Network-Oriented Modeling and Its Conceptual Foundations

Abstract

To address complexity of modeling the world’s processes, over the years in different scientific disciplines separation assumptions have been made to isolate parts of processes, and in some disciplines they have turned out quite useful. It can be questioned whether such assumptions are adequate to address complexity of integrated human mental and social processes and their interactions. In this paper it is discussed that a Network-Oriented Modeling perspective can be considered an alternative way to address complexity for modeling human and social processes.

Full text

1
Chapter 1
Network-Oriented Modelling
and its Conceptual Foundations
To address complexity of modelling the world’s processes, over the years different
strategies have been used. From these strategies isolation and separation assumptions are
quite common in all scientific disciplines and have often turned out very useful. They
traditionally serve as means to address the complexity of processes by some strong form of
decomposition. This also holds for classical disciplines such as Physics, where, for
example, for mechanical modelling for building construction only forces from objects on
earth are taken into account and not forces from all other objects in the universe that still do
have some effects as well. It is recognized that these distant effects from sun, moon, planets
and other objects do exist, but it assumed that they can be neglected. For such cases within
Physics such an isolation assumption may be a reasonable choice, but in how far is it
equally reasonable to address human complexity? Over the years within the Behavioural
and Social Sciences also a number of assumptions have been made in the sense that some
processes can be studied by considering them as separate or isolated phenomena. However,
within these human-directed sciences serious debates or disputes have occurred time and
time again on such a kind of assumptions. They essentially have the form of arguments pro
or con an assumption that some processes can be studied by considering them as separate
or isolated phenomena. Examples of such separation assumptions to address human
complexity concern:
 mind vs body
 cognition vs emotion
 individual processes vs collective processes
 nonadaptive processes vs adaptive processes
 temporal separation
Although within Physics isolation assumptions may be reasonable, it can be questioned
whether cognition can be studied while ignoring emotion, or mind while ignoring body, or
sensory processing in isolation from action preparation? Or, put more general, in how far
are these traditional means to address complexity by separation still applicable if the

2
complexity of human and social processes has to be addressed? Do we need to break with
such traditions to be able to make more substantial scientific progress in the area
addressing human and social behaviour? And, not unimportant, are there adequate
alternative strategies to address human complexity?
1.1 Addressing Human Complexity by Separation Assumptions
The position taken in this book is that indeed a number of the traditional habits followed in
order to address human complexity have to be broken to achieve more progress in scientific
development. Partly due to the strong development of Cognitive, Affective and Social
Neuroscience, in recent years for many of the issues mentioned above, a perspective in
which dynamics and integration are key elements has become more dominant: the
perspective in the direction of less separation and more interaction. Given this background,
for each of the issues listed above this will be discussed below in more detail. It will be
pointed out how in many cases separation assumptions as mentioned lead to some types of
discrepancies or paradoxes, and how as an alternative a dynamic and integrative approach
to address complexity can resolve such discrepancies and paradoxes.
Mind vs Body
Another isolation assumption that has some tradition is the assumption that the mind can be
studied in separation from the body. There has been debate about this since long ago.
Aristotle (350 BC) considered properties of ‘mind and desire’ as the source of motion of a
living being: he discusses how the occurrence of certain internal (mental) state properties
(desires) within the living being entail or cause the occurrence of an action in the external
world; see also (Nussbaum, 1978). Such internal state properties are sometimes called by
him ‘things in the soul’, ‘states of character’, or ‘moral states’. In that time such ‘things’
were not considered part of the physical world but of the ghost-like world indicated in this
case by ‘soul’. So, in this context the explanation that such a creature’s position gets
changed is that there is a state of the soul driving it. This assumes a separation between the
soul on the one hand, and the body within the physical world on the other hand. How such
nonphysical states can affect physical states remains unanswered. Over time, within
Philosophy of Mind this has been felt as a more and more pressing problem. The idea that
mental states can cause actions in the physical world is called mental causation (e.g., Kim,
1996, 1998). The problem with this is how exactly can nonphysical mental states cause
effects in the physical world, without any mechanism known for such an effect. Within
Philosophy of Mind a solution for this has been proposed in the form of a tight relation
between mental states and brain states. Then it is in fact not the mental state causing the

3
action, but the corresponding (physical) brain state. Due to this the separation is not
between the soul and the body, but between the brain and the body (Bickle, 1998; Kim,
1996, 1998).
However, this separation between brain and body also has been debated. More literature on
this from a wider perspective can be found, for example, in (Clark, 1998; Lakoff and
Johnson, 1999; Wilson, 2002). It is claimed that mind essentially is embodied: it cannot be
isolated from the body. One specific case illustrating how brain and body intensely work
together and form what is called an embodied mind is the causal path concerning feelings
and emotional responses. A classical view is that based on some sensory input, due to
internal processing emotions are felt, and based on this they are expressed in some
emotional response in the form of a body state, such as a face expression:
stimulus  sensory representation  felt emotion
 preparation for a body state  expressed emotion in body state
However, James (1884) claimed a different order in the causal chain (see also Damasio,
2010, pp. 114-116):
stimulus  sensory representation  preparation for a body state
 expressed emotion in body state  sensed body state
 representation of body state  felt emotion
The perspective of James assumes that a body loop via the expressed emotion is used to
generate a felt emotion by sensing the own body state. So, the body plays a crucial role in
the emergence of states of the brain and mind concerning emotions and feelings. Damasio
made a further step by introducing the possibility of an as-if body loop bypassing actually
expressed bodily changes (cf. Damasio, 1994, pp. 155-158; see also Damasio, 1999, pp.
79-80; Damasio, 2010):
stimulus  sensory representation  preparation for body state
 representation of body state  felt emotion
An as-if body loop describes a predictive internal simulation of the bodily processes,
without actually affecting the body, comparable to simulation in order to perform, for
example, prediction of action effects, mindreading or imagination; e.g., (Becker and Fuchs,
1985; Goldman, 2006; Hesslow, 1994, 2002, 2012). Damasio (1999, 2010) distinguishes an
emotion (or emotional response) from a feeling (or felt emotion). A brief survey of
Damasio’s ideas about emotion and feeling can be found in (Damasio, 2010, pp. 108-129).
According to this perspective emotions relate to actions, whereas feelings relate to
perceptions of own body states triggered by these actions:

4
‘… feelings are not a passive perception or a flash in time, especially not in the case of
feelings of joy and sorrow. For a while after an occasion of such feelings begins – for
seconds or for minutes – there is a dynamic engagement of the body, almost certainly in a
repeated fashion, and a subsequent dynamic variation of the perception. We perceive a
series of transitions. We sense an interplay, a give and take.’ (Damasio, 2003, pp. 91-92)
See further in Chapter 3, Section 3.2. This essentially shows a cyclic process involving both
mind and body that (for a constant environment) can lead to equilibrium states for both
emotional response (preparation) and feeling.
Cognition vs Emotion
Traditionally it is assumed that behaviour can be described in relation to cognitive states
such as beliefs and desires, while leaving affective states aside. The latter types of states
are considered as being part of a separate line of (affective) processes that produce their
own output, for example, in the sense of emotions and expressions of them. However, this
assumed separation between cognitive and affective processes has been questioned more
and more. Specific examples of questions about interactions between affective and
cognitive states are: how does desiring relate to feeling, and in how far do sensing and
believing relate to feeling? To assume that desiring can be described without involving
emotion already seems a kind of paradox, or at least a discrepancy with what humans
experience as desiring. Recent neurological findings suggest that this separation of
cognitive and affective processes indeed may not be a valid and fruitful way to go. For
example, Phelps (2006) states:
‘The mechanisms of emotion and cognition appear to be intertwined at all stages of
stimulus processing and their distinction can be difficult. (..) Adding the complexity
of emotion to the study of cognition can be daunting, but investigations of the neural
mechanisms underlying these behaviors can help clarify the structure and
mechanisms.’ (Phelps, 2006, pp. 46-47)
Here it is recognized that an assumption on isolating cognition from emotion is not
realistic, as far as the brain is concerned. Therefore models based on such an assumption
cannot be biologically plausible and may simply be not valid. Moreover, it is also
acknowledged that taking into account the intense interaction between emotion and
cognition ‘can be daunting’; to avoid this problem was a main reason for the isolation
assumption as a way to address complexity. However, Phelps (2006) also points at a way
out of this: use knowledge about the underlying neural mechanisms. In the past when there
was limited knowledge about the neural mechanisms this escape route was not available,

