1
Relevance of wide cognition for social intelligence. Key trends
Authors: Łukasz Afeltowicz, Nicolaus Copernicus University; Anna Borghi, University of Bologna &
Institute of Cognitive Sciences and Technologies, Italian Research Council, Rome; Stephen Butterfill,
University of Warwick; Robert Clowes, New University of Lisbon; Adam Gies, The New School
University; Victor Loughlin, University of Antwerp; Marek McGann, Mary Immaculate College ~
University of Limerick; Jakub Ryszard Matyja, Institute of Philosophy and Sociology, Polish Academy
of Sciences (Warsaw); Marcin Miłkowski, Institute of Philosophy and Sociology, Polish Academy of
Sciences (Warsaw, project coordinator); Roger O’Shea, University of Limerick; Aleksandra
Przegalińska, University of Warsaw; Joanna Rączaszek-Leonardi, University of Warsaw; Zuzanna
Rucińska, University of Hertfordshire; Fredrik Stjernberg, Linköping University; Luca Tummolini,
Institute of Cognitive Sciences and Technologies, Consiglio Nazionale delle Ricerche (CNR); Michael
Wheeler, University of Stirling; Witold Wachowski, Centre for Philosophical Research (Warsaw); Tad
Zawidzki (George Washington University).
Cover page design: © Jacek Seweryn Podgórski, 2013
2
Contents
Executive Summary
1. Wide Accounts of Cognition and Intelligence
1.1. Embodied and Grounded Cognition
1.2. Embedded and Situated Cognition
1.3. Extended Cognition
1.4. Enactive Cognition
1.5. Distributed Cognition
1.6. Interrelations
2. Applications in Research on Social Intelligence
2.1 Embodied Joint Action
2.2 Cognitive artifacts and ICT (Cloud-Tech, E-Memory and Virtual Identities)
2.2.1 E-Memory and Cloud Tech
2.2.2 Virtual Reality and avatars
2.3 Social Cognition and Culture
2.3.1. Mindreading
2.3.2. Two Case Studies: Imitation and Pretence
2.4 Social emotions
3. Challenges
4. Dangers
5. Summary
6. References
3
Executive Summary
Wide cognition approaches in cognitive sciences are relevant for studying social
intelligence, and they afford new ways of integrating interdisciplinary research. They offer
novel predictions and explanations by transcending methodological individualism.
Traditional cognitive science was methodologically individualist, and framed intelligent
behavior in terms of processing of internal representations of individuals. Embodied and
grounded cognition, the extended and scaffolded mind, enactivism and distributed cognition
all challenge the traditional approach in different ways. Social intelligence is accounted for in
terms of embodied interactions supported and extended by actively built cognitive niches.
Despite the variety of approaches we jointly dub “wide cognition”, they offer a new coherent
picture of cognition, as well make it possible to integrate and unify interdisciplinary
research.
In this white paper, we introduce basic frameworks of wide cognition and illustrate it with
promising developments in research on pathologies, child development, and mind reading.
Wide cognition offers novel predictions relevant for the study of social phenomena.
At the same time, owing to the multi-disciplinary nature of the approaches presented herein,
there is a danger of duplicating effort and disintegration of the field. For this reason, more
effort is needed to fully leverage the potential of the unification that is inherent in wide
cognition. Integrating different approaches of wide cognition will yield additional benefits
for the study of social intelligence, but that means that larger interdisciplinary networking
and research programs are needed in this field.
4
1. Wide Accounts of Cognition and Intelligence
By wide cognition we mean extended, embodied, enacted, embedded, and distributed
cognition, which are related but distinct concepts.
In contradistinction to traditional frameworks of cognitive science, they do not explain
cognitive phenomena solely with manipulation of (language-like) internal representations
but stress the fact that (1) minds can extend into the environment; (2) agents are cognitive
insofar as they are embodied; (3) their cognitive scaffolding is enacted, or constructed, in an
active fashion; (4) cognitive phenomena are always interactions with the environment; (5)
and that cognitive acts are not always but sometimes paradigmatically distributed among
multiple agents. There is a growing body of research that is directly relevant for
understanding social intelligence, in particular in its reliance on environmental design,
bodily interaction, shared cognitive and symbolic tools, and complex schemas of
collaboration.
In the Western tradition, cognition was understood as an internal process carried out in
proprietary representational language, fed by sensory apparatus, largely performed over
internal representations and finally used to produce action (Hurley, 2001). Mind was
considered quintessentially a private realm. Most traditional theorizing of the mind gives
credit for our cognitive abilities almost exclusively to the goings-on in our brains. This
outlook has been significantly challenged in recent years in both theoretical and empirical
research.
1.1. Embodied and Grounded Cognition
The claim of the embodied cognition (EC) is that the physical body of an agent is
constitutively relevant for cognition; in other words, cognitive processing involves more
than the brain. Core cognition, which essentially involves perception and action, depends
deeply on the features of the physical body.
In the last years, EC has become more and more widespread in all areas of cognitive science
– from neuroscience to cognitive psychology to cognitive linguistics to philosophy to
computer science and robotics. It is starting to penetrate also reflection on art, anthropology
and on gender issues. As recently stated by Lakoff (2012), “It may be hard to think back to a
time before the idea of embodied cognition, [but I was raised in that generation].” (p. 1). In a
recent paper on the future of psychology, Dendre Gentner (2010) has shown how embodied
cognition is challenging the classical approaches to knowledge representation and cognition
in general.
This success has led in the last 10-15 years to an impressive burst of experimental evidence
for the role of the body in cognition, in particular in psychology and neuroscience (for
reviews, see Barsalou, 2008; Fischer & Zwaan, 2008; Jirak et al., 2010; Toni et al., 2008;
Willems & Hagoort, 2007). Recent special issues have been published by the journals
5
Frontiers in Psychology (edited by Borghi & Pecher, 2011) and Topics in Cognitive Science
(edited by Davis & Markman, 2012). But this success has not always lead to terminological
precision. The term EC has been used in a variety of contexts and with different connotations
(see Goldman & De Vignemont, 2009), and more moderate and more radical embodied
positions are present. The unifying aspect of EC theories is the opposition to classical
propositional views, which held that cognition is defined exclusively in terms of
manipulation of amodal, abstract, and arbitrary codes. In contrast, according to EC theorists
cognition is constrained and enabled by our sensorimotor systems, by the specific
characteristics of our own brain-body system.
Some authors (e.g., Barsalou, 2008; Pezzulo et al., 2011) reduce the stress put on the role of
the body, and prefer to use the more general term “grounded cognition”, to emphasize the
fact that cognition is grounded not only in bodily states, but also in situations, situated
simulations, and so forth. On the neural side, some recent discoveries have given great
impetus to EC: it is the case, for example, of studies on how objects are represented in the
modal areas of the brain (see Martin, 2007, for a review), and of studies on the canonical and
mirror neuron systems (Arbib 2005, Rizzolatti & Craighero 2004).
Many research lines have been promoted. We do not pretend to be exhaustive, but among
the most important we would list the research areas on affordances (Gibson 1979, Chemero
2009), i.e., agent-related properties of the environment that offer the agent opportunities to
act, operationalised as motor activity elicited during observation of objects. There is also
research on the grounding of the concepts and language in perception, action, and emotional
systems; on intersubjectivity (also thanks to the renewed interest in the mirror neuron
system); on the sense of our own body, of the peripersonal space, and so forth.
An important part of EC is dynamical account of cognition (Kelso 1995), which has been also
applied to the study of social phenomena such as language (Rączaszek-Leonardi & Kelso
2008). One of the spectacular results in dynamical approaches was to find motor-based
explanations of apparently ‘cognitive’ deficits in infants. Piaget (1952) famously posited a
logical-symbolic deficit as the explanation of the error infants were making when choosing a
toy from an old location: Infants who have successfully uncovered a toy at location “A”
continue to reach to that location even after they watch the toy hidden in a nearby location
“B.” As it turns out, the phenomenon can be understood by the coupled dynamics of the
ordinary processes of goal-directed actions: looking, planning, reaching, and remembering
(Thelen et al. 2001).
Obviously EC has not been free from attacks and criticisms, and on some occasions debates
have been hot. For example, in a famous paper, Mahon and Caramazza (2008) have argued
that nobody questions the involvement of the motor system in conceptual and language
processing, but this involvement would be only epiphenomenal, and the process of
comprehension would not necessarily imply the motor system involvement. In contrast,
proponents of EC (at least in its strong version) argue that the involvement of motor system
is constitutive of the comprehension process. In recent years, some mixed approaches have
emerged that try to combine a propositional and an embodied approach (e.g., Dove 2009;
Chatterije 2010; Louwerse 2011). Among the most debated issues is the issue of
representation, i.e., whether representational and anti-representational, enactivist
approaches can coexist (e.g., Gallese & Sinigaglia 2011; Van Elk et al., 2010; Chemero 2009).
6
Another important issue concerns the way in which abstract concepts and words without a
perceptual object as referent (e.g., truth) are represented. There are several embodied
accounts of linguistic abstraction (for reviews, see Pecher, Boot & Dantzig 2011 and Wiemer-
Hastings & Barsalou 2005), and notwithstanding all differences, they share the focus on the
connection between abstract words and sensorimotor processing.
