|
The effects of natural sound breaks in news stories on ORIENTING and arousal
by
MARY BETH BRADFORD
Doctoral Student
The University of Alabama
College of Communication and Information Sciences
P.O. Box 870172
Tuscaloosa, AL 35487-0172
Phone: (205) 366-9768
E-mail: [log in to unmask]
AV Needs: Overhead projector
The Effects of Natural Sound Breaks in Television News Stories on Orienting
and Arousal
Television news operations convey information to the public through both
the video and audio channels of their stories. Both channels work together
to communicate a particular message. Although the television news industry
often covers a story because it has good video, the importance of the audio
channel to supplement the video should not be ignored. If someone has the
television on while cooking dinner, it would more likely be the audio
channel that causes the person to look up at the television. But beyond
that, Hesketh and York (1993) emphasize that "pictures without good sound
lose their impact" (p. 60). If a story on a building implosion is shown on
television, the sound of the crash with the video is important. News
photographers and editors use this natural sound to enhance the production
quality of news packages (Shook, 1989). Different ambient sounds from the
scene of the story, such as the shouting protestors at an event or the
tapping of a computer keyboard in an office, give the television news
viewer more of a sense of "being there." Sound adds interest to the
pictures of a news story (Shook, 1989). Livingston (1997) writes, "Sound
is the punctuation of visual storytelling. It also helps give the story
pace that will hopefully keep the viewer involved in your story" (p. 66).
Video editors also use natural sound to provide a means of transition from
one scene to another (Speake, 1996). An editor will either fade up the
sound from the news scene or "pop" a second or two of sound between
reporter tracks or interviews. These natural sound breaks consist of any
ambient audio from the news story's original environment that "breaks up"
the interview or reporter narration and briefly becomes the dominant
audio. For instance, if a news story starts at a press conference and
moves to video of violence in the Middle East, the editor will use a
natural sound break of a building falling or guns firing to aid in the
transition between visuals.
Because of the perceived benefits to both the news producer and consumer,
some trade organizations have tried to maximize the use of good sound to
accompany good video. The National Press Photographers Association has
advocated the heavy use of natural sound breaks in news packages (Murray,
1996). However, natural sound breaks interrupt the reporter or interview
track. It is possible that this interruption may affect how the viewers
cognitively process the news story.
Although researchers are interested in how structural features affect a
viewer's cognitive processing of media (Lang, 1990), the effect of natural
sound breaks in television news has not been researched. In addition, no
content analyses have looked at the use of natural sound breaks in news
packages. The closest analysis has been the use of sound effects in
television news magazine programs (Grabe, Zhou, & Barnett, 2001). Sound
effects were differentiated from natural sound in that they were created or
manipulated in post-production. After analyzing "Hard Copy" and "60
Minutes," the researchers found that "Hard Copy" used sound effects rather
frequently (one effect every 60.02 seconds).
This study intends to explore how natural sound breaks affect cognitive
processing. A limited capacity, information processing model of television
viewing will be employed to test these effects.
Literature Review and Hypotheses
Information Processing and Capacity Models
Information processing in psychology developed out of Shannon's
information theory of communication, importing ideas such as encoding,
channel capacity, and efficiency (Lachman, Lachman, & Butterfield,
1979). Like a computer, humans are information processors. Information
processing paradigms state that information from the environment is encoded
into symbols that are compared to other symbols in memory. Human
processors then may store those symbols in memory for later retrieval.
According to Shiffrin and Schnieder (1977), there are two types of
encoding—automatic detection and controlled search. Controlled search is
the traditional idea of "paying" attention to a stimulus. This requires
more cognitive effort and involves more of a person's
intentions. Automatic detection, on the other hand, can involve processing
that is well learned and occurs often without a person's
knowledge. Automatic detection can sometimes interrupt controlled search
(Shiffrin & Schnieder, 1977), such as when an orienting response occurs.
An orienting response is a signal that the processing system has noticed a
change in the stimulus (Sokolov, 1963). It can occur when the expectations
of a stimulus are violated, or when a participant is required to assess
something new in his or her environment (Kahneman, 1973; Graham, 1979;
Lang, 1994)1. Any information processing may be interrupted by an
orienting response (Reeves, Thorson, & Schleuder, 1986). The orienting
response is a key concept in information processing since it has the
ability to control to which stimuli we attend (Ohman, 1979). When an
orienting response occurs, the processing system calls for encoding
resources to attend to the eliciting stimulus (Lang, 2000). This call for
resources is indicated through, among other measures, physiological methods
such as alpha blocking in the EEG pattern, a momentary decrease in heart
rate, or an increase in skin conductance (Thorson & Lang, 1992; Lang, 2000).
