Available online at www.sciencedirect.com
ScienceDirect
Procedia Computer Science 00 (2019) 000–000
www.elsevier.com/locate/procedia
1877-0509 © 2019 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer-review under responsibility of the scientific committee of the 9th Annual International Conference on Biologically Inspired Cognitive
Architectures.
Postproceedings of the 9th Annual International Conference on Biologically Inspired Cognitive
Architectures, BICA 2018 (Ninth Annual Meeting of the BICA Society)
Computational Analysis of Gender Differences
in Coping with Extreme Stressful Emotions
S. Sahand Mohammadi Ziabari a
*
, Jan Treur a
a Vrije Universiteit Amsterdam, Amsterdam, The Netherlands
Abstract
In this paper a computational analysis is presented of differences between men and women in coping with extreme
emotions. This analysis is based on an adaptive temporal-causal network model. It takes into account the suppression
of connections between preparation states and sensory representations of action effects due to an extreme stressful
emotion. It is shown how this model can be used to represent the difference between males and females facing an
extreme emotion, thereby performing their own methods in coping with the extreme emotion, for males fight or flight
and for females tend-and-befriend.
© 2019 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer-review under responsibility of the scientific committee of the 9th Annual International Conference on Biologically Inspired
Cognitive Architectures.
Keywords: Adaptive Network; Rumination; Extreme Emotion; Gender
1. Introduction
When facing an extreme stressful emotion, generally speaking men and women have different strategies. More
specifically, men more often use a fight or flight strategy and women a tend and befriend strategy. Moreover, it has
been reported that women show a longer duration of rumination before a decision is made. In this paper these
* Corresponding author.
E-mail address: sahandmohammadiziabari@gmail.com, j.treur@vu.nl
2 Steve DiPaola et al. / Procedia Computer Science 00 (2019) 000–000
phenomena are analyzed based on an adaptive temporal causal network model that allows to display such gender
differences. The model takes into account suppression of connections between preparation states and sensory
representations of action effects due to an extreme stressful emotion, as described in [1].
Differences in gender in addressing stress and fear seem to start at an age as early as 9-12 year-old [2]. Up to one
third of these differences are estimated to go back to the genetic factors [5]. It is reported that females subscribe a
more massive violence to fear than males [3]. Epidemiological research implies that females are much more likely to
get anxiety disorders than males [3]. The neurological perspective in [4] shows that females have a weaker
Hypothalamic-pituitary-adrenal axis (HPAA) and autonomic reactivity than males.
In [5] it was found out that when females get stressed, the level of oxytocin will increase and it improves their
tendency of accompanying with others. In [6] it is contemplated that females have stronger feelings of worry whenever
they face threats. In [7] it is reported that females get a higher score on STAI (which is a cognitive and affective
describer of anxiety) than males. On the one hand, in [8] it is claimed that females have much more fear of physical
outcomes of anxiety, and on the other hand, in [9] it is reported that males have much more fear of the social outcomes
of anxiety. In [10] it is described that propensity to consider vague conditions as a threatening situation is an adaptive
method for females to maintain the safety of themselves and their offspring. In [11] it is noticed that females have
more activity in their cortex and orbitofrontal cortex in facing threats than males.
In [12] it is declared that males are more into individual problem solving and as such they tend to focus on coping
with emotion and anxiety in ways different from females. The more common male pattern of fight-or-flight response
has been declared in [13].
‘The sympathoadrenal system is a physiological connection between the sympathetic nervous
system and the adrenal medulla and is crucial in an organism’s physiological response to outside
stimuli. When the body receives sensory information, the sympathetic nervous system sends a signal
to preganglionic nerve fibers, which activate the adrenal medulla through acetylcholine. Once
activated, norepinephrine and epinephrine are released directly into the blood by postganglionic
nerve fibers where they act as the bodily mechanism for fight-or-flight response.’