1.2. Embedded and Situated Cognition
The situated and embedded approach to cognition holds that cognition should be
cashed out in terms of the interaction of the agent and its immediate surrounding. The
extra-bodily context constraints and enables cognition.
Many lines of inquiry offer support to embedded mind (and by extension situated cognition).
In this short section, I will identify only two (see Clark 2008, and Robbins and Aydede 2009,
for further examples). Behaviour-based robotics (BBR), as pioneered by Rodney Brooks
(1991), demonstrated that it was possible to build robots capable of performing simple tasks
despite those robots having no detailed, internal knowledge of the environments in which
they were operating. Such robots were designed to be “set up to be set off” by certain features
of their local surroundings, bypassing the need for complex, internal cognitive machinery.
The work of Brooks and many others showed that basic cognitive functions could be causally
dependent on movement and structures in the local environment (see Steels and Brooks
(eds), 1995, for an overview of these and related issues). As such, BBR arguably supports the
claims made by embedded mind and situated cognition.
Further evidence comes from work on what is called animate vision (Ballard, 1991). The
human eye only has high resolution in a small region near the optical axis (the fovea) yet
nonetheless encompasses a large field of view. This discrepancy between high resolution at
the centre and low resolution at the peripheries is compensated for by the fact that the eye is
in constant motion and mechanisms of gaze fixation ensure that the fovea can automatically
target whatever is of interest. Ballard (1991) argues that our ability to seamlessly fixate on
objects of interest simplifies our access to the environment since for many basic physical
tasks we do not need to build elaborate internal models of our local environment. Instead,
we just need to register or sense those aspects of the environment we deem immediately
relevant to the task in hand.
This is nicely illustrated in Ballard et al. (1997). Subjects were asked to copy coloured blocks
from one area into another on a computer screen. By tracking eye movements, it became
apparent that subjects were utilizing “minimal memory strategies” in order to complete the
task. That is, rather than construct internal models of the colour and location of the blocks
that needed to be copied and then using working memory in order to perform the task,
subjects instead used repeated eye saccades and accessed the colour and location
information only when and where it was required. Ballard et al. (1997, p. 739) conclude that
subjects were deploying “just-in-time” representations, that is, subjects were leaving
important information out in the world and only accessing it on a “just-in-time” or
immediate need-to-know basis.
Two things seem to follow from this. First, this supports the claim that the eye functions as a
deictic or pointing device which enables subjects to fulfill “do-it-where-I-am-looking”
7
strategies (ibid., p. 725). Second, the eye has a determinate computational role since its
“pointing movements bind objects in the world to cognitive programs” (ibid., p. 726). Deictic
computation offers support for claims about embedded mind and situated cognition. The
Ballard et al study demonstrates that subjects do not build detailed internal representations
of their local environments in order to fulfill physical tasks. Rather subjects exploit features
of their bodies and features of their environment in order to minimize reliance on internal
processing and accomplish computation ‘on the cheap’. There are thus demonstrable
complex causal dependencies between the subject and their environment for the completion
of certain tasks.
1.3. Extended Cognition
The extended mind is the idea that cognitive processes are not necessarily brain-bound and
can incorporate external resources such a tools, language, and external systems in order to
enhance or augment cognitive processes. The difference of this approach from situated
cognition is the emphasis that parts of what were traditionally considered the environment
should properly be understood as part of the agent’s mind.
Take the example from Andy Clark, one of the godfather’s of the idea of the extended mind
and extended cognition, reflecting on how the tools in his environment and his interactions
with them contribute towards the task of authoring a book chapter:
As I construct this chapter, for example, I am continually creating, putting aside, and
reorganizing chunks of text. I have files (both paper and online) which contain all
kinds of hints and fragments, stored up over a long period of time, which may be
germane to the discussion. As I (literally, physically) move these things about,
interacting first with one and then another and making new notes, annotations, and
plans, the intellectual shape of the chapter grows and solidifies. It is a shape that does
not spring fully developed from inner cogitations. Instead, it is the produce of a
sustained and iterated sequence of interactions between by brain and a variety of
external props. In these cases, I am willing to say, a good deal of actual thinking
involves loops and circuits that run outside the head and through the local
environment. (Clark, 1997, pp. 206-207)
Recent research in philosophy and cognitive science has endorsed Clark’s view and
emphasised the way artefacts and our manipulation of them can play as important a role in
cognition as anything that goes on inside our heads. This approach has been applied in a
variety of settings such as the cockpit of aeroplanes (Hutchins, 1995b), Jacobean theatre
(Tribble, 2005). The general idea is expressed in (Tribble and Sutton 2011): “’Mind’ is skilful
activity rather than a stock of knowledge: the analysis of mind must therefore be
fundamentally historical in character, because changing cultural artifacts, norms, and
institutions are not external but partly constitute it.” (p. 94).
To take the example of Jacobean Theatre, many renaissance scholars have expressed their
admiration for the prodigious feats of memory of the actors on Shakespeare’s stage. These
actors would often have a live repertoire of dozens of plays and would regularly learn a new
one each week followed by its performance while learning another. Tribble argues that the
8
memory load was considerably ameliorated by factors we tend to think of as external to the
subject and her memory capacities: the use and physical layout of the stage itself, set
patterns of entrances and exits, complex mnemonic strategies, and “parts” – written short
summaries of cues for the actors, which would allow the physical structure of the theatre
itself to remember for the actors. The general setting of the theatre production was also
“forgiving” in a special sense: certain minor slips of memory would never be noticed by the
public. “Remembering” the lines is a matter of producing something that contributes to the
production of the play in a fruitful way, not exclusively of getting the words exactly right.
These factors together indicate that the division between “internal” and “external” aspects of
memory is somewhat misguided.
This kind of use of the environment to memorize has parallels with the method known as the
“loci method” and has been investigated in ecological psychology to contrast traditional lab
experiments on memory with the real-life use of mnemonic methods (Neisser 1976). The
extended mind approach stresses that such methods and uses of material environment are as
much a part of human memory as anything that goes on inside the brain. Such use of the
environment is hence to be seen as a fully cognitive activity, and shows the existence of a
category that is neither fully internal nor fully external. Sutton characterizes the various uses
of mnemonic strategies in the following way: “They are cognitive even though they are not, in
a straightforwardly ancestral way, natural and biological; and they are extended even though
they are not literally external” (Sutton 2010, p. 209).
But if cognition needs to be understood as spread out over brains, instruments and their
social organisation, what of the mind itself? Might that too be spread and distributed beyond
the classically understood bounds of skin and skull? Andy Clark and David Chalmers (1998)
suggested that rather than necessarily being a private inner realm, the mind might also
stretch beyond bodily boundaries to incorporate a motley of artefacts, tools, and perhaps
even the brains of other persons.
According to the thesis of the extended mind, artefacts and instruments could – if used in
the right ways - be considered actual parts of our minds. Whether artefacts could be
profitably so considered would largely turn on their properties and the conditions of their
use. Three criteria were originally proposed to help demarcate which artefacts might count.
These were: constancy (we consult an artefact regularly and unthinkingly when appropriate),
facility (we use it with ease), trust and prior endorsement (we automatically trust
information retrieved from such resources). New generations of mobile information
technologies such as smart phones, tablets and cloud applications, and new patterns of use
suggest that an ever more tools might meet these criteria. But should widely socially
distributed technologies really count as proper parts of our minds?
Sterelny (2010) suggested an extra criterion: entrenchment and personalisation, which
holds that only those technologies that we as individuals customise to our own patterns of
usage should count as proper parts of our minds. It is not clear that this really helps us to
decide how to treat web-resources (such as Open Street Maps) that individuals might
customise for themselves while the technology itself simultaneously relies on the aggregation
of information provided by millions of users in order to work. Much seems to hang here on
our intuitions about whether shared resources should ever count as parts of individual minds
and thus raise important questions about social ontology: Can minds have proper parts
9
resources that are socially shared, or also count as part of other minds? What are the
implications of cognitively incorporating resources which are owned or run in the interests of
others? Under the control of external agencies? Or simply products of widely distributed
systems that the social web has made possible?
The practical implications of these ideas turn on exactly how and under what circumstances
external resources might integrate with internal or biological ones. What properties of
cognitive technologies might make them integrable? What properties of cognitive
technologies might make them integrable? Theorists differ in how they answer this question.
Some have endorsed an extended functionalist approach (Clark, 2008; Wheeler, 2010)
whereby it is the functional or causal role of the technology that makes it part of someone’s
mind. Others insist that it is the bodily manipulation of such technologies that integrates
them into a cognitive routine (Menary, 2006, 2010). A related but importantly distinct
approach is the complementarity framework (Menary, 2010a; Sutton, 2010) proposes that
we tend to integrate tools and artefacts into our cognitive processes when they provide
resources which complement our existing biological systems. Much research has been
carried out into the implications of these ideas around memory (Michaelian & Sutton, 2013)
but much further research is needed to understand which properties of external artefacts,
contexts of use, or cultures of usage imply that cognitive technologies really do complement
our mental processes. Popular discussion of related themes has suggested that the practical
implications of the cognitive integration of internet resources might be distracting and
decentring (Carr, 2010) rather than genuinely cognitively enhancing. Further work into the
properties of the sorts of hybrid systems human beings form with our technologies is
required to give a principled answer to this sort of challenge.
