This paper was presented at the Association for Education in Journalism and
Mass Communication in San Antonio, Texas August 2005.
If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
rakyat [ at ] eparker.org. For an explanation of the subject line,
send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").
(Feb 2006)
Thank you.
Elliott Parker
====================================================================
Manipulating Visual Information in the Digital Age
How Viewers React on Digitally Altered and Manipulative Captioned
Photographic Images
Results of a Quasi-Experimental Study
Kathrin Ziegler Klaus Forster
University of Munich
Contact:
Klaus Forster
Schleissheimer Str. 91
D-80797 Munich, Germany
Tel.: +49 89 123 58 36
Email: [log in to unmask]
[log in to unmask]
Paper Presented to the Visual Communication Division of the AEJMC for
the Annual Convention of the
Association for Education in Journalism and Mass Communication August
10 – 13, 2005 in San Antonio, Texas
Abstract
In our study we explore how digitally altered and manipulative
captioned pictures are perceived by
the viewers. We regard image perception as a first necessary step
toward any kind of possible effects.
Our findings suggest that viewers were not able to recognize
differences between differently manipulated
pictures with regard to their credibility on a cognitive level. On
the emotional level the viewers could
therefore be effectively manipulated with both captions and altered photos.
[page 1]
Introduction
Ever since the invention of photography, viewers had – and still have
– every right to question whether
a photographic image is authentic or not (Brugioni 1999; Carlebach
1992; Kobré 1995; Jaubert 1989).
Numerous examples of staged , composed, altered or false captioned
photos (Haus der Geschichte 2003,
Wheeler 2002) prove that common phrases like 'seeing is believing' or
'a picture says more than thousand
words' (Berger 1998; Brugioni 1999) are highly misplaced, especially
in the coverage of political events
and war reporting (Hörburger 2004; Jaubert 1989).
Moreover, since the advent of digital photo alteration the problem of
image manipulation seems to have
an enormously increasing relevance to both the audience and the
community of professional photojournalists
(Lester 2000, p. 222; Lowry 2001; Ritchin 1999). One most recent
example for the irritation a presumably
altered and simultaneously false captioned photo can stir is the case
of Lcpl. Boudreaux. In April 2004
a photo was distributed via the internet. It depicts Lance Corporal
Ted J. Boudreaux, a U.S. Marine with
two Iraqi boys, one of them holding a sign, saying „Lcpl. Boudreaux
killed my dad, th(en) he knocked up
my sister!" (see Figure 1). Boudreaux is looking bright, the boys are
smiling shyly.
Figure 1: First Version of the Boudreaux-Photo
Source: http://www.cair-net.org/images/lcpl11.jpg, 21.07.04
[page 2]
Nihad Awad, executive director of the Council on American-Islamic
Relations (CAIR), interpreted the image
as ridicule of the two boys and with them the Iraqi people. Boudreaux
said he was indeed the man in the
picture but insisted that the text on the sign had been manipulated.
As so many times before (e.g. Thomey
1996; Whelan 1999/2001; Willenbrock 2001), a war photo could have
been altered for propaganda purposes.
Shortly before the case was closed without conclusion, another
version of the picture appeared. This time
the sign said „Lcpl. Boudreaux saved my dad, th(en) he rescued my
sister!". The case was reopened but
there is still no definite conclusion whether the first caption or
the latter version is authentic. The
case was widely discussed on the Internet without a firm conclusion
either (Bonopartis 2004).
It is quite obvious that for the audience it is almost impossible to
decide what version of such a photo
is a 'true' depiction of reality and which is not, if the pictures
and its captions are the only relevant
clues for such a decision. But it is also a fact that at least one of
the two images must have been
manipulated. So the assessment whether there is "Phototruth or
Photofiction" (Wheeler 2002) depends
mostly on one's own preconceptions. In this study we deal with much
less controversial examples of
photo alterations but we nevertheless want to find out, how viewers
react to different forms of
manipulation with a special focus on manipulative captions.
1. Theoretical Foundation
While we learned from recent studies that viewers don't like to be
tricked, especially in the field of
hard news photography (Greer/Gosen 2002; Huang 2001), it is quite
unclear how manipulated photos are
perceived if there is no explicit hint at the alteration (Forster
2003). The fact that 'truth lies in
the eye of the beholder'– to cite again a common saying – is hardly
new to communication studies.
In Symbolic Interactionism this is acknowledged since the 1950s.
Herbert Blumer wrote in 1959:
"By now students of mass communication realize that effects cannot be
safely gauged from the 'manifest'
content of what is presented through a medium. It is necessary to
consider how people in the audience
are sensitized to the presentation and prepared to interpret it. Such
sensitivity and responsiveness
differ not only between people in the given audience but more
importantly in given people through time.
