Content uploaded by Peter Vorderer
Author content
All content in this area was uploaded by Peter Vorderer on Feb 03, 2014
Content may be subject to copyright.
Requests for reprints should be sent to Peter Vorderer, Department of Journalism and
Communication Research, Hannover University of Music and Drama,
Hohenzollernstrasse 47, 30161 Hannover, Germany. E-mail: peter.vorderer@hmt-
hannover.de
Does Entertainment Suffer From
Interactivity? The Impact of Watching an
Interactive TV Movie on Viewers’
Experience of Entertainment
Peter Vorderer
Department of Journalism and Communication Research
Hannover University of Music and Drama
Silvia Knobloch
Department of Communication Research
Dresden University of Technology
Holger Schramm
Department of Journalism and Communication Research
Hannover University of Music and Drama
To investigate whether interactivity enhances entertainment in
interdependence with individual factors, an experiment with a 3
2 between-
subject design was conducted. 427 participants aged between 14 and 49 were
randomly assigned to a 30-minute TV movie with three different interactivity
levels and two different introductions that should diversify feelings of
sympathy toward the protagonist. Dependent variables like empathy with the
protagonist, suspense, and movie evaluation were measured after exposure,
and some were measured during exposure. A number of personality scales
were also presented at the beginning of each session. The results indicate that
individuals with lesser cognitive capacity feel more entertained, that is, they
feel more empathic toward the protagonist, feel more suspense, and evaluate
the movie more positively when they watch it without any interactivity, in the
traditional passive manner. For individuals with greater cognitive capacity, it
is just the opposite: They can use interactivity to their advantage. An increase
in their entertainment experience was observed as a direct consequence of
their being able to influence the plot of the movie.
Most contemporary research dealing with media entertainment is based on, or at
least associated with, the so-called disposition theory of entertainment (see
MEDIA PSYCHOLOGY, 3, 343–363.
Copyright © 2001, Lawrence Erlbaum Associates, Inc.
3 4 4 VORDERER, KNOBLOCH, & SCHRAMM
Zillmann & Vorderer, 2000, for an overview). This theory sees the media users’
emotional relationships to the protagonists and antagonists of fictional stories
and the favorable outcome of the story as essential for any entertainment
experience (Zillmann, 1994, 1996). It predicts positive sentiments for media
users if they witness or predict positive outcomes for favorable, liked, and
beloved protagonists, and negative outcomes for unfavorable, disliked, and
resented antagonists. Consequently, it expects negative sentiments if viewers,
listeners, or readers witness negative outcomes for their favorites or positive
outcomes for the antagonists. The reason for the media users’ emotional
dependency on the fate of the characters lies in their ability and readiness to
empathize with their heroes (Zillmann, 1991). Exposure to TV drama may serve
as an example: Whenever TV viewers care about a fictional character in such a
drama, they not only hope for the protagonist’s positive outcome and fear a
negative ending, they also consequently share all emotions along with the
character. In a typical drama situation, when the character’s failure becomes
l i k e l y, they may even feel empathic stress, a rather negative emotional
experience, better known as suspense (Vorderer & Knobloch, 2000). Unpleasant
as it may be during exposure, when the conflicts are finally resolved and the
onlookers are presented with a positive outcome for their beloved protagonist,
former distress turns into relief. It is exactly this relief that, in retrospect, is often
described as the core of the viewers’entertainment experience (Vorderer, Wulff,
& Friedrichsen, 1996).
However, this feeling of suspense is only possible when viewers evaluate the
story’s protagonist positively, that is, when they like him or her (Zillmann &
Cantor, 1977). The same dependency on the sentiment toward the protagonist
applies to another feature of entertainment—the feeling of empathy with this
character: Empathy requires sympathy, or at least an absence of resentment
(Vorderer, 1996). Without this positive evaluation, empathy is replaced by
counterempathy toward a resented character (Zillmann, 1996). In other words,
the moral judgement of the story’s protagonist is the central prerequisite for the
type and intensity of the viewers’entertainment experience.
Despite the importance of such a positive judgment, very little empirical
research has been done to assess its actual impact on the media users’
entertainment experience, that is, on their feeling of empathy (with the
protagonist) and of suspense (due to the ongoing narrative), and on their
evaluation of the media offer itself. Therefore, before turning to the relationship
between entertainment and interactivity, we would first like to clarify this
relationship more precisely in the context of exposure to a rather traditional
narrative, a TV movie.
Hypothesis 1: Viewers who like the fictional character in the movie they
watch will feel better entertained (will be more empathic toward the
protagonist), will feel more suspense when he or she is in danger, and will
evaluate the movie more positively than those viewers who do not like the
character or who like this character less.
