Method Supporting the Adoption of Visual Stories in Judicial Systems

Full text

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 1
METHOD SUPPORTING THE ADOPTION OF VISUAL STO-
RIES IN JUDICIAL SYSTEMS
Research paper
Leite, Francisco, ISEC - Coimbra Institute of Engineering, Portugal, cbl@isec.pt
Antunes, Pedro, Victoria University of Wellington, New Zealand, pedro.antunes@vuw.ac.nz
Guimarães, Nuno, University Institute of Lisbon, Portugal, nuno.guimaraes@iscte-iul.pt
Pino, Jose A., DCC - Universidad de Chile, Chile, jpino@uchile.cl
Abstract
The submission of statements of claim is a critical procedure in civil judicial systems, where plaintiffs
plead the court to resolve conflicts with other parties. Under the assumption that plaintiffs seek to rep-
resent themselves, we are particularly concerned with the quality of the information exchange with the
court. To address this problem, we develop a method for constructing statements of claim that uses
visual stories. The research adopts design science research, focusing on method construction and eval-
uation in a set of experiments. The results suggest the method supports the construction of coherent
statements of claim and contributes to promote self-help and visual narrative in judicial systems.
Keywords: Narrative Theory, Visual Stories, Visual Factual Stories, Judicial Systems.
1 Introduction
The simplification of legal services is essential to increase the affordability and accessibility of judicial
systems (Goh, 2018; Rhode, 2015-2016). In particular, the concept of self-help, i.e. using information
systems (IS) to help citizens representing themselves, especially in the area of civil justice, seems very
compelling (Lupo & Velicogna, 2018; Procopiuck, 2018; Singh et al., 2018). However, by promoting
self-help, we may foster a problem with the quality of information exchange: without professional
legal assistance and knowledge about judicial processes, self-represented citizens may end up being
unfairly treated because of poor information exchange with the court (Rhode, 2015-2016).
Since judicial systems are very complex, we have to delimit the scope of this research. Firstly, we only
address the plaintiff’s viewpoint. Secondly, we only consider small civil claims (Cortes, 2013). Third,
we focus on a specific, albeit very important document, the statement of claim (Goh, 2018). This docu-
ment is critical in small claims, as it initiates and sets the stage for the whole process: if it has gaps or
unsupported claims, there is good reason for the court to immediately reject the case; and even if the
case is accepted, its low quality may negatively impact the outcomes. Finally, we address the problem
from a specific frame: visual narrative theory (Cohn, 2013, 2014; Cohn et al., 2012). Visual narrative
theory explains how people create and analyze visual stories (Cohn et al., 2012). Visual stories are ubiq-
uitous in society, helping people to understand events through visual cognition (Cohn, 2013). This the-
ory has already been successfully applied in teaching (Green & Myers, 2010), medicine (Green & Rieck,
2013), business process management (Antunes et al., 2013; Simões et al., 2018; Simões et al., 2016;
Simões et al., 2015), human-computer interaction (Sakamoto et al., 2007; Truong et al., 2006), and
software engineering (Haesen et al., 2010; Williams & Alspaugh, 2008).
We posit that visual narrative theory may also be useful in improving the quality of information
exchange in judicial processes, in particular when plaintiffs represent themselves submitting state-
ments of claim. We propose a method for self-constructing coherent statements of claim which uses

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 2
visual narrative theory. Visual narrative is an area that is unexplored in the judicial domain, which
emphasizes the originality of the proposal: no method like this has been proposed before.
Considering that the method is an IS artefact, and also that it is unique and innovative, we adopt Design
Science Research (DSR) as research paradigm (Hevner et al., 2004), which results in the following
research steps: 1) build a model representing statements of claim as visual stories; 2) build a template
to generate statements of claim as visual stories; 3) build a procedure to analyze the coherence of
statements of claim; and 4) experimentally evaluate the method using the abovementioned model,
template and procedure. This research contributes to explore the adoption of visual stories in judicial
systems. From a theoretical perspective, it contributes to extend visual narrative theory to the judicial
context. It also addresses the problem of the quality of information exchange in self-help judicial sys-
tems.
In the next section, we discuss the research methodology. Then, we discuss the theoretical foundations.
In sections 4 and 5 we discuss and evaluate the method. We conclude the paper with some points for
discussion and future research.
2 Research Methodology
Perimeter of research. In the origin of a judicial process in the civil domain, we can find a conflict.
The contradictory positions taken by those involved in the conflict determine that one of them decides
to resort to the court to resolve it. In general, civil processes are organized in pleadings, sanitation,
instruction, judgment, and sentence. In the first stage, the plaintiff submits a statement of claim to the
court, which is followed by a statement of defense submitted by the defendant. In the sanitation stage,
the plead is evaluated and condensed by a judge. If the process moves to the instruction stage, the parties
present evidence intended to prove the veracity of facts and arguments. In the judgment stage, the court
decides which facts are considered proven. Finally, the court decides and delivers the sentence. This
generic description helps understanding the complexity of the research context.
In this research, we are only interested in the first stage: construction of statements of claim by plaintiffs.
Implicit in our research is a scenario where the plaintiff is a single person facing a conflict with a big
company. Considering convenience and cost, the plaintiff would like to be self-represented. However,
the plaintiff faces the problem of submitting a coherent statement of claim. The different capacities of
the plaintiff and defendant in this scenario justify that our study is restricted to the statement of claim.
Even though it is a small part of a complex process, it is critical regarding the quality of information
exchange: if the claim contains inadequacies or inaccuracies, the court may consider it handicapped and,
worst-case scenario, may immediately close the process. If the claim is not handicapped but has low
quality, that may influence the process in favor of the defendant. Therefore, it is vital for plaintiffs to
submit good quality statements of claim.
Research approach. DSR concerns the design of innovative IS artifacts (Hevner et al., 2004). By arti-
fact we mean anything that is human made and has either a material (e.g. software tools) or abstract
nature (e.g. models, methods and procedures) (Gregor, 2006). This research is centered on method de-
sign. However, the method is a combination of other artefacts: model, template and procedure. The
iterative realization of model, template and procedure provides a rational and transparent pathway for
understanding the method design.
DSR is framed by two principles (Gregor & Jones, 2007; Hevner et al., 2004): rigor using existing and
generating new knowledge; and relevance to the wider research/practice environment in the form of
utility solving relevant problems. Considering rigor, all designed artifacts are based on kernel theory
providing justificatory knowledge for artifact construction (Gregor & Jones, 2007). Adopted as kernel
theory, we consider narrative theory (Labov, 1972, 1997), visual narrative theory (Cohn, 2013, 2014;
Cohn et al., 2012) and judicial narrative (Bex, 2011). These theoretical foundations rigorously bound
the artifacts to existing knowledge. Regarding relevance, we note the design concerns a generalized
problem (quality of information exchange), which clearly pertains to the IS field. Then, the research