Such questions imply that research on extended cognition has important ethical (as well as
legal and political) implications that have so far scarcely been explored. One example of such
questions: If our minds really do incorporate as proper parts technologies owned by others,
might deprivation of them count as a sort of enforced cognitive impairment? Such questions
might have very important implications for our usage, design and governance of cognitive
technologies in the future.
1.4. Enactive Cognition
The enactive approach to cognitive science recognises a crucial inter-dependency between
an autonomous agent and the world it inhabits. Cognitive activity is wholly defined neither
by the agents nor their environment, but it emerges from their interaction.
This core observation lies at the heart of the enactive approach, and forms the basis for the
mode of analysis and understanding central to enactive thinking (Di Paolo, De Jaegher &
Rohde, 2010; McGann, De Jaegher & Di Paolo, 2013; Thompson, 2007). For enactive
theorists, cognition is founded in a system's autonomous organisation (Di Paolo, 2005;
Thompson, 2005, 2007; Varela, 1979, 1997; Weber & Varela, 2002), which is to say that the
operation of the system's own components give rise to and support their own organisation.
This continual process of self-production provides the drive for the system's interaction with
its environment. Such self-determinant activity, emergent from the system's organisation, is
seen by some as the hallmark of agency and normativity for the enactive approach (Bickhard
10
& Terveen 1995; Weber & Varela, 2002; see also Barandiaran, Di Paolo & Rohde, 2009).
Other enactivists have contested whether the notion of normativity does indeed emerge from
merely biological basis; others claim that normativity appeals essentially to social practices
(Hutto & Satne, pending, Heras-Escribano et al. 2013). In addition, the necessity of
autopoiesis to be at the centre of enactivist approaches to mind has recently weakened; the
label has been applied more liberally. In particular, some accounts that ignore or downplay
autopoiesis but share the original theory’s emphasis on environmentally-situated bodily
coupling in place of internal representation were considered enactive as well (O Regan o
2001; Hutto & Myin 2013). However, for the purposes of brevity we will discuss the classical
approach to enactivism in this section.
Enactive autonomous systems are perforce embodied and instantiated in the physical and
social processes that make up their network identity (Di Paolo & Thompson, forthcoming,
Kyselo & Di Paolo, 2013). In order to maintain themselves autonomous agents must
coordinate their activity with their environment. Opportunities must be taken and
constraints or demands met, such that the system maintains its organisation, which will
inevitably be precarious, facing constant potential dissolution (Di Paolo, 2005; 2009). In
doing so, the relevant aspects of the environment are effectively incorporated into the on-
going actions and dynamics of the system. This coordination of the agent with its
environment is termed sense-making (Di Paolo et al., 2010; Thompson & Stapleton, 2009).
The concept of cognitive activity as sense-making challenges internalist modes of thought
and commits enactive theorists to a “wide” conception of cognition. To understand the mind
we must understand not just the agent, but the value-driven, normative interactions between
the agent and its environment.
Social interaction is a special case of such engagement with the environment, in which the
agent's activity is not simply coordinated with a physical world but with another agent.
Sense-making in such situations is not simply coordinated but negotiated, a process termed
participatory sense-making (De Jaegher & Di Paolo, 2007). Where effective social
coordination is achieved, participatory sense-making produces meaning and cognitive
activity that is shared across the participants, and must be understood as much in terms of
the conversation or interaction as a whole as the combined activities of the individuals
involved.
From an enactive perspective, then, the cognitive system is not a fully-formed, stable thing,
but a collection of tendencies that dynamically interact with the environment. Cognitive
activity is somewhat like a dance, or a handshake – something that only exists during its
enactment, not something that is kept stored or continually present, and then activated or
switched on as required. Viewing cognition as such a dynamic, almost ephemeral thing,
means that enactive research tends to draw heavily on the tools of dynamical systems theory,
and emphasise the manner in which behaviour is structured by skills (stable, reliable
patterns of goal-directed behaviour) – examining change over time, the processes of
emergence, transformation and disappearance, rather than attempting to catalogue or
categorise specific cognitive functions, or identify specific contents of cognition that might be
examined.
11
1.5. Distributed Cognition
Distributed cognition approach (DCog) does not focus on cognition as a property of an
individual organism/agent. Rather, it describes larger cognitive systems which may
encompass multiple individual agents and artefacts.
This approach distinguishes DCog from other approaches labeled here as ‘wide cognition’,
which treat an individual organism as a main unit of analysis, describing, for example, the
way an individual is embedded, connected to his/her environment or how his/her body
matters from the perspective of cognitive processes. In the case of DCog, the focus is on
heterogeneous elements (e.g., biological, material, discursive), which take part in sequential
or simultaneous processes of generation, transmission, and transformation of
representational states. DCog identifies those processes with cognition. It is also important
that DCog does not treat cognition as a passive representational process whose primary aim
is to create a model of the external world. Moreover, in DCog representational states are not
understood as mental states or other inner states of any individual agent. Examples of
representational states include meaningful gestures or poses of agents, written or spoken
information, visualizations displayed on screens, lines drawn on navigational chart and the
chart itself, but also non-symbolic cues which modify the behavior of agents (e.g.,
modifications of environment which make some actions more probable than others) (Zhang
& Norman, 1994, p. 87–122).
DCog may be perceived as a test of how far we can proceed in explaining cognition without
taking into account the central nervous system and all those processes which happen inside
the agent’s head. DCog is a response to internalism, which assumed that all important
cognitive processes take place inside the agent and that external factors associated with
cognition (including elements of the human social and material culture) may be taken into
consideration at later stages of research (Salomon, 1997). DCog shows that in many cases,
those excluded factors are responsible for a considerable reduction of complexity of cognitive
problems, which individual agents are faced with. They may also enhance human cognitive
capabilities or reduce the cognitive load.
In order to explain this, one may reflect on various cognitive functions of artifacts. First of
all, artifacts reduce the cognitive load of human memory – they may function as external
memory systems or automate selected actions. Furthermore, they may reduce complexity by
helping to integrate dispersed data, allowing the agent to focus only on the phenomenon’s
selected features or dimensions. As a result, they turn complex issues into more abstract
ones; that is, the agent may solve the problem performing fewer mental or manual
operations. Moreover, artifacts reify (or enact) innovations developed during the process of
evolution of larger, distributed cognitive systems.
Let us consider the example of modern nautical chart, a basic cognitive tool of pre-GPS-era
marine navigation. It is a device developed through decades of practice, which consolidates a
whole array of small local technical innovations introduced by successive generations of
seamen and navigators. It is also an ingenious “analog computer”, which integrates
navigational data and allows for computation-like operations, which take form of manual
manipulation and transformation of two-dimensional graphic representations. It is
important that all navigational operations could be performed without the chart and manual
12
manipulations performed thereon (one can imagine marine navigation as purely
mathematical computations), but this kind of approach would be far more complex from the
perspective of a human navigator and transmission of navigation skills would be hindered.
Hutchins (1995a) presents many other examples of analog computational devices.
In order to fully understand cognitive functions of an artifact, one should closely analyze
what part they take in processing of representational states, in particular larger cognitive
systems. DCog does not suppose any precise vision of larger cognitive systems. A particular
distributed system may encompass many individual agents or only one. It may include
material artifacts and symbolic media, but it is not determined what types of material objects
it should include or what their function is. Literature includes such examples of distributed
cognitive systems as teams of nautical navigators, financial brokers’ offices, air-traffic
controllers, teams of surgeons, and crime scene investigation teams. In each case, an
important part of a system consists in human personnel utilizing overlearned problem-
solving heuristics and means of work coordination, but technical equipment (tools, material
external representations, coordinating artifacts) also plays an important role. In distributed
cognitive systems, one cannot distinguish between the core and the periphery. A given object
(biological, material, cultural) will be considered a part of a system only if it fulfills an
important function.
Representatives of DCog stress that a larger cognitive system works in a different manner
and has different properties than individual agents who may be part of the system. In other
words, the effects of actions performed by a larger system is not a sum of the effects of
actions performed by the human and nonhuman “components” of this system. Let us
consider the Chinese room known from Searle’s famous thought experiment as a model of a
distributed cognitive system (Searle, 1980). In this case, an individual system is the human
agent inhabiting the room. The individual does not know Chinese. One cannot attribute the
property “knows Chinese” to the individual, nor to any object gathered in the room (e.g., the
instruction of symbol usage, baskets with symbols). But the room as a whole (agent +
objects) functions as an interlocutor who is of communicating with agents using Chinese.
DCog reinterprets in a radical way many basic categories of cognitive science and
epistemology, including cognition, representation, cognizing agent and its boundaries. For
instance, according to DCog, a process is not cognitive simply because it happens inside an
individual, nor is a process non-cognitive because it happens within the interactions among
many individuals. One can put forward a complaint that DCog abuses epistemological
categories which have a standard and consistent use. Some researchers, including R. Giere
(2004), suggest that claims of DCog can be expressed also in the frame of more traditional
approaches in the field of cognitive science or epistemology. Still, the reduction of complexity
of the cognitive task has to be accounted for, and this is what the traditional framework
usually ignores.