Although disposition theory has already been applied to different media and
to different genres (for an overview, see Zillmann & Vorderer, 2000), it implies
another important prerequisite: Within that theory, media users are consistently
considered merely witnesses of the events they perceive during exposure. It is
one of the basic underlying assumptions of affective disposition theory, that
media users are not capable of influencing the narrative they are exposed to. With
TV, this assumption is plausible, insofar as TV viewers can in fact only watch
and listen to the continuation of the story presented on the screen without
intervening. However, with interactive media, such as computer games, users do
have the ability to influence the continuation of the story; they may alter the
narrative while using the medium. In this case, the question arises that, given that
this experience traditionally requires a certain degree of passivity on the users’
side, what happens to their experience of entertainment?
The problem becomes even more relevant as the “traditional” media, like
television, have become more interactive over the last few years. There is not
only an increase in the variety TV offers viewers; TV also has begun to admit
onlookers’ influence on its contents (Rockwell & Bryant, 1999). The choice of
the camera perspective in what has already been called interactive television is
only the first step. Further and certainly more appealing developments will
follow. The possibility of choosing how a movie will continue throughout
exposure, for example, has been discussed as one of the far-reaching and
certainly much more attractive developments of truly interactive TV (Vorderer,
2000).
Based on the background of the earlier-mentioned prerequisites for feeling
entertained, we would like to address the following question: What exactly
happens to this entertainment experience, and in particular to suspense, empathy,
and to the evaluation of the media offering, when viewers have an opportunity to
use a rather traditional offering, such as a TV movie, in a nontraditional
(interactive) way? Two answers are conceivable: Vi e w e r s ’ e n t e r t a i n m e n t
experience will be intensified, because the media users become more involved in
the narrative while using it interactively, or the entertainment experience will be
less entertaining. The assumption that viewers’entertainment experience will be
somehow intensified is based on the observation of computer-game players, who
I M PA C T OF INTERACTIVE T V 3 4 5
3 4 6 VORDERER, KNOBLOCH, & SCHRAMM
obviously do not simply enjoy a game by witnessing the development or
unfolding of a story; they also seem to enjoy struggling with the task and
consistently trying to do their best in what is expected and demanded by the rules
of the game (Grodal, 2000). They often try to compete with others or with their
own previous performance and, hence, they may feel triumph if they are
successful, and disappoinment if they fail. The highest gratification, together
with a very strong sense of being entertained, is often felt when both conditions
are combined: when the players meet or even surpass the self-set requirements
and expectations of the game (as is the case in situations of flow;
Csikszentmihalyi, 1990), while at the same time the story itself presents a
beloved hero as being successful. If, however, only one of these two conditions
is given, entertainment may feel less intense, which should result in less
favorable assessments of the game.
Based on the background of affective-disposition theory, however, it may be
argued that media users are only distracted and overwhelmed by the possibility
of interacting with a narrative. Instead of purely witnessing the events in order to
develop sympathy, empathy, and suspense toward the hero, they have to struggle
with new technology and making decisions about the ongoing narrative of a
movie. The cognitive capacities and information processing abilities of media
users may be of central importance here, as they are in the realm of new media
in general. Taking both perspectives and assumptions into account, the following
hypothesis seems plausible:
Hypothesis 2: Viewers capable of processing information rather effortlessly
(and not overwhelmed or distressed by interactivity) will feel better
entertained (will feel more empathic toward the protagonist), will feel more
suspense when the protagonist is in danger, and will evaluate a movie more
positively when watching this movie interactively than when they cannot
interact with it. In contrast, viewers without this ability to process information
rather effortlessly will feel better entertained when watching this movie in a
traditional, noninteractive mode.
In addition to the viewers’cognitive capacity, there are certainly other factors
influencing entertainment experience while interacting with a movie. Age and
certain attitudes may be among them. For example, it seems to be likely that
young users with a strong affinity for new technology will be more eager to
entertain themselves with interactive TV than older people, who are more
skeptical toward new technologies. Previous experience with interactive media
should also be of importance here. Viewers who are used to interacting with
I M PA C T OF INTERACTIVE T V 3 4 7
computer games and/or with the Internet should be entertained by an interactive
movie more easily than those who have used the media primarily in a traditional
way. Finally, general preferences and personality factors, such as the ability and
readiness to empathize with others, could also influence the experience of
entertainment, particularly of suspense, while watching an interactive movie.
The question as to which of the earlier-mentioned factors is of more or less
relevance remains to be answered. Consequently, we formulate the following
research question:
RQ: How do users’age and previous experience with interactive media, their
generalized preference for suspenseful media contents, and their ability and
readiness to empathize with others influence their use of and their
entertainment experience with interactive movies?
METHOD
In order to test the two hypotheses, an experimental study with a 3 × 2 between-
subject design was conducted.
PARTICIPANTS
The 427 German participants were between the ages of 14 and 49, were
randomly selected, and their cognitive abilities and previous experience with
interactive media were controlled. All respondents were paid DM 20 (~USD
9.50) for their participation.