Broadbent (1958) studied information processing of audio and found that
when someone hears a separate audio message in each ear, the participant
can effectively process only one message at a time if the messages contain
a lot of information. His filter theory suggests that the cognitive system
can only handle a certain amount of information at a particular
moment. Attention can shift from one ear to another depending upon certain
factors, such as familiarity or some change in the stimulus that would
elicit an orienting response.
Kahneman (1973) described shifts in attention as partially a function of
selectivity, where an individual determines which message to which they
"pay" attention. However, there may be other mechanisms that affect our
attention. Kahneman's model suggests that how we allocate our attention
depends upon orienting responses (automatic detection), momentary
intentions (controlled search), evaluation of demands (i.e.- when two
activities require more attentional resources than we have available, we
only attend to one), and how much arousal is involved in the stimulus
(which Kahneman, 1973, argues may increase information processing
capacity). Our ability to attend to several stimuli in our environment
depends upon the cognitive effort required. Easier tasks require less
cognitive effort than complex ones. As the complexity of a stimulus
increases, more encoding resources are needed to process the message. If
there aren't enough encoding resources to process the message, the
processing system will borrow more resources from the storage and retrieval
subprocess in order to encode the message. This call for more processing
resources will breakdown the ability to store and retrieve information
(Kahneman, 1973; Lang, 2000).
Limited Capacity Processing of Electronic Media
The limited capacity model indicated by Kahneman (1973) has been adopted
to study how people process television messages (Lang, 2000). Similar to
the traditional information processing paradigms in psychology, the limited
capacity model of television viewing studies how different attributes of
television messages affect encoding, storage, and retrieval—three
subprocesses of information processing. A viewer must first encode the
message into understood representations, then store these representations
into a working memory. The viewer then retrieves a related representation
from long-term memory in order for the message to be fully
processed. Because television involves two modalities—audio and video—it
is psychologically complex.
Television messages vary in complexity in both structure and content
(Lang, 2000). A program could have complex content in that the verbal
message may be intellectually or conceptually difficult for the viewer. A
television program may also be structurally complex if it uses a lot of
graphics, edits, or sound effects. If a message is complex in both content
and structure, the demand for resources is high. Each subsequent orienting
response may cause an automatic call for resources to encoding. This may
cause the storage and retrieval subprocesses to suffer (Lang, 1995).
Orienting to Media Messages
The orienting response calls for more encoding resources to be
automatically allocated. This places more demand on the cognitive
load. Because the total amount of information an individual can process at
a given moment is limited (Kahneman, 1973), processing of television can
break down if too many resources are used for encoding.
Geiger and Reeves (1993) studied attention following cuts. Video that was
presented after the cut was either semantically related or unrelated to the
preceding video. Using secondary task reaction time, the researchers found
that more encoding resources were required for unrelated cuts than related
cuts. However, when measuring two seconds after the cut, the amount of
encoding resources required for unrelated and related cuts appeared to be
the same. Related cuts only resulted in an increase in demand for
cognitive resources between the onset of the cut and one second following
the cut. In other words, as the viewer sees a new frame of video, he or
she determines whether or not it is related to the last scene. If it is
related, resources required for encoding decreases since there is less
additional information to process.
Some structural features of television have been found to elicit orienting
responses (Shapiro & Lang, 1991), such as person movement (Lang, 1990),
graphics in televised lectures (Thorson & Lang, 1992), and cuts (Geiger &
Reeves, 1993; Lang, Geiger, Strickwerda, & Sumner, 1993). Voice changes in
radio have also been found to elicit orienting responses (Potter,
2000). All of these structural features introduce novelty into the
mediated environment.
A natural sound break is a segment of audio from a story's original news
environment that briefly interrupts the audio track. Because natural sound
breaks introduce a change in the mediated environment, it is hypothesized
that viewers will orient to it in the same way they do to other structural
features of media.
H1: A natural sound break in a television news package will elicit an
orienting response in viewers.
Recognition
The limited capacity model suggests that if an orienting response occurs,
it will affect cognitive processing of a message. If a viewer is not using
a lot of cognitive resources to process the message, he or she has the
ability to encode more information as a result of an orienting
response. However, if the viewer is using all of his or her resources
while processing the message, an orienting response will result in a
capacity overload, and the ability to store and retrieve information will
suffer (Thorson & Lang, 1992; Lang, 2000; Lang et al., 1993).
