Abstract

This article investigates how individuals' adventurousness (a personality characteristic) influences their experience with different types of computer text formats and whether expanding hypertext, a new hybrid hypertext format, addresses the most frequently cited hypertext problem, disorientation. Two hundred and one college students participated in an experimental study which evaluated three variations of computer text formats. Specifically, the study investigated how individuals with different levels of adventurousness exhibited different preferences for, and disorientation with, different computer text formats. Participant adventurousness was found to interact with text format on participant feelings of disorientation. The participants prefered the paged hypertext and the expanding hypertext over the scrolling text format.

Introduction

Thanks to the advent of new communication technologies, modern mass communication is no longer necessarily constrained by the traditional one-way, asymmetric communication medium. It is now possible to present communication materials that automate sophisticated interaction with individuals, thus extending the capacity of mass communication closer to symmetric communication. Based on a system's ability to adapt itself to each reader's unique characteristics, capacities, and preferences (Lee, Ferguson, & Tedder, 1999; Lee & Tedder, 2003, 2004), contemporary computer software technologies can project interactively-tailored content and formats according to the designer's intended purpose(s) (Lee & Tedder, 2004). This can be accomplished by means of electronic message presentation formats such as hypertext.

Hypertext is computer-mediated text in which highlighted words or titles serve as links to other excerpts or modes of supporting information (Lee, 1998, 2001; Lee et al., 1999; Lee & Tedder, 2003, 2004). By these means, individuals can choose their own orders and levels of detail, and even modalities of communication materials, forming their own cohesion and sequence of information (Lee & Tedder, 2004).

Despite the potential of hypertext to extend the effectiveness of computer-mediated communication, many have identified concerns, such as disorientation and fragmented knowledge, relating to the use of hypertext. Disorientation is the most frequently cited problem. Some assert that the nonlinear nature of hypertext tends to foster disorientation among users (Batra, Bisha & Donohue, 1993; Dias, Gomes, Correia, 1999; Hammond, 1989; Hammond & Allinson, 1989; Kim & Hirtle, 1995; McDonald & Stevenson, 1996,1999; Rouet, Levonen, Dillon, & Spiro, 1996; Smith & Wilson, 1993; Unz & Hesse, 1999).

To address this problem, Lee (2001) developed a new hybrid form of hypertext: expanding hypertext, an electronic text format with hyperlinks that, when clicked, cause the content pointed to (e.g., additional excerpts or documents) to be inserted into the same page as the hyperlink itself. In this article, expanding hypertext is investigated in terms of its usability to address the disorientation problem of earlier types of hypertext.

The most important potential of hypertext is that each user can interact with it differently (Lee, 2001; Lee & Tedder, 2003, 2004; Lee, Xie, & Tedder, in press). In this respect, unique characteristics of individuals that influence their experience should be investigated along with the development and refinement of new technologies.

One motivational characteristic of individuals that may influence how they perceive hypertext is need for sensation (Zuckerman, 1990) or activation (Donohew, Palmgreen, & Duncan, 1980). It has been argued that individuals exhibit different optimal levels of arousal or stimulation (Zuckerman, 1979) and that sensation seekers exhibit a higher need for arousal than do non-sensation seekers (Zuckerman, Persky, Hopkins, Murtaugh, Basu, & Schilling, 1966). Donohew et al. (1980) further argued that individuals process information selectively based on their activation needs. In this study, it was suspected that hyperlinks in different text structures might serve as stimuli, creating different reactions from individuals based on their different levels of need for activation or sensation.

Specifically, in this study, participants' adventurousness, one dimension of the sensation-seeking tendency (also known as the Thrill-Adventure Seeking Tendency), was suspected to influence their attitudes toward, as well as experience with, different text formats. Therefore, the purpose of this study is to investigate how individuals' adventurousness influences their preference for, and disorientation with, different computer text formats. To address this purpose, an experimental, post-test only group study was conducted with 201 participants.

Literature Review

Hypertext Systems for Effective Computer-Mediated Communication

Liu (1994) identified four major advantages of hypertext: nonlinearness, associativity, flexibility, and efficiency. By utilizing these advantages, many believe that hypertext holds much potential for optimizing the human-computer interaction experience. One example is hypertext's ability to present information in various sensory modes by using textual, graphic, video, and audio media (Lin & Davidson, 1996). This gives users not only the opportunity to manipulate the order of presentation (Jonassen, 1989) by providing a string of choices based upon previous choices (Fredin & David, 1998), but also the opportunity to modify the modality of information (i.e., visual versus auditory) based on their preferences and needs.

Because of the flexibility of hypertext and its potential to serve users individually, the relationship between different hypertext structures and users' experiences has been studied. A central focus has been put on nonlinearity or the networking of information units. Hypertext's flexibility allows nodes and links to be arranged in a variety of information structures (McDonald, 1998). These links allow multiple paths for the exploration of materials (Newmarch, 1998). Therefore, the design of hypertext materials always involves this link issue: decisions about how nodes or links should be interconnected must be made, and these decisions structure the presentation of content differently. Furthermore, it has been determined that the organization of content in a hypertext system does, in fact, alter learning outcomes (Dee-Lucase & Larkin, 1995; Lee, 1998, 2001; Lee & Tedder, 1999, 2003, 2004; Phillips, Watson-Paelis, Cook, Ming, & Tiancheng, 1992; Shapiro, 1998).

The most frequently tested hypertext structures are hierarchical and network structures (Pohl, 1998). Hierarchical structures connect nodes (excerpts or documents) in a hierarchical order in which a node at one level can be accessed only from nodes directly above or below it, while network structures connect a given node to any other related nodes (McDonald, 1998).