5
and therefore the isolation assumption may have made sense, although the validity of the
models based on that can be questioned. But, now neuroscience has shown a very strong
development, this provides new ways to get rid of the isolation assumption. Similar claims
about the intense interaction between emotion and cognition have been made by Pessoa
(2008). In experimental contexts different types of effects of affective states on cognitive
states have indeed been found; see, for example, (Eich, Kihlstrom, Bower, Forgas, and
Niedenthal, 2000; Forgas, Goldenberg, and Unkelbach, 2009; Winkielman, Niedenthal,
and Oberman, 2009; Frijda, Manstead, and Bem, 2000). Moreover, more specifically in the
rapidly developing area of cognitive neuroscience (e.g., Purves, Brannon, Cabeza, Huettel,
LaBar, Platt, and Woldorff, 2008; Gazzaniga, 2009) knowledge has been contributed on
mechanisms for the interaction and intertwining of affective and cognitive states and
processes (for example, involving emotion, mood, beliefs or memory); see, for example,
(Dolan, 2002; LaBar and Cabeza, 2006; Pessoa, 2008; Phelps, 2006; Storbeck and Clore,
2007).
Not only for desiring and believing the isolation assumption for cognition vs emotion is
questioned, but also for rational decision making. Traditionally, rationality and emotions
often have been considered each other’s enemies: decision making has often been
considered as a rational cognitive process in which emotions can only play a disturbing
role. In more recent times this has been questioned as well. For example, in (Loewenstein
and Lerner, 2003, p. 619) it is pointed at the positive functions served by emotions:
‘Throughout recorded human intellectual history there has been active debate about
the nature of the role of emotions or ‘passions’ in human behavior, with the dominant
view being that passions are a negative force in human behavior (…). By contrast,
some of the latest research has been characterized by a new appreciation of the
positive functions served by emotions’ (Loewenstein and Lerner, 2003, p. 619)
In particular, in decision making it may be questioned whether you can make an adequate
decision without feeling good about it. Decisions with bad feelings associated to them may
lack robustness. Many occasions may occur over time that trigger a temptation to change it
into a decision with a better associated feeling. So, human experience in rational decisions
and feelings about them is that they go in hand in hand and are not isolated. This indicates
another paradox or discrepancy between the isolation assumption and how real life is
experienced: emotions can be considered a vehicle for rationality (for more details, see
Chapter 6). A brief sketch of the alternative perspective is as follows. Decision making
usually considers a number of options for a choice to be made. Such a choice is often based
on some form of valuing of the options. In this valuing process emotions come in: some of

6
the options relate to a more positive feeling than other options. It has been found that such
valuations relate to amygdala activations (see, e.g., Morrison and Salzman, 2010; Murray,
2007; Salzman and Fusi, 2010). As valuing can be seen as a grounding for a decision, it
turns out that an emotional type of grounding is involved. Bad decisions are those that are
not solidly grounded by having a positive feeling about them. They may not last long, as
any opportunity to get rid of them will be a temptation to reconsider the decisions. Recent
neurological literature addressing this idea of emotional valuing and grounding of decisions
relates the notion of value to the amygdala; e.g., (Bechara, Damasio, and Damasio, 2003;
Bechara, Damasio, Damasio, and Lee, 1999; Montague and Berns, 2002; Janak and Tye,
2015; Jenison, Rangel, Oya, Kawasaki, and Howard, 2011; Morrison and Salzman, 2010;
Ousdal, Specht, Server, Andreassen, Dolan, Jensen, 2014; Pessoa, 2011; Rangel, Camerer,
Montague, 2008).
In Chapter 3 it is discussed how knowledge from Neuroscience can be used to find out how
the integration of emotions and cognitive processes can be modelled, illustrated for a
number of examples. In Chapters 4 and 5, more specifically the role of emotions in
generating dreams and learning during dreaming is discussed. In Chapter 6 the specific
case of emotions as a basis for rational decision making is addressed in more detail.
Individual vs Collective
Yet another isolation assumption concerns the distinction between processes within an
individual and social processes. The former are usually referred to the territory of
Psychology, whereas the latter are referred to the territory of Social Science. The idea then
is to study social processes as patterns emerging from interactions between individuals
thereby abstracting from the processes within each of the individuals. This easily leads to
some kind of paradoxes. For example, as persons in a group are autonomous individuals
with their own neurological structures and patterns, carrying, for example, their own
emotions, beliefs, desires and intentions, it would be reasonable to expect that it is very
difficult or even impossible to achieve forms of sharedness and collectiveness. However, it
can be observed that often groups develop coherent views and decisions, and, even more
surprisingly, the group members seem to share a positive feeling about it. In recent years
by developments in neuroscience new light has been shed on this seeming paradox of
individuality versus sharedness and collectiveness. This has led to the new discipline called
Social Neuroscience; e.g., (Cacioppo and Berntson, 2005; Cacioppo, Visser, and Pickett,
2006; Decety and Cacioppo, 2010; Decety and Ickes, 2009; Harmon-Jones and
Winkielman, 2007). Two interrelated core concepts in this discipline are mirror neurons
and internal simulation of another person’s mental processes. Mirror neurons are neurons

7
that not only have the function to prepare for a certain action or body change, but are also
activated upon observing somebody else who is performing this action or body change;
e.g., (Iacoboni, 2008; Pineda, 2009; Rizzolatti and Sinigaglia; 2008). Internal simulation is
internal mental processing that copies processes that may take place externally, for
example, in mental processes in another individual; e.g., (Damasio, 1994; Damasio, 1999;
Gallese and Goldman; 1998; Goldman, 2006; Hesslow, 1994, 2002, 2012). Mechanisms
involving these core concepts have been described that provide an explanation of the
emergence of sharedness and collectiveness from a biological perspective. This new
perspective breaks the originally assumed separation between processes within individuals
and processes of social interaction.
Temporal Separation
Another traditionally made separation assumption is that processes in the brain are
separated in time. For example, sensing, sensory processing, preparation for action and
action execution are assumed to occur in linearly ordered temporal sequences:
sensing → sensory processing → preparing for action → executing action
For the case of emotions it was already discussed that such linear patterns are not
applicable. But also more in general it can be argued that such linear patterns are too much
of a simplification, as in reality more cyclic patterns occur; often a form of internal
simulation takes place, as put forward, among others, by (Hesslow, 1994, 2002, 2012;
Damasio, 1994, 1999; Goldman, 2006; Barsalou, 2009; Marques and Holland, 2009;
Pezzulo, Candidi, Dindo, and Barca, 2013). The general idea of internal simulation that
was also mentioned above in the specific context of emotions, is that sensory representation
states are activated (e.g., mental images), which in response trigger associated preparation
states for actions, which, by prediction links, in turn activate other sensory representation
states for the predicted effects of the prepared actions:
sensory representation states  preparation states  sensory representation states
The latter states represent the effects of the prepared actions or bodily changes, without
actually having executed them. Being inherently cyclic, the simulation process can go on
indefinitely, as the latter sensory representations can again trigger preparations for actions,
and so on. Internal simulation has been used, for example, to describe (imagined) processes
in the external world, e.g., prediction of effects of own actions (Becker and Fuchs, 1985),
or processes in another person’s mind, e.g., emotion recognition or mindreading (Goldman,
2006) or (as discussed above) processes in a person’s own body by as-if body loops

8
(Damasio, 1994). This breaks with the tradition that there is a temporal separation of
processes such as sensing – internal processing – preparation for action.
Adaptive vs Nonadaptive Processes
Another assumption that sometimes is debated is that mental and social processes are not
adaptive. In reality processes usually have adaptive elements incorporated, but often these
elements are neglected and sometimes studied as separate phenomena. One example in
social networks is the following. Often a contagion principle in networks is studied,
describing how states of connected nodes affect each other, whereas the network does not
change over time. But in reality the networks themselves also change, for example based
on what is called the homophily principle: the more you are alike, the more you like (each
other); for example, see (Byrne, 1986; McPherson, Smith-Lovin, and Cook, 2001; Mislove,
Viswanath, Gummadi, Druschel, 2010). Another way of formulating this principle is: birds
of a feather flock together. It can often be observed that persons that have close
relationships or friendships are alike in some respects. For example, they go to the same
clubs, watch the same movies or TV programs, take the same drinks, have the same
opinions, vote for the same or similar parties. Such observations might be considered
support for the contagion principle: they were together and due to that they affected each
other’s states by social contagion, and therefore they became alike. However, also a
different explanation is possible based on the homophily principle: in the past they were
attracted to each other due to being alike. So, the cyclic relation between the states of the
members and the strength of their connection leads to two possible causal explanations of
being alike and being connected:
being connected  being alike (contagion principle)
being alike  being connected (homophily principle)
Such circular causal relations make it difficult to determine what came first. It may be a
state just emerging from a cyclic process without a single cause. For more discussion on
this issue, for example, see (Aral, Muchnik, and Sundararajan, 2009; Shalizi and Thomas,
2011; Steglich, Snijders, Pearson, 2010; Mundt, Mercken, Zakletskaia, 2012). This
phenomenon will be addressed in more detail in Chapter 11.
As another example illustrating how adaptivity occurs fully integrated with the other
processes, the function of dreaming is discussed. Usually dreaming is considered a form of
internal simulation of real-life-like processes serving as training in order to learn or adapt
certain capabilities. Dreaming makes use of memory elements for sensory representations
(mental images) and their associated emotions (learnt in the past) to generate ‘virtual