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 3
delivers a relevant design solution. In particular, we support the adoption of self-help, which is a current
challenge for innovating judicial systems (Goh, 2018). Finally, the utility of the design solution is
demonstrated through experimentation, which suggests the solution is viable in solving the problem.
3 Theoretical Foundations
A judicial process always materializes from a particular situation involving events and facts. As the facts
are unable to speak for themselves and often cannot be independently observed (as in a laboratory ex-
periment), the narrative reconstruction of events to determine the facts is a necessary and defining con-
stituent of a judicial process (Bex, 2011). Therefore, narrative is present in every judicial system. Judi-
cial narrative can then be defined as a description of events and properties of a situation of legal im-
portance (Almog, 2001; Schmid, 2010).
The construction of judicial narrative is based on stories submitted by the parties (Pennington & Hastie,
1992). Stories combine narrative with intent (Schmid, 2010). A story is a delimited, coherent and chron-
ologically planned sequence of events, mingled with contextual details highlighting the nature and im-
portance of events. Stories also involve discourse, which is the way in which the story is presented using
a language and a medium (e.g. text, speech, pictures, and animations).
A visual story is a particular form of story, using specific language and medium. Visual stories rely on
pictures with symbols, icons, shapes, text, and figures to convey meaning. Visual stories can be powerful
vehicles for the transmission and processing of events. Especially in low literacy communities, pictures
may have an important role illustrating, clarifying, simplifying, and reinforcing words. Visual stories
make the subject and intentions of the message more understandable and therefore significantly
strengthen the communicative effect (Boehme-Neßler, 2011; Kodagoda et al., 2012; Medhi et al., 2010).
For instance, the visual narratives created by the National Truth Commissions in Guatemala (Guatemala,
2000) to explain national violence are a compelling example that underlines the importance of explain-
ing complex events in simple, clear and effective ways.
However, introducing visual stories in judicial systems is a complex endeavor. We take a stepwise ap-
proach to discuss the role of visual stories in judicial systems. We start by analyzing the concept of story
in more detail.
3.1 Stories and factual stories
Narrative structure is implicit in the concept of story. Narrative structure is the organization of events,
links between events, and associated contextual information (Crossley, 2002; Labov, 1972; Sarbin,
1986). Links can be causal, temporal or spatial. Thoughts, memories and ideas help contextualizing the
events and contribute to understand the movement of characters and events in time and space (Jannidis,
2003). The development of narrative structure is commonly referred to as holistic (Bex, 2011). It is the
whole that determines the meaning of the different parts. In a judicial process, it is necessary to present
the facts, but this alone is not enough. An overall narrative structure is necessary to bind facts to events
and context. Sarbin (1986) points out that narrative structure brings logic to a set of events. Appan et al.
(2004) note that narrative structure helps reconstructing the events even when the story is incomplete
and unstructured, but from which the logical relationships can be identified and clarified.
Related to narrative structure, we find the notion of plot. The plot is the way of presenting events ac-
cording to a style (Abbot, 2007). Examples include presenting a story from begin to end, end to begin,
or using flashbacks. A story is held together by the plot (Ricoeur, 1991). Quesenbery and Brooks (2010)
note the plot helps the author to tell the story, helps the audience understanding the story, and helps the
story itself, to stand out as an intelligible account of the events. By controlling the plot, the narrator can
help internalizing the story, thus increasing its effectiveness (Quesenbery & Brooks, 2010).
Above narrative structure and plot, we consider meta-structure, which defines the overall framework on
which the narrator defines the plot. Meta-structure allows the plot to be more easily recognizable. Even