Different hypertext structures offer users different levels of control (McDonald, 1998). For example, hierarchical structures confine learners' movements and restrict their freedom to browse while network structures place few constraints on users' movements (McDonald, 1998). However, this freedom seems to create an additional cognitive burden and disorientation (McDonald, 1998). In addition, unstructured systems such as nonlinear or "paged" hypertext provide little information about topic relations. On the other hand, highly structured systems provide more information about topic relations (Shapiro, 1998) and might reduce possible disorientation while limiting the opportunity for individual freedom (Last, O'Donnell, & Kelly, 2001).

Lee and Tedder (2003, 2004) found that the linear text produced better recall scores than the other types of hypertexts. In addition, the participants who read the networked hypertext produced better recall scores than those who read the structured text. However, this difference disappeared when the participants' reading time was controlled. The researchers concluded that the different types of hypertext systems seem to influence how much time an individual reads a given material and in turn affects how much information is learned, suggesting an indirect causal relationship.

Hypertext Structures and Reading Manipulations

Hypertext structures have been typically studied in comparison with reading manipulations-methods of connecting the various passages in hypertext. There are two different types of reading manipulation: scrolling versus paging. Scrolling allows users to read the text as it is moved upward in a display window and reveals itself gradually from top to bottom. Paging displays the text in separate blocks, one at a time in the display window, somewhat like pages in a physical book.

The effects of hypertext structures with different reading manipulations vary depending on what tasks are measured. For example, based on 24 search tasks which had performance measurements for their ease of use and user satisfaction, a purely hierarchical hypertext with scrolling appeared to be more useful because it seemed to provide clear insight into the structure of the hypertext (Van Nimwegen, In Pouw, & Oostendorp, 1999). In fact, for search tasks, adding linearity did not help in increasing the usability of hypertext in terms of efficacy, ease of use, and user satisfaction when the reading manipulation consisted of scrolling as well as paging (Van Nimwegen, Pouw, & Oostendorp, 1999).

However, when comprehension was the goal, Piolat, Roussey, and Thunin (1997) found that paging showed more positive results over scrolling, while Lee and Tedder (2004) found scrolling superior over paging when memory was measured. In particular, Lee and Tedder (2004) again found the same results as their 2003 study of an indirect causal relationship in that the different types of hypertext systems seem to influence how much time an individual takes to explore a given system, and in turn influence how much information an individual remembers.

The effects of reading manipulation also vary based on readers' previous experience with similar systems. Tombaugh, Lickorish, and Wright (1987) found that paging is more usable than scrolling for experienced users, and that the reverse is true for inexperienced users. Lee and Tedder (2004) also found similar results.

Disorientation in Hypertext

Many hypothesize that individuals take an active role in finding information and in encountering different types of information in using hypertext (Bourne, 1990; Dee-Lucas & Larkin, 1995). However, there is considerable disagreement as to whether hypertext is beneficial for all individuals (Dee-Lucas & Larkin, 1995; Hammond & Allinson, 1989; Jonassen & Wang, 1990; McDonald, 1998; McDonald & Stevenson, 1996; Schroeder, 1994; Spiro & Jehng, 1990).

Many have pointed out that hypertext users often get lost or become disoriented (Batra et al., 1993; Hammond, 1989; Hammond & Allinson, 1989; Kim & Hirtle, 1995; McDonald & Stevenson, 1996, 1999; Rouet, Levonen, Dillon, & Spiro, 1996; Smith & Wilson, 1993; Unz & Hesse, 1999). It has been suggested that hypertext, by scrambling an author's intended order and causing sudden changes in context, disorients users (McDonald & Stevenson, 1996; Unz & Hesse, 1999). Furthermore, individuals tend to skip or omit crucial information, creating fragmented knowledge (Shneiderman, 1992).

Individuals can experience disorientation in two ways: First, users may experience difficulty achieving a coherent understanding of the content due to cognitive overload- known as cognitive disorientation. Another type of disorientation occurs simply when the users become lost in hyper-space-not knowing where they are or where to go next (Dillon, 1996; Edwards & Hardman, 1989; McKnight, 1996). Some studies have reported that hypertext users experienced navigation problems such as not remembering what they had or had not read and being uncertain about where to go next (Gray, 1990; McDonald & Stevenson, 1996), as well as perceived cognitive overload (Macedo-Rouet, Rouet, Epstein, & Fayard, 2003), while other studies found no significant effect of computer text formats on disorientation (Eveland & Dunwoody, 2001). One study even found the opposite result in that users in the linear mode were more disoriented than those in the nonlinear mode for navigation tasks (Baylor, 2001). These differences need to be resolved with careful scientific investigations. Therefore, in this study, an attempt was made to test empirically individuals' feelings of disorientation when using hypertext. In so doing, cognitive disorientation was operationalized and measured.

As noted above, it is believed that the nonlinear nature of hypertext tends to foster disorientation among users. In this study, Paged Hypertext (PH), which shows the additional text of each hyperlink on a separate page, is very similar to the widely-used hypertext format. Due to its nonlinear nature, PH was predicted to be the most disorienting format in contrast to scrolling text, in which information is presented in a traditional linear text format with a scroll bar on the right hand side that allows scrolling up and down to explore the given information.

H1: Participants who read Paged Hypertext (PH) will experience more disorientation than those who read Scrolling Text (ST).

Proposed Alternative Hypertext: Expanding Hypertext

In an effort to address the disorientation problem, a hybrid computer text format, expanding hypertext (EH), was developed (Lee, 2001). Unlike the most common format of hypertext (paged hypertext) in use today, expanding hypertext is an electronic text format with hyperlinks that, when clicked, cause the content pointed to (e.g., additional excerpts or documents) to be inserted into the same page as the hyperlink itself. This is in contrast to the more common action where clicking on a hyperlink (in paged hypertext) would replace the current page with a completely new page. In this way, the user can have additional details of information inserted at designated points when he or she clicks the related links, maintaining a linear flow of information presentation (Lee, 2001).