People are caught up in a world of moving, which foster new objects
of preoccupation, new lines of
judgment, and new orientation of feeling. As issues arise or subside,
as new interests emerge or
recede, as sophistication replaces naïveté, or in many other ways,
people shift in their sensitivity
to presentations and in their interpretation of them" (Blumer 1998, p. 185).
This quote is of importance for the study at hand in many ways. First
of all it says that images, as
any other media messages, are interpreted by the recipient and that
these interpretations can vary over
time and circumstances. Five years ago and without the Abu Ghraib
prison abuses the picture discussed
above would have been most likely
[page 3]
discarded as implausible from most of the audience. This had obviously
changed. The Symbolic Interactionist point of view also implies that
a manipulation is then, and only
then, successful – from the perspective of the manipulator – if the
recipient interprets the manipulated
image as intended. It shows us that it is at least possible for the
viewer to develop a certain sensitivity
for possible photo alterations which can be part of the so called
"visual literacy" (Messaris 1994, pp. 135;
Newton 2001, p. 183).
Similar to the Symbolic Interactionist position – and in part related
to it – is an approach originally
developed from Früh/Schönbach (1982), who named it
'dynamic-transactional'. Derived from this approach
is a certain kind of model of communication effects (see Figure 2),
which proved to be useful to resolve
theoretical and operational problems originating from our research interest.
Figure 2: Multi-Causal- /Multi-Effects Model
Source: Früh 2001, p. 28
If an image is altered to have a certain effect, in the
Multi-Causal-/Multi-Effects Model the internal
interpretation process is manipulated by the alteration to get the
intended result. According to the
dynamic-transactional approach, the interpretation process can be
viewed as a kind of internal dialogue
in which the cognitive and affective potentials for future effects
are shaped. The manipulation may then
influence the knowledge, opinions, attitudes etc. and finally the
behavior of a person. Subsequently,
the manipulation of this 'dialogue within' can even have an impact on
the public dialogue in a society.
Therefore media manipulations – and in our case photo alterations or manipula-
[page 4]
tive captions – are not only an ethical problem inside
photojournalism. They have the potential to
damage the public dialogue as a whole.
Unfortunately we know little of the reactions to, and the effects of
manipulated images. In the proposed
model the first step for having an effect at all, is the viewer's
interpretation of the picture. Therefore
we tried to find out how different forms of photo manipulations are
perceived, because if on this stage
the manipulation does not work, it will most likely not have the
intended effects or any effect at all.
Effects on perception can occur on a cognitive and an affective
level. In this study we related the
perceived credibility and the attracted attention of a picture to the
cognitive perception and the
perceived aesthetic quality of an image to the affective or emotional
perception.
2. Research Questions
The general research interest of this study is concerned with the
cognitive and affective perception of
photographic material. In a first step, three photographs were
digitally altered to get negative,
neutral, and positive versions of the pictures. In the second step
the neutral versions were provided
with a negative, neutral, and positive caption.
The general research interest can be divided into three more specific
research questions:
RQ 1: Do recipients perceive differences between negative, neutral,
or positive versions of the
presented pictures?
RQ 2: Do recipients perceive differences between pictures with
negative, neutral, or positive captions?
RQ 3: Do recipients perceive differences between the negative,
neutral, or positive versions of the
presented pictures and pictures with negative, neutral, or positive captions?
Because of the lack of empirically backed assumptions regarding our
topic, we didn't draw up specific
hypotheses. Nevertheless we expected significant differences between
the different versions of the photos
serving as treatment material, which can be statistically tested.
Despite our quantitative approach to
answer the research questions, this study can also be seen as an
exploratory project which seeks to help
generating hypotheses for future research (Bortz/Döring 1995, pp. 345).
[page 5]
3. Method
To answer the research questions we employed a quasi-experimental
design to check the differences between
various treatment conditions. Each subjects participant was asked to
rate three pictures with 16 bipolar
six-place adjective scales (as in a semantic differential), such as
'repulsive-appealing'. We tested
whether there is an impact of the independent variable 'mode of
manipulation' on the dependent variables,
which can be grouped in the dimensions 'perceived authenticity',
'degree of attention', 'aesthetic value',
and 'perceived tranquility'. Affiliation of the variables to the
dimensions has been calculated by a factor
analysis (main components; Varimax-rotation). Each dimension is
represented by an index variable. We
decided to use a design with repeated measures, meaning that every
subject was measured on several
treatment conditions (within subjects design). One advantage of this
design is a higher sensitivity so
that the existing treatment effects can be detected more easily
(Czienskowski 1996, p. 75). To avoid
carry-over effects, the subjects were allocated at random to a
variety of treatment sequences. We
therefore worked with a 'counterbalanced design' or 'rotation
experiment' (Campbell/Stanley 1966, pp.
50-51; Laatz 1993, p. 488).