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 4
though the narrator has freedom to define a plot, care should be taken to create a recognizable plot on
which to lay out the story. In particular, judicial processes depend on a specific meta-structure, which
concerns factual stories: the telling of events and other supporting information with the purpose to iden-
tify the facts. A factual story is usually oriented towards temporal order, which contributes to determine
temporal causality (Amsterdam & Bruner, 2002). Labov (1972) suggested a specific meta-structure for
factual stories, which consists of five consecutive elements (brackets indicate an element is optional):
Factual story = Orientation + Complication + Evaluation + Resolution + [Coda]
Orientation refers to the introductory part of the story. Factual stories always start with essential infor-
mation about the participants, setting, place, and temporal frame. The second element is complication.
The complication is an essential element of judicial narrative where the author describes a set of events
in an evolving conflict. The next element is evaluation. For Labov and Waletzky (1966), factual stories
involve points of personal interest, which are expressed in the evaluation. A plot that does not have
evaluation does not indicate the relative importance of the events and may not be effective. Another
element found in factual stories is resolution. The resolution answers the problem posed by the compli-
cation. The evaluation and resolution can be merged into one single element (Labov, 1972). Finally,
factual stories may have an optional element called coda, which signals that the story reached a close
from the narrator’s side (McCormack, 2004).
Even though this meta-structure constrains the story, the narrator is nevertheless allowed to develop
different plots. In particular, plots may be enriched with frames, prescriptions and interludes
(Quesenbery & Brooks, 2010). A frame is a sub-plot within a plot. The sub-plot does not change the
underlying narrative but creates a small story arc that moves out from the narrative and then comes back
to the same point. Sub-plots may be used to narrate what happens to characters (me-them-me), what
happens spatially (here-there-here), and what happens temporally (now-then-now). A prescription is
used to convey a causal explanation, which may be essential to deduce a fact. Quesenbery and Brooks
(2010) define a prescription using the following construct: “given [a context] and [some more context],
when [an event], then [an outcome] and [another result]”. Finally, an interlude, or break in the narrative,
enriches a story with contextual details, e.g. about places, time, culture, and geography.
3.2 Visual stories
The concepts of story and factual story apply to any type of story, including visual stories. It just happens
that visual stories use pictures instead of text. By pictures we mean a set of elements that must be un-
derstood through visual cognition, i.e. scanning shapes, symbols and characters (Goel, 1995). In fact,
pictures may include both visual and textual elements (Cohn et al., 2012). Multiple pictures may have
to be arranged to convey a visual story. As an example, a set of pictures may convey information about
a car accident, the location of intervening cars, and details about the intersection where the accident
occurred. The visual story may then use several pictures to tell how the car accident enfolded and where
it happened, showing e.g. a pedestrian crossing the road, then one vehicle trying to stop, and another
vehicle colliding with a motorcycle.
Visual stories also have narrative structure and plot. In the car accident example, while several pictures
narrate the events and the context in which the events occurred, the plot commands such information in
a particular way (Cohn, 2013). Sarbin (1986) says that, if people are presented with a set of pictures,
then they will try to combine them according to a known plot to build a story. The plot may be either
implicit or explicit in the visual story and depends on the knowledge and experience of readers (Cecez-
Kecmanovic, 2005).
Visual stories should also have a meta-structure, which helps understanding the plot. Cohn (2013) pro-
posed a meta-structure consisting of five consecutive elements (brackets indicate an element is optional):
Visual story = [Establisher] + [Initial [+ Prolongation]] + Peak + [Release]

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 5
Establisher is the element that establishes the action, without acting on it, while Initial is the element
that initiates the action, and prolongation extends the action according to a certain narrative arc. The
peak is the picture that marks the highest point of tension in the narrative (Cohn, 2013). Finally, release
is the element that releases the tension of the narrative. According to Cohn (2014), peak is the most
important and only essential element of a visual story, which tells the most important things that hap-
pened.
3.3 Coherence of factual stories
Stories in the judicial context need to be coherent, a concept referring to the way in which the elements
intervening in a story bind to each other to give sense to the narrative (Bex, 2011; Wintgens, 2005).
However, coherence is an umbrella term for more specific properties such as completeness, consistency
and credibility (Pennington & Hastie, 1992). Completeness assesses whether a story contains all neces-
sary elements, while consistency evaluates the existence of contradictions and credibility assesses
whether the narrative is realistic. Wintgens (2005) notes that consistency is not a necessary condition
for coherence; it is however a disabling property.
In the assessment of facts by a court, credibility can be checked by analyzing both the internal and
external narrative elements linking the story to the narrator and the socio-cultural reality (Arnauld &
Martini, 2015). Labov (1997) also noted that credibility is the extent to which the listeners believe that
the described events occurred as reported by the narrator. Placed in an area between the internal and the
external, the completeness of the description of a set of events also affects the credibility of a story
(Arnauld & Martini, 2015).
Additional properties of coherence can also be considered, such as temporal, causal, spatial, referential,
and thematic (Gernsbacher, 1991; Habermas & Bluck, 2000). Temporal coherence refers to consistency
in identifying the moments in which events occurred, while causal coherence refers to consistency in
the prescriptions suggesting explanations for a chain of events, spatial coherence refers to consistency
in identifying the places where the events occurred, referential coherence refers to the consistency about
whom or what is described, and finally, thematic coherence concerns the organization of a story around
a central theme.
Considering the connections between all these properties, we suggest that meta-structure promotes com-
pleteness and consistency through logical sequencing (in time and space), causal reasoning (in the chain
of events), and consideration for all elements required by judicial narrative.
4 Artefact Development
Model representing statements of claim as visual stories. We propose a model for representing state-
ments of claim which combines propositions from judicial narrative and visual narrative (Figure 1). We
take the view that meta-structure fosters coherent visual stories and therefore adopt elements proposed
by Labov (1972) and (Cohn, 2013) to define the model. A story starts by identifying the parties (plaintiff
and defendant) and objects (e.g. a product or a service) involved in the judicial process. We designate
this element as “constituents”. This element does not start the narration of the conflict; it just reflects
the assumption that a coherent factual story should describe who enters the action before describing the
action.
We then consider an element describing what initiates the conflict: the “initiation”. We consider two
alternative plots accomplishing this goal. The first plot describes one or more events that ignited the
conflict. Alternatively, the second plot describes the situation that ignited the conflict. These two alter-
natives reflect different ways of starting a factual story.
Then we move the story to the most extensive part, which we name “evolution”. The evolution element
uses various pictures to describe what happened with significant detail and contextual information. How-
ever, from the previous discussion, we also understand that factual stories need to be enriched with