DESIGN AND INDEPENDENT VARIABLES
All 427 participants were randomly assigned to the conditions of this 3
(interactivity) × 2 (parasocial relationship) between-subject design. Interactivity
was manipulated by giving the participants the possibility to intervene with the
ongoing narrative of a movie either three times (high level of interactivity), one
time (low level of interactivity), or not at all (no interactivity). The second
independent variable should influence feelings of sympathy toward the
protagonist by establishing a parasocial relationship between the onlookers and
the protagonist of the movie. Whereas one-half of the sample saw an introduction
to the movie that provided some information about the private background of the
protagonist, thereby encouraging sympathy (parasocial relation), the other half
followed an introduction that provided technical information about the movie
itself (no parasocial relationship).
3 4 8 VORDERER, KNOBLOCH, & SCHRAMM
EXPERIMENTALMATERIALS
The experimental movie was a thriller (Finke, 1992) produced for German public
television in 1991 (“Moerderische Entscheidung–Umschalten erwuenscht”
[Murderous Decision: Switching Required]). It was available in two versions:
one for each of the two public channels in Germany. Both versions were aired
simultaneously in 1991. This movie was used in this study only because it
provided these two different versions that made it ideal for allowing the different
options presented to the respondents in the two interactivity groups. The original
movie also had to be shortened in order to provide a 30-minute story that could
be digitalized and stored on a CD, from where it could be retrieved throughout
the experiment.
PROCEDURE
Each participant saw the 30-minute movie, which was interrupted three times
during exposure. During these interruptions, all respondents were asked how
much they could empathize with the protagonist. Following this question, for
participants in the no-interactivity condition and in the low-level-of-interactivity
conditions the movie continued. For respondents in the high-level-of-
interactivity condition, different options regarding how the movie might continue
were presented. For example, the movie was interrupted in the middle of a
conversation about money between two characters, one of them being the
protagonist. Participants in the high-level-of-interactivity condition could choose
whether they wanted to see the female character hand some money to the male
protagonist (first option), the female character kiss the male protagonist (second
option), or have the movie continue in the way the director thinks would be the
best version (third option). If they chose the third option, the respondents saw in
fact one of the first two versions, without knowing it. If they chose one of the
first two options, they did in fact see what they chose, but this presentation lasted
only for a few seconds, before returning to the movie that was the same for
everybody (again, of course, without knowing it). Participants in the low-level-
of-interactivity or no-interactivity conditions had the chance to alter the ongoing
development only at the end of the movie (low interactivity) or not at all (no
interactivity), but they were also asked about their feelings of empathy toward
the protagonist. With this procedure, two-thirds of the participants felt they
interacted with the movie, although all were exposed to almost the same movie.
DEPENDENT MEASURES
Most of the dependent measurements were obtained using a computer-aided
questionnaire; the questions were presented on a screen and could be answered
by the respondents using a computer mouse. This required an interface that was
developed with Visual Basic software. At the beginning of each session, the
respondents could practice using the mouse, in case they were unfamiliar with
one (which was rare). Then a number of personality scales were presented on the
screen, including a scale to measure the generalized preference for suspenseful
media content, and a scale to measure the ability and readiness to feel empathy
toward others. The participants were also given instructions, each according to
the earlier described experimental manipulations. The movie was then displayed
on the same screen in television quality. Neither a keyboard nor the computer
itself could be seen, so the participants were not reminded of the fact that they
were actually using a computer and not a television set. The questions presented
at the end of the movie investigated the feelings and moods of the respondents,
their evaluation of protagonists, technical or involving aspects of the movie, and
the possibility of intervening in the ongoing narrative of the movie.
In order to obtain information about the participants’ capability to process
information, two indicators were used. First, the sample was again divided into
two different groups, depending on whether they graduated from high school
(graduates) or not (nongraduates). It should be emphasized that the German
school system features three graduation levels and that by far not everybody
attains a high-school level (Abitur). As such, high school graduation served as
the first indicator of the respondents’ cognitive capacities. Another indicator,
usually considered as manifestation of the pace of information processing, was
response time, a cognitive correlate of intelligence (Neubauer, 1990; Schweizer,
1995). This information could be measured unobtrusively, in contrast to other
possible tests for mental ability that would have extended the comprehensive
session furthermore. These two measures for capability of information
processing are unsurprisingly interdependent but not equal (graduates had a
significantly shorter response time than nongraduates, M = 7.7 seconds vs. M =
8.6, t(423) = 2.9, p < .01, but the groups resulting from a split by response-time
median only overlap by 59% with the grouping along with formal education).
Because on-line measures were obtained throughout the experiment, it was
possible to take the time participants required to answer the questions displayed
on the screen as response times. However, prior experience with interactive
media may heavily influence these response times. It was necessary, therefore, to
adjust for such experiences with regard to the response times obtained in the
I M PA C T OF INTERACTIVE T V 3 4 9
experiment. As a second indicator of the respondents’ cognitive capacities,
therefore, average response times were measured prior to the experiment to take
into account the different levels of experience with interactive media.