9
simulations’; e.g., (Levin and Nielsen, 2007, pp. 499-500). Taking into account fear
emotions that often play an important role in dreams, strengthening of regulation of such
emotions is considered an important purpose of dreaming; see, for example, (Levin and
Nielsen, 2007; Walker and van der Helm, 2009; van der Helm, Yao, Dutt, Rao, Saletin, &
Walker, 2011; Gujar, McDonald, Nishida, and Walker, 2011; Deliens, Gilson, Peigneux,
2014; Pace-Schott, Germain, Milad, 2015; Sotres-Bayon, Bush, and LeDoux, 2004). To
this end in dreams adequate exercising material is needed: sensory representations of
emotion-loaden situations are activated, built on memory elements suitable for high levels
of arousal. The basis of what is called ‘fear extinction learning’ is that emotion regulation
mechanisms are available which are adaptive: they are strengthened over time when they
are intensively used. Fear extinction learning as an expression may sound a bit paradoxal; it
is not a form of unlearning or extinction of acquired fear associations, but it is additional
learning of fear inhibition connections in order to counterbalance the fear associations
which themselves remain intact (e.g., Levin and Nielsen, 2007, p. 507). Such a
strengthening of connections can be described by a Hebbian learning principle (Hebb,
1949); see also Chapter 2, Section 2.10. The processes of dreaming and the adaptive
elements involved in it are addressed in Chapters 4 and 5.
1.2 Addressing Complexity by Interaction in Networks Instead of by Separation
The separation assumptions to address complexity as discussed in Section 1.1 are strongly
debated as they all come with shortcomings. In this section it is discussed that in fact the
problem is not so much in the specific separation assumptions, but in the general idea of
separation itself. In social networks it is clear that the intense interaction between members
of the network based on their mutual and often interrelated cyclic connections, makes them
not very well suitable for any separation assumptions. And this does not only apply to
social processes but also to individual mental processes, as will be discussed in some more
detail here.
In the domain of neuroscience the structures and mechanisms found suggest that many
parts in the brain are connected by connections that often are part of cyclic paths, and such
cycles are assumed to play an important role in many mental processes (e.g., Bell, 1999;
Crick and Koch, 1998; Potter, 2007). As an example also put forward above, there is a
growing awareness, fed by findings in neuroscience that emotions play an important
mediating role in most human processes, and this role often provides a constructive
contribution, and not a disturbing contribution as was sometimes assumed. Usually mental
states trigger emotions and these emotions in turn affect these and other mental states. It

10
turns out that to address this type of circular effects, different views on causality and
modeling are required, compared to the traditional views in modeling of mental processes.
For example, Scherer (2009) states:
‘What is the role of causality in the mechanisms suggested here? Because of the
constant recursivity of the process, the widespread notion of linear causality (a
single cause for a single effect) cannot be applied to these mechanisms. Appraisal is
a process with constantly changing results over very short periods of time and, in
turn, constantly changing driving effects on subsystem synchronization (and,
consequently, on the type of emotion). (…) Thus, as is generally the case in self-
organizing systems, there is no simple, unidirectional sense of causality (see also
Lewis 1996).’ (Scherer, 2009, p. 3470)
Also in the domain of Philosophy of Mind this issue of cyclic causal connections is
recognized, for example, by Kim (1996). The idea is that a mental state is characterized by
the way it mediates between the input it receives from other states and the output it
provides to other states; this is also called the functional or causal role of the mental state.
For example, as a simplified example on the input side a mental state of being in pain is
typically caused by tissue damage and in turn on the output side it typically causes winces,
groans and escape behavior (Kim, 1996, p. 104).
Kim describes the overall picture as follows:
‘Mental events are conceived as nodes in a complex causal network that engages in
causal transactions with the outside world by receiving sensory inputs and emitting
behavioral outputs’ (Kim, 1996, p. 104)
For example, as a simplified example on the input side a mental state of being in pain is
typically caused by tissue damage and in turn on the output side it typically causes winces,
groans and escape behavior (Kim, 1996, p. 104); see Fig. 1.
Figure 1 Pain state with some of its causal relations
pain state
winces
groans
escaping
output:
impacts of
pain state
tissue damage
input:
impacts on
pain state

11
As input not only sensory input can play a role but also input from other mental states such
as in the pain example ‘being alert’. Similarly, as output not only behavioral output can
play a role but also other mental states can be affected, such as in the pain example feeling
distress and a desire to be relieved of it. Within Philosophy of Mind this is often considered
challenging:
‘But this seems to involve us in a regress or circularity: to explain what a given
mental state is, we need to refer to other mental states, and explaining these can
only be expected to require reference to further mental states, on so on – a process
that can go on in an unending regress, or loop back in a circle’ (Kim, 1996, pp.
104-105)
In Fig. 2 a fragment of such a cyclic causal network is depicted. Here mental state S1 has a
causal impact on mental state S2, but one of the states on which S2 has an effect, in turn
affects one of the input states for S1.
Figure 2 Mental states with their causal relations:
nodes in a complex, often cyclic causal network (see also Kim, 1996, p. 104)
This view from Philosophy of Mind is another indication that a modeling approach will
have to address networks with cycles well. To obtain an adequate understanding of such
cycles it is useful to take into account the temporal dimension of the dynamics of the
processes effectuated by the causal relations. In principle, this situation makes that an
endless cyclic process over time emerges. Note that in such a graph at each point in time
activity takes place in every state simultaneously (it is not that one state waits for the
other). The notion of state at some point in time used here refers to a specific part or aspect
of the overall state of a model at this point in time. Such an overall state can include, for
example, at the same time a ‘being in pain’ state, a ‘desire to get relieved’ state, and an
‘intention to escape’ state. The overall state at some point in time is the collection of all
state S2
input:
impacts on
state S2
output:
impacts
of state S1
output:
impacts
of state S2
loop back
state S1
input:
impacts
on state S1

12
states at that point in time. All the time the processes in the brain occur in parallel, in
principle involving all specific states within the overall state, mostly in an unconscious
manner. In this sense the brain is not different from any other part of the universe where
everywhere processes take place simultaneously, in parallel. During all this parallel
processing, any change in state S1 in principle will lead to a change in state S2, which in
turn will lead to another change in state S1, which leads to another change in state S2, and
so on and on. The state changes in such a process may become smaller and smaller over
time, and the cyclic process eventually may converge to an equilibrium state in which no
further changes occur anymore; but also other patterns are possible, such as limit cycles in
which the changes eventually end up in a regular, periodic pattern of changes (see also
Chapter 12). In this sense mental processes can show patterns similar to patterns occurring
in social networks, where cycles of connections are natural and quite common.
As intense interaction in networks as a way of modelling requires a dynamic, temporal
perspective, this will be discussed next.
1.3 The Dynamic Computational Modelling Perspective
The challenge to cope with a dynamical and cyclic picture of mental processes, imposes
certain requirements on a modeling approach. The modeling approach has to be able to
handle time and dynamics well. For example, in (van Gelder and Port, 1995) the symbolic
computational perspective is criticized as being not able to address the time-context of
cognitive processes in an adequate manner. In contrast they propose a perspective in which
cognition is considered as dynamics:
‘The alternative, then, is the dynamical approach. Its core is the application
of the mathematical tools of dynamics to the study of cognition. (…) But
the dynamical approach is more than just powerful tools; like the
computational approach it is a worldview. The cognitive system is not a
computer, it is a dynamical system. (…) The cognitive system is not a
discrete sequential manipulation of static representational structures; rather,
it is a structure of mutually and simultaneously influencing change.’ (van
Gelder and Port, 1995, p. 3).
They compare the dynamical perspective to the symbolic computational perspective as
described by Newell and Simon (1976)’s Physical Symbol System Hypothesis:
‘According to this hypothesis, natural cognitive systems are intelligent by
virtue of being physical symbol systems of the right kind. At this same

13
level of generality, dynamicists can be seen as embracing the Dynamical
Hypothesis: Natural cognitive systems are dynamical systems, and are best
understood from the perspective of dynamics. Like its computational
counterpart, the Dynamical Hypothesis forms a general framework within
which detailed theories of particular aspects of cognition can be
constructed’ (van Gelder and Port, 1995, p. 5)
It has taken a number of years before the dynamical perspective was adopted more
substantially in practical cognitive and neuroscientific modelling work; see for example:
‘Although the idea of applying dynamical systems theory to the study of
neural and cognitive mechanisms has been around for at least two decades
(Beer, 2000; Kelso, 1995; Thelen and Smith, 1994; van Gelder, 1998), the
dynamical systems approach has only recently begun to figure prominently
in neuroscience (…).’ (Schurger and Uithol , 2015)
The notion of state-determined system, adopted from Ashby (1960) is taken by van Gelder
and Port (1995) as a definition of what a dynamical system is:
‘A system is state-determined only when its current state always determines a
unique future behaviour. Three features of such systems are worth noting.
First, in such systems, the future behaviour cannot depend in any way on
whatever states the system might have been in before the current state. In other
words, past history is irrelevant (or at least, past history only makes a difference
insofar as it has left an effect on the current state).
Second, the fact that the current state determines future behaviour implies the
existence of some rule of evolution describing the behaviour of the system as a
function of its current state. For systems we wish to understand we always hope
that this rule can be specified in some reasonable succinct and useful fashion. One
source of constant inspiration, of course, has been Newton’s formulation of the
laws governing the solar system.
Third, the fact that future behaviours are uniquely determined means that state
space sequences can never fork.’ (van Gelder and Port, 1995), p. 6.
Ashby (1960) emphasizes the importance of the identification of state-determined systems
in a wide variety of scientific domains; for more details on this notion, see Section 2.8
below. This perspective on mental systems as state-determined dynamical systems put
forward by Ashby (1960) and van Gelder and Port (1995) can be viewed as a further
extension of the world view for the universe as developed much earlier, for example, by
Descartes. As also discussed in (Treur, 2007, Sections 2.1 and 2.2, pp. 58-59), Descartes