Lang et al. (1993) found that related cuts in television messages required
less processing than unrelated cuts, and that recognition memory improved
for video and audio information surrounding the related cuts. Recognition
memory for audio and video information after the unrelated cut was worse
than memory following related cuts. Visual memory after the unrelated cut
suffered more than audio memory.
In Potter's (2000) study of voice changes in radio, two-minute messages
with a medium number of voice changes (10-15 voice changes) were compared
to those with a high number of voice changes (more than 20 voice
changes). Because a voice change elicited an orienting response, a
recognition test showed that memory for information immediately after the
voice change was less than memory for information 3 seconds preceding and 3
seconds proceeding the voice change. The study found that listeners
appeared to have "borrowed" information-processing resources from storage
and retrieval in order to respond to a voice change.
Lang et al. (1999) found a non-linear relationship with structural
complexity in a study of television messages. Recognition for messages
with a moderate number of cuts (4-6 cuts in 30 seconds) was higher than
messages with a slow pace (0-1 cut) and a fast pace of cuts (11 or more in
30 seconds). Viewers were unable to keep up in the fast-paced condition
because more resources were required to encode the information, causing the
processing system to become overloaded and resulting in a breakdown of
storage. However, when calm and arousing content were compared,
recognition increased for calm messages as pace increased. With arousing
messages, recognition decreased as pace increased. This study found that
structural pacing and arousing content increased storage ability to a
certain point, then suffered as the demand for encoding resources
overloaded. With calm messages, there was not as much demand for
resources, so the viewer was able to process the faster-paced messages
(Lang et al., 1999).
Although Lang, Potter, and Bolls (1999) found that encoding visual
information can be more of an automatic process than encoding audio,
studies have found that visual recognition also can suffer due to complex
structure (Geiger & Reeves, 1993) or negatively arousing content (Lang,
Newhagen, & Reeves, 1996). If a natural sound break has an effect on
recognition similar to other structural features, it is predicted that the
recognition data will produce a similar pattern.
H2: Viewers will have significantly less visual recognition memory for
information during a natural sound break compared to information before and
after the natural sound break.
The Role of Arousal
Arousal causes more resources to be allocated to encoding and storage
(Bradley, Greenwald, Petry, & Lang, 1992). Structural complexity has been
found to increase viewer arousal. Lang et al. (1999) varied the pacing of
cuts in arousing and calm messages to see the effect on the processing
system. Cuts were defined as a shift from one scene to a completely
different scene (defined as "unrelated" in Lang et al., 1993; and Geiger &
Reeves, 1993). As the pace of the cuts increased, skin conductance levels
increased, indicating that arousal increased.
A related study differentiated cuts from edits (Lang, Zhou, Schwartz,
Bolls, & Potter, 2000). Edits were different shots within the same scene
and cuts were defined as different shots that were semantically related
(defined as "related cuts" in Geiger & Reeves, 1993), but outside of the
same visual scene. Edits were determined to be different than cuts in that
edits do not introduce as much new information. The study found that even
with edits, faster-paced messages elicited more arousal (self-reported and
physiological) than the slow and medium-paced messages (Lang et al., 2000)2.
Grabe, Zhou, Lang, and Bolls (2000) looked at how viewers processed and
evaluated tabloid and standard versions of news stories. Tabloid stories
were created by using various effects such as flash frames, sound effects,
and slow motion. Viewers that saw the tabloid version of news stories
reported feeling more aroused and exhibited higher physiological arousal
than those who saw the standard versions.
Potter and Callison's (2000) study of simple and complex promotions found
that the complex messages increased listener's self-reported
arousal. Because the complex messages had significantly more structural
features, it was suggested that more processing resources were allocated to
encoding and storage. The researchers hypothesized that multiple, repeated
orienting led to higher arousal. This has since been replicated using both
self-report and physiological data (Potter, Choi, Yu, Kim, & Carpentier, 2002).
If natural sound breaks add to the structural complexity of a news story,
it is posited:
H3: Participants will experience higher levels of skin conductance for
packages with natural sound breaks.