Table 1: Treatment Sequences
Set-Nr. Treatment Sequences Set-Nr. Treatment Sequences
1 F1 G2 M3 7 F4 G5 M6
2 F2 G3 M1 8 F5 G6 M4
3 F3 G1 M2 9 F6 G4 M5
4 M2 F1 G3 10 M5 F4 G6
5 M3 F2 G1 11 M6 F5 G4
6 M1 F3 G2 12 M4 F6 G5
Three photographs were chosen to provide answers to the research
questions. These pictures were digitally
altered to get a negative, a neutral, and a positive version of each
photo. In addition, the neutral
version of each image was provided with a negative, a neutral, and a
positive caption. In effect we
have six versions of each picture.
Figure 3: Photograph 1 - Sleeping Girl Digitally Altered
Fig. 3.1: Neutral Original Fig. 3.2: Positive Alteration Fig. 3.3:
Negative Alteration
[page 6]
The sleeping girl was framed with fresh bananas and melons to get a
more positive image. Her cheeks got a
rosy complexion. Shining colors and fresh fruit create a warm and
friendly atmosphere (see Figure 3.2).
The negative version of the photo contains dark spots in the face of
the girl to simulate bruises and a
bandage on the right foot with bloodstains (see Figure 3.3).
Figure 4: Photograph 1 - Sleeping Girl with Different Captions
Fig. 4.1: Neutral Caption:
A short nap after a well done job. Kim Sue and her siblings are
helping their father in his small shop
every day.
Fig. 4.2: Positive Caption:
Everyone is sleeping at noon at the Russian Market in Phnom Penh.
Fig. 4.3: Negative Caption:
Domestic violence is a wide spread problem in Cambodia. The full
range of this problem is still unknown.
Because of fear and shame most women remain silent about the violence
they and their daughters have to
endure at home.
For the treatments shown in Figure 4 we used the neutral version of
the picture and added three different
captions. A difference in the perception is therefore solely caused
by the text accompanying the picture.
We used captions with a neutral, a positive, and a negative
connotation. The negative caption hints on a
well known problem in Cambodia (Ray 2000, p. 75).
Figure 5: Photograph 2 – Cambodian Woman Digitally Altered
Fig.5.1: Neutral Original Fig.5.2: Positive Alteration Fig.5.3:
Negative Alteration
[page 7]
The positive appearance of the Cambodian woman in Figure 5 was
enhanced by employing a variety of digital
cosmetics. The scar on the nose was removed and the spot softened, as
well as the wrinkles on the forehead,
around the eyes and the mouth. The lips got a soft reddish tone, the
eyes were brightened. In the end the
whole picture was lightly softened (see Figure 5.2). In the negative
alteration scars and wrinkles were
sharpened, the eyes were narrowed and the skin was dotted with dark
spots (see Figure 5.3).
The neutral version of the picture with the Cambodian woman was
captioned with three different texts
again. The neutral caption is simply a description of the photo (see
Figure 6.1). The positive caption
gives the woman a name and an honorable occupation in caring for the
wildlife, especially in a country
where tigers and leopards could soon become extinct (see Figure 6.2;
Ray 2000, p. 70). The negative
caption is directly related to the genocide under the Khmer Rouge
regime and should still be remembered
enough to give the picture a dark context (see Figure 6.3).
Figure 6: Photograph 2 – Cambodian Woman with Different Captions
Fig. 6.1: Neutral Caption
Market woman with a traditional Krama-Schal.
Fig. 6.2: Positive Caption
The good soul of the Kirirom National Park in the Elephant Mountains:
Chao Saya is engaged in the breeding
of young tigers and leopards.
Fig. 6.3: Negative Caption
The dreadful past of their country is written in the faces of many
Cambodians. This farmer's wife barely
escaped the Pol Pot regime. Her family died on the Killing Fields
This time the original picture served as the positive version. The
picture was digitally altered to get a
neutral and a negative version. These versions are "composite
image(s)" known as "photomontage(s)"
(Brugioni 1999, p. 20). In the alteration process the heads of two
sad looking children in another photo
(see Figure 7.4) were copied into the original version to replace the
two happy boys (see Figure 7.1).
[page 8]
Figure 7: Photograph 3 – Man with Boys Digitally Altered
Fig.7.1: Neutral Alteration Fig.7.2: Positive Original Fig.7.3:
Negative Alteration
Fig.7.4: Source of the boys in the Neutral Alteration
In the negative version the face of the man was altered to get a more
depressing scenario. So his smile
isn't quite bright anymore and his teeth are with a yellow tone now
(see Figure 7.3). The neutral
alteration was again captioned in three different ways (see Figure 8).
Figure 8: Photograph 3 – Man with Boys and Different Captions
Fig. 8.1: Neutral Caption
A Farmer living with his wife and children in the region of
Rattanakiri close to the jungle.
Fig. 8.2: Positive Caption
In spite of the primitive living conditions in the rural regions of
Cambodia the people are showing a
peaceful contentedness not often found in wealthy Europe.
Fig. 8.3: Negative Caption
Orphanage in the Rattanakiri province at the edge of the jungle.