14
(1634) introduced a perspective on the world that sometimes is called the clockwork
universe. This perspective claims that with sufficiently precise understanding of the
world’s dynamics at some starting time, the future can be predicted just by applying a set
of ‘laws of nature’. He first describes how at some starting time matter came into existence
in a diversity of form, size, and motion. From that time on, dynamics continues according
to these laws of nature.
‘From the first instant that they are created, He makes some begin to move in one
direction and others in another, some faster and others slower (or indeed, if you
wish, not at all); thereafter, He makes them continue their motion according to the
ordinary laws of nature. For God has so wondrously established these laws that,
even if we suppose that He creates nothing more than what I have said, and even if
He does not impose any order or proportion on it but makes of it the most confused
and most disordered chaos that the poets could describe, the laws are sufficient to
make the parts of that chaos untangle themselves and arrange themselves in such
right order that they will have the form of a most perfect world, in which one will
be able to see not only light, but also all the other things, both general and
particular, that appear in this true world.’ (Descartes, 1634, Ch 6: Description of a
New World, and on the Qualities of the Matter of Which it is Composed )
Descartes emphasizes that after such a starting time nothing (even no God) except the laws
of nature determines the world’s dynamics:
‘Know, then, first that by "nature" I do not here mean some deity or other sort of
imaginary power. Rather, I use that word to signify matter itself, insofar as I
consider it taken together with all the qualities that I have attributed to it, and under
the condition that God continues to preserve it in the same way that He created it.
For from that alone (i.e., that He continues thus to preserve it) it follows of
necessity that there may be many changes in its parts that cannot, it seems to me, be
properly attributed to the action of God (because that action does not change) and
hence are to be attributed to nature. The rules according to which these changes
take place I call the "laws of nature".’ (Descartes, 1634, Ch 7: On the Laws of
Nature of this New World)
This view on the world’s dynamics is often compared to a clockwork. The view assumes
that systematic relationships (laws of nature) are possible between world states over time,
in the sense that (properties of) past world states entail (properties of) future world states.
The clockwork universe view has been developed further by Newton, Leibniz, Laplace and
others. The following quotation taken from Laplace (1825) sketches how an intellect could

15
be able to determine (by means of ‘a single formula’) future world states from a present
world state, that by itself is the effect of past world states:
‘We may regard the present state of the universe as the effect of its past and the
cause of its future. An intellect which at any given moment knew all of the forces
that animate nature and the mutual positions of the beings that compose it, if this
intellect were vast enough to submit the data to analysis, could condense into a
single formula the movement of the greatest bodies of the universe and that of the
lightest atom; for such an intellect nothing could be uncertain and the future just
like the past would be present before its eyes.’ (Laplace, 1825)
The worldview of Descartes and others described above in principle focuses on the
physical universe. As such it applies to all physical and also biological processes in the
universe, for example, those in the brain. The dynamical perspective on cognition put
forward by Ashby (1960) and van Gelder and Port (1995) can be viewed as an extension of
the above worldview from the physical world to the mental world. As this dynamical
worldview already is assumed to apply to the physical processes in the brain, it is an
advantage that also assuming such a worldview for mental processes will make it easier to
relate mental and neural processes, as discussed in Chapter 2, Section 2.3.
1.4 Network-Oriented Modelling
This chapter started in Sections 1.1 and 1.2 by some reflection on traditional means to
address complexity by assuming separation and isolation of processes, and the
shortcomings, discrepancies and paradoxes entailed by these assumptions. In Section 1.2
the circular or cyclic and distributed character of many processes (with interacting sub-
processes running in parallel) was identified as an important challenge to be addressed, and
it was recognized that a temporal or dynamic perspective is needed for this. In Section 1.3
such a dynamic perspective was discussed in some more detail. However, it was not
discussed in how far a dynamical perspective can also be used to address complexity,
which was the problem considered from the start. Will a dynamical modelling perspective
not translate the complexity of the world into similar complexity within models,
represented as long lists of difference or differential equations? How can such complex
models be handled conceptually, mathematically and computationally? It is easy to criticize
certain strategies to address complexity, but if alternatives also have shortcomings in the
way they address complexity, such critics will not have much impact.
In this section a network-oriented modelling perspective is proposed as an alternative way
to address complexity. This perspective takes the concept of network as a basis for

16
conceptualization and structuring of any complex processes. To be clear, network-oriented
modelling is not modelling of (given) networks, but modelling any (complex) processes as
networks, or by networks. It is useful to keep in mind that the concept network is a just a
mental concept and this is used as a conceptual structuring tool to conceptualize any
processes that exist in reality. The concept of network is easy to visualize on paper, on a
screen or mentally and as such provides a good support for intuition behind a model.
Moreover, as the network-oriented modelling approach presented here (see Section 1.5)
also incorporates a temporal dimension enabling interpretation of connections as temporal-
causal connections, the mental concept of causal graph also provides support for the
intuition behind dynamical models.
The scientific area of networks has already a longer tradition, of more than 40 years,
starting for social contexts within the Social Sciences, but it has developed further and
within many other disciplines, such as Biology, Neuroscience, Mathematics, Physics,
Economics, Informatics or Computer Science, Artificial Intelligence, and Web Science;
see, for example (Boccalettia, Latorab, Morenod, Chavez, Hwanga, 2006; Valente, 2010;
Giles, 2012). These developments already show how processes in quite different domains
can be conceptualized as networks. Within such literature often graphical representations of
networks are used as a means of presentation. After getting accustomed to such
conceptualisations of processes that exist in the real world as networks, a belief may occur
that these networks actually exist in reality (as neural networks, or computer networks, or
social networks, for example), and modelling as networks could be phrased alternatively as
modelling networks. But still the concept ‘network’ is a mental concept used as a tool to
conceptualise any type of processes. To make this distinction more clear linguistically, the
phrase network-oriented modelling is a more adequate indication for modeling as networks.
Network-oriented modelling offers a conceptual tool to model complex processes in a
structured, intuitive and easily visualisable manner, and the approach described here also
incorporates the dynamics of the processes in these models. Using this approach, different
parts of a process can be distinguished, but in contrast to the separation and isolation
strategy to address complexity, a network-oriented approach does not separate or isolate
these parts, but emphasizes and explicitly models the way how they are connected and
interact. Moreover, by adding a temporal dimension to incorporate a dynamic perspective,
it is explicitly modelled how they can have intense and circular causal interaction.

17
1.5 Network-Oriented Modelling: a Temporal-Causal Network Modelling Approach
As discussed above both the internal mental processing and the processing within social
networks involve multiple cyclic processes. This has implications for the type of modelling
approach to be used. The dynamic models considered have to integrate such cycles, and
also allow adaptive processes by which individuals can change their connections within
such cycles. Similarly, at the level of a social network many cyclic and dynamic
connections occur. To model such dynamics, a dynamical modelling perspective is needed
that can handle such combinations of cycles. Given the cyclic character of networks, within
network-oriented modelling as discussed here the dynamic perspective has to be
incorporated as well: a temporal dimension is indispensable. This is what has been
achieved in the temporal-causal modelling approach described in Chapter 2; see also
(Treur, 2016).
Temporal-causal network models can be represented at two levels: by a conceptual
representation and by a numerical representation. These models can be used not only to
display interesting network pictures, but also for numerical simulation. Furthermore, they
can be analysed mathematically and validated by comparing their simulation results to
empirical data. Moreover, they usually include a number of parameters for domain, person,
or social context specific characteristics. To estimate values for such parameters, a number
of parameter tuning methods are available.
The temporal-causal network modelling approach is a generic, dynamic AI modelling
approach based on networks of causal relations (e.g., Kuipers and Kassirer, 1983; Kuipers,
1984; Pearl, 2000), but that in addition incorporates a continuous time dimension to model
dynamics. As discussed above, this temporal dimension enables causal reasoning and
simulation for cyclic causal graphs or networks that usually inherently contain cycles, such
as networks of mental or brain states, or social networks. The modelling approach can
incorporate ingredients from different areas, for example, ingredients that are sometimes
used in specific types of (continuous time, recurrent) neural network models, and
ingredients that are sometimes used in probabilistic or possibilistic modelling. It is more
generic than such methods in the sense that a much wider variety of modelling elements are
provided, enabling the modelling of many types of dynamical systems, as described in
(Chapter 2, Section 2.9).
A description of a conceptual representation of a dynamical model in the first place
involves representing in a declarative manner states and connections between them that
represent (causal) impacts of states on each other, as assumed to hold for the application

18
domain addressed. What else is needed to describe processes in which causal relations play
their role? In reality not all causal relations are equally strong, so some notion of strength
of a connection is needed. Furthermore, when more than one causal relation affects a state,
in which manner do these causal effects combine? So, some way to aggregate multiple
causal impacts on a state is needed. Moreover, not every state has the same extent of
flexibility; some states may be able to change fast, and other states may be more rigid and
may change more slowly. Therefore, a notion of speed of change of a state is used. These
three notions are covered by elements in the temporal-causal network modelling approach,
and are 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 weight value X,Y representing
the strength of the connection, often between 0 and 1, but sometimes also below 0
(negative effect) or above 1.
 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.
How exactly does one impact on a given state add to another impact? In other words, what
types of combination functions can be considered? The more general issue of how to
combine multiple impacts or multiple sources of knowledge occurs in various forms in
different areas, such as the areas addressing imperfect reasoning or reasoning with
uncertainty or vagueness. For example, in a probabilistic setting, for modelling multiple
causal impacts on a state often independence of these impacts is assumed, and a product
rule is used for the combined effect; e.g., (Dubois and Prade, 2002). In practical
applications, this assumption is often questionable, or difficult to validate. In the areas
addressing modelling of uncertainty also other combination rules are used, for example, in
possibilistic approaches minimum- or maximum-based rules are used; e.g., (Dubois and
Prade, 2002). In another different area, addressing neural network modelling yet another
way of combining effects is used often. In that area, for combination of the impacts of
multiple neurons on a given neuron usually a logistic sum function is used: adding the
multiple impacts and then applying a logistic function; e.g., (Grossberg, 1969; Hirsch,
1989; Hopfield, 1982, 1984; Beer, 1995).