A 5-point Likert scale was applied to the statement, “Overall, the movie was
suspenseful,” and displayed on the screen at the end of the movie in order to
measure suspense. In addition, participants were asked to continuously rate their
level of suspense throughout exposure by using a device that resembled a rather
small box with a little button on top that could be moved up or down. Of course,
the participants were instructed prior to exposure to move the button upward
whenever they felt more suspense, and downward whenever suspense decreased.
They could rate their level of suspense in five different degrees. Next to the
screen on which the movie was displayed was a light signal, visible to the
respondents, indicating the direction in which they had moved the button.
Personal conversations with the participants after the experiment, together with
analyses of the information provided by using these tools (for example, about the
frequency of changing the position of the button throughout reception) did not
indicate any problems relating to their usage. The on-line measure of suspense
yielded measures for a period of 30 min, a total of 1,851 sec.
Empathy with the protagonist was measured four times by interrupting the
movie during exposure, and once again following the end of the movie. The item
was constructed as a rating on a 5-point Likert scale, in answer to the following
question: “Right now, do you feel for Stefan?” (who was the protagonist).
In addition to using various scales to measure the evaluation of the movie, the
protagonist and interactivity, an additional item was used to cover the overall
impression of the movie (again on a 5-point Likert scale): “How did you find the
movie in general?”
RESULTS
Hypothesis 1: The impact of sympathy for the protagonist on entertainment
This hypothesis predicts a strong impact of viewers’ feelings for the protagonist
on their experience of entertainment, that is, on their empathy with the
protagonist, on their feeling of suspense, and on their evaluation of the movie. In
order to prove this hypothesis, analyses of variance (ANOVA) were conducted
with parasocial relationship as an independent factor, and empathy, suspense,
and evaluation as dependent measures. Prior to this, a manipulation check was
applied to control the influence of the experimental manipulation of parasocial
relationship on the sympathy toward the protagonist. As Table 1 shows,
3 5 0 VORDERER, KNOBLOCH, & SCHRAMM
participants who viewed an introduction to the movie that provided information
about the private background of the protagonist (parasocial relationship
condition) found the protagonist to be significantly more sympathetic than those
who received an introduction providing only technical information about the
movie itself, t(426) = 4.28; p < .01. In contrast to this, the other characters of the
story did not rate any more or less sympathetic according to manipulation (Table
1). Therefore, it can be assumed, that the two experimental conditions (with or
without parasocial relationship) did in fact lead to different feelings of sympathy
toward the protagonist.
TABLE 1
Sympathy Toward Characters by Either Receiving or Not Receiving Information
About the Protagonist
Treatment
Information Information
About the Protagonist About the Movie
Sympathy To... M SD M SD
Character 1 (protagonist) 4.25 .88 3.88 .91
Character 2 3.03 .89 3.07 .80
Character 3 2.09 .99 2.09 .95
However, this effect did not last very long. When measuring how much
sympathy the respondents felt for the protagonist at the end of exposure, all
differences had disappeared. Watching the movie for 30 minutes obviously had
a stronger impact on the participants’ evaluation of the protagonist than did the
introduction. We therefore decided to take this measure of post-exposure
sympathy into account instead of the manipulated conditions. Based on this, the
influence of sympathy for the protagonist on empathy, suspense, and evaluation
was analyzed by dividing the total sample (split-half) into two groups
(respondents who felt more sympathy for the protagonist and respondents who
felt less sympathy). As Figure 1 shows, participants felt significantly more
empathy (M = 3.37 vs. M = 2.34; t(426) = –11.81, p < .01), more suspense (M =
3.42 vs. M = 2.65; t(426) = –7.19, p < .01), and evaluated the movie more
positively (M = 3.10 vs. M = 2.38; t(426) = –7.80; p < .01) when they liked the
protagonist than when they liked him less.
I M PA C T OF INTERACTIVE T V 3 5 1
Figure 1. Influence of sympathy on empathy, suspense, and movie evaluation.
Hypothesis 2: The impact of interactivity and cognitive capacity on
entertainment
This hypothesis predicts that respondents with greater cognitive capacity will
feel better entertained when watching the movie interactively than when they
cannot interact with a movie. In contrast, viewers with less capacity should feel
better entertained when watching the movie in a traditional noninteractive mode.
In order to test this hypothesis, ANOVAs were conducted with “interactivity”
and cognitive capacities, indicated by graduates versus nongraduates or by
average response time as independent factors, and empathy, suspense, and
evaluation as dependent measures. The resulting six groups comprising three
levels of interactivity and two levels of cognitive capacity consisted of
approximately 70 respondents each.