Method
This experiment was a 2 (version) x 4 (news stories) x 4 (order) mixed
design. Version was within subjects with two levels, natural sound breaks
or no breaks. A natural sound break is defined to be a short audio segment
from a story's original news environment that is either faded up or edited
into the news package to briefly interrupt the audio track. For example, a
natural sound break in a hot air balloon story could be the hiss of a
propane tank between reporter tracks. Each participant viewed four feature
news stories—two with natural sound breaks and two with the breaks edited
out. Order was the only between subjects factor. Across four orders, each
story appeared twice with natural sound breaks and twice without
breaks. When analyzing version, the two stories a participant saw with
natural sound breaks were compared with the two seen without natural sound
breaks.
Materials
The stimulus stories were features originally broadcast in a major
television market outside of the participants' viewing area. The stories
were general features on a hot-air balloon launch, a local gymnast, a
cruise on a tall ship, and an agricultural contest. Each story had one or
two sentences edited out in order to create recognition test foils. Each
feature had between 5-15 natural sound breaks in the original version of
the story and ran no longer than two and a half minutes in length. The
natural sound breaks were edited out of each story to produce two different
versions of the features—the original and the no natural sound version.
A recognition tape was produced in order to measure encoding. Six frames
of video (1/5 of a second) were chosen from the feature stories at certain
points of interest. One snippet of video was taken 3 seconds before a
natural sound break. Another snippet was the video occurring at the onset
of the natural sound break. A third represented video 3 seconds after the
natural sound break. Other snippets that acted as foils were taken from
similar news stories or from the original news package, where a sentence
had been edited out. There were 50 total recognition items—30 from the
stimulus material and 20 foils. Each snippet was followed by three seconds
of video black. In order to control for fatigue effects, two recognition
orders were produced.
Dependent Variables
For this study, the orienting response (OR) was indicated through heart
rate analysis. All participants in the experiment had their cardiac
activity monitored and the time between heartbeats recorded (called the
inter-beat interval, or IBI). There are two types of orienting
responses—monophasic and biphasic. A monophasic OR will produce a decrease
in heart rate for about 6 seconds, then returns to baseline levels after 10
seconds. A biphasic OR will show a dip in heart rate for 2-3 seconds after
the onset, an increase in heart rate until about 7 seconds, then a leveling
off to baseline levels (Graham, 1979; Velden & Schumacher, 1979; Lang,
1990; Lang, 1994).
Arousal was measured through skin conductance. Skin conductance measures
the activity of the autonomic nervous system and has been used as a valid
physiological measure of sympathetic nervous system activation, which
indicates arousal (Hopkins & Fletcher, 1994; Grabe, Lang, Zhou, & Bolls,
2000; Lang et al., 1999; Lang et al., 2000). Skin conductance levels were
collected from participants 50 times per second during the media
presentation.
Recognition was assessed using paper and pencil methods. Participants
watched the recognition tape and recorded whether or not they had seen the
frames of video in the stimulus material.
Participants
Forty-one participants from an undergraduate university communication
course volunteered for this study. Each participant either received extra
credit for participation in the study or fulfilled a course
requirement. Participants were randomly assigned to one of the four orders
of presentation.
Apparatus
The stimulus and recognition tapes were shown on a Sharp 26" television
and a Panasonic AG1980 VCR. Volume levels were standard. Physiology was
collected using a Compaq Prolinea Net 1/25S computer with a Labmaster A/D
D/A board. Heart rate was measured by placing two Beckman standard AG/AGCL
electrodes on the participant's forearms. A ground electrode was also
placed on the participant's non-dominant forearm. Heart rate was recorded
in milliseconds between beats using a Coulbourn bio-amplifier with bandpass
filter.
Skin conductance was measured by placing two Beckman standard AG/AGCL
electrodes on the participant's non-dominant hand. In order to control
hydration, the participant's hand was washed with distilled water prior to
electrode application. The signal was passed to a Coulbourn SC module,
which provides a constant measurement voltage of .5v.
Procedure
Each participant took part in the experiment individually. After each
participant signed an informed consent, the participant was taken into the
lab and seated approximately 6 feet from the television and VCR. The
participant removed any jewelry from his or her hands and wrists. The
researcher acquainted the participant with the physiology electrodes and
the necessary preparation for physiological recording, then cleaned the
participant's non-dominant hand and both forearms with distilled
water. Two electrodes were placed on the participant's non-dominant hand,
and three electrodes were placed on the non-dominant forearm.
The participants were told that they would be watching four different news
stories. After each story played, the tape would stop in order for them to
fill out an attitude scale that is not reported here. Each participant was
told to sit comfortably, but to stay as still as possible during the
stimulus portion of the experiment. Before the start of each news story,
at least 15 seconds of baseline physiology data were collected.