19
So, there are many different approaches possible to address the issue of combining multiple
impacts. The applicability of a specific combination rule for this may depend much on the
type of application addressed, and even on the type of states within an application.
Therefore the temporal-causal network modelling method incorporates for each state as a
kind of parameter a way to specify how multiple causal impacts on this state are
aggregated. For this aggregation a number of standard combination functions are made
available as options and a number of desirable properties of such combination functions
have been identified (see Chapter 2, Sections 2.6 and 2.7), some of which are shown in
Table 1.
Table 1 Overview of some standard combination functions
name
description
formula c(V1, …, Vk) =
ssum(..)
Scaled sum
(V1 + … + Vk)/ with  > 0
product(..)
cproduct(..)
Product
Complement product
V1 * … * Vk
1 – (1 - V1) * …* (1 – Vk)
min(..)
max(..)
Minimal value
Maximal value
min(V1, …, Vk)
max(V1, …, Vk)
slogistic,(..)
Simple logistic sum
1/(1+ e - (V1+..+Vk -)) with ,  ≥ 0
alogistic,(..)
Advanced logistic sum
[(1/(1+e - (V1+..+Vk -)) - (1/(1+e))] (1+e -) with ,  ≥ 0
These options cover elements from different existing approaches, varying from approaches
considered for reasoning with uncertainty, probability, possibility or vagueness, to
approaches for recurrent neural networks; e.g., Dubois, Lang, and Prade, 1991; Dubois and
Prade, 2002; Giangiacomo, 2001; Zadeh, 1978, Grossberg, 1969; Hirsch, 1989; Hopfield,
1982, 1984; Beer, 1995). In addition, there is still the option to specify any other (non-
standard) combination function, preferably taking into account the desired properties.
A conceptual representation of temporal-causal network model can be transformed in a
systematic or even automated manner into a numerical representation of the model as
follows (Treur, 2016):
 at each time point t each state Y in the model has a real number value in the interval
[0, 1], denoted by Y(t)
 at each time point t each state X connected to state Y has an impact on Y defined as
impactX,Y(t) = X,Y X(t) where X,Y is the weight of the connection from X to Y
 The aggregated impact of multiple states Xi on Y at t is determined using a
combination function cY(..):
aggimpactY(t) = cY(impactX1,Y(t), …, impactXk,Y(t))= cY(X1,YX1(t), …, Xk,YXk(t))
where Xi are the states with connections to state Y

20
 The effect of aggimpactY(t) on Y is exerted over time gradually, depending on speed
factor Y:
Y(t+t) = Y(t) + Y [aggimpactY(t) - Y(t)] t
or dY(t)/dt = Y [aggimpactY(t) - Y(t)]
 Thus, the following difference and differential equation for Y are obtained:
Y(t+t) = Y(t) + Y [cY(X1,YX1(t), …, Xk,YXk(t)) - Y(t)] t
dY(t)/dt = Y [cY(X1,YX1(t), …, Xk,YXk(t)) - Y(t)]
Summarizing, the temporal-causal network modeling approach described here provides a
complex systems modeling approach that enables a modeler to design high level conceptual
model representations in the form of (cyclic) graphs or connection matrices, which can be
systematically transformed in an automated manner into executable numerical
representations that can be used to perform simulation experiments. The modeling
approach makes it easy to take into account on the one hand theories and findings about
complex networks from any domain from, for example, biological, psychological,
neurological or social sciences. This applies, among others, to mental processes based on
complex brain networks, which, for example, often involve dynamics based on interrelating
and adaptive cycles, but equally well it applies to social networks and their dynamics. This
enables to address complex phenomena such as the integration of emotions within all kinds
of cognitive processes, of internal simulation and mirroring of mental processes of others,
and dynamic social interaction patterns.
1.6 Application Areas and Achievements
Concerning the scope of applicability, it has been shown mathematically that any smooth
continuous state-determined system or any dynamical system described by a set of first
order differential equations can also be modeled by this temporal-causal network modeling
approach, by choosing suitable parameters such as connection weights, speed factors and
combination functions (see Chapter 2, Section 9). In this sense it is as general as modeling
approaches put forward, for example, in (Ashby, 1960; Forrester, 1973, 1987; Thelen and
Smith, 1994; Port and van Gelder, 1995; van Gelder and Port, 1995; Beer, 1995; Kelso,
1995; van Gelder, 1998), and other network-oriented approaches such as described, for
example in (Grossberg, 1969; Hopfield, 1982, 1984; Hirsch, 1989; Funahashi and
Nakamura, 1993).
To facilitate applications, dedicated software is available supporting the design of models
in a conceptual manner, automatically transforming them into an executable format and

21
performing simulation experiments. A variety of example models that have been designed
illustrates the applicability of the approach in more detail, for example, addressing
integration of phenomena such as:
 Embodiment, as-if body loops, mindfulness (Chapter 3)
 Imagination, visualisation and dreaming as internal simulation (Chapter 4)
 Mirroring of other minds (Chapter 7)
 Integration of affective and cognitive processes (Chapters 3, 6, 7, 10)
 Fear extinction learning (Chapter 5)
 Emotions as a basis for rationality (Chapter 6)
 Empathic understanding (Chapters 7, 9)
 Emergence of shared understanding and collective action (Chapter 7)
 Group processes and crowd behavior (Chapter 7)
 Prior and retrospective ownership of actions (Chapter 8)
 Social responsiveness (Chapter 9)
 Joint decisions (Chapter 10)
 Social contagion (Chapters 7, 11)
 Social network evolution (Chapter 11)
The topics addressed have a number of possible application areas. An example of such an
application is to analyse the spread of a healthy or unhealthy lifestyle in society. Another
example is to analyse crowd behaviour in emergency situations. A wider area of
application addresses socio-technical systems that consist of humans and devices, such as
smartphones, and use of social media. For such mixed groups, in addition to analysis of
what patterns may emerge, also for the support side the design of these devices and media
can be an important aim, in order to create a situation that the right types of patterns
emerge. This may concern, for example, safe evacuation in an emergency situation or
strengthening development of a healthy lifestyle. Other application areas may address, for
example, support and mediation in collective decision making and avoiding or resolving
conflicts that may develop.
1.7 Overview of the Book
Most of the material of the book has been used and positively evaluated by undergraduate
and graduate students and researchers in cognitive and AI domains. The book is composed
of six parts:

22
I. Introduction of the Modelling Approach
II. Emotions all the Way
III. The Others
IV. Analysis Methods
V. Philosophical, Societal and Educational Perspectives
Part I Introduction of the Approach
This part is the introduction to the book, both conceptually and in a more technical sense. It
consists of the current introduction Chapter 1, and a next Chapter 2 in which the temporal-
causal modeling perspective is introduced.
Part II Emotions all the Way
In Part II a number of models are discussed that address single agents and the way in which
emotions are integrated in practically all mental processes.
In Chapter 3 it is discussed how within cognitive, affective and social neuroscience more
and more mechanisms have been found that suggest how emotions relate in a bidirectional
manner to many other mental processes and behaviour. Based on this, in this chapter an
overview of neurologically inspired models for the dynamics and interaction of emotions is
discussed. Thus an integrative perspective is obtained that can be used to describe, for
example, how emotions relate to beliefs, experiences, decision making, and to emotions of
others, and also how emotions can be regulated. It is pointed out how integrated
computational models of such mental processes incorporating emotions can be obtained.
In Chapter 4 it is discussed how emotions play a role in generating dream episodes from a
perspective of internal simulation. Building blocks for this internal simulation are memory
elements in the form of sensory representations and their associated emotions. In the
presented model, under influence of associated feeling levels and mutual competition,
some sensory representation states pop up in different dream episodes. As a form of
emotion regulation the activation levels of both the feelings and the sensory representation
states are suppressed by control states. The model was evaluated by example simulation
experiments.
In Chapter 5 it is discussed how dreaming is used to learn fear extinction. Here fear
extinction has been found not to involve weakening of fear associations, as was assumed
longer ago, but instead it involves the strengthening of fear suppressing connections that

23
form a counter balance against the still persisting fear associations. So, to regulate fear
associations neural mechanisms are used that take care of strengthening these suppressing
connections, as a form of learning of emotion regulation. The presented model addresses
dreaming as internal simulation incorporating memory elements in the form of sensory
representations and their associated fear, as in Chapter 4. But this time it is modelled how
the regulation of fear that takes place during dream episodes, is strengthened. This
adaptation or learning process is modelled as Hebbian learning. The model was evaluated
by a number of simulation experiments for different scenarios.
Chapter 6 addresses the role of emotions in rational decision making. Traditionally it has
been assumed that emotions can only play a disturbing and non-rational role in decision
making. However, more recently it has been found that neurological mechanisms
involving emotions play an important role in rational decision making. In this chapter an
adaptive decision model based on predictive loops through feeling states is presented,
where the feeling states function in a process of valuing of decision options. Hebbian
learning is considered for different types of connections in the decision model. Moreover,
the model is analysed from the perspective of rationality. To assess the extent of rationality,
a measure is introduced reflecting what would be rational for a given environment’s
behaviour. Simulation results and the extents of rationality of the different models over
time are discussed. It is shown how during the adaptive process this model for decision
making achieves higher levels of rationality.
Part III The Others
Part III focuses on agents functioning in a social context. Given that ach agent has his or
her own beliefs, desires, intentions, emotions and still more mental states, it might be
expected that social coherence is not often achieved. However, the fact that still often
social coherence is observed presents a kind of paradox. This paradox can only be
understood by assuming that some neurological mechanisms are responsible for this, and
by analyzing more in detail how through such mechanisms influences from the social
context affect internal mental processes.
First, in Chapter 7 an overview is presented of a number of recent findings from Social
Neuroscience, that form an explanation of how agents can behave in a social manner. For
example, shared understanding and collective power are social phenomena that serve as a
form of glue between individual persons. They easily emerge and often involve both
cognitive and affective aspects. As the behaviour of each person is based on complex