The average intensity of feeling empathic with the protagonists was reported
to be quite moderate throughout the reception. For the total sample, M = 2.92 on
a scale from 1 (not at all) to 5 (very much), SD = 1.02. However, it was in fact
heavily influenced by the respondents’ cognitive capacities. Using the first
indicator, ANOVA showed that viewers who were high school graduates scored
significantly lower (M = 2.81) than those who were nongraduates, M = 3.04,
t(423) = 2.32, p < .05. In addition, a statistical interaction was found, indicating
that graduates felt more empathy when they were watching the movie in an
3 5 2 VORDERER, KNOBLOCH, & SCHRAMM
3.1
3.37
2.34
3.42
2.65
2.38
2
2.4
2.8
3.2
3.6
low sympathy high sympathy
Empathy
Suspense
Movie-Evaluation
interactive way: M = 2.94 for the low-interactivity condition, and M = 2.81 for
the high-interactivity condition, in contrast to M = 2.66 for the no-interactivity
condition: Nongraduates felt more empathy when they were exposed to the
movie without interactivity, M = 3.21, in contrast to M = 2.91 and M = 2.98,
respectively; F(2, 419) = 3.02; p < .05 (see Figure 2).
Figure 2. Influence of graduate status and interactivity on empathy.
The average level of suspense yielded M = 3.08 (SD = 1.17) for the total
sample, with neither cognitive capacities nor interactivity individually having
any influence. However, when both factors were taken into consideration
simultaneously, a statistical interaction was found: Graduates felt the movie was
significantly more suspenseful when they had a chance to intervene with it (M =
3.19 for the high-interactivity condition, and M = 3.24 for the low-interactivity
condition, in contrast to M = 2.83 for the no-interactivity condition), whereas
nongraduates found it to be more suspenseful when they saw the movie in a
noninteractive way (M = 3.23, in contrast to 3.07 and 2.97 for the two
interactivity conditions). Hence, for both graduates and nongraduates, it wasn’t
relevant how often they could intervene with the movie, but whether they could
intervene at all, F(2, 419) = 3.05; p < .05 (see Figure 3).
I M PA C T OF INTERACTIVE T V 3 5 3
2.98
3.21
2.91
2.81
2.66
2.94
2.4
2.6
2.8
3
3.2
3.4
0 1 3
Interactivity
nongraduates
graduates
Figure 3. Influence of graduate status and interactivity on suspense.
A similar pattern was found when using the second indicator cognitive
capacities. Again, differences between the six groups (three levels of interactivity
and two levels of cognitive capacy) were observed, this time as a result of
dividing the sample (median split) into two subgroups. One half answered the
computer-aided questionnaire rather quickly (greater capacities), one half doing
so rather slowly (lesser capacities), and prior experience in dealing with
interactive media was taken into account as a covariate.
With regard to suspense, neither cognitive capacity nor interactivity
individually showed any influence on the average level of suspense, M = 3.08,
SD = 1.17. However, when both factors were taken into consideration
simultaneously, a statistical interaction could be found. Participants with greater
capacity (shorter response times) felt the movie was significantly more
suspenseful when they had a chance to intervene (M = 3.06 for the low-
interactivity condition, and M = 3.19 for the high-interactivity condition, in
contrast to M = 2.74 for the no-interactivity condition), whereas respondents with
lesser capacity (longer response times) found it to be more suspenseful when
they saw the movie in a noninteractive way (M = 3.31, in contrast to 3.23 and
2.95 for the two interactivity conditions). Once again, the combination of
cognitive capacity and interactivity led to significant differences, F(2, 419) =
4.22; p < .05 (see Figure 4).
3 5 4 VORDERER, KNOBLOCH, & SCHRAMM
2.97
3.23
3.07
3.19
2.83
3.24
2.7
2.9
3.1
3.3
0 1 3
Interactivity
nongraduates
graduates
Figure 4. Influence of cognitive capacity and interactivity on suspense.
The general evaluation of the movie was tentatively negative across all
participants, M = 2.79, SD = 1.01. Neither their formal education nor the
different number of interventions possible with the ongoing narrative of the
movie influenced this evaluation. The same holds true for the combination of
both factors: No interaction effect could be found. However, when response
times were taken into account (as second indicator for cognitive capacities), a
significant interaction effect was found. Participants with lesser cognitive
capacities evaluated the movie positively when they saw it in the no-interactivity
condition (M = 3.07 vs. M = 2.82 for the low-interactivity condition, and M =
2.71 for the high-interactivity condition), whereas participants with greater
capacities evaluated the highly interactive version positively (M = 2.94 vs. 2.55
for the no-interactivity condition, and M = 2.56 for the low-interactivity
condition; F(2, 419) = 5.07, p < .01; see Figure 5).
As a matter of fact, the possibility of interacting with a movie has an opposite
effect on suspense and on the evaluation of the movie within the two different
groups of participants: Individuals with greater cognitive capacities (as indicated
both by their graduate status and by their average response times) feel more
suspense with characters and evaluate the movie positively when they have a
chance to intervene with the narrative. On the other hand, for individuals with
lesser cognitive capacities, it is just the opposite. They feel more suspense and
evaluate the movie positively when they do not have to bother with interactivity.
The two hypotheses can therefore be considered supported for the most part.