After the participant viewed all four stories, the participant took part
in an unrelated study of radio messages. This distraction task took
approximately 45 minutes. After the distraction task, the sensors were
removed from the participants.
The researcher then gave the participant a recognition packet. The
participants were instructed to view the recognition tape and answer on the
packet whether or not they had seen the snippet of video. The participants
were told to answer as quickly as possible since the time between snippets
was short. After the recognition test was completed, the participants took
part in the recall of the radio experiment. During the cued recall portion
of the radio experiment, the participants were asked if they knew the
nature of the television study. None of the participants knew the
manipulation of the independent variable. The participants were debriefed,
thanked, and dismissed.
Data Editing and Reduction
Four points of interest were analyzed in each story. The points were
those where a natural sound break occurred in the natural sound version of
the story. For the stories that did not have natural sound breaks, the
data were taken from the video point where the natural sound break would
have occurred. Order did not have a significant effect on any analysis,
and the results are not reported here.
Cardiac data were cleaned and edited using VPM Event (Cook, 1999). Because
the inter-beat interval of an average individual ranges between 500-1200
milliseconds (Lang, 1990), any inter-beat interval that was less than 15
milliseconds was automatically added to the previous interval using VPM
Event. All other data that fell outside the average range was edited
manually. For instance, if the data read two consecutive inter-beat
intervals of 350 milliseconds, the two intervals were added together. If
an interval read 1300 milliseconds, the interval was split. Heart rate
data were then converted from milliseconds to beats per minute per second.
One participant's data were removed from all physiology analyses due to
excessive noise in the data. During data collection, there were five cases
where the onset of a particular story was unclear due to researcher
error. For the orienting response analysis, these cases were replaced by
the means for that data point in the same point of interest in the same
story and order. A total of 40 participants were used in the orienting
response analysis.
For skin conductance, one participant's data were removed from the analysis
for a total of 39 participants. Of these 156 data segments, four were
replaced with the mean skin conductance levels for that story and
order. Because the amount of baseline data collection varied, the baseline
levels were standardized to be the last 10 seconds of data prior to the
story onset. Skin conductance levels were converted to change scores based
on the final second of baseline before each story (similar to the analysis
of Lang et al., 1999). The scores were averaged into five-second intervals
for analysis.
Recognition data were converted from simple percent-correct to
non-parametric measures (a') (Grier, 1971). Recognition data were analyzed
using signal detection by converting the data into percentages of hits
(items the participant has seen and participant reports to have seen),
misses (items the participant has seen and participant has not reported to
have seen), correct rejections (items the participant has not seen and
reports not having seen), and false alarms (items the participant has not
seen and participant reports to have seen). Sensitivity, or the accuracy
of memory, was then calculated with a formula using the percentage of hits
and false alarms (Grier, 1971; Shapiro, 1994)3. The greater the a', the
more accurate (or sensitive) the participant is to the recognition items
(Grabe, Lang, Zhou, & Bolls, 2000). Two participants did not keep up with
the recognition tape and therefore missed items, and one participant could
not complete the test because the recognition tape was not properly
cued. A total of 38 participants' recognition data were therefore used for
analysis.
Results
Hypothesis 1—Orienting to Natural Sound Breaks
Hypothesis 1 predicted that natural sound breaks would elicit an orienting
response in viewers. An OR was operationally defined as a significant
decrease in heart rate over time that also significantly followed a cubic
trend (Graham, 1979). The test of H1 began by constructing a cardiac
response curve (CRC) that graphed participants' heart rate level for each
of 10 seconds following the onset of the natural sound in both the first
and second natural sound stories. That CRC is shown in Figure 1.
[--- WMF Graphic Goes Here ---]
A 2 (Story) x 10 (Time) mixed design analysis of variance (ANOVA) showed a
main effect for Time with a cubic trend, F(1,36)=38.054, p<.001. However,
the Story x Time cubic interaction approached significance, F(1,36)=2.511,
p=.122.
Because the news stories also contained visual changes at the moment
natural sound was used, the possibility existed that the ORs in Figure 1
were not due to natural sound breaks, but to the change in picture. To
examine that possibility, two other CRCs were constructed showing the exact
time points where the natural sound break would have occurred in the
no-natural sound versions of the stories. All four CRCs are shown in Figure 2.