24
internal mental processes involving, for example, own goals, emotions and beliefs, it would
be expected that such forms of sharedness and collectiveness are very hard to achieve.
Apparently, specific neurological mechanisms are required to tune the individual mental
processes to each other in order to enable the emergence of shared mental states and
collective behaviour. Having knowledge about these mechanisms provides a basis to
modelling corresponding mechanisms in a computational setting. From a neurological
perspective, mirror neurons and internal simulation are core concepts to explain the
mechanisms underlying such social phenomena. In this chapter it is discussed how based
on such neurological concepts computational mechanisms can be identified to obtain social
agent models. It is discussed how these social agent models indeed are an adequate basis to
simulate the emergence of shared understanding and collective power in groups of agents.
Within a social context the notion of ownership of actions is important. Chapter 8
addresses this notion. It is related to mechanisms underlying self-other distinction, where a
self-ownership state is an indication for the self-relatedness of an action and an other-
ownership state to an action attributed to someone else. The computational model
presented in this chapter generates prior and retrospective ownership states for an action
based on principles from recent neurological theories. A prior self-ownership state is
affected by prediction of the effects of a prepared action as a form of internal simulation,
and exerts control by strengthening or suppressing actual execution of the action. A prior
other-ownership state plays a role in mirroring an observed action performed by another
agent, without imitating the action. A retrospective self-ownership state depends on
whether the sensed consequences of an executed action co-occur with the predicted
consequences, and is the basis for acknowledging authorship of actions in social context. It
is shown how a number of known phenomena can be obtained as behaviour by the model.
For example, scenarios are shown for vetoing a prepared action due to unsatisfactory
predicted effects, Moreover, it is shown how poor action effect prediction capabilities can
lead to reduced retrospective ownership states, as in persons suffering from schizophrenia.
This can explain why sometimes the own actions are attributed to others or actions of
others are attributed to oneself: reduced self-other distinction.
Chapter 9 addresses how in social interaction between two persons usually each person
displays understanding of the other person. This may involve both nonverbal and verbal
elements, such as bodily expressing a similar emotion and verbally expressing beliefs about
the other person. Such social interaction relates to an underlying neural mechanism based
on a mirror neuron system. Differences in social responses of individuals can often be
related to differences in functioning of certain neurological mechanisms, as can be seen, for
example, in persons with a specific type of Autism Spectrum Disorder (ASD). This chapter

25
presents a computational model capable of showing different types of social response
patterns based on such mechanisms, adopted from theories on the role of mirror neuron
systems, emotion integration, emotion regulation, and empathy in ASD. The mechanisms
may show different variations over time. This chapter also addresses this adaptation over
time. It includes a computational model capable of learning social responses, based on
insights from Social Neuroscience.
Chapter 10 addresses joint decision making. The notion of joint decision making as
considered does not only concern a choice for a common decision option, but also a good
feeling about it, and mutually acknowledged empathic understanding. In this chapter a
social agent model for joint decision making is presented addressing the role of mutually
acknowledged empathic understanding in the decision making. The model is based on
principles from recent neurological theories on mirror neurons, internal simulation, and
emotion-related valuing. Emotion-related valuing of decision options and mutual contagion
of intentions and emotions between agents are used as a basis for mutual empathic
understanding and convergence of decisions and their associated emotions.
In Chapter 11 it is discussed how dynamical models can be designed from a social network
perspective. This perspective simplifies agents to just one state and expresses the
complexity in the structure of the network. The states can represent an agent’s emotion, a
belief, an opinion, or a behaviour. Two types of dynamics are addressed: dynamics within a
given network, and dynamics of a network. Dynamics within a network means that the
network stays the same, but states (nodes) in the network may change their activation level
over time due to the influence they get from the network (social contagion). In the case of
dynamics of a network, the network connections change, for example their weights may
increase or decrease, or connections are added or removed. Both types of dynamics can
also occur together. Different types of dynamic networks are addressed, based on different
principles: the homophily principle assuming that connections strengthen more when the
agents are more similar in their state (the more you are alike, the more you like each other),
and the more becomes more principle assuming that agents that already have more and
stronger connections also attract more and stronger connections. Moreover, it is discussed
how dynamics of networks can be modelled when (empirical) information over time is
available about actual interaction between agents (both in the sense of frequency and of
intensity). Based on such information connection weights can be modelled in an adaptive
manner.

26
Part IV Analysis Methods
Models can be analysed by performing simulation experiments in a systematic manner. For
example, it can be found out that under certain conditions a certain state always gets a
certain activation level, and how fast it reaches that level. Moreover, during such
experiments values for the parameters of a model can be identified by hand such that for
these parameter values the model shows a certain type of behavior. For more complex
models such processes may be difficult. In this part some techniques are discussed to
achieve this by analysis of the model in different ways.
Chapter 12 addresses the mathematical analysis of properties of a model such as:
 whether some values for the variables exist for which no change occurs (equilibria),
and how such values may depend on the values of the parameters of the model and/or
the initial values for the variables
 whether certain variables in the model converge to some limit value (attracting
equilibria) and how this may depend on the values of the parameters of the model
and/or the initial values for the variables
 whether or not certain variables will show monotonically increasing or decreasing
values over time (monotonicity)
 how fast a convergence to a limit value takes place (convergence speed)
 whether situations occur in which no convergence takes place but in the end a
specific sequence of values is repeated all the time (limit cycle)
Such dynamic properties of models can be analysed by systematically conducting
simulation experiments. These types of properties can often also be analysed in an analytic
mathematical manner, without performing any simulation, and afterwards the results may
be compared to outcomes of simulations. Such comparisons can be used for verification of
the (implemented) model. If discrepancies are found, then probably there is something
wrong in the implementation of the model. In this chapter some methods to analyse such
properties of dynamical models will be described and illustrated for the Hebbian learning
model, and for dynamic connection strengths in social networks. The properties analysed
by the methods discussed cover equilibria, increasing or decreasing trends, and recurring
patterns (limit cycles).
Chapter 13 discusses dynamic properties of processes that emerge over time, and how they
can be verfied in a systematic manner. A process often generates patterns over time that
can be described in a temporally more global manner, by expressing temporal relations
over longer time periods, in contrast to model descriptions that specify local mechanisms
over small time durations. Such patterns can be considered as emergent phenomena, and it

27
is often an interesting challenge to analyse whether they occur and if so, how their
occurrence relates to the local descriptions of underlying mechanisms and their
characteristics. Such properties have in common that within them references occur to
different time points and order relations between time points such as ‘before’ and ‘after’.
Moreover, quantifiers over time are used such as expressed by ‘eventually’, ‘always’,
‘during’, ‘for some time point...’, or ‘for all time points...’. Such dynamic properties can be
expressed logically in informal, semiformal and formal ways. Expressing them in a formal
format makes it possible to verify whether they hold in some given empirical or simulated
scenario in a systematic or even automated manner. This can be helpful in particular if
many such checks have to be done, for example by analysing the effects of a systematic
variation of initial values and/or parameters in a simulation experiment.
In Chapter 14 it is discussed how a personalised model can be obtained that fits well to a
specific person’s characteristics. A model is a close approximation, but always a form of
abstraction of a real world phenomenon. Its accuracy and correctness mainly depend on the
chosen abstracting assumptions and the values of the parameters in the model. Depending
on the complexity of the model, the number of its parameters can vary from just a couple to
thousands. These parameters usually represent specific characteristics of the modelled
phenomenon, for example, for modelling human processes person-specific characteristics.
No values for such parameters are given at forehand. Estimation of parameters for a given
model is a nontrivial task. There are many parameter estimation methods available in the
literature. In this chapter a number of these methods are briefly discussed.
Part V Philosophical, Societal and Educational Perspectives
In Part V some wider perspectives are addressed. It is discussed how the computational
modeling approach relates to historical and philosophical developments concerning
dynamics, how it fits in current trends in societal development, and how such modeling
perspective can play a crucial role in an integrative multidisciplinary academic curriculum.
In Chapter 15 it is discussed how dynamics has been a challenging issue since long ago; for
example Greek philosophers already wondered what it is in a given state that is driving the
change to a next state. This issue has been addressed since that time for different domains,
in Physics but also in Mathematics, Cognitive Science and Philosophy of Mind. In the
development of Physics this issue has played an important role and has led to notions such
as velocity, momentum, kinetic energy and force that drive motion in mechanics. The issue
of dynamics is still out there today, for example, in the domain of Cognitive Science and
Philosophy of Mind concerning the physical realism of assumed but not directly physically

28
observable mental states such as desires and intentions that are supposed to drive
(physically observable) behaviour. Four cases of dynamics within different disciplines
(Cognitive Science, Physics, Mathematics, Computer Science) are discussed in this
chapter. Among them are concepts like desire and intention in Cognitive Science,
momentum, kinetic energy, and force in Physics, and derivatives of a function and Taylor
approximations in Mathematics. Similarly, it is shown how causal graphs and transition
systems (often used in Computer Science) can be interpreted from a perspective of
dynamics. This provides a unified view on the explanation of dynamics across different
disciplines which is directly related to the basic assumptions underlying the temporal-
causal modelling perspective.
Chapter 16 outlines the strong societal development to the integration of more and more
smart devices in all aspects of life. Scientific areas addressing this development have
names such as Ambient Intelligence, Ubiquitous Computing, Pervasive Computing,
Human-Aware Computing or Socially aware Computing. This development in society
often results in integrated complex systems involving humans and technical equipments,
also called socio-technical systems. In this chapter it is discussed how in such systems
often not only sensor data, but also more and more dynamic computational models based
on knowledge from the human-directed sciences such as biomedical science, neuroscience,
and psychological and social sciences are incorporated. These models enable the
environment to perform in-depth analyses of the functioning of observed humans, and to
come up with well-informed interventions or actions. It is discussed which ingredients are
important to realize this view in a principled manner, among which computational models,
and how frameworks can be developed to combine these ingredients to obtain the intended
type of systems.
Chapter 17 discusses the design of a curriculum with main focus on human-oriented
scientific knowledge and how this can be exploited to develop support for humans by
means of advanced devices in the daily environment. The aim was to offer a study path for
those students with exact talents with an interest mainly in human functioning and society.
The curriculum was designed from a problem-oriented perspective in relation to societal
problem areas. From human-oriented disciplines scientific knowledge for human
functioning in such problem areas was obtained. Computational modelling for such human
processes plays a central role as an integrating factor in the curriculum. Elements from
Ambient Intelligence, Artificial Intelligence, and Informatics are included for design of
support systems.