It should be noted that the previous analyses in the field of DCog have usually focused on the
function of human collectives equipped with more or less advanced artifacts performing
routine cognitive functions. Their goal was to solve well structured problems with maximum
reliability and high accountability. DCog is very often associated with actor-network theory
(Latour & Woolgar 1979, Latour 2010, 2011) and with social engineering, understood as
building artificial, stable and reproducible social systems. The majority of the systems known
from the DCog literature are characterized by high redundancy and a kind of “prodigality”
manifested in the fact that they contain a number of control loops compensating for the
errors of the subsystem and ensuring not only confidence in the obtained results, but also the
13
possibility that eventual errors will be used for improving the system (it can be noted that the
process of learning in the case of a broader cognitive system is not identical to staff acquiring
knowledge) (see, e.g., Hutchins, 1995b, p. 265-288). However, one can specify DCog systems
that have been designed not for reliability, but for speed or for creativity. Then utilized areas
of application of DCog are design evaluation of digital environments, human-computer
interaction, information visualization processes, human factors in aviation, multimodal
interaction (see: Perry, 2003, p. 193-223; Woods, 2003, p. 37–53; Ross et al., 2007).
14
1.6. Interrelations
Real cognitive episodes can be seldom profitably understood in terms of isolated goings-on
in brains alone. Instead, an analysis of how brains, bodies, and equipment are deployed in
wider settings is required.
Even if one can analytically distinguish extended, embodied, situated, enacted, and
distributed approaches, they are not mutually exclusive, and a large body of research exhibits
properties specific for many of the above approaches. For example, a dynamical and
distributed account of linguistic activities may link the ideas of extended cognition and
enactivism (Cowley 2009). Language is then not couched in terms of symbolic processing
but of “languaging”–an ongoing social, distributed activity of multiple agents.
It should be noted that, for example, DCog cannot be automatically incorporated into the 4E
approach (embodied, embedded, extended, and enacted cognition) that takes a biological
organism as its focus. Notably, a biological account of DCog, which relates it to the tradition
of biosemiotics, like in the work of Stephen Cowley, allows for such integration. In other
words, to link DCog with other 4E frameworks, one needs to build a bridge between a more
biological account of cognition and cultural cognition. Below, in section 2.3.1, we describe
mindreading as a multi-faceted phenomenon. One of the ways to frame mindreading is to
understand it in terms of mindshaping (Zawidzki 2013), which has both institutional and
biological underpinnings. In this respect, Zawidzki’s account of mindshaping blends easily
with a more cultural-level based account of minds as artifacts of social institutions
(Castelfranchi 2013).
At the same time, some argued that distributed cognition can usually be subsumed under the
umbrella of extended mind (Dror & Harnad, 2008), although some of research may also be
seen as focusing on joint or shared cognition, in which the whole cognitive system can no
longer be analysed on the level of individual agents (the unit of analysis is different). In such
cases, the distinction between DCog and the extended mind seems to be altogether blurred.
For example, the sociologist Donald MacKenzie (2009) argues that the locus of decision-
making is usually not an individual trader, nor a group of traders working for a single
company, nor even a group of traders working for a single company together with their
computer models, but a group of traders working for a single company with their computer
models and with their competitors. Information about market events, trends and
opportunities is passed from traders at one company to traders from another through
informal contacts and personal networks, and this information then informs decision-
making by all of them. MacKenzie’s work is based on an ethnographic study he undertook
with a hedge fund. Similar results have been obtained by Buenza and Stark (2008).
That it is easy to integrate various wide approaches and blur the boundaries between them is
obviously not a sign of the weakness of wide approaches; rather, it seems that they easily
cluster together, and that they can be further developed to form a more extensive framework.
There are already some attempts to integrate, for example, embodied and extended
approaches (e.g., Clark, 2008; Borghi & Cimatti, 2010; Borghi et al. 2013; Walter, 2013).
Although a great deal of theoretical work has been developed in this area there remains great
scope to extend the analysis beyond the range of scenarios in which these ideas have been
applied.
15
2. Applications in Research on Social
Intelligence
2.1 Embodied Joint Action
Joint action may be roughly defined as “any form of social interaction whereby two or
more individuals coordinate their actions in space and time to bring about a change in the
environment” (Sebanz et al. 2006b). The body plays an important role in the cognitive
mechanisms that enable two or more agents to coordinate their actions in performing joint
actions.
The importance of understanding of the principles of joint action is at least twofold. Firstly,
joint action and the mechanisms underlying it may be considered as the cornerstones of
social cognition as such (e.g., Sebanz et al. 2006b). Secondly, embodied approach to joint
action (introduced below) has already proven its therapeutic utility in treatment of children
with autism and affective disorders (e.g., Kim et al. 2008; Koelsch et al. 2010).
Standardly, joint action is characterised in ways that ignore the body. Accordingly, standard
attempts to say what distinguishes joint actions from parallel but merely individual actions
focus on structures of amodal, propositional representation or commitment (e.g. Bratman
1992; Gilbert 1992; Searle 1990; Tomasello et al. 2005). By contrast — and in line with
current embodied turn in cognitive science (see Gibbs 2005 for review) – the body plays a
role in much research on the cognitive mechanisms that enable two or more agents to
coordinate their actions in performing joint actions (recent reviews include Knoblich et al.
2010; Vesper et al. 2010). For instance, coordination in joint action may depend on
interpersonal synergies (Ramenzoni et al. 2011) and may exploit joint affordances
(Richardson et al. 2007). The aim of this white paper is to explore and underline the role of
the body in joint action.
Accordingly, further role for the body in research on joint action depends on motor
representations, which are representations with a bodily, non-propositional format. There is
evidence that in joint action, there can be motor representations concerning not only one’s
own but also one’s partner’s actions (Kourtis et al. 2010 , 2013). As this suggests, in joint
action it is sometimes necessary to inhibit planning or performing another’s action (Sebanz
et al. 2006a). In general, performing one action while observing a complementary action
produces interference (e.g. Kilner et al. 2003). Given this, we might guess that where an
agent is engaged in joint action involving complementary actions and motor representations
concerning the others actions occur, these representations would interfere with the agents
performance and so hinder coordination.
However, it turns out that the opposite is true, at least in some cases. For a variety of joint
16
actions including passing an object, ballistic actions and playing a piano duet, motor
representations concerning another’s actions can facilitate interpersonal coordination
(Kourtis et al. 2013 , p. 9; Loehr and Palmer 2011; Novembre et al. 2013; Vesper et al. 2013;
Vesper et al. 2013). But just here we should be cautious in drawing conclusions linking motor
representation with joint action. Sartori and Cavallo (2011) provide evidence that, in
competitive actions, there are motor representations concerning an opponent's action as well
as for ones own. So is there a distinctive role for motor representation in joint action, or does
motor representation enable interpersonal coordination in ways that are insensitive to
whether the context is competitive or collaborative? That motor representation does play a
distinctive role in joint action is indicated by findings that in joint action there is sometimes
a single outcome which is represented motorically by each agent (Loehr et al. 2013; Ménoret
et al. tted; Tsai et al. 2011). This supports the further hypothesis that when an agent is
involved in joint action, there is sometimes a single planning-like motor process for both her
own and the other agents (or agents) actions. If this is right, it may be necessary to appeal to
motor representation not only in explaining how two or more agents coordinate their
actions, but also in giving an account of what, exactly, joint action is. At this point we face a
challenge. In what ways, if any, should standard characterisations of joint action be modified
in the light of these and other findings about the role of the body in the coordination of joint
action?
2.2 Cognitive artifacts and ICT (Cloud-Tech, E-Memory and
Virtual Identities)
Internet-mediated technologies offer the potential to change the way that knowledge is
represented and the new possibilities for interaction they make possible.
Donald orman defines cognitive artifacts as “artificial devices that maintain, display, or
operate upon information in order to serve a representational function and that affect
human cognitive performance.” ( orman 1990; for a useful overview of perspectives on
cognitive artifacts, see Heersmink 2013). There is little doubt that the new regimes of
internet-mediated technologies offer the potential to change the way that knowledge is
represented and the new possibilities for interaction they make possible.
Although cognitive technologies have been with us for a long time, from at least the
invention of writing they have had, and continue to have important implications for the way
we think, who we think we are, and how we organise our societies. Historically speaking,
cognitive technologies have:
1. Increased and made more various human cognitive capacities and enhanced human
efficiency on many tasks.
2. Affected how people and society go about achieving their goals, indeed arguably
changed what those goals are.
3. Highlighted and transformed how we human beings view ourselves and our goals.
4. Changed the way we cognize both individually and in groups.
5. Given rise to new forms of cognition (such as distributed cognition), perceptions and
mental states that are either distributed across or even embodied in cognitive
technology.
17
Arguably the deployment of these technologies has also caused dramatic shifts in how
human beings think of themselves and have sometimes made possible revolutionary societal
change sometimes annihilating cultural practices that have been central for centuries or even
millennia. Oral Poetry, once one of the central mechanisms for cultural transmission is
effectively annihilated by the development and widespread dissemination of writing. Indeed
even certain forms of cognition such as the ability to memorize, improvise and perform
thousands of lines of poetry is now beyond contemporary cognitive abilities for almost all
living humans.