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 6
narrative elements beyond the sequence of events. For that reason, we consider the inclusion of multiple
frames, interludes and prescriptions.
We then consider one single picture appearing after the evolution phase and singling out a particular
event as the peak of the factual story. In a statement of claim, it corresponds to the point where the
plaintiff explicitly identifies the conflict.
Figure 1. Model representing statements of claim as visual stories
We finally consider that a factual story may end with a component combining evaluation and resolution.
This element is set as a follow-up to the peak. Regarding the statement of claim, the follow-up may
combine three types of information: 1) evaluation, expressing a final opinion about the conflict; 2) result,
identifying the consequences of the conflict; and 3) solution, stating how the plaintiff would like the
conflict to be resolved.
Figure 2. Template for creating statements of claim as visual stories

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 7
Template for creating statements of claim as visual stories. We are aware of many tools that can be
used to create generic visual stories, such as Witty Comics, Strip Generator, Make Beliefs, Write Com-
ics, Chogger, Cartoon Story Maker, and ComicStripCreator. However, none specifically facilitates the
creation of statements of claim. To accomplish our specific goals, we created a template in Microsoft’s
PowerPoint, which is illustrated in Figure 2. The template makes the proposed model explicit to the
users by defining placeholders for inserting the constituents, initiation, evolution, peak, and follow-up
of a conflict. The template also contains generic pictures showing typical commercial events such as
buying a product, paying a product, discussing with the seller, etc., which can be used to compose the
story. Pictures show either one or two characters, which can represent both the plaintiff and the defend-
ant. The characters are shown along with labels and dialogue bubbles. The labels are used to provide
references to time, space and the character’s identities. The dialogue bubbles can then be used to put
words in the character’s mouths. Most pictures also have larger text fields used to narrate the portrayed
events in more detail. The visual elements shown in pictures were taken from publicly available artwork,
e.g. a house, a container, a plant or a court image.
The template does not require the narrator to draw a story. Instead, a story is composed by picking
pictures from the library, placing them in a placeholder provided by the template, and then configuring
labels, dialogue bubbles and text fields.
Coherence
Elements
Questions
Mandatory
elements
Mandatory elements = constituents, initiation, evolu-
tion, peak, follow-up
Has the statement of claim all manda-
tory elements?
Element
types
Constituents = parties AND objects
Initiation = events XOR situation
Evolution = events, frames, interludes, prescriptions
Peak = event
Follow-up = evaluation, result, solution
Each element in the statement of
claim has the required type?
Order of ele-
ments
Order = constituents ® initiation ® evolution ®
peak ® follow-up
Evolution = events, frames, interludes, prescriptions
Each element is defined in the re-
quired order?
Frames, interludes and prescriptions
are defined within the evolution ele-
ment?
Table 1. Analytical procedure for analyzing structural coherence
Procedure for analyzing statements of claim. We now define a procedure to check the coherence of
statements of claim. This procedure is an important part of our research and is necessary to evaluate the
experimental data. We divide the procedure in two phases considering structural analysis and substan-
tive analysis. In the first phase, we check if a statement of claim complies with the model (Table 1) by
looking at the form without considering the actual content. The model defines some mandatory elements,
such as the constituents, initiation, evolution, peak, and follow-up. If any mandatory element is missing,
then the statement of claim cannot be considered coherent. The model also defines element types. For
instance, the initiation must contain either events or a situation. If any element does not conform to the
defined types, then the statement of claim is not coherent. Finally, we also check the order of elements.
If the order does not adhere to the model, the statement of claim is not coherent.
Structural analysis provides a useful and meaningful assessment of the statement of claim, which nev-
ertheless must be followed by an analysis of the substance of the claim. Therefore, in the second phase,
we focus on the substance of the pictures provided in the statement of claim. This involves checking the
details, i.e. what dialogues are expressed by the characters, how they relate to time and space, and how
they relate to contextual information provided in text fields. To structure this analysis, we consider the
requirements suggested by Gernsbacher (1991), Pennington and Hastie (1992) and Habermas and Bluck
(2000). The procedure is summarized in Table 2.