Despite of the confidence of their
attendant the children have to live under the most miserable conditions.
The negative caption is again a (fictional) description of a farmer
and his family (see Figure 8.1).
The positive caption emphasizes the peaceful and friendly appearance
of the three people (see Figure 8.2)
and the negative caption makes the father an attendant in an
orphanage and depicts the two boys as orphans
living under most humble conditions. It can be noted that virtually
all these manipulative alterations
were following some kind of stereotypes, because we know since
Lippmann (1997, pp. 53) that these
[page 9]
manipulations will be the most effective ones, if the pictured
stereotypes are corresponding with our
own expectations.
The experiment was conducted in the summer of 2004. Altogether 158
subjects individuals (79 female, 79 male)
were subjected to the experimental treatments. Their age varied from
17 to 81. 52 participants had no high
school degree, 54 participants had an high school and 52 an academic
degree. Participants were randomly
allocated to the different treatment sets. All in all 474 ratings of
the different pictures and the
different captions were obtained.
4. Results
4.1 Influence of the Photo Alterations on the Viewers' Perception
Our first research question deals with the viewer's perception of the
digitally manipulated pictures:
RQ 1: Do recipients perceive differences between negative, neutral,
or positive versions of
the presented pictures?
The idea here was that a negative image should be perceived
differently than a neutral or a positive
image. The tricky part for the manipulator is to maintain a high
degree of the 'perceived authenticity'
of a picture on a cognitive level while manipulating the perceptions
on an affective, respectively
emotional level. In the case with the sleeping girl this seems to
work quite well. While participants
rated the authenticity of all versions of the picture as relatively
high, other perceptions show
significant differences according to the direction of the photo
alteration (see Table 2).
Table 2: Ratings of the Pictures with the Sleeping Girl
Version Negative (N = 27) Neutral (N = 25) Positive (N = 29)
Dimensions Mean s* Mean s* Mean s*
Perceived Authenticity 4,91 0,83 5,32 0,61 4,72 1,19
Degree of Attention** 3,78 0,89 4,42 0,72 3,78 0,99
Aesthetic Value** 3,09 1,14 3,39 0,98 4,30 0,94
Perceived Tranquility** 4,17 0,98 4,22 1,05 4,86 0,93
Dimensional Reduction: Varimax-rotated Factor Analysis.
* s = Standard Deviation
** Significant differences of means (p = .05) (ANOVA).
Participants rated the 'aesthetic value' of the negative version
lower than that of the neutral version.
The positive version had the highest 'aesthetic value' in the
participants' eyes. 'Perceived tranquility'
is again highest with the positive altered picture,
[page 10]
while negative alteration and neutral version differ little.
Interestingly though, the 'degree of
attention' is highest in the neutral version of the photo (see Table 2).
The ratings of the 'perceived authenticity' of the photos with the
Cambodian woman are again on a
satisfyingly high level, albeit the 'degree of attention' that all
three pictures were able to generate,
is not very distinct here. The manipulation worked in part for the
affective perception of the photos,
but only in the negative direction: 'Aesthetic value' and 'perceived
tranquility' of the negative
version of the picture are rated significantly lower than the neutral
and the positive versions.
There are little to no differences between the latter (see Table 3).
Table 3: Ratings of the Pictures with the Cambodian Woman
Version Negative (N = 23) Neutral (N = 26) Positive (N = 28)
Dimensions Mean s* Mean s* Mean s*
Perceived Authenticity 4,77 0,73 4,95 0,69 4,80 1,07
Degree of Attention 3,13 0,93 3,46 1,01 3,42 1,05
Aesthetic Value** 3,01 0,63 4,27 0,77 4,24 0,78
Perceived Tranquility** 4,26 0,74 4,90 0,68 5,04 0,68
Dimensional Reduction: Varimax-rotated Factor Analysis.
* s = Standard Deviation
** Significant differences of means (p = .05) (ANOVA).
The photos of the group of one man and two boys were rated as highly
authentic but, again, didn't
attract very much attention. This time the manipulation worked only
in the positive direction in
case of the assigned 'aesthetic value': The positive version of the
picture is rated significantly
higher than negative or neutral versions of the photo. No significant
differences could be observed
for the 'perceived tranquility', which is quite high in all three
images (see Table 4).
Table 4: Ratings of the Pictures with the Man and Boys
Version Negative (N = 25) Neutral (N = 22) Positive (N = 30)
Dimensions Mean s* Mean s* Mean s*
Perceived Authenticity 4,62 0,98 4,36 1,02 4,85 0,95
Degree of Attention 3,63 1,00 3,33 1,17 3,51 0,99
Aesthetic Value** 4,51 0,93 4,41 0,80 5,11 0,70
Perceived Tranquility 4,48 1,10 4,36 0,69 4,58 0,72
Dimensional Reduction: Varimax-rotated Factor Analysis.