Expanding hypertext is designed to preserve hypertext's flexibility to inter-link related information while maintaining a linear presentation of information (Lee, 2001; Lee & Tedder, 2004). It allows users to explore a given system based on their needs and preferences while better facilitating their cognitive processes by minimizing cognitive overload caused by the non-linearity of hypertext. Therefore, the following hypothesis was tested.

H2: Participants who read Expanding Hypertext (EH) will experience less disorientation than those who read PH.

Since EH was developed to address PH's disorientation problem by combining the linking ability of hypertext with linearity of presentation, it was suspected that those who read EH would be less disoriented than those who read PH.

Theory of Information Processing and Sensation-seeking

Because computer-mediated communication is influenced by both the characteristics of the individuals using it and the structural and interface design of the systems (MacGregor, 1999), without understanding both aspects, the development and utilization of this new technology will be greatly limited (Lee, 2001; Lee & Tedder, 2003, 2004; Lee, Xie, & Tedder, in press).

Dillon and Watson (1996), based on their extensive review of research on individual differences, concluded that understanding individual differences in human-computer interaction would lead to better system designs. Chen, Czerwinski, and Macredie (2000) also argued that studies on individual differences in human-computer interaction could offer effective and flexible ways to accommodate those differences through ongoing research and development.

One motivational characteristic of individuals that may impact how they process hypertext information is need for sensation (Bardo, Donohew, & Harrington, 1996; Zuckerman, 1994). Marvin Zuckerman, in 30 years of research on sensation-seeking, found that individuals vary in levels of predisposition toward sense-arousing stimuli (Zuckerman, Kolin, Price, & Zoob, 1964). Sensation-seeking is defined as a trait characterized by the need for "varied, novel, and complex sensations along with experience and the willingness to take physical and social risks for the sake of such experience" (Zuckerman, 1979, p. 10).

One of the four dimensions identified by Zuckerman (1971) was Thrill and Adventure Seeking (TAS), also known as adventurousness¹ (Ferguson, Valenti, & Melwani, 1991). Adventurousness is associated with self-reports of enjoyment of risk such as new and exciting experiences, including novel information (Zuckerman, 1990). Sensation seekers tend to exhibit a higher need for arousal than others. In this study, different text structures are suspected to stimulate individuals differently.

Donohew et al. (1980) postulated an activation model of exposure to information. This reflects how individuals choose which information to process, based on their cognitive and activation needs. Attention depends on an individual's need for stimulation by an information source (Zillmann & Bryant, 1985). Donohew, Lorch, and Palmgreen (1998) suggested that the effectiveness of messages depends on target audiences' sensation-seeking tendencies. They found that high-sensation seekers tended to pay more attention to messages with high levels of suspense and tension and responded more positively than did low-sensation seekers (Donohew, Lorch, & Palmgren, 1991; Palmgreen, Donohew, Lorch, & Rogus, 1991). Therefore, they concluded that the sensation value of information was an important determinant of a message effect, particularly for sensation seekers. This provides evidence that materials can produce different results, based on the characteristics of target audiences. Therefore, characteristics of both the hypertext and the user can render one form of hypertext more or less disorienting than, or superior to, another.

In this study, expanding hypertext was tested for its usability, particularly in terms of individuals' preference for, and disorientation with, the given text format. It was speculated that hyperlinks in different text structures might serve as stimuli and could be beneficial for certain individuals by providing uncertain or novel stimulation. For this reason, individuals' sensation-seeking tendency was expected to play a role in their exploration of computer-mediated materials. Therefore, an interaction effect of text formats and participants' adventurousness was predicted in terms of how individuals experience disorientation with different types of computer texts.

In the literature addressing need for sensation or activation, it is speculated that, among individuals with lower levels of need for sensation, PH might provide too much stimulation due to its nonlinearity, resulting in feelings of disorientation or frustration. Therefore, the varying degrees of nonlinearity represented in different text formats might be perceived very differently by individuals with different levels of adventurous sensation-seeking tendency. If this is the case, it is also possible that individuals with higher levels of need for sensation may become easily frustrated when given environments that do not provide appropriate levels of stimulation. Therefore, the following interaction effect was hypothesized.

H3: Among participants who are low in adventurousness, those who read PH will experience more disorientation than those who read ST, but those who are high in adventurousness will not show the same anticipated text format effect.

H4: Among participants who are high in adventurousness, those who read EH will experience less disorientation than those who read ST.

Donohew et al. (1998) suggested that messages high in sensation value should be more attractive to high sensation seekers. In this study, it is assumed that presentation of hyperlinks can be perceived as a stimulating factor ("high sensation value"). If so, it is logical to assume that high sensation values in information presentation style such as hyperlinks should engender a more positive reaction from high-sensation seekers.

H5: Among highly adventurous participants, those who read PH will show higher scores on liking of presentation style than those who read ST.

H6: Among highly adventurous participants, those who read EH will show higher scores on liking of presentation style than those who read ST.

Method

An experiment involving a post-test only model with three variations of computer text formats was conducted. Two hundred and one (N=201) undergraduate students attending a university in the northwestern United States were tested. Of these, 94% of the participants were between 19 and 23 years old. Seventy-five percent (N=150) of the participants were females while 25% (N=51) were males.

Before the participants were randomly assigned to one of the conditions, they were asked to answer a questionnaire about their adventurous tendencies, refined from one used by Ferguson et al. (1991).

The computer manipulation involved presenting participants with three unrelated articles² in one of three formats: scrolling text, expanding hypertext, and paged hypertext. After the participants finished reading the articles, they were asked to answer some questions about their preference for and disorientation with the given text formats.

Procedure

Participants were randomly assigned to one of three conditions, involving the same text content organized differently. The researcher explained the procedures at the beginning of the study and the participants were encouraged to read the text completely before clicking a "Finished" button. To control for a possible article order effect, the order in which the three articles were presented was randomized. In addition, the amount of time each participant spent on each screen was recorded. The experiment was conducted in a computer lab, each session taking approximately 35 to 50 minutes.