Rumination is explained as thinking more, and more about consequences of a worriness symptom rather than
graceful outcomes [14]. For instance, when your teenagers are out late you may be worried that something bad happens
to them, when your parents are ill you may be worried about how they manage without your help. In [15] it is found
out that gender differences in rumination exist in a way that females are more likely to focus on a disturbing and
negative results of the anxiety and emotion and it is more difficult to control the negative emotions rather than males.
Rumination can be seen as a result of the suppression of connections that play a role in decision making in stressfull
situations [12]. This phase of suppression may last longer in females than in males; e.g., [16].
The paper is organized as follows. Section 2 describes the scenarios addressed. As a basis for the computational
analysis we use an adaptive causal network model which was presented in [17]. In Section 3 this adaptive temporal-
causal network model is defined and illustrated by simulation of an example scenario. In Section 4 the simulation
results of the model are discussed. Here we consider the empirical information presented in [16] as the basis for the
gender-related duration differences of rumination and for the specific decisions made.
2. The scenarios addressed
For the computational analysis, the following scenario is addressed here, based on the behaviors of psychiatric
nurses. Acute stress and threat are common stimuli occurring during working as a psychiatric nurse. The impact of
an effect of the extreme emotion on decision making on what to do is modeled. An important factor is to decide which
action should be chosen to prevent any incorrect behavior. Hebbian learning of connections that represent action effect
prediction is considered here as a role of adapting the decision making to experiences over time. The suppression of
the connections between the preparation state and sensory representation of predicted effect due to an extreme emotion
during a first phase is considered to allow better conditions, as a kind of reset, for renewed construction of connections
[1] after this phase.
Steve DiPaola et al. / Procedia Computer Science 00 (2019) 000–000 3
In this paper, two scenarios are considered based on the gender of the nurses facing extreme emotions coming from
psychiatric disorders of their patients. The differences in decision making about the actions between males and females
are addressed in [18]. In [18] it is declared that men are more efficient in coping with violence, threats or risks of
damage than females and male nurses achieve this capacity through their physical strength. Also, in [19] it is noted
that this efficiency comes from the male body in the sense that the physical ability of males leads to some form of
dominance in the situation.
Based on the description given before, in our model male nurses are considered as individuals who can cope with
the effect of the impact of the extreme emotion by using a fight or flight strategy to cope with the situation, where
females use another strategy named tend-and-befriend which is more related to socialization. The scenario consists of
two psychiatric nurses working in a mental hospital. Two different genders are considered here, male and female. The
male nurse confronts a patient who always does the disgusting things like making her room messy, fighting with other
individuals, etc. In this condition, one day the patient refuses to swallow her drugs and tries to take them out of her
mouth and start fighting with the nurse.
The male nurse first tries to force her to swallow her drugs but nothing happens and the behavior of the patient
does not change and this make her more aggressive, thereby causing extreme stress. Therefore, after a short period of
time the male nurse (based on predicted consequences of different actions) reconsiders his decision. As not much
rumination takes place, the amount of time that the male nurse uses to think about different consequences is less than
for the female nurse. After this short period of time, the male nurse finds a way to cope with this situation and tries to
convince the patient by calming down her and talking to her based on his knowledge about behaving disorder
individuals. In contrast, in the scenario the female nurse takes more time for rumination. After which she starts the
tend-and-befriend strategy and socialize with other nurses and disorder relatives to find a solution for the stressful
situation. Therefore, like male nurses, she starts to cope with the situation but with a different strategy.
3. The adaptive temporal-causal network model
First the Network-Oriented Modelling approach used to model this process is briefly explained. As discussed in
detail in [20], [21] (see also [22]) this approach is based on temporal-causal network models which can be represented
at two levels: by a conceptual representation and by a numerical representation. A conceptual representation of a
temporal-causal network model in the first place involves representing (mental) states and connections between them
that represent (causal) impacts of states on each other. The states are assumed to have (activation) levels that vary
over time. The following three notions form the defining part of a conceptual representation of a temporal-causal
network model:
Strength of a connection X,Y Each connection from a state X to a state Y has a connection weight X,Y representing
the strength of the connection, usually between 0 and 1, but sometimes also below 0 (negative impact).