I M PA C T OF INTERACTIVE T V 3 5 5
2.95
3.31
3.23
3.19
2.74
3.06
2.6
2.8
3
3.2
3.4
0 1 3
Interactivity
low capacity
high capacity
Figure 5. Influence of cognitive capacity and interactivity on evaluation of movie.
Empathy and Suspense as Processes
However, the analyses of suspense were based on a single postexposure measure
only, and the indicator of empathy is limited to the mean of four different
measures conducted during and after exposure. It may be argued that these
measures, particularly those of suspense, do not give a clear picture of the
participants’experience. This experience may be considered to come about as a
process and not so much as a state in which viewers happen to be after exposure.
If empathy and suspense are, in fact, a process rather than a state, they may
change significantly throughout reception. Thus, what respondents usually rate
as suspenseful may be nothing but an overall and perhaps strongly biased
estimation of how the process looks in retrospect. It therefore seems to be
necessary to expand and to broaden the measures by taking into account what
happens to suspense during exposure.
As previously mentioned, the participants in this experiment were asked to
indicate the level of suspense they felt continuously, that is, on-line. Given these on-
line measures, it is possible to study the feeling of suspense over a certain period of
time so that one does not have to rely exclusively on single measures after exposure.
Time-series analyses lend themselves to scrutinize the data base in great detail, but
due to the complexity of the method and the large number of respondents, such an
analysis goes beyond our focus.We instead only inspect the obtained data visually.
But before doing so, we first refer to the four different measures of empathy.
3 5 6 VORDERER, KNOBLOCH, & SCHRAMM
2.82
3.07
2.71
2.562.55
2.94
2.5
2.7
2.9
3.1
0 1 3
Interactivity
low capacity
high capacity
As far as empathy is concerned, the overall measures (for the total sample)
look quite usual (M1 = 2.90, M2 = 2.81; M3 = 2.62, M4 = 3.33), so empathy was
almost stable. With only a slight decrease during reception until the very end of
the movie when it turned around and increased. It should be kept in mind that the
fourth and last measure of empathy was obtained after exposure. A significantly
higher score of empathy after the movie might in fact indicate a biased estimation
by the respondents, because ex-post assessments frequently differ from reports
during reception (Fredrickson & Kahneman, 1993).
The same can be shown for suspense. Again, the level of suspense felt
throughout exposure was lower at any point than stated by the respondents in the
single postexposure measure (reported earlier). If one calculates the mean for the
total sample, based on 1,851 seconds, M = 2.32, whereas the postexposure
measure (again for the total sample) was M = 3.08 (see earlier). Looking at the
process itself, there is a slight increase of suspense over the duration of the
movie, and it is possible to identify “mini suspense and resolution episodes” in
the sense of Brewer (1996) that actually fit well with certain scenes in the movie.
Not surprisingly, participants rated their suspense the highest during the
showdown at the end of the movie (between the third and the final empathy
measure).
Along the line of differentiation, we again distinguished between the different
groups described earlier (using the first indicator of cognitive capacities: high
school graduation), this time applying the procedure to the on-line measures of
suspense and the repeated measures of empathy. In order to make it easier to
follow the figures, we only include the extreme groups, those with either a high
level of interactivity or with no interactivity. Figures 6 and 7 also show the points
in time at which the viewers could choose how the movie should continue (given
they were in the high-interactivity condition) and the repeated measures of
empathy, respectively. Figure 6 describes the respondents with greater cognitive
capacities, using graduate status as indicator, Figure 7 those with lesser
capacities.
As can be seen, the course of suspense looks pretty similar when comparing
the graduates in the high-interactivity condition to those in the no-interactivity
condition (Figure 6). When they differ from each other, the highly interactive
respondents felt the movie to be more suspenseful than those who didn’t interact.
Again, empathy was higher among the highly educated when they could interact,
for the first and fourth measures in particular (Figure 6).
I M PA C T OF INTERACTIVE T V 3 5 7
3 5 8 VORDERER, KNOBLOCH, & SCHRAMM
Figure 6. Suspense and empathy of graduates by level of interactivity.
More significant differences can be seen for those who did not graduate from
high school (Figure 7). Interestingly enough, both courses of suspense are
practically identical up to the first point in time when both groups were
interrupted. Briefly after this interruption, the high-interactivity group (those
who actually had a choice) felt the movie was less suspenseful than those in the
no-interactivity condition (i.e., those who saw the movie interrupted, but only
because they were asked how much they empathized with the protagonist and not
because they could choose the ongoing developments). The same holds true for
empathy. The difference between the two groups developed only after the first
interruption. Interactivity seems to distract these respondents from empathizing,
with those who did not have a choice about the ongoing narrative empathizing
more (see Figure 7).
The on-line measures clearly confirm the results of the ANOVA and even
demonstrate, more precisely, that viewers with lesser cognitive capacities feel the
same about a movie as long as the movie is not interrupted and as long as they
have not been asked to contribute to the development of the movie. Those who
are asked, however, immediately begin to feel different in terms of empathy and
suspense. The differences between the groups are not a consequence of the
interruptions themselves; nor can they be experimental effects. It is the way
movies are perceived that makes onlookers’feelings change.