In Condition 1, each article was presented in a linear text format. A scroll bar on the right hand side was presented, allowing scrolling up and down (Scrolling Text-ST) to explore the given information. In Condition 2, each article was presented in an expanding-hypertext format (EH) where hyperlinks³ allowed the reader to access more detailed information while avoiding losing the linear flow of information. Clicking a link added new lines of information (paragraphs, sentences, or something equivalent to a page) to the currently displayed text (or page) at the point of the original link. As in Condition 1, there was a scroll bar on the right hand side. In Condition 3, each article was presented in a paged-hypertext (PH) format. The texts were presented with each excerpt on its own screen. The participant was able to navigate between excerpts by clicking on highlighted, underlined terms serving as links. However, no particular structure was presented: Wherever a term existed in the article for which an elaborating excerpt was available, a link was presented with a "Back" button for the reader to return to previous screens. This created a web-like, user-navigated architecture of links (also known as "networked" organization) where looping was possible (Lee & Tedder, 2003, 2004) (see Figure 1).

After the participants finished reading each article, they clicked on a button labelled "Finished" that automatically led them to a set of questions measuring format preference, disorientation, and level of comfort with the presentation style.

Creating Indices

To measure different degrees of the participants' adventurousness, feeling of disorientation, and liking of text formats, indices were created. To create the indices, a principal axis factor analysis with oblimin rotation was conducted with related items to reveal factor loading. After a reliability test, a summed-index of factor scores was created for each variable.

Adventurousness

For measuring participants' adventurousness, an index was created out of four 10-point Likert—type items loaded on one factor—"adventurousness"⁴ (69% of variance explained, Cronbach's = .78). The items are "I like adventure," "I welcome new and exciting experiences, even if they are a little frightening," "I like challenges," and "I enjoy or would enjoy skydiving." These items were taken from Ferguson et al. (1991) and Lee and Ferguson (2003). A summed-index of factor scores was created based on the factor analysis.

Dependent Variables

Disorientation

To measure the degree of disorientation the participants experienced while they read the articles, five 10-point Likert-type questions were developed and used to create an additive index (81% of variance explained, Cronbach's = .86) for each article. The items are: "While I was reading the article, I sometimes got frustrated," "While I was reading the article, I often felt lost, not knowing where I was," "While I was reading the article, I had a hard time understanding it," "While I was reading the articles, I had a hard time keeping track of what I just read," and "While I was reading the articles, I often had a hard time deciding which link to click next." All three indices were added together to represent the degree of disorientation experienced.

Liking of Presentation Style

To measure the participants' degree of liking related to presentation style, an index was created out of four 10-point Likert-type items loaded on one factor (57% of variance explained, Cronbach's = .75). The questions were the following: "The way the articles were presented was stimulating," "It was easy to read through information with the way the articles were presented," "I like the idea that I had more choices to click," and "I like the way the articles were presented to me." All three indices were added to represent how much the participants liked the given presentation style.

Initially, correlation tests were conducted between the participants' feeling of disorientation and liking of the presentation style. As anticipated, disorientation was negatively correlated with the liking of presentation style, r (201) = -.27, p < .01.

Results

To test the effects of the different text formats in relation to participants' adventurousness on format preference and feeling of disorientation, the General Linear Model⁵ was used.

Hypotheses about Disorientation

There was no significant main effect of the text formats on disorientation, F (2, 201) = .10, p = .90. Therefore, H1 and H2 were not supported. However, as anticipated, there was a significant interaction effect between the text formats and adventurousness, F (2, 201) = 2.9, p = .05. To analyze the interaction effect, adventurousness was split into three groups and later contrasted between the upper and lower thirds as "low" and "high".⁶ Further analysis revealed that among those who were low in adventurousness, participants who read the scrolling text experienced less disorientation (M =53, SD = 17.5) than those who read the expanding hypertext (M = 72.8, SD=25.8), t (56) = -3, p < .01. Paged hypertext fell in between the other two formats (M = 62.5, SD = 21). Independent samples t-tests showed a near-significant difference between the paged hypertext and the expanding hypertext, t (46) = 1.5, p = .08, or the scrolling text, t (42) = -1.6, p = .06.

Furthermore, for participants high in adventurousness, a significant difference was found between the scrolling text and the paged hypertext, t (46) = 1.9, p < .05. Those who read the scrolling text reported the most disorientation (M = 60.31, SD = 24) and those who read the paged hypertext reported the least disorientation (M = 45.4, SD = 28.7) (Figure 2). Therefore, H3 was supported. However, H4 was not supported even though the anticipated direction was detected in that EH fell in between ST and PH.

In addition, to further examine any linear relationship between adventurousness and disorientation in each condition, a regression model was used. The regression model showed that while there was no significant correlation between the subject's levels of adventurousness and disorientation in Condition 1 (ST), F (1, 67) = 1.47, p = .24, there was a negative linear correlation between the two variables in Condition 2 (EH), F (1, 69) = 3.98, p < .05. A negative growth relationship between adventurousness and disorientation was detected among those who read the Paged Hypertext (Condition 3), F (1, 65) = 4.64, p < .05. Among those who read either EH or PH (but not those who read ST), the more adventurous they were, the less likely they were to experience disorientation while reading the text. To further test whether this pattern was consistent in all three articles, separate analyses were conducted for each article and found a similar pattern in all three articles, except that, in one of the articles, no significant linear relationship between adventurousness and disorientation for EH was observed.

Hypotheses about Liking of Presentation Style

There was a main text format effect on the participants' liking of presentation style, F (2, 201) = 9.8, p < .01. There was no significant interaction effect between hypertext format and adventurousness. Further analysis revealed that those who read the paged hypertext (M =50.8, SD=14) or the expanding hypertext (M = 50, SD=15) showed higher scores in liking of presentation styles than those who read the scrolling text (M = 40.3, SD=15.6) (Table 1).

