyu, H., & Brown, S. (1995). Learner-control: the effects of
learning a procedural task during computer based videodisk
instruction. International Journal of Instructional Media, 22, 217-231
Siegel, M., Derry, S., Kim, J.B., Steinkuehler, C., Street, J.,
Canty, N., Faasnacht, C., Hawson, K., Hmelo, C., & Spiro, R. (2000).
Promoting teachers' flexible use of the learning sciences through
case-based problem solving on the WWW: A theoretical design approach.
Paper presented at the International Conference of the Learning
Sciences, Ann Arbor, MI.
Singer, J. (2002). Information trumps interaction in local papers'
online caucus coverage. Newspaper Research Journal, 23, 91-96.
Singer, J. (2003). Campaign Contributions: Online Newspaper Coverage
of Election 2000. Journalism and Mass Communication Quarterly, 80, 39-56.
Southwell & Lee, (2004). A pitfall of new media? User controls
exacerbate editing effects on memory. Journalism and Mass
Communication Quarterly, 81, 643-656.
South, J. (1999, June 11). Web staffs urge the print side to think
ahead. Online Journalism Review. Retrieved May 14, 2004 from
Spiro, R. & Jehng, J. (1990). Cognitive flexibility and hypertext:
theory and technology for the nonlinear and multidimensional
traversal of complex subject matter. In Don Nix and Rand Spiro
(eds.), Cognition, education and multimedia: Exploring ideas in high
technology. Hillsdale, N.J.: Lawrence Erlbaum Associates.
Spiro, R. J., Feltovich, P.J., Jacobson, M.J. and Coulson, R.L.
(1991). Cognitive flexibility, constructivism, and hypertext: random
access instruction for advanced knowledge acquisition in
ill-structured domains. Educational Technology, 31, 24-33.
Sweller, J. (1988). Cognitive load during problem solving: Effects on
learning, Cognitive Science, 12, 257-285.
Tuchman, G. (1978). Making the News: A Study in the Construction of
Reality. New York: The Free Press.
Turow, J. (1997). Media systems in society: Understanding industries,
strategies and power. Longman: New York.
Valentino, N.A., Buhr, T. A. and Beckmann, M.N. (2001). When the
frame is the game: revisiting the impact of strategic campaign
coverage of citizens information retention. Journalism and Mass
Communication Quarterly, 78, 93-112
Van Nimwegen, C. Pouw, M. and Van Oostendorp, H. (1999). The
influence of structure and reading-manipulation on usability of
hypertexts. Interacting with Computers, 12, 7-21.
Zaichowsky, J.L. (1986). Concepualizing Involvement. Journal of
Advertising, 15, 4-14, 34.
Figure 1. The CFT news report.
[--- ??? Graphic Goes Here ---]
Figure 2. The nonlinear news report.
[--- ??? Graphic Goes Here ---]
Figure 3. The linear news report.
[--- ??? Graphic Goes Here ---]