0.0
0.4
0.8
1.2
1.6
2.0
2.4
2.8
3.2
3.6
time
Suspense, Empathy
suspense, no interactivity suspense, high interactivity
interventions empathy, no interactivity
empathy, high interactivity
Figure 7. Suspense and empathy of nongraduates by level of interactivity.
Research Question: What influences the respondents’use and evaluation
of interactivity?
The research question addressed the issue of whether age, previous experience
with interactive media, generalized preference for suspenseful media contents,
and the respondents’ ability and readiness to empathize with other people have
an influence on their usage and evaluation of interactivity, as well as on their
entertainment experience. In order to answer this question, the following
analyses were restricted to those respondents who had the opportunity to
influence the ongoing development of the movie in the experimental condition
high interactivity (N = 142). Regression analyses were conducted to identify
factors that have an impact on the participants actually using the possibility to
intervene, as well as to check these factors’ impact on the participants’
entertainment experience. All factors mentioned in this research question were
taken into account (method = forward). Dependent measures (criteria) were,
again, empathy, suspense, and evaluation of the movie, together with the number
of possible interventions used and the evaluation of this possibility. Table 2
shows standardized β values and coefficients of the various regression models.
The results clearly show that previous experience with interactive media is the
most relevant factor for the actual usage of the interventions possible, as well as
for the evaluation of this possibility: The more previous experience participants
reported having had with interactive media, the more they chose a specific
narrative development of the movie (instead of voting for the director’s choice),
I M PA C T OF INTERACTIVE T V 3 5 9
0.0
0.4
0.8
1.2
1.6
2.0
2.4
2.8
3.2
3.6
time
Suspense, Empathy
suspense, no interactivity suspense, high interactivity
interventions empathy, no interactivity
empathy, high interactivity
and the more positively they assessed receiving the chance to do it. In contrast to
this, previous experience with interactive media did not influence the
p a r t i c i p a n t s ’ experience of entertainment. Empathy, suspense, and movie
evaluation appear to be independent from this expertise. Of importance for their
experience of entertainment, however, was the degree to which respondents
prefer suspenseful offers in general and whether they were able and willing to
empathize with other persons (see Table 2).
TABLE 2
R e g ression Coefficients of Various Factors for Using and Experiencing
Interactivity
Usage of Evaluation of Movie
Interactivity Interactivity Empathy Suspense Evaluation
Experience with
Interactive Media .16
~
.24** –.01 .06 –.03
Graduation .12 –.01 –.10 .10 –.01
Response Time .01 .06 –.01 –.11 –.09
Age .01 –.09 –.06 .01 –.07
Preference of
Suspense –.10 .02 .05 .11 .22*
Ability to
Empathize –.08 –.01 .39** .09 .07
R .22 .27 .41 .23 .27
adj. R2 .01 .03 .14 .01 .03
F 1.10 1.73 4.69** 1.25 1.83
~
N 142.00 142.00 142.00 142.00 142.00
**p < .01, *p < .05,
~
p < .10
DISCUSSION
The two hypotheses could be confirmed: Viewers who liked the movie’s
protagonist felt better entertained, that is, they empathized more with him, felt
more suspense when he was in danger, and evaluated the movie more positively
than those who liked the protagonist less. Also, viewers with lesser cognitive
capacities to process information empathized more with the protagonist when
they watched the movie in a traditional noninteractive mode. When they saw the
movie in an interactive manner, the difference reversed: In this case, the
graduates empathized more, the nongraduates less. It seems to be plausible that
3 6 0 VORDERER, KNOBLOCH, & SCHRAMM
with interactivity the nongraduates felt distracted from the movie and could not
concentrate on the protagonist as easily. Their attention may have been turned
away from the content of the movie toward themselves, as they prepared for the
next situation in which they would have to make a choice. For viewers with
greater cognitive capacities, it seems as if this choice made them feel even more
involved with the story and this, in turn, may have led them to concentrate even
more on the protagonist, thus feeling more intensely for and with him. The
additional mental effort necessary for interactivity supports a more intense
entertainment experience for those who can handle this amount of information
easily.
As far as suspense goes, again, the graduates felt more suspense when they
could make choices about the ongoing developments, whereas the nongraduates
felt less. This shows that interactive television can indeed be entertaining, but not
for everybody. Similar results were found for the evaluation of the movie.
Viewers with lesser cognitive capacity evaluated the interactive movie more
negatively than they did the same movie when they were traditionally exposed to
it. In contrast, those with greater capacity evaluated the movie more positively
when they watched it interactively. Furthermore, personality factors such as the
ability and readiness to empathize, together with a general preference for
suspenseful contents, were also of relevance for being empathic toward the
protagonist, as well as for a positive evaluation of the movie.