Furthermore, among those who were highly adventurous, there was a significant difference between PH and ST on liking of presentation style, t (46) = -2.8, p < .01. Those who read the paged hypertext (M = 57.2, SD = 14.2) exhibited higher scores than those who read the scrolling text (M = 42.3, SD = 18.9). Therefore, H5 was supported.

Regarding EH, among highly adventurous participants, even though the mean score (M = 49.7, SD = 17.4) of the participants who read the expanding hypertext was greater than the average score for the scrolling text participants (M=42.3, SD =18.9), as expected, the statistical analysis was not significant at .05, which makes it difficult to draw conclusions with great confidence (Table 1). Therefore, H6 was not supported.

Discussion and Limitations

In this study, an attempt was made to test empirically the disorienting effect of hypertext through an experiment. Since this effect has been widely discussed in the literature, the fact that H1 was not supported was somewhat unexpected. This might be because much of the discussion occurred when hypertext was a relatively young technology, but now familiarity with the technology is rampant. Many of the students in the study were already familiar with hypertext (PH), and it is therefore possible that they had already developed appropriate orienting skills for utilizing hypertext. It is also possible that the feeling of disorientation occurs on a macro level, such as getting lost in hyper-space, rather than at a micro level.

One interesting finding was that participants' adventurousness exhibited an interaction with the text formats on disorientation. Among those who were low in adventurousness, both PH and EH were considered more disorienting than ST, while the results were the opposite for the participants who were highly adventurous. This might be because highly adventurous readers, exhibiting higher sensation needs, lose their attention quickly with a plain text presentation. Indeed, the negative correlation found between participants' adventurousness and disorientation in both hypertext conditions (PH and EH) indicates that presenting hyperlinks in different text structures may benefit those who are highly adventurous by providing uncertain or novel stimulation. Further research is necessary to better substantiate this finding and to attempt an explanation of this phenomenon.

The participants' liking of the presentation style was affected by different text formats regardless of their adventurous tendencies. The participants liked the paged hypertext and the expanding hypertext more than the scrolling text, as expected. The participants who read expanding hypertext fell in between. In particular, the highly adventurous participants exhibited the same pattern with a stronger preference for PH in comparison with the other groups.

These findings provide strong evidence that individual characteristics such as adventurousness influence participants' experience with different types of hypertext. It may well be that individual differences in sensation-seeking lead to different patterns of communication experience or perhaps information-seeking. Thus it can be argued that communication experience may be enhanced if materials are tailored based on individual needs for sensation, such as presenting materials in their most favorable formats. Therefore, we propose that hypertext should be designed in a manner that is tailored according to individuals' characteristics (Lee, 2001; Lee & Tedder, 2003, 2004).

The fact that expanding hypertext was perceived as less disorienting than, and was preferred to, ST by the highly adventurous participants suggests a positive outlook for further research and development. However, many of the participants expressed unfamiliarity with EH, which in turn makes it difficult to compare with the more familiar ST and PH styles. This is understandable since expanding hypertext is a new technology. This may explain why, among those who were low in adventurousness, the participants who reported the most disorientation read the expanding hypertext, rather than the paged hypertext. Further research needs to be done in this area. It will be helpful to test the effects of EH after participants have developed familiarity with this new technology. In this respect, a longitudinal study (such as a panel study) may provide valuable information for further refinement of this technology.

Much of the earlier discussion of the disorientation problem was attributed to the nonlinear nature of hypertext. However, the fact that no significant disorientation effect was detected across the conditions, and expanding hypertext seemed to pattern similarly with PH, indicates that hyperlinks, serving as a stimulus rather than fostering nonlinearity, seem to influence individuals' feeling of disorientation, depending on the individuals' varying degrees of adventurousness.

Furthermore, it is also suspected that type of content may have an influence. It was noticed that the content of an article dictates the design of hypertext materials to a certain degree. The content of materials tends to contain logical breaking points that determine where hyperlinks can occur, pointing from excerpt to excerpt or between related documents. In some cases, a particular order must be maintained for coherence, such as with the description of a chronological event or process. In such cases, breaking the natural presentation order might cause disorientation. In fact, even though the original focus of the current study was to test text format effects across different content types, the fact that the separate analyses revealed an exception regarding EH in one of the articles suggests a possible three-way interaction among formats, content type, and individuals' adventurousness. More research needs to be conducted in this area.

Some limitations of the present study should be taken into consideration. First, even though the scrolling and paged text were supposed to simulate what was most commonly found in computer application help systems and web pages, the manipulations of text formats are not necessarily the same as would be found online. For example, in the experiment, the links within the three documents led directly to other excerpts of the same document, whereas those on the Internet might lead to entirely unrelated content. However, the main focus of the study was to test how different types of computer text formats interact with individuals' level of adventurousness, and the findings should be helpful for designing more effective computer-mediated communication materials targeting individuals with different levels of adventurousness.

Second, the study setting, as is typical in an experimental study, was somewhat artificial in that the participants were encouraged to read the text carefully. In a naturalistic setting, individuals are rarely encouraged to explore the given materials in any specific way. However, this was done in the present study to ensure that the subjects used the features of each format for which they were being measured. It might be useful to investigate separately how an individual's need for sensation would affect his or her exploration of the material without encouragement to be thorough. User purposes of learning, finding, and leisurely browsing could be analyzed.

Finally, an established measure of adventurousness was employed in this study, since the sensation-seeking tendency is believed to be a stable human trait that is mainly based on "inherited differences in the structure and biochemistry of the central nervous system" (Zuckerman, 1990, p. 313). However, a reliable measure for sensation-seeking (adventurousness) in the context of computer-mediated communication should be developed in the near future. It would be particularly helpful to identify specific online behaviors, such as browsing activities and navigation patterns, as indicators of individuals' adventurous tendency. This avenue is worth exploring for the development and refinement of sophisticated designs of computer-mediated communication materials in general as well as hypertext technologies themselves.