* s = Standard Deviation
** Significant differences of means (p = .05) (ANOVA).
[page 11]
Interim results
It seems to be clear that participants were not able to spot many
differences between the pictures with
regard to their 'perceived authenticity', at least not on a
statistically significant level. Therefore
there is no evidence in the data at hand that viewers are able to
detect photo alterations themselves.
Nevertheless, the tested photo alterations were not very successful
in attracting attention.
However, on the affective level of perception the alterations worked
fairly well in the expected negative
and positive directions, albeit in two cases only in one direction:
The negative version of the picture
with the Cambodian woman reduced the mean 'aesthetic value'
significantly and the positive version of
the group of man and boys was perceived as significantly more
aesthetic than the other two photos.
4.2 Influence of the Captions on the Viewers' Perception
The second research question deals with the problem of the
manipulative captioning of pictures.
RQ 2: Do recipients perceive differences between pictures with
negative, neutral, or positive captions?
This is a particularly interesting question, because in contrast to
altered photographs, manipulative
captions cannot be detected without additional information. While it
is at least possible – though not
very likely – that viewers are able to detect photo manipulations
just from viewing the manipulated photo,
manipulative captioning can – in most cases – be revealed only with
the help of other information than
the photo itself.
Table 5: Ratings of the Picture with the Sleeping Girl according to
different captions
Version Negative (N = 24) Neutral (N = 29) Positive (N = 24)
Dimensions Mean s* Mean s* Mean s*
Perceived Authenticity 4,56 1,04 4,78 0,93 4,86 0,89
Degree of Attention 3,87 1,19 3,82 1,00 3,93 1,04
Aesthetic Value** 3,34 1,05 4,15 0,98 4,13 0,85
Perceived Tranquility 4,46 1,03 4,83 0,81 4,81 0,72
Dimensional Reduction: Varimax-rotated Factor Analysis.
* s = Standard Deviation
** Significant differences of means (p = .05) (ANOVA).
The results for the picture with the sleeping girl show no
significant differences between the negative,
neutral and positive captions in the dimensions 'perceived
authenticity', 'degree of attention', and
'perceived tranquility'. Only the 'aesthetic value' is reduced
statistically significant through the
negative caption (see Table 5).
[page 12]
Table 6: Ratings of the Picture with the Cambodian Woman according to
different captions
Version Negative (N = 24) Neutral (N = 28) Positive (N = 29)
Dimensions Mean s* Mean s* Mean s*
Perceived Authenticity 5,15 0,67 4,71 0,86 4,98 0,74
Degree of Attention** 3,79 0,92 3,17 1,01 3,88 0,87
Aesthetic Value 3,90 0,80 4,05 0,54 4,37 0,76
Perceived Tranquility 4,67 0,97 4,86 0,86 4,75 0,87
Dimensional Reduction: Varimax-rotated Factor Analysis.
* s = Standard Deviation
** Significant differences of means (p = .05) (ANOVA).
The negative and the positive caption of the photo of the Cambodian
woman attracted significantly more
attention to the picture than the neutral caption. Tendency
reflecting – but not statistically
significant – are the differences between the ratings of the
'aesthetic value' in showing the
presumed effects: The photo with the negative caption is rated
lowest, the one with the positive
caption highest in the dimension of the perceived aesthetic quality.
Interestingly, the 'perceived
authenticity' is highest with the negative caption and lowest with
the neutral caption, but again
these differences are not statistically significant. The 'perceived
tranquility' is fairly equal with
all three versions (see Table 6).
The best results for manipulative captions were realized with the
photo of a man and two boys. Tendency
reflecting the 'perceived authenticity' increased from negative to
positive captions. The 'degree of
attention' was highest again with the negative caption – while
differences in the first two dimensions
are not statistically significant. In the aesthetical dimension the
manipulation of viewers through
captions worked nearly perfect: The 'aesthetic value' of the image
with the negative caption is
statistically significant lower than the 'aesthetic value' of the
pictures with neutral or negative
captions. This applies to the 'perceived tranquility' as well, albeit
here is no difference between
neutral and positive captioned pictures (see Table 7).
Table 7: Ratings of the Picture with the Man and Boys according to
different captions
Version Negative (N = 25) Neutral (N = 28) Positive (N = 28)
Dimensions Mean s* Mean s* Mean s*
Perceived Authenticity 4,56 1,11 4,69 1,23 5,10 0,85
Degree of Attention 4,16 0,89 3,63 0,83 3,89 1,02
Aesthetic Value** 3,47 0,90 4,45 0,67 4,91 0,56
Perceived Tranquility** 3,82 0,88 4,73 0,67 4,72 0,70
Dimensional Reduction: Varimax-rotated Factor Analysis.
* s = Standard Deviation
** Significant differences of means (p = .05) (ANOVA).
[page 13]
Interim results
The 'perceived authenticity' of all stimuli remains on a high level
regardless of the used captions.