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 8
Table 2. Analytical procedure for analyzing substantive coherence
5 Evaluation
To evaluate the method, a research question is formulated: Considering the context where plaintiffs do
not have specific knowledge about judicial processes, can they create coherent statements of claim in
the form of visual stories?
For the evaluation, we formulate the equation R = f (P, C, M) where R is the result of applying the
procedure for analysing a statement of claim; P is a participant; C is the case describing a conflict; and
M is the method used to create a statement of claim. To assess R, we control P and C, and use M and R
as independent and dependent variables, respectively.
Two cases of legal conflict between citizens and companies were selected (C1, C2). The cases describe
real claims taken from sentences published by courts of first instance. Relevant elements from the sen-
tences were identified and transposed to construct descriptions of conflicts from the plaintiffs’ perspec-
tives. These descriptions were sufficiently detailed to allow the participants to elaborate a coherent state-
ment of claim. However, three different narratives of the events were developed for each case: C(o)
written narrative of events in chronological order; C(u) unordered written narrative of the events; and
C(v) video narrating the events in chronological order. The purpose of C(o) and C(u) is to moderate
effects caused by how the narrative structure of the cases presented to the participants in the experiment
affects the experimental results.
Method comparison. The creation of statements of claim involved two methods: written (Mw); and
visual (Mv). Mw is the baseline for comparative evaluation, as it corresponds to the typical case where a
self-represented plaintiff submits a written statement of claim to the court. Mv is the treatment we seek
to evaluate, as it corresponds to the case where a self-represented plaintiff creates a statement of claim
as a visual story, using the proposed method.
Coherence
criteria
Elements
Requirements
Questions
Temporal
Initiation,
evolution,
peak
Events must be temporally anno-
tated
Events must be presented according
to temporal order
Are all events temporally annotated?
Is it possible to understand which event
follows a given event?
Spatial
Initiation,
evolution,
peak
Events must be spatially annotated
Events must be related to reported
places
Are all events spatially annotated?
Is it possible to understand which place
follows a given place?
Causal
Initiation,
evolution,
peak
Follow-up must be causally related
to peak
Prescriptions must have cause-ef-
fect links to events
Is peak related to initiation?
Is follow-up related to peak?
Are prescriptions, if any, related to events?
Thematic
All
Elements must refer to a central
theme
Is the set of events referring to the initia-
tion?
Complete-
ness
All
Elements must provide all neces-
sary elements to understand the
conflict
Is the set of events complete sufficient to
understand the conflict?
Does the evolution provide enough contex-
tual information?
Con-
sistency
All
Elements must not provide any in-
ternal contradictions
Are there any contradictions among
events?
Credibility
All
Elements must fit the sociocultural
reality
Is the information in elements feasible?

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 9
For Mv, we used the template for creating statements of claim as visual stories (Figure 2). For Mw, we
had to define a way for creating written statements of claim, which could be compared to the developed
template. The adopted approach is a textual template which presents the users with the following sec-
tions: introduction, narrative, and conclusion. This structure follows the structure we commonly find in
written statements of claim.
Operationalization of experiments. The evaluation involved four sets of participants (P1-4) and four
experiments (E1-4). The participants consisted of twenty people between 19 and 54 years old, either
graduated or pursuing undergraduate studies, selected by direct contact. The participants had to comply
with two constraints: not having specific knowledge about judicial processes; and knowledgeable about
Microsoft’s PowerPoint.
In E1, P1 members were asked to create visual claims for case 1 (P1,C1(o),Mv), and P2 were asked to
create written claims for the same case (P2,C1(o),Mw). In this experiment, the first treatment represents
the plaintiff who, to resolve a conflict, appeals to the court using a visual statement of claim and the
second treatment represents the plaintiff who appeals to the court using a written claim. For E2, to min-
imize any learning effects, a different case was used and P1 and P2 were asked to create claims using the
alternative treatment (P1,C2(o),Mw) and (P2,C2(o),Mv). In E3, a different group of participants created
written and visual claims from cases narrated with video (P3,C1(v),Mw) and (P3,C2(v),Mv). Finally, in E4
another group of participants created written and visual claims from unordered textually narrated cases
(P4,C1(u),Mw) and (P4,C2(u),Mv).
Each experiment started by showing the participants an example case (C3) and a coherent statement of
claim about the case in the form required by the experiment (either written or visual). Then the partici-
pants were asked to create a similar output for the assigned case. No further instructions were provided.
Data processing. The procedure for analyzing statements of claim was used to assess the data. To trans-
late the results into quantities, two indicators were defined: 1) Ac, average coherence of a structural
element in the statement of claim; and 2) Aw, average coherence of the whole statement of claim. Ac is
calculated in two steps. In the first step, for a particular structural element, we assess all possible coher-
ence criteria shown in Tables 1 and 2. In the second step, we average all the coherence measures for the
structural element under consideration. Each coherence check returns a value that is either 0 (not coher-
ent) or 1 (coherent).
To illustrate, when evaluating the temporal coherence of a statement of claim narrated as a visual story,
each picture belonging to the initiation, evolution and peak must be analyzed. If in these elements it is
verified that all dates are well defined and that from these dates it is possible to understand the temporal
sequence of events, then 1 is assigned to these three elements, regarding temporal coherence. Of course,
the pictures in the statement of claim must also be assessed regarding the spatial, causal, thematic, and
other properties.
Table 3 summarizes the results for Aw in the various treatments. The results for Ac are not shown for
conciseness. These results show that the participants created more coherent visual claims than coherent
textual claims. These are aggregate results considering both structural and substantive coherence. How-
ever, we note that the differences between treatments result from substantive coherence: 80% visual
claims were not coherent in a substantive way, while 50% textual claims were either not complete or
consistent in a substantive way.
To illustrate the summary provided in Table 3, in the evaluation of the visual claim from participant 16,
it was identified that, in the initiation and evolution elements, an essential event for understanding the
conflict was missing. Therefore, the statement of claim had no temporal, no spatial and no causal coher-
ence, was incomplete, was not consistent, and was not credible. In this claim, each of the six specific
coherence properties was left with a value of 0 for initiation and evolution elements. Considering the set
of elements on which the six properties applies (Table 2), this claim had: Ac = 33% for temporal, spatial
and causal coherence; Ac = 60% for completeness, consistency and credibility; Ac = 100% for thematic
coherence; and the overall coherence was Aw= 52%.