The general conclusion might be drawn that interactive entertainment may in
fact be even more attractive than regular entertainment if the right audience is
addressed. This is in line with what we know about television and the Internet in
general: Regular TV is most attractive to those with a lower level of education,
whereas the Internet, being the most interactive medium used nowadays,
attracted initially primarily highly educated users with vast experience in using
other interactive media. Amid this background, we suspect that the future of
interactive media will be successful only with a portion of the general audience.
Of course, the crucial prerequisite is not the high school degree, nor is it response
time. These are only indicators of specific cognitive capacities that are
worthwhile of study in more detail. For future research, more refined measures
for mental efforts and agility as well as for decision making should be applied.
Furthermore, it is desirable to study higher levels of interactivity that go beyond
interactive television, for example in computer games. However, this topic
contains even more sophisticated problems for experimental research and the
investigation of causal relationships. The days when we could make confident
statements, or even predictions, about an overall audience seem to be over.
I M PA C T OF INTERACTIVE T V 3 6 1
3 6 2 VORDERER, KNOBLOCH, & SCHRAMM
REFERENCES
Brewer, W. F. (1996). The nature of narrative suspense and the problem of
rereading. In P. Vorderer, H. J. Wulff, & M. Friedrichsen (Eds.), Suspense:
Conceptualizations, theoretical analyses, and empirical explorations (pp.
107–127). Mahwah, NJ: Lawrence Erlbaum Associates.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New
York: Harper & Row.
Finke, R. (1992). “Moerderische Entscheidung—Umschalten erwuenscht.” Ein
einmaliges Fernsehspiel? [“Murderous decision—switching required”: A
unique television game?] Unpublished master’s thesis, University of
Dortmund, Germany.
Fredrickson, B. L., & Kahneman, D. (1993). Duration neglect in retrospective
evaluations of affective episodes. Journal of Personality & Social
Psychology, 65(1), 45–55.
Grodal, T. (2000). Video games and the pleasures of control. In D. Zillmann &
P. Vorderer (Eds.), Media entertainment. The psychology of its appeal (pp.
197–213). Mahwah, NJ: Lawrence Erlbaum Associates.
Neubauer, A. C. (1990). Speed of information processing in the Hick paradigm
and response latencies in a psychometric intelligence test. Personality and
Individual Differences, 11(2), 147–152.
Rockwell, S. C., & Bryant, J. (1999). Enjoyment of interactivity in an
entertainment program for children. Medienpsychologie, 11, 244–259.
Schweizer, K. (1995). Kognitive Korrelate der Intelligenz [Cognitive correlatives
of intelligence]. Göttingen, Germany: Hogrefe.
Vorderer, P. (1993). Audience involvement and program loyalty. Poetics. Journal
of Empirical Research on Literature, Media and the Arts, 22, 89–98.
Vorderer, P. (1996). Toward a psychological theory of suspense. In P. Vorderer,
H. J. Wulff, & M. Friedrichsen (Eds.), Suspense: Conceptualizations,
theoretical analyses, and empirical explorations (pp. 233–254). Mahwah,
NJ: Lawrence Erlbaum Associates.
Vorderer, P. (2000). Interactive entertainment and beyond. In D. Zillmann & P.
Vorderer (Eds.), Media entertainment: The psychology of its appeal (pp.
21–36). Mahwah, NJ: Lawrence Erlbaum Associates.
Vorderer, P., & Knobloch, S. (2000). Conflict and suspense in drama. In D.
Zillmann & P. Vorderer (Eds.), Media entertainment. The psychology of its
appeal (pp. 59–72). Mahwah, NJ: Lawrence Erlbaum Associates.
Vo r d e r e r, P., Wu l ff, H. J., & Friedrichsen, M. (Eds.). (1996). S u s p e n s e :
Conceptualizations, theoretical analyses, and empirical explorations.
Mahwah, NJ: Lawrence Erlbaum Associates.
Zillmann, D. (1991). Empathy: Affect from bearing witness to the emotions of
others. In J. Bryant & D. Zillmann (Eds.), Responding to the screen:
Reception and reaction processes (pp. 135–168). Hillsdale, NJ: Lawrence
Erlbaum Associates.
Zillmann, D. (1994). Mechanism of emotional involvement with drama. Poetics.
Journal of Empirical Research on Literature, Media and the Arts, 23, 33–51.
Zillmann, D. (1996). The psychology of suspense in dramatic exposition. In P.
Vo r d e r e r, H. J. Wu l ff, & M. Friedrichsen (Eds.), S u s p e n s e :
Conceptualizations, theoretical analyses, and empirical explorations (pp.
199–231). Mahwah, NJ: Lawrence Erlbaum Associates.
Zillmann, D., & Cantor, J. R. (1977). Affective responses to the emotions of a
protagonist. Journal of Experimental Social Psychology, 13, 155–165.
Zillmann, D., & Vo r d e r e r, P. (Eds.). (2000). Media entertainment. The
psychology of its appeal. Mahwah, NJ: Lawrence Erlbaum Associates.
I M PA C T OF INTERACTIVE T V 3 6 3