It seems as if negative as well as positive captions have some
potential regarding the attraction of
attention. But – as in the case of photo alterations – captions work
best at the emotional level of
perception. Therefore participants rated the 'aesthetic value' of the
same picture clearly according to
the negative and positive captions, at least in the case of the third
picture – the image of the man with
boys.
These findings correspond with previous research in cognitive
psychology: There is substantial evidence for
an influence of additional verbal respectively semantic information
on the perceptual processing of
pictures (e.g. Wiseman/MacLeod/Lootsteen 1985;
Gauthier/James/Curby/Tarr 2003).(Footnote 1)
Footnote 1:
Although it may be useful for future research to take the differences
between verbal and visual
information processing (e.g. Anderson 1996, pp. 103) into account, we
omitted this research tradition
deliberatively in the present study. The reason for this is of
pragmatic nature mostly: The exploratory
character of our research does not require a detailed catalogue of
hypotheses, derived – for example –
from cognitive theory and research. Besides, the discussion of
general models of information processing
(e.g. 'dual coding') would not lead to verifiable hypotheses in our
field anyway and the results of the
studies mentioned above are too specific to guide re-search on image
manipulations at the present stage.
Nevertheless, the manipulative potential of captions may be limited
in reality, because viewers could tend
to ignore the captioning far more often than they read it. In an
experimental setting, where participants
were required to read the captions carefully, the manipulative
potential can be explored, but its impact
on perceptual processes in a 'real world' situation must be assessed
very cautiously.
However, the findings raise an interesting question: Why should
anyone invest time and/or money to alter
photos, when she or he can get the same manipulative effects with a
well crafted caption? The results
for the third research question should provide some answers.
[page 14]
4.3 Influence of Alterations and Captions in Comparison
The third research question deals with the assumption that it could
be possible to achieve the same or
similar effects with well constructed captions instead of time
consuming photo alterations.
RQ 3: Do recipients perceive differences between the negative,
neutral, or positive versions
of the presented pictures and pictures with negative, neutral, or
positive captions?
First of all, the differences between the negative, neutral, and
positive versions of the photos and
the captions are quite small – and on a relatively high level –
regarding the assessed authenticity.
The pictures and the captions used as treatments in the
quasi-experimental design were obviously not
identified as photo manipulations or manipulative captions by the
participants. Therefore the first
step to manipulate the viewer had been made: Only if manipulative
material is considered as authentic
on the cognitive level of perception, manipulative effects are likely
to occur and to be measured.
In this respect the design of this study was effective (see Figure 9).
Figure 9: Perceived Authenticity of Different Pictures and Captions
The next important step for a successful manipulator would be to call
attention to the manipulation.
This step can be seen as located on the cognitive level of perception as
[page 15]
well and it seems to be a hurdle, more difficult to be taken than the
'creation' of authenticity.
Whereas generally rated as highly authentic, the 'degree of
attention' directed to the treatment material
is at least in part significantly lower (see Figure 10).
Figure 10: Degree of Attention of Different Pictures and Captions
In particular the images of the Cambodian woman and the group of the
man with the boys seem to generate
interesting results regarding positive and negative captioning: Both
types of captions generate more
attention than neutral captions or different versions of the pictures
(see Figure 10). In the field of
attracting attention on the cognitive level of perception,
spectacular captions – regardless of being
positive or negative – could be extremely useful for manipulative
purposes, because without diverting
attention to manipulative material it is unlikely to have much effect
on the affective perception.
The assigned 'aesthetic value' of a picture is the main measurement
of affective effects of manipulative
pictures and captions in this study. At this point the potential of
effective photo manipulation as well
as its possible restraints are beginning to take shape. The digitally
altered negative and positive
versions of the sleeping girl are fitting perfectly to the assumption
of strong and calculable effects
of photo alterations, but this is of limited applicability for the
two other motives: For the Cambodian
Woman the nega-
[page 16]
tive version only had significant manipulative power and for the
group of man and boys
only the positive version proved to be effective on a statistical
significant level. However, for the
picture of the group the different captions were clearly effective.
They caused that one and the same
image was assessed completely different, regarding the perceived
aesthetic quality of the photo
(see Figure 11).
Figure 11: Aesthetic Value of Different Pictures and Captions
Because it is unusual for published photos to be unaccompanied by
captions, these results should be
regarded at least as somehow unsettling. If for specific motives
photo alteration as well as manipulative
captioning can be applied to achieve desired and calculable effects
on the affective level of perception,
no one should be very surprised to find more manipulative efforts
combining fake pictures and fake captions.
Given the long tradition of such combinations – "the first faked
picture and caption combination"
(Lester 1991, p. 92) is from 1840 – it seems to be overdue to examine
those combinations in depth.