Combining multiple impacts on a state cY(..) For each state (a reference to) a combination function cY(..) is chosen
to combine the causal impacts of other states on state Y.
Speed of change of a state Y For each state Y a speed factor Y is used to represent how fast a state is changing
upon causal impact.
Combination functions can have different forms, as there are many different approaches possible to address the
issue of combining multiple impacts. Therefore, the Network-Oriented Modelling approach based on temporal-causal
networks incorporates for each state, as a kind of label or parameter, a way to specify how multiple causal impacts on
this state are aggregated by some combination function. For this aggregation a number of standard combination
functions are available as options and a number of desirable properties of such combination functions have been
identified; see [21], Ch 2, Sections 2.6 and 2.7. In Fig. 1 the conceptual representation of the temporal-causal network
model used here is depicted. A brief explanation of the states used is shown in Table 1.
4 Steve DiPaola et al. / Procedia Computer Science 00 (2019) 000–000
Table 1. Conceptual explanation of the states in the model
Next, the elements of the conceptual representation shown in Fig. 1 are elucidated in some more detail. The state
srss depicts the sensory representation of stimulus s from the world. This sensory representation influences the
activation level of the preparation state of one or more of the three actions a1, a2, or a3.
Figure 1. Conceptual representation of the network model
X1
srss
Sensory representation of stimulus s
X2
fsee
Feeling state for extreme emotion ee
X3
srse1
Sensory representation of (predicted) action effect e1
X4
srse2
Sensory representation of (predicted) action effect e2
X5
srse3
Sensory representation of (predicted) action effect e3
X6
psa1
Preparation state for action a1
X7
psa2
Preparation state for action a2
X8
psa3
Preparation state for action a3
X9
cs1
Control state for timing of suppression of connections
X10
cs2
Control state for suppression of connections
Steve DiPaola et al. / Procedia Computer Science 00 (2019) 000–000 5
The sensory representation of the (predicted) effects of the preparation states (prediction) psa1, psa2, and psa3 of the
actions are represented by srse1, srse2, and srse3 respectively. In addition, fsee indicates the feeling state associated to
the extreme emotion. Control state cs2 stands for controlling suppression of connections from preparation state to
sensory representation of predicted effect of all three actions. The control state cs1 stands for a control state to limit
this suppression in time. The connection weights ωi in Fig.1 are as follows. All preparation states psa1, psa2, and psa3
have three incoming connections from (srss, srse1, psa2, psa3), (srss, srse2, psa1, psa2), and (srss, srse3, psa1, psa2), with
weights (ω1, ω3, ω9, ω12), (ω2, ω4, ω10, ω14), (ω21, ω5, ω11, ω13), respectively. The connection weights ω1, ω2, ω21 are for
eliciting actions a1, a2, a3 during the time that nurse A is working. Connection weights ω7, ω8, ω6 stand for the
connections for affecting the sensory representation of the feeling effect associated to actions a1, a2, a3. The negative
connection weights ω15, ω10, ω11, ω12, ω13, ω14 enable to specify that the three actions exclude each other. Connection
weight ω7, ω8, ω6 are the weights of the connections to predict the sensory representation of the feeling influence of
effect of the preparation state of actions a1, a2, a3, respectively. States srse1, srse2, and srse3 have two incoming
connections from (psa1, fsee), (psa2, fsee), and (psa3, fsee) with connection weights (ω7, ω15) and (ω8, ω16), (ω6, ω17),
respectively. The connections from cs2 to cs1 and vice versa for timing the suppression have weights ω18, ω19 and ω20
(negative) which means that after some period of time the suppression of the connections will be released by
suppressing cs2.