[--- WMF Graphic Goes Here ---]
As can be seen in the figure, the deceleration was greater for the natural
sound versions than the no natural sound versions. A repeated measures
2(Story) X 2(Natural Sound) X 10(Time) mixed design analysis of variance
(ANOVA) showed a significant Nat Sound X Time interaction in the cubic
trend analysis, F(1,36)=15.742, p<.001. Because cardiac activity
significantly decreased following natural sound breaks and because that
decrease was significantly greater than in response to visuals alone, H1 is
confirmed.
Hypotheses 2—Recognition and Natural Sound
Hypothesis 2 predicted that the recognition would be worse for information
during the natural sound break than the information before and after the
natural sound break. A 3 (Position) x 2 (Natural Sound) mixed design
analysis of variance (ANOVA) showed a main effect for Position,
F(2,68)=11.173, p<.001. Participants recognized less of the video items
that occurred during the natural sound break than those before or after the
break. In addition, the stories with natural sound breaks appeared to have
a higher overall recognition than the ones without breaks. However, the
Position x Natural Sound analysis was not significant, F(2,68)=.117,
p<.890. Therefore, the analysis showed that there was an interaction in
the recognition data, but natural sound breaks did not cause this
interaction. The results are shown in Figure 3. H2 is not supported.
[--- WMF Graphic Goes Here ---]
Hypotheses 3—Natural Sound Breaks and Arousal
In analyzing the skin conductance levels, a repeated measures 2 (Natural
Sound) x 2 (Story) x 10 (Time) mixed design analysis of variance (ANOVA)
was conducted. Although the main effect for natural sound breaks was not
significant, F(1,35)=1.561, p=.220, the Natural Sound x Time interaction
was significant, F(9,315)=3.945, p<.001. This interaction can be seen in
Figure 4. Because the skin conductance levels in the first natural sound
package increased within the first 10 seconds of the story, the levels
remained much higher and resulted in a less pronounced decrease in arousal
through the duration of the news story. Hypothesis 3 is supported.
[--- WMF Graphic Goes Here ---]
Discussion
Members of the National Press Photographers Association have been pleading
to their news directors and reporters about the importance of using natural
sound breaks in news stories (Murray, 1996). However, there has not been
much academic research investigating how natural sound breaks can affect
storytelling. This study intended to take the first step in investigating
the importance of natural sound by looking at how it affects viewer
attention and arousal.
The limited capacity model has been implemented to test the effects of
various structural features (Lang, 1990; Lang, 1991; Lang et al., 2000;
Potter, 2000; Potter & Callison, 2000) and arousing content (Lang et al.,
1999; Lang et al., 1996; Grabe, Lang, et al., 2000; Grabe, Zhou, et al.,
2000) on the processing of mass-mediated messages. This study tested this
model using natural sound breaks in news stories and found that natural
sound is another feature that can affect how viewers process television
news.
The results of the orienting response data show that natural sound breaks
do produce an orienting response in viewers. However, participants did
orient to the video that appeared at the natural sound break, whether a
natural sound break was present or not. What is important to note is that
the OR was significantly more pronounced when the natural sound break was
added. Participants also seemed to have a greater, though not significant,
orienting response to natural sound breaks during the first natural sound
story than the second natural sound story. It is possible that
participants may have habituated to the natural sound breaks by the time
the second natural sound story was shown. This suggests that during a
typical newscast with various types of stories, stories with natural sound
breaks airing later in a broadcast may not have as much of an effect on
cognitive processing as those stories airing earlier in the broadcast.
It is possible that if more participants were included in this experiment,
the difference in the recognition data could have reached significance. In
addition, further research should add more than 50 recognition items, since
those recognition items had to be assigned to a particular story and
limited the power of the analysis. Some stories had more available
recognition items and therefore had a more robust recognition analysis. It
would be better to balance the recognition items equally, so each story
would have the same number of recognition items. Further research could
also analyze memory through the use of free or cued recall to see if
natural sound breaks caused participants to remember the stories more.
Further research should also test audio recognition. This study measured
natural sound's effect on video recognition, but it would be interesting to
see if participants' audio recognition would follow the same
pattern. Whereas we've seen effects of arousing video content on verbal
memory (Lang et al., 1999), further research should see if arousing audio
could have the same effect.
Because the data collection of the physiology was not precise, the onset
of the orienting response may not have been precise. Data collection may
have started after the onset of a particular story, which would cause the
data not to be synchronized with the stimulus. Time code in the stimulus
and data collection would correct this problem.