The second measurement of emotional effects of the treatment material
was named 'perceived tranquility'
and means respectively the calmness and the peacefulness a picture
exudes. In the case of the tested
material little difference could be found here. But the positive
version of the sleeping girl and the
negative version of the Cambodian
[page 17]
woman seemed to have a distinct influence of the perception of the
participants. Only the negative caption
of the group picture had a significant negative effect on the
'perceived tranquility' (see Figure 12).
Figure 12: Perceived Tranquility of Different Pictures and Captions
All in all, we may conclude that there are signs of the potential
effectiveness of both altered photos
and manipulative captions. Why these effects are not constant under
all treatment conditions is open for
speculation. One possible explanation is the difference of the
implemented alterations. The first motive
– the sleeping girl – was manipulated very extensively whereas the
second motive – the Cambodian woman –
was only cosmetically altered in both directions, positive and
negative. The main difference is between
the negative and the two other versions, probably because of the fact
that cosmetic alterations in a
positive direction are to a certain degree more acceptable than other
forms of photo manipulation
(Huang 2001, p.167; Forster 2003, p. 88). The alterations of the
third motive – the group with man
and boys – were probably too weak as well in its positive and
negative expressions, to cause stronger
effects. Similar reasons could be responsible for the different
performance of the captions. Obviously
the captions of the third motive had the strongest formulation and
therefore the most significant effects.
[page 18]
5. Discussion
The manipulation of photographic images and the manipulative
utilization of photo- accompanying captions
have both a long and inglorious tradition, and with digitalization
they are entering a new age. However,
what makes faked photographic images so disturbing is not the
easiness of the forgery in the times of
Photoshop, nor the unethical behavior of photo reporters and editors,
which can be dealt with (Ang 2000,
pp. 137; Kobré 2000, pp. 300; National Geographic Magazine 1995).
What makes photo alteration a problem
is first and foremost the supposedly high potential of manipulated
pictures to deceive the viewer.
These presumed negative effects are central to the condemnation of
the "photo fakery" (Brugioni 1999).
Unfortunately our knowledge of the potential effects of altered
images is still next to nothing. This
study sought to reduce our gap of knowledge in this specific field of
media effects research and can
only be regarded as a first step in the right direction.
Despite earlier findings (Forster 2003, p. 97), there is no evidence
in the results of this study that
viewers are able to recognize photo manipulations. Furthermore, the
effects on the affective perception
of the pictures are calculable to a certain degree. It seems also
plausible to acknowledge that
manipulative captions can have effects on viewers' perceptions as
well. Therefore the most unsettling
development would be an increase of combined picture and caption
manipulations, with presumably
devastating effects on journalistic credibility if detected. However,
the presented results could have
been biased to some extent by the employed quasi-experimental design
which assured that the participants
not only viewed the photo but read the caption as well. Therefore the
potential of manipulative captions
to have effects in real viewing situations might not be as high as in
this study. Otherwise, the initially
cited example of Lcpl. Boudreaux shows that it is easy to place a
caption directly in the image so that
it can hardly be overlooked. Finally, further research on the
perception and the effects of altered and
manipulative captioned images seems to be necessary.
[page 19]
References
Anderson, John R.: Kognitve Psychologie [Cognitive Psychology and its
Implications]. Heidelberg:
Spektrum ²1996.
Ang, Tom: Picture Editing. Oxford: Focal Press ²2000.
Berger, Arthur Asa: Seeing is Believing. An Introduction to Visual
Communication. Mountain View:
Mayfield Publishing ²1998.
Blumer, Herbert: Symbolic Interactionism. Perspective and Method.
Berkeley: Univ. of Calif. Press
[1969] 1998.
Bonopartis, Nikolas: Naked eye can´t tell if it´s a lie – Military
scandal shows doctoring images is
easy. 2004.
(http://www.cs.dartmouth.edu/~farid/press/poughkeepsie04/poughkeepsie04.html;
20.07.04)
Bortz, Jürgen/Döring, Nicola: Forschungsmethoden und Evaluation
[Research Methods and Evaluation].
Berlin: Springer ²1995.
Brugioni, Dino A.: Photo Fakery. The History and Techniques of
Photographic Deception and Manipulation.
Dulles: Brassey's 1999.
Buell, Hal: Moments. The Pulitzer Prize Photographs. A Visual
Chronicle of Our Time. New York:
Black Dog & Leventhal 1999.
Campbell, Donald T./Stanley, Julian C.: Experimental and
Quasi-Experimental Designs for Research.
Boston: Houghton Mifflin 1966.
Carlebach, Michael L.: The Origins of Photojournalism in America.
Washington: Smithsonian Institution 1992.
Chapnick, Howard: Truth Needs No Ally. Inside Photojournalism.
Columbia: Univ. of Missouri Press 1994.
Czienskowski, Uwe: Wissenschaftliche Experimente: Planung.
Auswertung, Interpretation
[Scientific Experiments. Design, Analysis, Interpretation]. Weinheim:
Psychologie Verlags Union 1996.
Forster, Klaus: Rezeption von Bildmanipulationen [Perception of
Manipulated Pictures].