We are now going through a phase of much change led by development and rapid
innovations in the field of ICT which may be having similarly fundamental implications the
basic media of mind that human beings use. Much of the new technology is Cloud-Tech,
technology that makes use of an ever-present internet to provide data system and services
via mobile networks making possible lightning fast interactivity with vast warehouses of data
behind it. Alongside the interactive possibilities this makes available are also unprecedented
possibilities for surveillance of the individual. This surveillance has obviously problematic
dimensions for the privacy of the individual but can also offer novel cognitive affordances as
the technology becomes ever more closely fitted to our individual cognitive profiles. This
intimate and ever-present nature of much Cloud-Tech – packaged now through our smart
phones and shortly by wearable technologies like Google Glass — means that ICT technology
is bound ever more closely to our organic cognitive faculties. This is likely to be already
having important cognitive implications. It is often easier to consult our gadgetry to
“remember” a fact by Googling than scratching our heads and wracking our brains.
A broad class of the mobile devices we now use on an everyday basis, such as megapixel
cameras, smart-phones, iPods and iPads and wearable devices provide E-Memory
technology, that is to say functions which have broadly similar functions to the memory
architecture of our brains. However the functions do not simply replace our organic
functions but augment, extend and in some circumstances may diminish those cognitive
functions. (We will not discuss in detail the heated question of cognitive enhancement and
diminishment here but it is clearly one of central interest to the future of our societies.)
Rather we concentrate on some particular properties and direct implications of the
widespread usage of the new Cloud-Tech and especially E-Memory.
18
2.2.1 E-Memory and Cloud Tech
E-Memory (and Cloud) technologies have been argued to present (Clowes, 2013) the
following collection of novel properties: Totality, Incorporability, Autonomy and
Entanglement. Individually and collectively these properties promise (or threaten) important
cognitive implications as we increasingly adopt and incorporate them into our cognitive
lives. Consideration of them helps us consider how human cognitive potential is likely to
change over the coming years. Let us consider the properties in a little detail:
1. Totality: E-Memory promises to record our everyday activities on a scale and with a
fidelity and completeness that would have been practically unimaginable under
previous regimes of memory technology. Cloud Tech brings to this picture an ever-
present repository of “memory traces” that can be retrieved and brought to bear on
ongoing cognition.
2. Practical Cognitive Incorporability: E-Memory and Cloud-Tech are rapidly
becoming a constant context for many cognitive processes. The devices which present
this technology increasingly possess a transparency-in-use that makes them ready
competitors (or complements) with certain of our internal resources. It is often as
easy, or easier, to rely upon these technologies to carry out certain cognitive tasks
than internal organic resources. They are thus poised for deep and pervasive
integration with our organic cognitive systems.
3. Autonomy: E-Memory repositories increasingly do not merely store data but
actively process it. Thanks to tagging, indexing and AI systems resident on our
devices and in the cloud, we can expect E-Memory systems to not merely store and
retrieve memory-traces and other information, but process, restructure and re-
present them in ways that are independent of, but tightly coupled to, our native
cognitive profile.
4. Entanglement: E-Memory often tracks interactions between people (or people and
organisations). Many of the cloud services we use rely on collective data-storage and
interactions collected from a mass of users. However it is presented back as
narrowcast and tightly coupled to the activities of us as individuals. The data that
composes many E-Memory stores and cloud services is thus inherently two-sided in
relying on the interaction of these collective and individual dimensions.
To focus on E-Memory and totality it is already claimed in some quarters that trends
toward life-logging can already be regarded as important cognitive enhancements that
change the nature of human memory (Bell & Gemmell, 2009) making possible new forms of
self-knowledge and self-understanding. Others have argued that the property of totality, the
19
capacity in some sense to “remember everything” may already be changing and perhaps
undermining certain aspects of our common humanity (Mayer-Schönberger, 2011).
Without prejudging the cognitive and societal implications of such technologies it is difficult
not to treat technologies such as Google Glass as being a departure in the history of human
memory. A wearable technology that can pipe HD video directly from a owner’s viewpoint to
the cloud and then allows that video to be automatically time and location stamped, and in
other ways categorized, is sure to have societal implications. Because of the practical
incorporability of these wearable artefacts there are likely cognitive implications which
are every bit as profound cognitively as the societal effects (really predicted effects) which
have been much in the news in recent times. Mobile Internet available tablets and smart-
phones are becoming increasingly ubiquitous throughout Europe and wearable devices like
(or very different) from Google Glass are sure to follow. Such technologies have caused
concern in some quarters which fear that as we rely ever more on these technologies our
biological resources are placed under threat (Carr, 2010). However there has been so little
real research, either practical or theoretical, into these implications of these technologies
that a proper assessment is currently not possible.
Some of this research needs to focus on the deep philosophical and theoretical implications
of this technology. In fact the philosophy of cognitive science has already developed
significant resources which may help us analyse these developments. As this white paper has
discussed, human beings tend to extend their cognitive powers by incorporating technologies
into themselves. Under some circumstances cognitive technologies can even be considered
proper parts of our minds (Clark & Chalmers, 1998). However there are interesting
theoretical differences about how exactly the particularly cognitive contribution of
technologies should be considered. Technologies like Wikipedia and perhaps Google Search
might be considered a sort of cognitive commons (Dror &Harnard, 2008); that is, a kind of
collective cognitive resources maintained by and contributing to the cognitive abilities of a
mass of users. Other might be so intimate to individual users, so tailored to, and incorporate
in individual cognitive processes that they should be considered as proper parts of individual
human minds.
Theoretical questions about when exactly a technology should be considered part of an
individual’s (or collective’s) mind may start to have very practical implications.
2.2.2 Virtual Reality and avatars
Improvement of human abilities, in the bodily, mental and social dimension, can be
accomplished in several ways. Under the right conditions an avatar in a virtual world can
20
substantially enhance the abilities of the user – the person who owns and operates it.
As William Bainbridge argues in Transhumanist Reader, avatars demonstrate one way in
which enhancement is not merely a matter of increasing the effectiveness of a person in
taking action, but also can mean an altered form of consciousness that expands
opportunities for new social experiences and, in some cases, escape from conventional
system of moral constraints (in More and Vita More 2013: 91).
There already exists a strong body of research showing that virtual beings affect our bodily
and sensorimotor abilities, as well as reshape our body scheme. Many studies of first-person
shooters (see: Ash 2013, Galloway 2006) show that operating an avatar not only equips the
player with new affordances, but also affects his/her memory and speed of volitional
attention switches. Moreover, as Amorim showed in his What is my avatar seeing?: The
coordination of "out-of-body" and "embodied" perspectives for scene recognition across
view, using an avatar for coordination of “out-of-body” and “embodied” perspectives is
extremely helpful in relation to social perception, in order to understand what another
individual sees.
This brings us to the issue of how avatars may affect our sense of identity and social presence
in various ways. Sherry Turkle once said that the roles played by a constructed identity in
virtual role-playing games are a very important aspect of personality development because
they provide the opportunity to express unexplored parts of the Self. As Turkle claimed in
Constructions and Reconstructions of Self in Virtual Reality: Playing in the MUDs,
involvement in the virtual world may invoke “second chances" for adults to resolve their
identity issues as well as other complicated issues related to their involvement in various
kinds of social relations (1994:159). Turkle argued that virtual life could lead to changes in
the understanding of and involvement in social relationships, as well as influence changes in
social practices in the real world.
However, a lot still remains to be done in this area. For instance, De Jaegher, Di Paolo, and
Gallagher (2010: 442) suggest that virtual beings could in the future provide excellent
material for neurophenomenological models of how we connect with others on the social
level. They suggest close analysis of experiences of virtual characters or socialized robots.
They consider it promising for further studies of participatory sense making, as well as the
role of imagination and mirroring behaviours in the social sphere.
One central question is: How and to what extent will the new ICT technologies have novel
cognitive potentialities, and what might we expect from the potentialities they make
available? What role and what degree of agency do we have in the deployment and use of
these technologies? Ultimately perhaps what kinds of human beings will we be when we
21
transfer a big part of our life to Virtual Reality or when the availability and our familiarity
with technologies like Google Glass are as everyday as current use of paper and pencil.
2.3 Social Cognition and Culture
2.3.1. Mindreading
In philosophy, as well as in developmental and comparative psychology, social cognition
has traditionally been understood in terms of “mindreading”. This is a term of art that is
often used interchangeably with “mentalizing” and “theory of mind”.
The guiding concept behind the notions of mindreading, mentalizing, and theory of mind is
that an individual’s success at predicting the behavior of and coordinating with her
conspecifics depends on understanding the unseen psychological causes responsible for their
behavior. On this orthodox conception of human social cognition, differences in the
sophistication of social interaction, e.g., cooperation and coordination, across phylogenetic
and ontogenetic time, can be traced to the sophistication of interactants’ theories of mind.
For example, whereas nonhuman primates are restricted to an understanding of
conspecifics’ goals, perceptions, emotional states, and behavioral dispositions, adult humans
can also attribute beliefs, intentions, and other propositional attitudes, arranged in rational
plans causally responsible for their interactants’ behaviors.
Recently, this orthodox, consensus understanding of social cognition has come under
considerable pressure. For example, Hutto (2008) argues that non-human, human infant,
and even typical, quotidian adult human social cognition does not depend on interactants
understanding the unseen psychological causes of each other’s behavior via sophisticated
theories of mind. Rather, he argues, most social cognition involves an embodied
understanding of perspectives–an ability to track relations between whole organisms and
salient features of the environment,–what Hutto calls “intentional attitudes”. Understanding
of hidden psychological causes, especially propositional attitudes, depends on prior
enculturation in culturally specific narrative practices, and is useful only in making sense of
deviations from or exceptions to typical behavioral patterns.