Because the research had a limited number of available packages that used
natural sound in its storytelling, the stories varied in length. The
shortest package used was 54 seconds in length, so the tonic heart rate
analysis could only analyze the first 54 seconds of all the stories, even
though some stories ran well over two minutes. In the future, stories that
are similar in length should be used, and more stories should be used.
Further research should also look into varying the number of natural sound
breaks. Natural sound breaks are encouraged by the NPPA, but too many are
discouraged (Murray, 2001). Therefore, stimulus material could have a
high, medium, and low number of natural sound breaks to see the effect on
arousal, tonic attention, and recognition (similar to Lang et al.,
2000). In addition, further research could use other cognitive models to
investigate if natural sound "proves" a reporter's words and enhances
comprehension of the news story (Murray, 1998).
The results of this study could be of interest to the television news
industry. Regardless of story content, participants were somewhat affected
by the use of natural sound breaks in news stories. General managers and
news directors who are interested in setting their station apart from
others may benefit from using natural sound breaks. As the competition for
audiences continues to increase, news organizations would benefit from
learning what viewers find to be more engaging without sacrificing the
audience's capacity to obtain information.
Notes:
1. A novel stimulus is not the only condition in which an orienting
response can occur (Kahneman, 1973). It can also occur when a stimulus
matches something that has been previously primed as significant to the
person, such as a person's name (Ohman, 1979).
2. Self-reported arousal and physiological arousal for slow and medium
edits were lower than fast and very fast-paced edits. However, the
relationship is not linear between medium and fast-paced messages. The
levels of arousal appear to "step up" during the fast-paced messages.
3. The researcher used the following formula for calculating a' (Grier, 1971):
A'=1/2 + (hits-false alarms)(1 + hits –false alarms)/4(hits)(1-false alarms)
Another formula for a' was also calculated in a post hoc analysis, but the
results were similar. The following formula was used in the post-hoc analysis:
A'=1 – ¼((false alarms/hits + (1-hits)/(1-false alarms))
References
Bradley, M. M., Greenwald, M. K., Petry, M. C., & Lang, P. J.
(1992). Remembering pictures: Pleasure and arousal in memory. Journal of
Experimental Psychology, 18, 379-390.
Broadbent, D. E. (1958). Perception and Communication. Oxford: Pergamon.
Cook, E. W. I. (1999). VPMEVENT (Version 6.4). Birmingham, AL.
Geiger, S., & Reeves, B. (1993). The effects of scene changes and semantic
relatedness on attention to television. Communication Research, 20(2),
155-175.
Grabe, M. E., Lang, A., Zhou, S., & Bolls, P. D. (2000). Cognitive access
to negatively arousing news. Communication Research, 27(1), 3-27.
Grabe, M. E., Zhou, S., Lang, A., & Bolls, P. (2000). Packaging television
news: The effects of tabloid on information processing and evaluative
responses. Journal of Broadcasting & Electronic Media, 44, 581-598.
Grabe, M. E., Zhou, S., & Barnett, B. (2001). Explicating sensationalism
in television news: Content and the bells and whistles of form. Journal
of Broadcasting & Electronic Media, 45, 635-655.
Graham, F. K. (1979). Distinguishing among orienting, defense, and startle
reflexes. In H. D. Kimmel, E. H. Van Olst, & J. F. Orlebeke (Eds.), The
orienting reflex in humans (pp137-167). Hillsdale, NJ: Lawrence Erlbaum.
Grier, J. B. (1971). Nonparametric indexes for sensitivity and
bias: Computing formulas. Psychological Bulletin, 75(6), 424-429.
Hesketh, B., & Yorke, I. (1993). An introduction to ENG. Oxford,
England: Focal.
Hopkins, R., & Fletcher, J. E. (1994). Electrodermal
measurement: Particularly effective for forecasting message influence on
sales appeal. In A. Lang (Ed.) Measuring Psychological Responses to Media
(pp. 113-132). Hillsdale, NJ: Erlbaum.
Kahneman, D. (1973). Attention and effort. Englewood Cliffs,
NJ: Prentice-Hall.
Lachman, R., Lachman, J. L., & Butterfield, E. C. (1979). Cognitive
psychology and information processing: An introduction. Hillsdale,
NJ: Lawrence Erlbaum.
Lang, A. (1990). Involuntary attention and physiological arousal evoked by
structural features and emotional content in tv commercials. Communication
Research, 17(3), 275-300.