In our scenario, psa1 is considered as a preparation state for performing an action when a nurse does not face an
extreme emotion. Furthermore, psa2 is considered as a preparation state for performing an action after rumination for
a male nurse to cope with the stressful situation which is considered as a fight or flight pattern. Similarly, psa3 is
considered a preparation state for an action for female nurses which in this case is a tend-and-befriend action; they
use this method rather than psa2 which is more specialized for men.
The conceptual representation was transformed into a numerical representation based on [20], [21] as also shown
in [17].
4. Simulation of the scenarios
The first scenario considered here is for females. An example simulation of this scenario for a female nurse is
shown in Figure 2. Table 2 shows the connection weights used for females, where the values for ω6, ω7, ω8 are initial
values as these weights are adapted over time. The time step was t = 0.25. The scaling factors for the states with
more than one incoming connection are also depicted in Table 2.
Table 3. Connection weights and scaling factors for the example simulation for females
weight
2
3
4
5
8
9
6
7
8
19
20
21
value
6
0.25
1
1
1
(0) = 0.7
(0) = 0.75
8(0) = 0.6
-0.1
-0.1
-0.1
-0.1
-0.1
-0.1
-0.01
0.1
0.3
0.4
0.7
-0.9
0.4
The primary stimulus for performing one of the actions is the sensory representation state srss (also denoted by X1),
which has value 1 all the time. At the early time of working, there is a convenient condition in the mental hospital.
Therefore, as can be seen in the Fig. 2, fsee (also denoted by X2) has a low value, which, however strongly increases
after some time (around time point 30) when the disturbances develop. Sensory representation state srse1 (X3) shows
that up to that time the female nurse initially has a good feeling for the effect of action a1 for the convenient situation,
which strengthens the preparation psa1 (X6) for this action. However, after time point 30 when the disturbance develops
and the psychiatric patient starts aggressive behavior, this begins to change. Both the feeling srse1 for the action effect
and the preparation psa1 for the action a1 drop. Furthermore, after this time point control state cs2 (X10), which stands
for the control state for suppression of the connections, starts to go up but after some longer time the other control
state cs1 (X9) in turn begins to play a role in suppressing cs2. Thus, the actual suppression of the connection weights
ω6, ω7, ω8 mainly takes place between time points 48 and 182. During that time of rumination the connections 6, 7
and 8 are suppressed by control state cs2 which is depicted in the graphs of these two connection weights in Fig. 3.
state
X3
X4
X5
X6
X7
X8
X10
1
1
1.3
1.6
1.1
1.4
1.1
6 Steve DiPaola et al. / Procedia Computer Science 00 (2019) 000–000
After some time, the suppression is released, and by the Hebbian learning the female nurse reconsiders the decision
making. This is illustrated by increasing 8 (for the connection X8-X5) in Fig. 3, in contrast to the decrease of 6 and
7 (for connection X6-X3 and X7-X4), and by increased activation of preparation state psa3 (X8) in Fig. 2. Due to this,
now the preparation for action a3 is dominating (a tend-and-befriend type of action).
Figure 2. Simulation results of working under an extremely stressful condition: states for females
Figure 3. Simulation results for suppression and Hebbian learning
for 8 (for the connection X8-X5) and 6 and 7 (for connection X6-X3 and X7-X4) for females
The second scenario which is considered here is for males. An example simulation of this process for males is
shown in Figure 2. Table 3 shows the connection weights used for males, where the values for are initial values as
these weights are adapted over time. The time step was again t = 0.25. The scaling factors for the states with more
than one incoming connection are also depicted in Table 3. The primary stimulus for performing one of the actions is
the sensory representation state srss (also denoted by X1), which has value 1 all the time. At the early time of working,
there is a convenient condition in the mental hospital. Therefore, as can be seen in the Fig. 3, fsee (also denoted by X2)
has a low value, which, however strongly increases after some time (around time point 30) when the disturbances
start.