Hutto’s (2008) alternative to orthodox conceptions of social cognition has a number of
motivations. For Hutto, there is little evidence that the non-linguistic behavior of both non-
humans and humans is caused by propositional attitudes, as these have been understood by
the philosophical tradition: discrete, content-bearing, representational mental states that are
causally responsible for intelligent behavior. Hence, there is little reason to think that
attributing such mental states is necessary for successful coordination. Hutto also raises
problems for how concepts of such states could possibly have arisen, in phylogeny or
ontogeny, prior to the linguistic, narrative practices that make them salient for human
children. Finally, he notes that much successful coordination can be explained by appeal to
the attribution of intentional attitudes, rather than hidden mental states: interactants
succeed by tracking each other’s embodied, perceptually available, whole-organism relations
to salient features of the environment.
This alternative understanding of human social cognition draws on the phenomenological
tradition in philosophy, according to which minded behavior is not typically a matter of
representing facts about the world, and hence, tracking minded behavior should not involve
22
attributing such representational capacities, in the form of mental states like propositional
attitudes (Gallagher and Hutto 2008). One advantage of this perspective is that it
undermines the traditional assimilation of successful social cognition to solving the
apparently intractable, philosophical “other minds” problem, i.e., determining whether one’s
fellows have mental states, and if so, which ones, based only on observable behavior.
However, one does not need to endorse the whole phenomenological program, along with
wholesale anti-representationalism about human cognition, in order to raise problems for
the orthodox, mindreading-centered conception of social cognition.
It is widely recognized that propositional attitudes, like belief and desire, bear very tenuous
relations to observable circumstances and behavior. The reason is holism: even on orthodox
views, particular propositional attitudes lead to behavior only against a background of
indefinitely many other propositional attitudes (Morton 1996; 2003; Bermudez 2003; 2009;
Apperly 2011). For example, a belief that it is raining will issue in umbrella retrieval only
when conjoined with a desire to stay dry that is stronger than competing desires, appropriate
beliefs about the location of an umbrella and the relative costs of retrieving it, etc. This
feature of propositional attitudes raises significant problems for the orthodox conception of
social cognition as dependent on mindreading, and especially for the widely held assumption
that successful navigation of the social world requires attributing propositional attitudes.
The reason is that it raises issues of computational tractability. Most successful social
interactions take place seamlessly and dynamically at relatively short time scales. It seems
unlikely that successful interpreters search the immense space of propositional attitude
attributions compatible with brief, observed bouts of behavior, in time to arrive at
attributions accurate enough to support successful interaction. For this reason, a number of
theorists who share neither Hutto’s general skepticism about representationalism nor his
allegiance to the phenomenological tradition, have recently raised similar worries about the
orthodox conception.
Most prominently, Apperly & Butterfill (2009) and Apperly (2011) have argued for a
distinction between System 1 and System 2 mindreading. The distinction is inspired by
theories of rationality (Stanovich 2011; Evans 2003) and numeracy (Feigenson et al. 2004)
that distinguish between two kinds of cognitive response to similar tasks. One kind of
response is rapid, unconscious, automatic, beyond deliberate control, phylogenetically old,
and available very early in ontogeny. For example, infants and non-human animals are
sensitive to differences in the numerosity of small groups of objects, and can process such
information extremely rapidly to succeed in experimental tasks. Likewise, human adults
come to very rapid, unreflective judgments in reasoning tasks, which often violate norms of
rationality, yet lead to adaptive responses in ecologically plausible contexts. The second kind
of response is slow, conscious, deliberate, reflective, phylogenetically very new (because it is
dependent on enculturation in appropriate cognitive practices), and available only in late
childhood, after formal schooling. For example, older children and adults can also come to
reflective and much more precise judgments about the numerosity of groups of objects of
unlimited size, i.e., by counting. And older children and adults can come to reason in
normatively acceptable ways, when trained in appropriate techniques, and given time to
apply rules reflectively. The first kind of response is thought to be mediated by “System 1”
number cognition and rationality; the second kind by “System 2” number cognition and
rationality. Apperly and Butterfill suggest that there are likewise two systems of
mindreading: System 1 mindreading supports successful social cognition in rapidly evolving,
23
dynamic quotidian contexts, and does not require the attribution of full-blown propositional
attitudes. This kind of mindreading mediates the socio-cognitive feats of non-human
animals, very young infants, and older children and adults engaged in everyday interactions.
System 2 mindreading involves the attribution of propositional attitudes as such, and
includes an appreciation of holism, and thus, the tenuous connection between propositional
attitudes and behavior. This kind of mindreading is available only in late childhood and
adulthood, and unavailable to nonhuman species.
Apperly and Butterfill’s hypothesis is extremely promising because it suggests a plausible
explanation of recent findings in developmental and comparative psychology; however, it
also raises a number of difficult questions. The developmental psychology of social cognition
has recently undergone a revolution with the introduction of non-verbal versions of the false
belief task. The standard false belief task, used to gauge children’s theory of mind
competence for several decades, depends on verbal probes of children’s knowledge about a
target’s mental states. Based on such studies, it was widely held that children do not
understand that interpretive targets can have false beliefs until they reach the age of four.
However, employing looking time as a measure of preverbal infants’ expectations, Onishi
and Baillargeon (2005) showed that infants as young as 15 months of age appear to
anticipate that interpretive targets will act on false beliefs. This result has proven robust: all
find congenial results in infants below two years of age (Baillargeon et al. 2010). On the most
sanguine interpretations of this data, infants master the concepts of belief and other
propositional attitudes long before they can even speak; in fact, many theorists claim this
data shows the concept of belief to be innate, and the delay until age four or later in its fully
competent deployment to be due to performance limitations caused by relatively delayed
maturation of domain general capacities like attention and memory (Leslie 2005). However,
as Apperly (2011) argues, these experiments do not show mastery of the full-blown concept
of belief, of the kind employed in System 2 mindreading. In fact, neither do the long-used,
verbal versions of the false belief task. Children can pass such tests without understanding
that particular propositional attitudes yield behavior only relative to an indefinitely broad
background of other propositional attitudes. And they pass such tests long before
understanding the opacity of propositional attitude attribution: the fact that individuals can
represent and believe the same facts under individually variable modes of presentation
(Apperly 2011). The distinction between System 1 and System 2 mindreading nicely handles
the recent experimental results: very young infants employ System 1 mindreading, as do
some non-human species, as well as adults in everyday, embodied social cognition (Apperly
2011).
While promising, this proposal raises a number of issues. First and foremost, there is the
question of how to characterize System 1 mindreading. System 2 mindreading is the
competence that has long been taken for granted in philosophy and psychology - what
Apperly terms the “normative account” (2011). This is the understanding of full-blown
propositional attitudes as such: concrete, mental causes of behavior, with content,
represented via individually variable modes of presentation, and tenuous, holistically
mediated connections to observable circumstances and behavior. What features of this
theory of mental states are lacking in System 1 mindreading? The best way to answer this is
to determine the minimal competence required for the kinds of socio-cognitive feats of which
very young infants, some nonhuman species, and adults in unreflective, dynamic, quotidian
24
interactions are capable. It is not immediately clear how to characterize this capacity.
Apperly and Butterfill (2013) suggest that such social cognition is mediated by an
understanding of what they term “registrations” - unobservable mental states, with causal
control over behavior and some form of representational content, yet with much more direct
connections to observable circumstances and behavior than full-blown propositional
attitudes, i.e., not holistically constrained by indefinitely broad networks of background
propositional attitudes, and no individually variable modes of presentation. But once one
goes this far, why understand System 1 mindreading as any kind of mindreading at all? Why
assume that infants, non-human primates, and adults in unreflective, dynamic, quotidian
interactions have to represent the hidden causes responsible for behavior? Perhaps Apperly
and Butterfill’s “registrations” can be understood as more like relations between interpretive
targets and worldly states of affairs, rather like Hutto’s intentional attitudes (Zawidzki 2011).
Another important question raised by the hypothesis of two mindreading systems concerns
their evolution. System 1 mindreading is supposed to be phylogenetically old, and largely
shared with some nonhuman species. But given how effective System 1 mindreading can be,
it is unclear why System 2 mindreading - the attribution of full-blown propositional attitudes
as such - would ever have evolved. Why go to all the trouble if our ancestors were already
adept at predicting and coordinating with each other’s behavior using System 1
mindreading? This is especially puzzling if we hypothesize some evolutionary divergence
between our ancestors and other primates even at the System 1 level. It is plausible that,
unlike, say, chimpanzees, humans have System 1 mindreading components devoted to
various forms of joint action, of the kind that support our unique cooperative and
coordinative feats. Such behavior occurs seamlessly and rapidly, in dynamic contexts; so it is
unlikely that it requires careful, reflective, System 2 mindreading. In fact, Butterfill (2013)
suggests that System 1 capacities for joint action may have actually made possible later,
effective System 2 mindreading. Furthermore, Zawidzki (2008; 2013) argues that human
populations are distinguished from our closest non-human relatives by an array of effective
“mindshaping” techniques, aimed at making potential interactants more alike, and hence
easier to interpret. For example, only humans engage in “overimitation” ( ielsen
Tomaselli 2010), that is, copying each other’s fine-grained, apparently non-functional
behaviors. Only humans engage in constant and pervasive pedagogy. Only humans set up
elaborate normative regimes, and linguistic, narrative constructs, and then pressure each
other to conform to them. Such practices tend to make human populations easier to interpret
using relatively simple, System 1 mindreading systems. And this, again, raises the question,
why bother evolving System 2 mindreading?