Conclusion

One problem with the conventional approach to studying new technologies is that considerable effort has been invested in the traditional media framework without understanding the characteristics and potentials of the Internet (Lee, 2001). It is no longer necessary, however, to find a single structure or system that maximizes its effects for all. New technologies now enable us to engage in more specialized interactions with different individuals, making it possible to approach more closely the ideal of maximizing communication effects for everyone on an individual basis (Lee, 2001; Lee & Tedder, 2004).

One can find many current examples of this trend in the business sector. For example, an increasing number of websites come to "know" individual readers by tracking their browsing activities. This is done by requiring a reader to log in or simply by observing reader sessions tracked by the address/identification of the reader's computer. While perhaps the most common use of tailoring web content to the individual is to be found in some sales-focused websites, there is no reason why similar techniques could not be applied to other kinds of message delivery, educational and health communication applications being two possibilities.

The findings of the present study have practical applications in various settings. In the traditional media framework, they might provide guidance in targeting a specifically-segmented audience online. For example, a sports-related website might reasonably expect more adventurous readers and therefore should maintain a preference for paged and/or expanding hypertext. Also, when targeting sensation seekers through computer-mediated health campaigns, hyperlinks might be considered as a viable option.

This is not to say that the designers of materials should create different versions of the hypertext documents based on the presumed level of adventurousness or other characteristics of individuals. The fact that we now have contemporary computer software technologies that can project interactively-tailored content and formats according to the designer's intended purpose(s) (Lee, 2001; Lee & Tedder, 2004) opens new possibilities and in turn requires more sophisticated investigations of human-computer interactions for the refinement of existing technologies and future innovation of new technologies. The key to a more effective approach to advancing computer-mediated communication lies in the development of a deeper understanding of multi-dimensional aspects of human and computer interaction. Effective and flexible ways of tailoring communication materials and systems for individual needs and characteristics ought to be developed along with the technology. Among the many questions that need to be addressed is how best to make use of hypertext potentials (Shierhorn, Wearden, Schierhorn, Tabar, & Andrews, 1998). If the goal is to identify effective uses of hypertext in computer-mediated communication, individuals' characteristics that influence the success of communication need to be carefully investigated.

Notes

1. Ferguson, Valenti, and Melwani (1991), focusing on individual behavioral (not attitudinal) characteristics of risk-takers, refined the Thrill-Adventure Sensation seeking as Adventurousness.
2. The three unrelated articles used in this study were from previous studies: Two were from Lee (1998), and one was from Lee (2001). All three articles were pre-tested in terms of gender neutrality and level of interest among undergraduate students, where the most interesting and gender neutral articles were selected. One article was entitled, "How much do you know about our Latin American neighbors?" (1032 words), from the World Almanac and Book of Facts 1995, edited by Robert Famighetti. Another article was entitled, "The battle against segregation in America" (613 words), from How to Prepare for the GRE. The third article was entitled, "Ethology versus Behaviorism" (679 words), from How to Prepare for the GRE. For more detailed description, see Lee (2001) and Lee and Tedder (2004).
3. Hyperlinks were presented in the standard format that is commonly used on the Internet, in that words were presented in a different color (i.e., blue) and underlined.
4. As a tendency to enjoy challenges and exciting experiences, adventurousness seemed closely related to individuals' exploration of information in hypertext systems.
5. The General Linear Model was selected in this study because this model allows us to test the effects of two different kinds of independent variables (text formats—a categorical variable, and adventurousness—a continuous variable).
6. Based on a review of the predicted regression lines between adventurousness and disorientation for all three conditions, the decision to use a three-level categorization was made in order to examine the interaction effect while best preserving the original regression lines.

References

Bardo, M. T., Donohew, R. L., & Harrington, N. G. (1996). Psychobiology of novelty seeking and drug seeking behavior. Behavioral Brain Research, 77 (1), 23-43.

Batra, S., Bishu, R. R., & Donohue, B. (1993). Effect of hypertext topology on navigation performance. Advances in Human Factors and Ergonomics, 19, 175-180.

Baylor, A. L. (2001). Perceived disorientation and incidental learning in a web-based environment: Internal and external factors. Journal of Educational Multimedia & Hypermedia, 10 (3), 227-251.

Bourne, D. E. (1990). Computer-assisted instruction, learning theory, and hypermedia: An associative linkage. Research Strategies, 8 (4), 160-171.

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). Hillside, NJ: Lawrence Erlbaum Associates.

Chen, C., Czerwinski, M., & Macredie, R. (2000). Individual differences in virtual environments - Introduction and overview. Journal of the American Society for Information Science, 51 (6), 499-507.

Conklin, E. J. (1987). Hypertext: An introduction and survey, IEEE Computer, 20 (9), 17-41.

Dee-Lucas, D., & Larkin, J. H. (1995). Learning from electronic texts: Effects of interactive overviews for information access. Cognition and Instruction, 13 (3), 431-468.

Dias, P., Gomes, M. J., & Correia, A. P. (1999). Disorientation in hypermedia environments: Mechanisms to support navigation. Journal of Educational Computing Research, 20 (2), 93-117.

Dillon, A. (1994). Designing Usable Electronic Text: Ergonomic Aspects of Human Information Usage. Philadelphia, PA: Taylor and Francis.

Dillon, A. (1996). Myths, misconceptions, and an alternative perspective on information usage and the electronic medium. In J. Rouet, J. J. Levonen, A. Dillon, R. J. Spiro (Eds.), Hypertext and Cognition (pp. 25-42). Hillside, NJ: Lawrence Erlbaum Associates.

Dillon, A., & Watson, C. (1996). User analysis in HCI - The historical lessons from individual differences research. International Journal of Human-Computer Studies, 45 (6), 619-637.