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
020 40 60 80 100 120 140 160 180 200 220
X1 X2 X3 X4 X5 X6 X7 X8 X10
0.0
0.2
0.4
0.6
0.8
1.0
1.2
020 40 60 80 100 120 140 160 180 200 220
X6_X3 X7_X4 X8_X5
Steve DiPaola et al. / Procedia Computer Science 00 (2019) 000–000 7
Table 3 Connection weights and scaling factors for the example simulation for males
weight
2
3
4
5
8
9
value
6
0.1
1
1
1
(0) = 0.6
(0) = 0.6
8(0) = 0.9
-0.1
-0.1
weight
6
7
8
19
20
21
value
-0.1
-0.1
-0.1
-0.1
-0.0001
0.6
0.1
0.4
0.7
-0.9
0.16
Sensory representation state srse1 (X3) shows that at that time the male nurse initially has a good feeling for the
effect of action a1 in that convenient situation, which strengthens the preparation psa1(X6) for this action. However,
after time point 30 when the disturbance occurs and psychiatric individual starts aggressive behavior, this begins to
change. Both the feeling srse1 for the action effect and the preparation psa1 for the action a1 drop. Furthermore, after
this time point control state cs2 (X10), which stands for the control state for suppression of the connections, starts to go
up but after some time the other control state cs1 (X9) in turn begins to play a role in suppressing cs2. Thus, the actual
suppression of the connections mainly takes place between time points 60 and 150. During that time due to the
rumination the connections 6, 7 and 8 are suppressed by control state cs2 which is depicted in the graphs of these
two connection weights in Fig. 4. After some time, the suppression is released, and by the Hebbian learning the male
nurse reconsiders the decision making. This is illustrated by an increased activation of preparation state psa2 (X7) in
Fig. 3 and by increasing 7 (for the connection X7-X4) in Fig. 5, in contrast to the decrease of 6 and 8 (for connection
X6-X3 and X8-X5). Due to this, now the action a2 is dominating (of a fight-or-flight type).
Figure 4. Simulation results of working under an extremely stressful condition: states for males
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
020 40 60 80 100 120 140 160 180 200 220
X1 X2 X3 X4 X5 X6 X7 X10 X8
state
X3
X4
X5
X6
X7
X8
X10
1
1
1.3
1.6
1.1
1.4
1.1
8 Steve DiPaola et al. / Procedia Computer Science 00 (2019) 000–000
Figure 5. Simulation results for suppression and Hebbian learning
for 8 (for the connection X8-X5) and 6 and 7 (for connection X6-X3 and X7-X4) for males
For better comparison the behavior between males and females the values between males and females for both the
connection weights and the speed factors have been shown in Tables 4 and 5. The most important difference which is
also considered in the scenario based on the difference in confronting the extreme emotion for females is the time
duration for rumination. This difference was achieved by a much lower speed factor for cs1 for females (0.035) which
than for males (0.5). This value makes the simulation correct based on the time of the rumination. This difference
relates to empirical data about gender differences in duration shown in [16]. They found out the differences in time
duration of rumination of males and females. The adolescence period of both males and females were simulated here.
They have shown that the rumination time duration ratio for males and females is (females=2.77 / males=1.90)
1.4578947368. Therefore, as we considered the rumination duration for males between 60 and 150 which the
rumination time will be 90, the rumination time for females should be considered as 1.4578947368 * 90 which will
be equal to 131.210526 which indeed is achieved by considering the speed factor 0.035 for control state cs1. This leads
to a rumination time duration from around 48 to 182. There is not any other difference in selecting speed factors for
males and females. In selecting connection weights for males and females there are some subtle differences which
have been considered based on the difference in the genders in choice of action under stress. There are such differences
among females and males as indicated in [12] in different periods of time of their lives. Moreover, due to the tendency
to perform different types of actions, the action psa2 has been considered for males and psa3 for females, the related
connections weights have been set accordingly. For example, the suppression of the state from feeling state to the
sensory representation state of the emotions srse1and srse2 should be considered differently (6, 7) because of having
different actions for males and females as discussed in the scenarios.