One possibility explored by Zawidzki (2013), is that System 2 mindreading, i.e., the
attribution of full-blown propositional attitudes as such, evolved to play a justificatory,
rather than a predictive function. That is, rather than help predict the behavior of potential
interactants, for which System 1 mindreading supplemented by mindshaping was sufficient,
attribution of full-blown propositional attitudes evolved as a tool for situating interactants in
a normative space - as committed and entitled to various discursive and non-discursive
moves, given prior such moves. Inspired by Brandom’s (1994) theory of discursive practice,
this idea finds some support in social psychology as well. Bruner (1990) points out that many
everyday uses of propositional attitude attributions are triggered by behavior that deviates
from a canonical cultural pattern, and function to excuse or at least make sense of the
25
behavior, e.g., when a person appears to break some norm, as when one fails to fulfil a
promise, sanctions can be mitigated if one can provide an excuse in terms of what one was
thinking. There is some empirical evidence in favor of this hypothesis: Malle et al. (2007)
showed that, when explaining behavior, adults are much more likely to appeal to reason
explanations, i.e., propositional attitudes, when they are motivated to make the behavior
look good. Without such motivation, they tend to explain behavior in terms of situational
factors, like character traits.
The mindshaping hypothesis, and the role of propositional attitude attribution in justifying
apparently counter-normative behavior to one’s interactants, suggest a very precise sense in
which human social cognition is “wide”: much of the work required to predict our fellows
involves not intracranial socio-cognitive resources, like so-called “theory of mind”, but
rather, external cultural practices aimed at making our likely interactants more familiar and
easy to anticipate. Culture seems to function as a classic example of “epistemic action” (Kirsh
1996; Clark 1997). Rather than tackle the seemingly computationally intractable task of
predicting our conspecifics by building ever more complex, intracranial computational
capacity, natural selection seems to have developed systematic means of structuring the
social environment in ways that make it easier to predict using relatively simple intracranial
resources. The variety of mindshaping techniques employed by a given culture -
overimitation, pedagogy, norm construction and enforcement, the use of linguistic narrative,
e.g., myths, to construct “virtual” models for all to imitate - can all be seen as forms of wide
social cognition, aimed at regimenting the potentially unruly and intractable social
environment. If this hypothesis is on the right track, it suggests that the study of culture and
its effects on cognition must play a central role in the sciences of social cognition.
In fact, over the past fifteen years there have been many new researches exploring the impact
of culture on the development of human cognitive structures. Areas of knowledge that deals
with this subject are cross-cultural psychology, neuroantropology and cultural neuroscience.
What unites the above-mentioned disciplines is the emphasis put on socio-cultural aspects of
human cognitive abilities. This position is sometimes called as bio-cultural constructivism
[or bio-cultural co-constructivism (Baltes, Rosler, Reuter- Lorenz, 2006)] and it states, that
“that brain and culture are in a continuous, interdependent, co-productive transaction and
reciprocal determination” (Baltes, Rosler, Reuter- Lorenz, 2006, pp.3). Basic assumption is
that the structure of the human brain is not programmed a priori, but rather co-shaped by
widely understood sensory stimulation: the socio-cultural experience of the entity, its
environment, etc. Generalizing - the process of ontogenetic development is stimulated
genetically, environmentally and culturally. Consequently, this leads to the abandonment of
radical genetic, neuronal or cultural determinism or environmentalism and highlights the
simultaneous impact of all the above factors on human ontogeny and evolution.
The main challenge that lies in front of the researchers associated with this way of thinking is
the precisely distinction between “software” and “hardware”. In other words: we need to
answer to the question of which aspects of our cognitive determinants are the result of
evolutionary adaptation or cultural adaptation, which of them stem from the experience of
our species, and which are associated with the experience of a individual subject.
26
2.3.2. Two Case Studies: Imitation and Pretence
Pretence and imitation play a crucial role in human children’s social development. In both
philosophy and psychology, the current consensus emphasizes the way both abilities
contribute to children’s acquisition of mindreading abilities.
A core limitation of this mindreading-centered approach is that it focuses almost exclusively
on the role of these capacities in learning to interpret behavior, while tending to
systematically underplay social, cultural, and dynamic factors. In particular, both imitation
and pretence are ways of learning culturally specific types of behavior by actively
participating in simple social relationships. Below, we provide a brief overview of how the
mindreading consensus has tended to understand imitation and pretence, explain its
limitations for each kind of ability, and outline more embodied and extended approaches
that offer finer grained accounts of just these cultural, social, and dynamic factors.
Case Study 1: Imitation in Human Phylogeny and Development
Imitation is part of a broad network of social learning strategies that involve acquiring a new
behavior by observing another agent perform it. Generally speaking, imitation is distinct
from other social learning strategies such as mimicry and emulation because it requires an
appreciation of the relationship between means and goals (see Tomasello and Carpenter
2005 for discussion). On the one hand, mimicry involves the rote replication of another
creature’s behavior without an underlying appreciation of its goal. On the other, emulation
learning consists in learning about an environmental affordance by observing the outcome of
another agent’s behavior, but does not include an appreciation of how the agent’s specific
means contribute to bringing it about. In contrast to both mimicry and emulation, imitation
consists in reproducing another agent’s behavior by taking it as a model for how to bring
about a goal.
By the end of their first year of life, human children are sophisticated imitators of adults’
behavior. They are able to learn new behaviors both by faithfully reproducing fine-grained
aspects of an adult’s behavior and by identifying alternative means for bringing about the
same goal. Both aspects are helpfully illustrated by a famous study by Gergely, Bekkering,
and Kiraly (2002). In the study, 14-month-old children each watched a human adult do
something quite unusual—she touched her head to a panel on a table, and it made the panel
light up. Children in one group watched the adult perform this action while her hands were
occupied because she was holding a blanket around herself, while those in a second group
saw the adult perform the same action while her hands were unoccupied. Children in the
first group responded by simply using their hands to turn on the light panel, but those in the
second group carefully matched the adult’s behavior by using their heads to turn on the light.
Gergely and colleagues interpret this finding to show that young children are sensitive to the
contextual constraints under which an adult performs an action and imitate accordingly:
they use their hands when the adult’s hands are occupied because that is a more efficient way
of achieving the same goal, but they use their heads when the adult’s hands were unoccupied,
assuming that this was the point of the action. A complementary finding by Andrew Meltzoff
(1995) further underscores the flexibility of young children’s capacities for imitation. When
18-month-olds observe an adult perform a failed action with a novel artifact, such as failing
27
to pull a toy dumbbell apart because her fingers keep slipping off, they will imitate the
behavior by performing a successful version of the action, rather than simply mimicking the
adult’s failed action. This suggests that they are sensitive to the goal-related features of even
unsuccessful actions.
The mindreading-centered approach has exerted an outsized influence on research into the
phylogeny and ontogeny of human imitation. Consider an influential account of the
phylogeny of imitation developed by Michael Tomasello and his Leipzig research team (e.g.,
Tomasello 1999; Tomasello and Carpenter 2005; Tomasello et al. 2008). On their account,
human imitation is different from the social learning strategies of other social species
because it is closely linked to mindreading abilities in two main ways. First, they claim,
human imitators’ sensitivity to both the means and goals of another agent’s behavior reflects
a sophisticated understanding of other people as intentional agents, whose doings are
manifestations of hidden mental states, such as intentions. Second, human imitators have
species-distinctive motivations to identify with conspecifics as intentional agents like
themselves. These two factors combined allow human beings, from an early period of
development, to ‘see through’ the surface aspects of a conspecific’s behavior, identify the
intention behind it, and reproduce it appropriately in light of a tacit appreciation that one is,
like the other person, an intentional agent. In this way, Tomasello and colleagues argue,
imitation is a central driver of the evolution other sophisticated cognitive capacities, most
crucially cultural transmission and innovation. The capacity to replicate behavior faithfully
provides a basis for cultural transmission, while the ability to identify the intentions behind
it as well as alternative means for achieving them is a basis for innovation.