29
Part VI Discussion
Chapter 18 is a discussion
References
Aral, S., Muchnik, L., and Sundararajan, A. (2009). Distinguishing Influence Based
Contagion From Homophily Driven Diffusion in Dynamic Networks. Proceedings of
the National Academy of Sciences (USA) 106:2, 1544-1549.
Aristotle (350 BC). Physica (translated by R.P. Hardie and R.K. Gaye)
Ashby, W.R. (1960). Design for a Brain, Chapman and Hall, London (second extended
edition). First edition, 1952.
Ashby, W.R. (1960). Design for a Brain, Chapman and Hall, London (second extended
edition). First edition, 1952.
Barsalou, W. (2009). Simulation, situated conceptualization, and prediction Lawrence.
Phil. Trans. R. Soc. B (2009) 364, 1281–1289
Bechara, A., Damasio, H., and Damasio, A.R. (2003). Role of the Amygdala in Decision-
Making. Ann. N.Y. Acad. Sci. 985, 356-369 (2003)
Bechara, A., Damasio, H., Damasio, A.R., and Lee, G.P. (1999). Different Contributions
of the Human Amygdala and Ventromedial Prefrontal Cortex to Decision-Making.
Journal of Neuroscience 19, 5473–5481 (1999)
Becker, W. & Fuchs, A.F.: Prediction in the Oculomotor System: Smooth Pursuit During
Transient Disappearance of a Visual Target. Experimental Brain Research 57, 562–
575 (1985)
Beer, R.D. (1995). On the dynamics of small continuous-time recurrent neural networks.
Adaptive Behavior, 3, 469-509.
Beer, R.D. (2000). Dynamical approaches to cognitive science. Trends in Cognitive
Sciences, 4, 91-99.
Bell, A. (1999). Levels and loops: the future of artificial intelligence and neuroscience.
Phil. Trans. R. Soc. Lond. B, 354, 2013–2020.
Bell, A. (1999). Levels and loops: the future of artificial intelligence and neuroscience.
Phil. Trans. R. Soc. Lond. B, 354, 2013–2020.
Bickle, J. (1998). Psychoneural Reduction: The New Wave. Cambridge, Mass: MIT Press.
Boccalettia, S., Latorab, V., Morenod, Y., Chavez, M., Hwanga, D.-U. (2006). Complex
networks: Structure and dynamics. Physics Reports 424 (2006) 175 – 308.
Byrne, D. (1986). The attraction hypothesis: Do similar attitudes affect anything? Journal
of Personality and Social Psychology, 51(6), 1986, 1167-1170.
Cacioppo, J.T., Berntson, G.G. (2005). Social neuroscience. Psychology Press (2005)
Cacioppo, J.T., Visser, P.S., & Pickett, C.L. (2006). Social neuroscience: People thinking
about thinking people. Cambridge, MA: MIT Press (2006)

30
Clark, A. (1998). Being there: Putting brain, body, and world together again. MIT Press.
Crick, F., & Koch, C., (1998). Constraints on cortical and thalamic projections: the no-
strong-loops hypothesis. Nature, 391, 245-250.
Damasio, A.R. (1994). Descartes’ Error: Emotion, Reason and the Human Brain.
Papermac, London.
Damasio, A.R. (1999). The Feeling of What Happens. Body and Emotion in the Making of
Consciousness. New York: Harcourt Brace.
Damasio, A.R. (2003) Looking for Spinoza. Vintage books, London
Damasio, A.R. (2010). Self comes to mind: constructing the conscious brain. Pantheon
Books, NY (2010)
Decety, J., and Cacioppo, J.T. (eds.) (2010). Handbook of Social Neuroscience: Oxford
University Press (2010)
Decety, J., and Ickes, W. (2009). The Social Neuroscience of Empathy. MIT Press (2009)
Deliens, G., Gilson, M., Peigneux, P. (2014). Sleep and the processing of emotions Exp
Brain Res (2014). 232:1403–1414 DOI 10.1007/s00221-014-3832-1
Descartes, R. (1634). The World or Treatise on Light. Translated version by M.S.
Mahoney:
URL: http://www.princeton.edu/~hos/mike/texts/descartes/world/world.htm
Dolan, R.J. (2002). Emotion, Cognition, and Behavior. Science, 298, 1191-1194.
Dubois, D., Lang, J., and Prade, H. (1991), Fuzzy sets in approximate reasoning, Part 2:
Logical approaches, 1991. Fuzzy Sets and Systems 40, 203-244 North-Holland
Dubois, D., and Prade, H. (2002), Possibility Theory, Probability Theory and Multiple-
valued Logics: A Clarification, Annals of Mathematics and Artificial Intelligence 32,
35–66, (2002)
Eich, E., Kihlstrom, J.F., Bower, G.H., Forgas, J.P., & Niedenthal, P.M. (2000). Cognition
and Emotion. New York: Oxford University Press.
Forgas, J.P., Goldenberg, L., & Unkelbach, C. (2009). Can bad weather improve your
memory? An unobtrusive field study of natural mood effects on real-life memory.
Journal of Experimental Social Psychology, 45, 254–257.
Forrester, J.W. (1987). Lessons from System Dynamics Modeling. System Dynamics
Review, 3(2), 136-149.
Forrester, J.W., 1973. World Dynamics, (2d ed.). Waltham, MA: Pegasus
Communications. 144 pp. F
Frijda, N.H., Manstead, A. S. R., & Bem, S. (2000b). The influence of emotions on beliefs.
In: Frijda, N.H., et al. (eds.) (2000a). Emotions and beliefs: how feelings influence
thoughts. Cambridge University Press, pp. 1-9.
Funahashi, K. & Nakamura, Y. (1993). Approximation of dynamical systems by
continuous time recurrent neural networks. Neural Networks, 6, 801-806.
Gallese, V., Goldman, A. (1998). Mirror neurons and the simulation theory of mind-
reading. Trends in Cognitive Sciences 2:493–501 (1998)
Giangiacomo, G. (2001). Fuzzy logic: Mathematical Tools for Approximate Reasoning,
Kluwer Academic Publishers, Dordrecht, 2001

31
Giles, J., Computational Social Science: Making the Links. Nature, vol 488, 2012, pp. 448-
450
Goldman, A.I. (2006). Simulating Minds: The Philosophy, Psychology, and Neuroscience
of Mindreading. New York: Oxford Univ. Press. (2006)
Grossberg, S. (1969). On learning and energy-entropy dependence in recurrent and
nonrecurrent signed networks. Journal of Statistical Physics, 1, 319-350.
Gujar, N., McDonald, S.A., Nishida, M., and Walker, M.P. (2011). A Role for REM Sleep
in Recalibrating the Sensitivity of the Human Brain to Specific Emotions. Cerebral
Cortex 21,115-123 (2011
Harmon-Jones, E., and Winkielman, P. (eds.) (2007). Social neuroscience: Integrating
biological and psychological explanations of social behavior. New York: Guilford
(2007)
Hebb, D. (1949) The Organisation of Behavior. Wiley (1949)
Hesslow, G. (1994). Will neuroscience explain consciousness? J. Theoret. Biol. Vol. 171
(1994), 29–39.
Hesslow, G. (2002). Conscious thought as simulation of behaviour and perception. Trends
Cogn. Sci. 6, 242–247 (2002)
Hesslow, G. (2012). The current status of the simulation theory of cognition. Brain Res.
1428, 71–79. doi: 10.1016/j.brainres.2011.06.026 H
Hirsch, M. (1989). Convergent activation dynamics in continuous-time networks. Neural
Networks, 2, 331–349.
Hopfield, J.J. (1982). Neural networks and physical systems with emergent collective
computational properties. Proc. Nat. Acad. Sci. (USA), 79, 2554-2558.
Hopfield, J.J. (1984). Neurons with graded response have collective computational
properties like those of two-state neurons. Proc. Nat. Acad. Sci. (USA), 81, 3088-
3092.
Iacoboni M. (2008a). Mirroring People: the New Science of How We Connect with Others.
New York: Farrar, Straus & Giroux (2008)
Janak, P.H., and Tye, K.M. (2015). From circuits to behaviour in the amygdala. Nature,
vol. 517, 2015, pp. 284-292
James, W. (1884). What is an emotion, Mind 9, 188–205 (1884)
Jenison, R.L., Rangel, A., Oya, H., Kawasaki, H., and Howard, M.A. (2011). Value
Encoding in Single Neurons in the Human Amygdala during Decision Making. The
Journal of Neuroscience, vol. 31, 2011, 331–338.
Kelso, J.A.S. 1995. Dynamic patterns: The self-organization of brain and behavior.
Cambridge: MIT Press.
Kim, J. (1996). Philosophy of Mind. Westview Press.
Kim, J. (1998). Mind in a Physical world: an Essay on the Mind-Body Problem and Mental
Causation. MIT Press, Cambridge, Mass.
Kuipers, B.J., (1984). Commonsense reasoning about causality: Deriving behavior from
structure, Artificial Intelligence 24 (1984) 169-203.
Kuipers, B.J. and Kassirer, J.P., (1983). How to discover a knowledge representation for
causal reasoning by studying an expert physician. In: Proceedings Eighth