Donohew, L., Lorch, E. P., & Palmgreen, P. (1991). Sensation seeking and targeting of televised anti-drug PSAs. In L. Donohew, H. E. Sypher, & W. J. Bukoski (Eds.), Persuasive Communication and Drug Use Prevention (pp. 209-226). Hillside, NJ: Lawrence Erlbaum Associates.

Donohew, L., Lorch, E. P., & Palmgreen, P. (1998). Applications of a theoretic model of information exposure to health interventions. Human Communication Research, 24 (3), 454-468.

Donohew, L., Palmgreen, P., & Duncan, J. (1980). An activation model of information exposure. Communication Monographs, 47 (4), 295-303.

Edwards, D., & Hardman, L. (1989). Lost in hyperspace: cognitive mapping and navigation in a hypertext environment. In R. McAleese (Ed.), Hypertext: Theory into Practice (pp. 105-125). Oxford: Intellect.

Eveland. W. P., & Dunwoody, S. (2001). User control and structural isomorphism or disorientation and cognitive load?: Learning from the web versus print. Communication Research, 28 (1), 48-78.

Ferguson, M. A., Valenti, J. M., & Melwani, G. (1991). Communicating with risk takers: A public relations perspective. In L. A. Grunig & J. E. Grunig (Eds.), Public RelationsResearch Annual: Vol. 3 (pp. 195-224). Hillside: NJ: Lawrence Erlbaum Associates.

Fredin, E. S. & David, P. (1998). Browsing and the hypermedia interaction cycle: A model of self-efficacy and goal dynamics. Journalism and Mass Communication Quarterly, 75 (1), 35-54.

Gray, S. H. (1990). Using protocol analysis and drawing to study mental model construction during hypertext navigation. International Journal of Human-Computer Interaction, 2 (4), 359-377.

Hammond, N. (1989). Hypermedia and learning: Who guides whom. In H. Maurer (Ed.), Computer Assisted Learning. Lecture Notes in Computer Science, Vol. 360 (pp. 167-181). New York: Springer-Verlag.

Hammond, N., & Allinson, L. (1989). Extending hypertext for learning: An investigation of access and guidance tools. In A. Sutcliffe and L. Macauley (Eds.), Proceedings of the Fifth Conference of the British Computer Society, Human Computer Interaction Specialist Group on People and Computers V (pp. 293-304). Cambridge, UK: Cambridge University Press.

Jonassen, D. H. (1989). Hypertext/Hypermedia. Englewood Cliffs, New Jersey: Educational Technology Publications.

Jonassen, D. H., & Wang, S. (1990, October). Acquiring Structural Knowledge From Semantically Structured Hypertext. Paper presented at the annual meeting of the Association for the Development of Computer-Based Instructional Systems, San Diego, CA.

Kim, H., & Hirtle, S. C. (1995). Spatial metaphors and disorientation in hypertext browsing. Behaviour and Information Technology, 14 (4), 239-250.

Last, D. A., O'Donnell, A. M., & Kelly, A. E. (2001). The effects of prior knowledge and goal strength on the use of hypertext. Journal of Educational Multimedia and Hypermedia, 10, 3-25.

Lee, M. J. (1998, August). The effect of hypertext on readers? Recall based on gender. Poster Presented at the Annual Convention of the Association for Education in Journalism and Mass Communication, Baltimore, MD.

Lee, M. J. (2001). Effective Tailored Communication: Expanding Hypertext, Working Memory and Risk Taking Tendency. Unpublished Ph.D. dissertation. University of Florida, Florida.

Lee, M. J., Ferguson, M. A. & Tedder, M. C. (1999, November). The effect of hypertext on readers' recall and comprehension based on working memory span. Paper presented at the National Communication Association, 1999 Annual Convention, Chicago.

Lee, M. J., & Tedder, M. C. (1999, November). The effects of three different computer texts on readers' recall (based on working memory, risk-taking tendencies, and hypertext familiarity and knowledge). Paper presented at the annual convention of the National Communication Association, Chicago, IL.

Lee, M. J., & Tedder, M. C. (2003). The effects of three different computer texts on readers' recall (based on working memory, risk-taking tendencies, and hypertext familiarity and knowledge). Computers in Human Behavior, 19 (6), 767-783.

Lee, M. J., & Tedder, M. C. (2004). Introducing expanding hypertext based on working memory capacity and the feeling of disorientation: Tailored communication through effective hypertext design. Journal of Educational Computing Research, 30 (3), 171-195.

Lee M. J., Xie, G., & Tedder, M. C. (In press). Effective computer text design to enhance readers' recall: Text formats, individual working memory capacity and content type. Journal of Technical Writing and Communication.

Lin, C., & Davidson, G. (1996). Effect of linking structure and cognitive style on students' performance and attitude in a computer-based hypertext environment. Journal of Educational Computing Research, 15 (4), 317-329.

Liu, M. (1994). Hypermedia-assisted-instruction and second language learning: A semantic-network-based approach. Computers in the Schools, 10 (3/4), 293-312.

Macedo-Rouet, M., Rouet, J., Epstein, I., & Fayard, P. (2003). Effects on online reading on popular science comprehension. Science Communication, 25 (2), 99.

MacGregor, S. K. (1999). Hypermedia navigation profiles: Cognitive characteristics and information processing strategies. Journal of Educational Computing Research, 20 (2), 189-206.

McAllese, R. (1991). The acquisition and representations of domain specific knowledge using notecards. In R. MaAleese (Eds.), Hypermedia Courseware Structures of Communication and Intelligent Help (pp. 11-18). Berlin: Springer-Verlag.

McDonald, S. (1998). Effects of text structure and prior knowledge of the learner on navigation in hypertext. Human Factors, 40 (1), 18 -27.

McDonald, S., & Stevenson, R. I. (1996). Disorientation in hypertext: The effects of three text structures on navigation performance. Applied Ergonomics, 27 (1), 61-68.