Table 5. Difference between connection weights and speed factors for the two genders
weight
2
3
4
5
8
9
6
7
8
19
20
21
Females
value
6
0.1
1
1
1
(0)=0.7
(0)=0.75
8(0)= 0.6
-0.5
-0.1
-0.1
-0.1
-0.1
-0.1
-0.01
0.1
0.3
0.4
0.7
-0.9
0.4
Males
value
0.6
0.25
1
1
1
(0)=0.6
(0)=0.6
8(0)=0.9
-0.15
-0.1
-0.1
-0.1
-0.1
-0.1
-0.0001
0.6
0.1
0.4
0.7
-0.9
0.16
0.0
0.2
0.4
0.6
0.8
1.0
1.2
020 40 60 80 100 120 140 160 180 200 220
X6_X3 X7_X4 X8_X5
States
X1
X2
X3
X4
X5
X6
X7
X8
X9
X10
Speed Factors Females
0
0.02
0.5
0.5
0.5
0.5
0.5
0.5
0.035
0.02
Speed Factors Males
0
0.02
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.02
Steve DiPaola et al. / Procedia Computer Science 00 (2019) 000–000 9
5. Conclusion
In this paper a computational analysis of the different genders, males and females, facing an extreme emotion was
discussed, in particular differences in rumination and decision making. An adaptive temporal-causal network model
was used as a basis [20], [21], [22]. The extreme emotion causes rumination in females more than in males. Due to
this the connections between preparation state for doing an action sensory representation of that preparation (which is
considered as a rumination is psychology) are suppressed by a control state to prevent doing any action for a period
of time until the suppression itself is suppressed by another control state. Males generally have their own methods to
cope with such a situation named ‘fight or flight’ which means confronting an extreme emotion or run away from it.
Females generally have their own policy called ‘tend-and-befriend’ which means they consult the tough situation with
others to find a better result in facing with the extreme emotion and acute stress.
A number of simulations were performed some of which were presented in the paper. Findings from Neuroscience
were taken into account in the design of the adaptive model. This literature reports findings for patterns under stress-
induced conditions as addressed from a computational perspective in the current paper. For validation of the model,
in an ideal world suitable numerical empirical data would be best. However, these were not found. In a non-ideal
world a second best option then is validation with respect to qualitative empirical information about observed patterns.
Such empirical information was available (e.g., [16]) and has been used to validate the model.
Also, a precise mathematical analysis [23] has been done to verify that behavior of our model is as expected. This
model can be used as the basis of a virtual agent model to get insight in such processes and to consider certain support
or treatment of individuals to handle rumination when they have to work in a stressful context condition like hospitals
for nurses and prevent some stress-related disorders that otherwise might develop. In further research, the combination
of this scenario and the adaptive temporal causal model in an extreme emotion condition which suppression of
connections happened between sensory representation and preparation state can be take into account.
References
[1] Sousa, N., Osborne, F.X. Almeida. (2012) “Disconnection and reconnection: the morphological basis of (mal)adaptation to stress.” Trends
in Neurosci 35(12): 742-51.
[2] Ollendick, T.H. (1983) “Reliability and validity of the Revised Fear Survey Schedule for Children (FSSC-R).” Behaviour Research and
Therapy 21(6): 685−692.
[3] Arrindell, W. A. & Luteijn, F. (2000) “Similarity between intimate partners for personality traits as related to individual levels of
satisfaction with life.” Personality and Individual Differences 28(4): 629-637.
[4] Kajantie E, Phillips DI. (2006). “The effects of sex and hormonal status on the physiological response to acute psychosocial stress.”
Psychoneuroendocrinology 31(2): 151-78.
[5] Kendler, K.S., Neale, M.C., Kessler, R.C., Heath, A.C., & Eaves, L.J. (1992). “The genetic epidemiology of phobias in women: The
interrelationship of agoraphobia, social phobia, situational phobia, and simple phobia.” Archives of General Psychiatry 49(4): 273−281.