In: Knieper, Thomas/Müller, Marion (Eds.): Authentizität und
Inszenierung von Bilderwelten
[Authenticity and Staging of Image Worlds] Köln: Halem 2003, 66-101.
Früh, Werner: Gewaltpotentiale des Fernsehangebots. Programmangebot
und zielgruppenspezifische
Interpretation [Potentials of Violence in TV Programs]. Wiesbaden:
Westdeutscher Verlag 2001.
Früh, Werner/Schönbach, Klaus: Der dynamisch-transaktionale Ansatz.
Ein neues Paradigma der
Medienwirkungen (The Dynamic-Transactional Approach. A New Paradigm
of Media Efects]. In:
Publizistik Vol. 27, 1982, 74-88.
Gauthier, Isabel/James, Thomas W./Curby, Kim M./Tarr, Michael J.: The
Influence of Conceptual
Knowledge on Visual Discrimination. In: Cognitive Neuropsychology
Vol. 20, 2003, 507-523.
Greer, Jennifer D./Gosen, Joseph D.: How Much is Too Much? Assessing
Levels of Digital Alteration
as factors in Public Perception of News Media Credibility. In: Visual
Communication Quarterly. Vol.
9, No. 3, Summer 2002, 4-13.
Haus der Geschichte der BRD [House of the History of the FRG]: Bilder
die Lügen [Pictures That Lie]
Bonn: Bouvier 2003.
[page 20]
Hörburger, Christian: Kriegsbilder oder Wandel des Entsetzlichen
[Pictures of War or the Change
of the Dreadful]. In: Büttner, Christian/Gottberg, Joachim
von/Metze-Mangold, Verena (Eds.):
Der Krieg in den Medien [The War in the Media]. Frankfurt: Campus 2002, 29-37.
Huang, Edgar Shaohua: Reader's Perception of Digital Alteration in
Photojournalism. In: Journalism
& Communication Monographs, Vol. 3, No. 3, Autumn 2001, 149-182.
Jaubert, Alain: Fotos die lügen. Politik mit gefälschten Bildern
[Photos That Lie. Politics With
Faked Pictures. Frankfurt a.M.: Athenäum 1989.
Kobré, Kenneth: Photojournalism. The Professionals' Approach. Boston:
Focal Press 42000.
Kobré, Kenneth: The Long Tradition of Doctoring Photos. In: Visual
Communication Quarterly Vol.
2, No. 2, Spring 1995, 14-15.
Laatz, Wilfried: Empirische Methoden. Ein Lehrbuch für
Sozialwissenschaftler. [Empirical
Methods. A Textbook for Social Scientists]. Frankfurt/Main: Verlag
Harri Deutsch 1993.
Lester, Paul Martin: Visual Communication. Images with Messages.
Belmont: Wadsworth ²2000.
Lester, Paul: Photojournalism. An Ethical Approach. Hillsdale:
Lawrence Erlbaum Asssociates 1991.
Lippmann, Walter: Public Opinion. New York: Free Press [1922] 1997.
Lowrey, Wilson: Normative Conflict in the Newsroom: The Case of
Digital Photo Manipulation.
Paper presented at the AEJMC Convention, Washington 2001.
Messaris, Paul: Visual Literacy. Image, Mind, & Reality. Boulder:
Westview 1994.
National Geographic Magazine: Photography Alteration Protocol, June 7, 1995.
Newton, Julianne H.: The Burden of Visual Truth. The Role of
Photojournalism in Mediating Reality.
Mahwah, NJ: LEA 2001.
Ray, Nick: Travel Handbuch Kambodscha [Travel Handbook Cambodia].
Berlin: Loose 2000.
Ritchin, Fred: In Our Own Image. The Coming Revolution in
Photography. How Computer Technology
is Changing Our View of the World. New York: Aperture ²1999.
Thomey, Tedd: Immortal Images. A Personal History of Two
Photographers and the Flag Raising
on Iwo Jima. Annapolis: Naval Institute Press 1996.
Whelan, Richard: Robert Capa. The Defintive Collection. New York: Phaidon 2001.
Whelan, Richard: Robert Capa in Spain. In: Aperture (Ed.): Heart of
Spain. Robert Capa's
Photographs of the Spanish Civil War. New York: Aperture 1999.
Wheeler, Tom: Phototruth or Photofiction? Ethics and Media Imagery in
the Digital Age.
Mahwah: LEA 2002.
Willenbrock, Harald: Bilder, Lügen und Legenden [Pictures, Lies, and
Legends]. In:
Süddeutsche Zeitung, Nr. 166, 21./22.07.2001, III.
Wiseman, Sandor/MacLeod, Colin M./Lootsteen, P. J.: Picture
Recognition Improves with Subsequent
Verbal Information. In: Journal of Experimental Psychology: Learning,
Memory and Cognition.
Vol. 11, No. 3, July 1985, 588-595.
|