32
International Joint Conference on Artificial Intelligence, IJCAI’83, Karlsruhe, F.R.G.
(William Kaufman, Los Altos, CA, 1983).
LaBar, K.S., & Cabeza, R. (2006). Cognitive neuroscience of emotional memory. Nature
Reviews: Neuroscience, 7, 54-64.
Lakoff, G., Johnson, M. (1999) Philosophy in the flesh: The embodied mind and its
challenge to western thought. Basic Books
Laplace, P.S. (1825). Philosophical Essays on Probabilities. Springer-Verlag, New York,
1995. Translated by A.I. Dale from the 5th French edition of 1825.
Levin, R., and Nielsen, T.A. (2007). Disturbed dreaming, posttraumatic stress disorder, and
affect distress: A review and neurocognitive model. Psychological Bulletin 133, 482–
528 (2007)
Loewenstein, G. , & Lerner, J. (2003). The role of emotion in decision making. In R. J.
Davidson , H. H. Goldsmith , & K. R. Scherer (Eds.), The handbook of affective
science. Oxford, England: Oxford University Press, pp. 619–642
Marques, H.G., and Holland, O. (2009). Architectures for functional imagination.
Neurocomputing, 72, 2009, pp. 743-759.
McPherson, M., Smith-Lovin, L., Cook, J.M. (2001). Birds of a feather: homophily in
social networks. Annu. Rev. Sociol. 27, 415–444 (2001).
Mislove, A., Viswanath, B., Gummadi, K.P., Druschel, P. (2010). You Are Who You
Know: Inferring User Profiles in Online Social Networks. Proc. WSDM’10, February
4–6, 2010, New York City, New York, USA, pp. 251-260
Montague, P.R., Berns, G.S.: Neural economics and the biological substrates of valuation.
Neuron 36, 265-284 (2002)
Morrison, S.E., and Salzman, C.D. (2010). Re-valuing the amygdala. Current Opinion in
Neurobiology 20, 221–230 (2010)
Mundt, M.P., Mercken, L, Zakletskaia, L.I. (2012). Peer selection and influence
effects on adolescent alcohol use: a stochastic actor-based model. BMC pediatrics, 12,
115.
Murray, E.A. (2007). The amygdala, reward and emotion. Trends Cogn Sci 11, 489-497
(2007)
Newell, A., and Simon, H.A. (1976). Computer science as empirical inquiry: Symbols and
search, Commun ACM 19 (3) (1976) 113-126.
Nussbaum, M. (ed.) (1978). Aristotle's De Motu Animalium. Princeton: Princeton
University Press.
Ousdal, O.T., Specht, K., Server, A., Andreassen, O.A. Dolan, R.J., Jensen, J. (2014). The
human amygdala encodes value and space during decision making. Neuroimage, vol.
101 (2014) 712–719.
Pace-Schott, E.F., Germain, A., Milad, M.R. (2015). Effects of sleep on memory for
conditioned fear and fear extinction. Psychol Bull, vol 141(4), 2015, 835-857.
Pearl, J., (2000). Causality. Cambridge University Press, 2000.
Pessoa, L. (2008). On the relationship between emotion and cognition. Nature Reviews:
Neuroscience, 9, 148-158.

33
Pessoa, L. (2011). Emotion and cognition and the amygdala: From “what is it?” to “what's
to be done?”. Neuropsychologia, 49, 2011, pp 681-694
Pezzulo, G., Candidi, M., Dindo, H., Barca, L. (2013). Action simulation in the human
brain: Twelve questions. New Ideas in Psychology, vol. 31, 2013, pp. 270–290
Phelps, E.A. (2006). Emotion and Cognition: Insights from Studies of the Human
Amygdala. Annu. Rev. Psychol., 57, 27–53.
Pineda, J.A. (ed.) (2009). Mirror Neuron Systems: the Role of Mirroring Processes in
Social Cognition. Humana Press Inc. (2009)
Port, R.F., & van Gelder, T. (1995). Mind as motion: Explorations in the dynamics of
cognition. Cambridge, MA: MIT Press.
Potter S.M. (2007). What can Artificial Intelligence get from Neuroscience? In: Lungarella,
M., Bongard, J. & Pfeifer, R. (eds) Artificial Intelligence Festschrift: The next 50
years. Berlin: Springer-Verlag.
Purves, D., Brannon, E.M., Cabeza, R., Huettel, S.A., LaBar, K.S., Platt, M.L., Woldorff,
M.G. (2008). Principles of Cognitive Neuroscience. Sunderland MA: Sinauer
Associates Inc., 2008.
Rizzolatti, G, and Sinigaglia, C. (2008). Mirrors in the Brain: How Our Minds Share
Actions and Emotions. Oxford University Press (2008)
Salzman, C.D., and Fusi, S. (2010). Emotion, Cognition, and Mental State Representation
in Amygdala and Prefrontal Cortex. Annu. Rev. Neurosci. 33, 173–202 (2010)
Scherer, K.R. (2009). Emotions are emergent processes: they require a dynamic
computational architecture. Phil. Trans. R. Soc. B, 364, 3459-3474
Schurger, A. and Uithol , S. (2015). Nowhere and Everywhere: The Causal Origin of
Voluntary Action Rev. Phil. Psych., vol. 6, 2015, pp 761-778. DOI 10.1007/s13164-
014-0223-2
Shalizi, C.R., and Thomas, A.C. (2011). Homophily and Contagion are Generically
Confounded in Observational Social Network Studies. Sociological Methods &
Research 40(2) 211–239
Sotres-Bayon, F., Bush, D.E., and LeDoux, J.E. (2004). Emotional perseveration: an update
on prefrontal-amygdala interactions in fear extinction. Learn. Mem. 11, 525–535
(2004).
Steglich, C.E.G., Snijders, T.A.B., Pearson, M. (2010). Dynamic networks and behavior:
separating selection from influence. Sociol Methodol 40, 329–393. Yoo, S.S., Gujar,
N., Hu, P., Jolesz, F.A., Walker, M.P. (2007). The human emotional brain without
sleep – a prefrontal amygdala disconnect. Curr. Biol. 17, R877–R878 (2007)
Storbeck, J., & Clore, G.L. (2007). On the interdependence of cognition and emotion.
Cognition and Emotion, 21, 1212-1237.
van der Helm, E., Yao, J., Dutt, S., Rao, V., Saletin, J.M. & Walker, M.P. (2011). REM
Sleep Depotentiates Amygdala Activity to Previous Emotional Experiences. Current
Biology 2011; 21(23):1-4.
Walker, M.P., van der Helm, E. (2009). Overnight therapy? The role of sleep in emotional
brain processing. Psychol Bull. 135, 731-748 (2009)

34
Thelen, E., and L. Smith. (1994). A dynamic systems approach to the development of
cognition and action. Cambridge: MIT Press.
Treur, J. (2007). Temporal Factorisation: a Unifying Principle for Dynamics of the World
and of Mental States. Cognitive Systems Research Journal, vol. 8, 2007, pp. 57-74.
Treur, J. (2016) Dynamic Modelling Based on a Temporal-Causal Network Modelling
Approach. Biologically Inspired Cognitive Architectures Journal, 2016, to
appear. URL:
https://www.researchgate.net/publication/289193241_Dynamic_Modeling_Based_on
_a_Temporal-Causal_Network_Modeling_Approach
Valente, T.W. (2010). Social Networks and Health: Models, Methods, and Applications.
New York, Oxford University Press, 2010.
van Gelder, T. (1998). The dynamical hypothesis in cognitive science. Behavioral and
Brain Sciences 21, 1998, pp. 615–665.
van Gelder and Port (1995). It's about time: An overview of the dynamical approach to
cognition. In: Port, R.F., & van Gelder, T. (1995). Mind as motion: Explorations in
the dynamics of cognition. Cambridge, MA: MIT Press, pp. 1-43.
Wilson, M. (2002). Six views of embodied cognition. Psychonomic Bulletin & Review 9,
2002, pp 625-636.
Winkielman, P., Niedenthal, P.M., & Oberman, L.M. (2009). Embodied Perspective on
Emotion-Cognition Interactions. In: Pineda, J.A. (ed.), Mirror Neuron Systems: the
Role of Mirroring Processes in Social Cognition. Humana Press/Springer Science,
2009, pp. 235-257.
Zadeh, L. (1978). Fuzzy Sets as the Basis for a Theory of Possibility, Fuzzy Sets and Systems, 1, 3–
28, (1978). (Reprinted in Fuzzy Sets and Systems 100 (Supplement): 9–34, 1999)