[6] Craske, M.G. (2003) “The origins of phobias and anxiety disorders: Why more women than men.” Oxford, England: Elsevier Science.
[7] Endler, N.S., Hunt, J.M., & Rosenstein, A.J. (1962) “An S–R inventory of anxiousness.” Psychological Monographs: General and
Applied 76(17): 1-33.
[8] Stewart, S. H., Taylor, S., & Baker, J.M. (1997) “Gender differences in dimensions of anxiety sensitivity.” Journal of Anxiety Disorders
11: 179−200.
[9] Foot, M., & Koszycki, D. (2004) “Gender differences in anxiety-related traits in patients with panic disorder.” Depression and Anxiety
20(3): 123−130.
[10] Wood,W., & Eagly, A.H. (2002) “A cross-cultural analysis of the behavior of women and Men. Implications for the origins of sex
difference.” Psychological Bulletin 128: 699−727.
[11] McClure, E.B., Monk, C.S., Nelson, E.E., Zarahn, E., Leibenluft, E., Bilder, R.M., et al. (2004) “A developmental examination of gender
differences in brain engagement during evaluation of threat.” Biological Psychiatry 55(11): 1047−1055.
[12] McLean CP, Anderson ER. (2009) “Brave men and timid women? A review of the gender differences in fear and anxiety.” Clin Psychol
Rev. 29(6): 496-505.
[13] Christensen, N.J. (1991) “The biochemical assessment of sympathoadrenal activity in man.” Clinical Authonomic Research 1(2): 167-172.
[14] Nolen-Hoeksema, S., & Jackson, B. (2001) “Mediators of the gender difference in rumination.” Psychology of Women Quarterly 25(1):
37−47.
[15] Ewalds-Kvist SB, Hirvonen T, Kvist M, Lertola K, Niemelä P. (2005) “Depression, anxiety, hostility and hysterectomy.” J Psychosom
Obstel Gynaecol. 26(3): 193-204.
10 Steve DiPaola et al. / Procedia Computer Science 00 (2019) 000–000
[16] Birditt, S. & Fingerman, L. (2003) “Age and Gender difference in adult’s descriptions of emotional reactions to interpersonal problems.”
Journal of gerontology, Series B: Psychological Sciences 58B: 237-245.
[17] Treur, J., and Ziabari, S.S.M. (2018) “An Adaptive Temporal-Causal Network Model for Decision Making Under Acute Stress.” In:
Nguyen, N. T., Trawinski, B., Pimenidis, E. & Khan, Z. (eds.), Computational Collective Intelligence: Proc. of the 10th International
Conference ICCCI 2018. Springer/Verlag, vol. 2, p. 13-25 Lecture Notes in Computer Science 11056 Springer.
[18] Ewald-Kvist, B., Algotsson, M, Bergstrom, A., Lutzen, K. (2012) “Psychiatric Nurses’ Self-Rated Competence.” Issues in Mental Health
Nursing 33: 469-479.
[19] Kumpula, E., & Ekstrand, P. (2009) “Men and masculinity in forensic psychiatric care: An interview study concerning male nurses’
experiences of working with male caregivers and male patients.” Issues in Mental Health Nursing 30, 538–546.
[20] Treur, J. (2016) “Dynamic Modeling Based on a Temporal-Causal Network Modeling Approach.” Biologically Inspired Cognitive
Architectures 16: 131-168.
[21] Treur, J. (2016). “Network-Oriented Modeling: Addressing Complexity of Cognitive, Affective and Social Interactions.” Springer
Publishers.
[22] Treur, J. (2018) “The Ins and Outs of Network-Oriented Modeling: from Biological Networks and Mental Networks to Social Networks
and Beyond.” Transactions on Computational Collective Intelligence, Springer Publishers, in press. Paper for Keynote lecture at the 10th
International Conference on Computational Collective Intelligence, ICCCI'18.
[23] Treur, J. (2016) “Verification of temporal-causal network models by mathematical analysis.” Vietnam J Comput Sci 3: 207–221.