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 10
Analysis of results. The experimental results show that plaintiffs may indeed be able to create coherent
statements of claim – both using written and visual claims. However, the participants created a higher
number of coherent visual claims than coherent written claims. These results suggest the method pro-
motes the quality of information exchanged between self-represented citizens and judicial systems.
We carried out statistical tests to determine if the experimental results from the two samples (Mw and
Mv) were statistically different. First, normality tests were performed using the Shapiro-Wilk test. With
a probability of error of 5%, it was concluded that the sample distributions were not normal. Then, the
samples were analyzed using the Wilcoxon matched-pairs (Wilcoxon signed-ranks test) nonparametric
test. In the interpretation of the test results a significance level of α = 0.05 was used. For the conditions
specified, the Wilcoxon test statistic is W = 6 and the critical value is W * = 10. In this case, as W < W
*, the result led to the rejection of the null hypothesis. Thus, statistically, there is sufficient evidence to
state that, on average, visual claims are more coherent that written claims.
Exp.
(E)
Part.
(P)
Coherence of visual claims (Mv)
Coherence of textual claims (Mt)
Case
Struc-
tural
Substan-
tive
AW
(%)
Case
Struc-
tural
Substan-
tive
AW
(%)
E1,
E2
1
C1
1
1
100
C2
1
1
100
2
1
1
100
1
0
88
3
1
1
100
1
0
88
4
1
0
76
1
0
88
5
1
0
88
1
1
100
6
C2
1
1
100
C1
1
1
100
7
1
1
100
1
1
100
8
1
1
100
1
1
100
9
1
0
88
1
0
88
10
1
1
100
1
0
84
E3
11
C1
1
1
100
C2
1
1
100
12
1
1
100
1
0
68
13
1
1
100
1
0
76
14
1
1
100
1
1
100
15
1
1
100
1
0
88
E4
16
C1
1
0
52
C2
1
0
28
17
1
1
100
1
1
100
18
1
1
100
1
0
16
19
1
1
100
1
1
100
20
1
1
100
1
1
100
100.0%
80.0%
95
100%
50.0%
86
Table 3. Experimental results (1 means the claim is coherent, while 0 means it is not)
We also evaluated if there were significant statistical differences between the number of coherent visual
claims and the number of coherent written claims. For the two samples, normality tests were performed
using the Shapiro-Wilk test. With a probability of error of 5%, it was concluded that in the two samples
the distribution is not normal. Thus, the analysis used the Wilcoxon nonparametric test for paired data.
For the conditions specified, the Wilcoxon test statistic is W = 4.5 and the critical value is W * = 5. In
this case, as W < W *, the result led to the rejection of the null hypothesis. Therefore, with a 5% error
probability, we can statistically infer that the number of coherent visual claims was higher than the
number of coherent written claims.

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 11
Due to the nature of judicial processes, it is important that statements of claim are coherent. Therefore,
these results provide preliminary empirical evidence that, for plaintiffs lacking technical knowledge
necessary to create a coherent statement of claim, visual narrative can facilitate the creation of a coherent
description of events. The results also support the claim that participants using a visual method perform
better than using a textual method for generating a statement of claim.
Looking into the results with more detail, we note that structural coherence was similar in both visual
and textual claims. These results may be related to the way we operationalized the experiments, using
in both treatments templates that foster users to provide all structural elements required by a statement
of claim. Differences between visual and textual claims therefore result from substantive coherence,
where the template for creating statements of claim as visual stories seems to provide an advantage.
Since the template makes the temporal and logical relationships between the elements of a statement of
claim more explicit to users, we suggest that the observed differences may in fact be attributed to the
added value provided by the proposed model.
6 Discussion and Conclusions
The adoption of visual narrative in the judicial domain has not been previously researched. This study
provides a first step in this research area, which is necessarily delimited in scope and exploratory in
nature. However, we focus on a problem that is very relevant for the IS field, which concerns the quality
of information exchanged between self-represented citizens and judicial systems. Contributions in this
area may have significant impact in society, as self-representation and self-help may contribute to ex-
pedite processes, reduce costs, and increase convenience, accessibility, fairness, and openness.
Our contribution is a method supporting the creation of statements of claim by self-represented plain-
tiffs. The uniqueness of this method is adopting visual narrative theory to create visual stories. Further-
more, the method supports the creation of coherent stories. The method includes a model and a template.
The model defines the meta-structure of visual stories in the judicial domain. The template then supports
the creation of statements of claim with model support (self-help). The model extends narrative theory
and visual narrative theory to the particularities of the judicial context. Regarding coherence, we address
this challenging requirement from multiple dimensions, including temporal, spatial, causal, thematic,
credibility, consistency, and completeness.
The juxtaposition of coherence and narrative structure, one orienting another, is essential to our approach
to the problem of quality of information exchange. Considering the close relationships between narrative
structure and coherence, the proposed model articulates a set of structural elements and types of rela-
tionships from which the narrator can build a coherent statement of claim. On the one hand, narrative
structure supports the creation of stories conforming to well-known referential schemes used in judicial
narrative. On the other hand, the adoption of common principles of coherence facilitates the construction
and analysis of statements of claim, which in turn contribute to avoid certain deficiencies, inaccuracies
and ambiguities in portrayed events, and which ultimately contribute to judicial narrative.
Considering the ambitious, albeit exploratory, nature of this research, the challenges and problems val-
idating it were considerable. To evaluate the method, we had to develop a detailed and systematic pro-
cedure for checking the coherence of statements of claim. Then, we had to devise and implement a set
of controlled experiments. Finally, we had to devise a strategy and appropriate measures for comparing
the method against a baseline.
The empirical results provide some evidence that the method supports the creation of coherent state-
ments of claim; and also provide some evidence that using the method can give better results than simply
writing a statement of claim.
In a broader perspective, we note our results not only foster the adoption of self-help in judicial systems
but also suggest several opportunities for digital institutionalization in the judicial environment. For
instance, we envisage the method integration as a service in online judicial platforms, which may then