Lang, A. (1991). Emotion, formal features, and memory for televised
political advertisements. In F. Biocca (Ed.) Television and political
advertising: Vol. 1. Psychological processes (pp. 221-243). Hillsdale,
NJ: Erlbaum.
Lang, A. (1994). What can the heart tell us about thinking? In A. Lang
(Ed.) Measuring Psychological Responses to Media (pp. 99-111). Hillsdale,
NJ: Erlbaum.
Lang, A. (2000). The limited capacity model of mediated message
processing. Journal of Communication, 50(1), 46-71.
Lang, A., Bolls, P. D., Potter, R. F., & Kawahara, K. (1999). The effects
of production pace and arousing content on the information processing of
television messages. Journal of Broadcasting and Electronic Media, 43,
451-475.
Lang, A., Geiger, S., Strickwerda, M., & Sumner, J. (1993). The effects of
related and unrelated cuts on television viewers' attention, processing
capacity, and memory. Communication Research, 20(1), 4-30.
Lang, A., Newhagen, J., & Reeves, B. (1996). Negative video as
structure: Emotion, attention, capacity, and memory. Journal of
Broadcasting and Electronic Media, 40, 460-477.
Lang, A., Potter, R. F., & Bolls, P. D. (1999). Something for nothing: Is
visual encoding automatic? Media Psychology, 1, 145-163.
Lang, A., Zhou, S., Schwartz, N., Bolls, P. D., & Potter, R. F.
(2000). The effects of edits on arousal, attention, and memory for
television messages: When an edit is an edit can an edit be too
much? Journal of Broadcasting and Electronic Media, 44(1), 94-109.
Livingston, S. (1997). Scott Livingston: Livingston offers top 10
storytelling tips. News Photographer, 52(7), 66-67.
Murray, J. (1996). How to please the judges: Advice for the NPPA
television news photography contest. News Photographer, 51(12), 30-31.
Murray, J. (1998). 'Textbook examples' from a photographer's
reporter. News Photographer, 53(12), 34-38.
Murray, J. (2001). Knox, KARE, Menell tops in annual TV judging. News
Photographer, 56(5), 25-35.
Ohman, A. (1979). The orientation response, attention, and learning: An
information processing perspective. In H. D. Kimmel, E. H. Van Olst, & J.
F. Orlebeke (Eds.), The orienting reflex in humans (pp.
443-472). Hillsdale, NJ: Lawrence Erlbaum.
Potter, R. F. (2000). The effects of voice changes on orienting and
immediate cognitive overload in radio listeners. Media Psychology, 2, 147-177.
Potter, R. F., & Callison, C. (2000). Sounds exciting!: The effects of
auditory complexity on listeners' attitudes and memory for radio
promotional announcements. Journal of Radio Studies, 7, 29-51.
Potter, R. F., Choi, J., Yu, H., Kim, J., Carpentier, F. (2002). Increased
memory for structurally complex radio messages: Could arousal be the
mechanism?. Psychophysiology, 39(S1), 67 (Abstract).
Reeves, B., Thorson, E., & Schleuder, J. (1986). Attention to
television: Psychological theories and chronometric measures. In J.
Bryant & D. Zillmann (Eds.), Perspectives on Media Effects (pp.
251-279). Hillsdale, NJ: Erlbaum.
Shapiro, M. A. (1994). Signal detection measures of recognition
memory. In A. Lang (Ed.) Measuring Psychological Responses to Media (pp.
133-148). Hillsdale, NJ: Erlbaum.
Shapiro, M. A., & Lang, A. (1991). Making television
reality. Communication Research, 18, 685-706.
Shiffrin, R. M., & Schneider, W. (1977). Controlled and automatic human
information processing: Perceptual learning, automatic attending, and a
general theory. Psychological Review, 84, 127-190.
Shook, F. (1989). Television field production and reporting. White
Plains, NY: Longman.
Sokolov, Y. N. (1963). Perception and the conditioned reflex. New
York: Pergamon.
Speake, K. (1996). Planning, flexibility make Michaelsen's work stand
out. News Photographer, 51(7), 62-63.
Thorson, E., & Lang, A. (1992). The effects of television video graphics
and lecture familiarity on adult cardiac orienting responses and
memory. Communication Research, 19(3), 346-369.
Velden, M., & Schumacher, R. (1979). Orienting and defensive cardiac
responses. In H. D. Kimmel, E. H. Van Olst, & J. F. Orlebeke (Eds.), The
orienting reflex in humans (pp 199-218). Hillsdale, NJ: Lawrence Erlbaum.
|