McDonald, S., & Stevenson, R. J. (1999). Spatial versus conceptual maps as learning tools in hypertext. Journal of Educational Multimedia and Hypermedia, 8 (1), 43-64.

McKnight, C. (1990). A comparison of linear and hypertext formats in information retrieval. In R. McAleese & C. Green (Eds.), Hypertext: State of the Art (pp. 10-19). Oxford: Intellect.

McKnight, C. (1996). What makes a good hypertext? In H. Van Oosterndorp & S. De Mul (Eds.), Cognitive Aspects of Electronic Text Processing (pp. 213-238). Norwood, NJ: Ablex.

Newmarch, J. (1998, November). An Investigation Into the Effect of Hypertext Structure on Student Use of Courseware. Paper was presented at WebNet 98, Orlando, FL.

Nielsen, J. (1990). Hypertext and Hypermedia. San Diego, CA: Academic Press.

Palmgreen, P., Donohew, L., Lorch, E. P., & Rogus, M. (1991). Sensation seeking, message sensation value, and drug use as mediators of PSA effectiveness. Health Communication, 3 (4), 217-227.

Phillips, T. L., Watson-Paelis, G., Cook, J., Ming, L., & Tiancheng, L. (1992, February). The Effects of Structure Aids on Level of Learning and Efficiency in an Instructional Hypermedia Environment. Paper presented at the annual conference of the Association for Educational Communications and Technology, Washington, DC.

Piolat, A., Roussey, J. Y., and Thunin, O. (1997). Effects of screen presentation on text reading and revising. International Journal of Human-Computer Studies, 47 (4), 565-589.

Pohl, M. (1998, June). Hypermedia as a Cognitive Tool. Paper presented at 98 World Conference on Educational Multimedia and Hypermedia & World Conference on Educational Telecommunications, Freiburg, Germany.

Rouet, J., Levonen, J. J., Dillon, A., & Spiro, R. J. (1996). An introduction to hypertext and cognition. In J. Rouet, J. J. Levonen, A. Dillon, & J. Spiro (Eds.), Hypertext and Cognition (pp. 3-8). Mahwah, NJ: Lawrence Erlbaum Associates. Schierhorn, C., Wearden, S., Schierhorm, A., Tabar, P., & Andrews, S. C. (1998, August). Digital Formats For the Future: The Web vs. Paper vs. Vertical-Screen, Page-Based Design. Paper presented at annual convention of Association of Education in Journalism and Mass Communication, Baltimore, MD.

Schroeder, E. E. (1994). Navigating Through Hypertext: Navigational Technique, Individual Differences, and Learning. Paper presented at the nation convention of the Association for Educational Communications and Technology, February 16-20, Nashville, TN.

Shapiro, A. (1998). Promoting active learning: The role of system structure in learning from hypertext. Human-Computer Interaction, 13 (1), 1-35.

Shneiderman, B. (1992). Designing the User Interface: Strategies for Effective Human-Computer Interaction. Boston, MA: Addison-Wesley.

Smith, P. A., & Wilson, J. R. (1993). Navigation in hypertext through virtual environments. Applied Ergonomics, 24 (4), 271-278.

Spiro, R. J. & Jehng, J. C. (1990). Cognitive flexibility and hypertext: Theory and technology for the nonlinear and multidimensional traversal of compels subject matter. In D. Nix & R. J. Spiro (Eds.), Cognition, Education, and Multimedia: Exploring Ideas in High Technology. Hillside, NJ: Lawrence Erlbaum.

Stevenson, R. J., & Palmer, J. A. (1994). Learning: Principles, Processes, and Practices. London: Cassell.

Tombaugh, J., Lickorish, A., and Wright, P. (1987). Multi-window displays for readers of lengthy texts. International Journal of Man-Machine Studies, 26 (5), 597-616.

Unz, D. C., & Hesse, F. W. (1999). The use of hypertext for learning. Journal of Educational Computing Research, 20 (3), 279-295.

Van Nimwegen, C., Pouw, M., & van Oostendorp, H. (1999). The influence of structure and reading-manipulation on usability of hypertexts. Interacting with Computers, 12 (1), 7-21.

Zillmann, D. & Bryant, J. (1985). Selective Exposure to Communication. Hillsdale, NJ: Lawrence Erlbaum Associates.

Zuckerman, M. (1971). Dimensions of sensation seeking. Journal of Consulting and Clinical Psychology, 36 (1), 45-52.

Zuckerman, M. (1979). Sensation Seeking: Beyond the Optimal Level of Arousal. Hillside, NJ: Lawrence Erlbaum.

Zuckerman, M. (1990). The psychophysiology of sensation seeking. Journal of Personality, 58 (1), 313-345.

Zuckerman, M. (1994). Behavioral Expression and Biosocial Bases of Sensation Seeking. New York: Cambridge University Press.

Zuckerman, M., Kolin, E. A., Price, L., & Zoob, I. (1964). Development of a sensation-seeking scale. Journal of Consulting Psychology, 28 (6), 477-482.

Zuckerman, M., Persky, H., Hopkins, T. R., Murtaugh, T., Basu, G. K., & Schilling, M. (1966). Comparison of stress effects of perceptual and social isolation. Archives of General Psychiatry, 14 (4), 356-365.

About the Author

Moon J. Lee (Ph. D., University of Florida) is an assistant professor at Washington State University. Her research focuses primarily on the development and evaluation of new media technologies, health communication campaigns, and information processing and decision-making process of individuals. She serves as a PI and co-PI on several grant projects that include the development and evaluation of "Self-voicing Test" (a computer test system for the blind) and "Generator of Accessible Tests" at Educational Testing Service, and "Expanding Hypertext," a new text presentation technology.
Address: Murrow 241BB PO Box 642520, Pullman WA 99164-2520 USA