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 12
support and oversee the construction of statements of claim. Our strict separation between method, tem-
plate and tool allows the method to be integrated with different templates and tools and using different
levels of technological support in the construction of individual stories. We also envisage the method
being provided by independent organizations supporting citizens with difficulties accessing the law.
Even if such organizations face financial constraints in developing and maintaining complex judicial
services, the proposed method offers a low-cost alternative to self-help support, which may consist in
just offering a collection of templates for the most frequent claims.
This study also opens several avenues for future research. Firstly, by providing a discussion of prelimi-
nary evidence and existing limitations, the study offers a foundation for more thorough experimentation
with visual stories in the judicial domain. In particular, we consider strengthening the evaluation by
conducting further experiments with more controlled conditions, e.g. regarding differences between vis-
ual and textual construction of statements of claim, and potential impact of templates in the results. We
also consider conducting experiments with incomplete narratives, as omissions may also affect the re-
sults. Future research should also consider combining the assessment of coherence with other evaluative
dimensions, such as understandability (for citizens) and utility (for citizens and courts).
Secondly, future experiments may consider comparing self-representation/self-help with the more tra-
ditional approach to communicating with the court through legal experts. Another avenue to consider
arises from the combination of the proposed method with decision support systems. Currently, the tem-
plate used to create statements of claim is based on a common presentation tool. If instead based on a
decision support system, the template may offer additional functionality, in particular self-assessment
of the coherence of statements of claim. A validated model for judicial narrative also provides a funda-
mental trigger for the development of other computer-based tools that could help users to better under-
stand judicial processes.
Finally, future research may also explore the adoption of visual stories in other stages of the judicial
process requiring communication with the court, which may also consider bi-directional communica-
tion. As noted earlier, such communication is critically dependent on the quality of the information that
is exchanged between the different stakeholders, and this research provides a contribution to address
this relevant problem.
References
Abbot, P. (2007). Story, plot, and narration. In D. Herman (Ed.). The Cambridge Companion to
Narrative. Cambridge: Cambridge University Press.
Almog, S. (2001). As I Read, I Weep-In Praise of Judicial Narrative. Oklahoma City University Law
Review, 26(2), 471-501.
Amsterdam, A., & Bruner, J. (2002). Minding the Law. Cambridge, MA: Harvard University Press.
Antunes, P., Simões, D., Carriço, L., & Pino, J. (2013). An End-User Approach to Business Process
Modeling. Journal of Network and Computer Applications, 36(6), 1466-1479.
doi:10.1016/j.jnca.2013.03.014
Appan, P., Sundaram, H., & Birchfield, D. (2004). Communicating everyday experiences. Paper
presented at the Proceedings of the 1st ACM workshop on Story representation, mechanism and
context, New York, NY, USA.
Arnauld, A., & Martini, S. (2015). Unreliable Narration in Law Courts. In V. Nünning (Ed.). Unreliable
Narration and Trustworthiness. Intermedial and Interdisciplinary Perspectives. (pp. 347-370).
Berlin, Boston: De Gruyter.
Bex, F. (2011). Arguments, Stories and Criminal Evidence, A Formal Hybrid Theory. Dordrecht
Heidelberg London New York: Springer.
Boehme-Neßler, V. (2011). Pictorial Law - Modern Law and the Power of Pictures. Heidelberg:
Springer.

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 13
Cecez-Kecmanovic, D. (2005). A Sensemaking Theory of Knowledge in Organizations and Its
Application. In J. Davis, E. Subrahmanian, & A. Westerberg (Eds.). Knowledge Management.
Organizational and Technological Dimensions (pp. 55-74). Heidelberg: Physica-Verlag
Cohn, N. (2013). Visual Narrative Structure. Cognitive Science, 34, 413–452. doi:10.1111/cogs.12016
Cohn, N. (2014). The architecture of visual narrative comprehension: the interaction of narrative
structure and page layout in understanding comics. Frontiers in Psychology, 5, 680.
Cohn, N., Paczynski, M., Jackendoff, R., Holcomb, P., & Kuperberg, G. (2012). (Pea)nuts and bolts of
visual narrative: Structure and meaning in sequential image comprehension. Cognitive
Psychology, Volume 65(1), 1-38.
Cortes, P. (2013). Small Claims in Ireland and the EU: The Need for Synergy between National Courts
and Extrajudicial Redress. In N. Neuwahl & S. Hammamoun (Eds.). The Philosophy of Small
Change: Transnational Litigation in the EU and Beyond: Éditions Thémis.
Crossley, M. (2002). Introducing narrative psychology. Narrative, Memory and Life Transitions.
Huddersfield: University of Huddersfield Press.
Gernsbacher, M. (1991). Cognitive Processes and Mechanisms in Language Comprehension: The
Structure Building Framework. Psychology of Learning and Motivation, Volume 27, 217–263.
doi:10.1016/S0079-7421(08)60125-5
Goel, V. (1995). Sketches of Thought. Cambridge, MS, USA: The MIT Press.
Goh, J. (2018). The self-represented litigants’ challenge: A case study. Alternative Law Journal, 43(1),
48-50.
Green, M., & Myers, K. (2010). Graphic medicine: use of comics in medical education and patient care.
BMJ, 340. doi:10.1136/bmj.c863
Green, M., & Rieck, R. (2013). Missed It. Annals of Internal Medicine, 158(5_Part_1), 357-361.
doi:10.7326/0003-4819-158-5-201303050-00013
Gregor, S. (2006). The Nature of Theory in Information Systems. MIS Quarterly, 30(3), 611-642.
Gregor, S., & Jones, D. (2007). The Anatomy of a Design Theory. Journal of the Association of
Information Systems, 8(5), 312-335.
Guatemala, A. d. (2000). Memoria, Verdad y Esperanza (Versión popular del informe REMHI:
Guatemala: Nunca Más)
Habermas, T., & Bluck, S. (2000). Getting a life: The emergence of the life story in adolescence.
Psychological Bulletin, 126(5), 748-769.
Haesen, M., Meskens, J., Luyten, K., & Coninx, K. (2010). Draw Me a Storyboard: Incorporating
Principles and Techniques of Comics to Ease Communication and Artefact Creation in User-
Centred Design. Paper presented at the 24th BCS Proceedings of the Conference on Human
Computer Interaction, Dundee, UK.
Hevner, A., March, S., Park, J., & Ram, S. (2004). Design Science in Information Systems Research.
Management Information Systems Quarterly, 28(1), 75-105.
Jannidis, F. (2003). Narratology and the Narrative. In T. Kindt & H. Müller (Eds.). What is
narratology?: questions and answers regarding the status of a theory (Vol. 1): Gruyter.
Kodagoda, N., Wong, B., Rooney, C., & Khan, N. (2012). Interactive visualization for low literacy
users: from lessons learnt to design. Paper presented at the Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems, Austin, Texas, USA.
Labov, W. (1972). Language in the inner city: Studies in the Black English vernacular (Vol. 3):
University of Pennsylvania Press.
Labov, W. (1997). Some Further Steps in Narrative Analysis. Journal of Narrative and Life History, 7,
1-4.
Labov, W., & Waletzky, J. (1966). Narrative analysis: Oral versions of personal experience. In J. Helm
(Ed.). Annual Spring Meeting of the American Ethnological Society (pp. 12-44): University of
Washington Press.
Lupo, G., & Velicogna, M. (2018). Making EU justice smart? Looking into the implementation of new
technologies to improve the efficiency of cross border justice services delivery. Smart
Technologies for Smart Governments (pp. 95-121). Cham: Springer.

Leite et al. / Visual Stories in Judicial Systems
Twenty-Seventh European Conference on Information Systems (ECIS2019), Stockholm-Uppsala, Sweden. 14
Medhi, I., Menon, S., Cutrell, E., & Toyama, K. (2010). Beyond strict illiteracy: abstracted learning
among low-literate users. Paper presented at the Proceedings of the 4th ACM/IEEE
International Conference on Information and Communication Technologies and Development,
London, UK.
Pennington, N., & Hastie, R. (1992). Explaining the evidence: Tests of the Story Model for juror
decision making. Journal of Personality and Social Psychology, 62(2), 189-206.
Procopiuck, M. (2018). Information technology and time of judgment in specialized courts: What is the
impact of changing from physical to electronic processing? Government Information Quarterly,
35(3), 491-501.
Quesenbery, W., & Brooks, K. (2010). Storytelling for User Experience. Crafting Stories for Better
Design. Brooklyn, NY: Rosenfeld Media.
Rhode, D. (2015-2016). What we know and need to know about the delivery of legal services by
nonlawyers. SCL Review, 67(329), 429-442.
Ricoeur, P. (1991). Life in Quest of Narrative. In D. Wood (Ed.). On Paul Ricoeur: Narrative and
Interpretation (Vol. Warwick studies in philosophy and literature, pp. 20-33). New York:
Routledge.
Sakamoto, R., Sumi, Y., & Kogure, K. (2007). Hyperlinked Comic Strips for Sharing Personal Contexts.
International Journal of Information Technology & Decision Making, 06(03), 443-458.
doi:doi:10.1142/S0219622007002563
Sarbin, T. (1986). Narrative Psychology. The Storied Nature of Human Conduct. Westport: Praeger
Publishers.
Schmid, W. (2010). Narratology - An Introduction. Berlin, New York: De Gruyter.
Simões, D., Antunes, P., & Carriço, L. (2018). Eliciting and Modelling Business Process Stories: A Case
Study. Business & Information Systems Engineering, 60(2), 115–132.
Simões, D., Antunes, P., & Cranefield, J. (2016). Enriching Knowledge in Business Process Modelling:
A Storytelling Approach. In L. Razmerita, G. Phillips-Wren, & L. Jain (Eds.). Innovations in
Knowledge Management: The impact of social media, semantic web and cloud computing (Vol.
95, pp. 241-267). Heidelberg: Springer.
Simões, D., Thuan, N., Jonnavithula, L., & Antunes, P. (2015). Modelling Sensible Business Processes.
In T. Dang, R. Wagner, J. Küng, N. Thoai, M. Takizawa, & E. Neuhold (Eds.). 2nd International
Conference on Future Data and Security Engineering (FDSE). Ho Chi Minh City, Vietnam (Vol.
9446). Heidelberg: Springer.
Singh, M., Sahu, G., Dwivedi, Y., Rana, N., & Tamilmani, K. (2018). Success Factors for e-Court
Implementation at Allahabad High-Court. Paper presented at the Twenty-Second Pacific Asia
Conference on Information Systems, Japan.
Truong, K., Hayes, G., & Abowd, G. (2006). Storyboarding: an empirical determination of best
practices and effective guidelines. Paper presented at the Proceedings of the 6th conference on
Designing Interactive systems, University Park, PA, USA.
Williams, A., & Alspaugh, T. (2008, 9-9 Sept. 2008). Articulating Software Requirements Comic Book
Style. Paper presented at the Third International Workshop on Multimedia and Enjoyable
Requirements Engineering, Barcelona, Spain.
Wintgens, L. (2005). Making Sense of Coherence. The Level Theory of Coherence. In M. Moens & P.
Spyns (Eds.). Legal Knowledge and Information Systems (pp. 23-24). Amsterdam: IOS Press.