Mindreading-centered accounts of imitation have also had a profound influence on
developmental studies of young human children’s acquisition of abilities for interpreting
other people’s behavior. There are two broadly different ways of working out this idea. The
main difference between them is how each account explains the abilities that underlie
mature adult mindreading. On one account, mature mindreading involves using a ‘Theory of
Mind’, which enables interpreters to explain and predict behavior by drawing on tacit
theoretical generalizations. Andrew Meltzoff’s ‘Like Me’ hypothesis presents early human
imitation as a platform for acquiring a theory of mind (e.g., Meltzoff 2005, 2011). On his
account, imitation leads to later abilities for mindreading because it enables infants and
young children to recognize other people as ‘like me’, which provides a basis for generalizing
from one's own experience as an intentional agent to that of others. For instance, Meltzoff
(2005) claims that his 1995 finding on accidental actions can be explained on the hypothesis
that children generalized from their own experiences of failure and came to appreciate that
the adult was trying to pull the dumbbell apart. On the second kind of account, mature
mindreading is underwritten by capacities for mental simulation, which consists in
imaginatively putting oneself into another’s ‘mental shoes’ and using this identification with
the other person’s point of view to interpret, explain, and predict his or her behavior. Alvin
Goldman (2005, 2006) argues that mental simulation is the common cognitive basis for both
early human imitation and later mindreading. On his account, imitation requires simulating
the intentions behind another’s behavior, and this basic capacity for imaginative
identification is later expanded to include simulating more complex mental states, such as
beliefs and desires. For instance, Goldman explains Meltzoff’s 1995 finding by claiming that
children adopt the adult’s point of view, mentally simulate his repeated efforts, and come to
recognize that his or her intention is to pull the toy dumbbell apart.
28
There are two critical limitations to this general mindreading-centered approach to
imitation, one theoretical and one empirical. The theoretical problem is closely connected to
issues about holism and underdetermination: mindreading is cognitively demanding because
many mental states, including intentions, bear only a tenuous connection to behavior. With
respect to imitation studies, the problem is how young children manage to identify the
intention behind a novel behavior that they have never seen before, especially when it
involves the manipulation of a novel artifact, such as a toy dumbbell. In different ways, both
Meltzoff and Goldman try to explain Meltzoff’s 1995 finding by appealing to young children’s
own experiences of failure. But given that the children had neither seen the toy dumbbell
being used before nor ever used it themselves, it is difficult to understand how they could
identify the adult’s intention accurately (Zawidzki 2013, p. 189). Specifically, how did they
know the adult’s intention was to pull the dumbbell apart, rather than to do something else
with it, such as break it? The core problem is that recognizing that an action has failed does
not necessarily provide a basis for positively determining what the agent intended to do.
The second, empirical limitation is that mindreading-centered views of imitation do not
appear to offer adequate resources for explaining a recent set of findings on human
children’s dispositions to overimitate. Overimitation consists in faithfully replicating
arbitrary features of an adult’s behavior, including ineffective means or ones simply causally
irrelevant to its goal (see Lyons 2009 for discussion). For instance, standardly developing
children will imitate arbitrary tapping gestures that do not contribute to opening a box, both
after they have learned through individual trial and error how to open it efficiently without
tapping (Nielsen and Tomaselli 2010) and when they are already familiar with how to open
the box (Marsh et al. 2013). In these cases, children appear to be disposed to replicate fine-
grained aspects of behavior that are not strictly relevant to the goal-related intention behind
the action. Nielsen and Tomaselli (ibid.) explain this by hypothesizing that overimitation is
specialized for cultural initiation because it disposes children to imitate the contingent
features of other people’s culturally specific behaviors.
In terms of alternatives to mindreading-centered approaches, one promising hypothesis is
that infants and young children identify the goals of agent’s actions through a simple
“teleological stance” (e.g., Gergely and Csibra 2003). On this account, infants do not identify
the goal of another agent’s actions by imputing a hidden mental intention. Instead, they
presume that agents act efficiently and use this presumption to identify the way that the
observable features of a behavior contribute as means to an observable outcome. A key
strength of this account is that it avoids the issues related to holism and the
underdetermination of behavior because it does not present young children’s capacities for
understanding goals in terms of a further appreciation of inner psychological causes. A
closely related, and more recent suggestion by Zawidzki (2011, 2013) develops Gergely and
Csibra’s idea about a ‘teleological stance’ in order to better accommodate the findings on
overimitation. On this alternative account, imitation is a mechanism for mindshaping rather
than mindreading. According to Zawidzki, overimitation is best explained on the hypothesis
that young children are intrinsically motivated to match fine-grained features of an adult’s
behavior, thereby coming to conform to culturally specific styles of behaving. This
uniformity, in turn, ultimately makes behavior easier to interpret. On this alternative,
imitation does not require sophisticated mindreading at all, but is jointly underwritten by
29
capacities for parsing surface features of a bout of behavior into means and ends and
standing dispositions to treat other agents’ behavior as a model for one’s own.
In sum, the core shortcoming of mindreading-centered approaches is their inability to
explain overimitation and the closely connected problem of how young children succeed in
interpreting and reproducing novel behaviors with unfamiliar artifacts. The alternative
approaches offer simpler mechanisms than mindreading for explaining how imitators
interpret instrumental behaviors and more nuanced accounts for explaining imitators’
dispositions to replicate non-instrumental, culturally specific features of behavior. More
broadly, the core philosophical insight of these alternatives is that mindreading approaches
are too narrow because a great deal of human agents’ behaviors are shared by members of a
group, rather than being manifestations of an individual agent’s distinctive, ‘inner’
psychological motivations. This insight requires a conception of human social cognition as
‘wide’, extended, and embodied, one that closely links capacities for interpreting behaviors
with capacities for producing them and that gives a central role to processes of social
conformity in making behavior interpretable.
Case Study II: Pretence and Social Cognition
Pretending is an activity that adults and children routinely engage in. In developmental and
clinical psychology, it is one of the benchmarks of a healthy social development. For
example, we see underdeveloped spontaneous pretend play in children with Autism
Spectrum Disorder (Baron-Cohen 1995, Currie and Ravenscroft 2002).
In the philosophical tradition, the current consensus holds that pretending is a consequence
of the ability to represent an alternative scenario. In this sense, “pretence is similar to false
belief in that actions stemming from both mental states are directed at situations that do not
actually obtain” (Harris Kavanaugh 1993). On this view, pretence is assumed to be a type
of mental state one is in, where one needs to use symbolic capacities to act as if one thing
(e.g., a banana) was another (e.g., a phone). Hence, on the consensus view, pretence is
simply defined as requiring complex representational capacities, i.e., imagination,
conceptual manipulations or double knowledge of what is real and not real.
A distinguishing feature of this consensus view is that it closely links capacities for pretence
with mindreading, but there is disagreement about explanatory mechanisms that achieve it.
Leslie’s (1987) classic theory of pretence proposes that pretending involves
metarepresenting, that is, representing a representation (e.g., “I pretend that ‘banana is a
phone’”). When engaging in pretence, young children manipulate the concept ‘pretend’ and a
set of propositions that are not true. Leslie also argues that engaging in pretence depends on
the capacity to understand pretence in others, which in turn depends on an ability to
understand another’s mental state. In this way, according to Leslie, pretending requires a
theory of mind. Another popular theory that reflects the consensus view but stands against
Leslie’s metarepresentational model conceives of pretence in terms of behaving-as-if (Harris
1994, 2000, Lillard 1994, Nichols and Stich 2000). For instance, Lillard (1994) holds that
behaving-as-if consists in forming a mental representation of the pretence scenario, which
30
must be held in the pretender’s mind and projected onto reality for successful pretence
behavior to occur. According to Harris (1994, 2000), behaving-as-if is a precursor to
simulative abilities because it involves interpreting and producing behavior in light of
projecting an imaginative scenario, and is thereby a precursor to the ability to put oneself in
another’s shoes.
Against both explanatory models that rely on the consensus view of human social cognition,
the Embodied and Enacted Cognition (EEC) account proposes an alternative conception of
human social cognition: it is to be understood in terms of embodied agency and direct
responses to others (Gallagher and Hutto 2008, De Jaegher and Di Paolo 2009, Hutto and
Myin 2013). The most important difference between the consensus view and the EEC view is
that the latter holds that social cognition extends beyond the boundaries of biological
individuals, incorporating environmental factors and dynamically emerging interactions that
are structured by shared norms. The EEC account is a promising alternative to the
traditional cognitivist accounts of mind because it incorporates the social world into the
explanation of individual cognition, which is an improvement to the traditional theories
considering the fact that human beings are social animals from birth (Reddy 2009).
This new approach provides motivation for exploring an alternate account of pretence,
focusing on the role of embodiment, enaction and intersubjectivity at its core. The key
advantages of the EEC account of pretence is that it broadens our understanding of the
possibilities and limitations of imaginative capacities, suggesting that the external world is
actively involved in shaping the imaginative capacities of children: their creativity is to be
understood as a result of active engagement with the world, not passive workings of a mental
mechanism. It follows that a novel approach to explaining lack of engagement in pretence
play of children with ASD may be in order. To explain creativity, usually symbolic
representational capacities are invoked, but it is unlikely that children of 18 months of age
possess such higher-cognitive capacities. In the spirit of ecological validity and Occam’s
razor, EEC can provide a simpler, more basic account of pretence that is in line with current
developmental findings.
The application of EEC to pretend play shows in the following way. It makes use of notions
such as object affordances (the role of play props), social affordances (role of play
participants), and wider social context (role of narratives) in order to explain the
development and shaping of pretence capacities of children and adults. Affordances,
understood possibilities of action, are a dynamically construed relationships between the
object’s properties and the actor’s history of interactions with it. Such interactions suggest
novel possibilities of play. Taking the example of a banana, its physical shape and size can
afford certain ways of playing to the child (a “phone” when held to the ear, a “hat” when
placed on the head, etc.), yet, limit other kinds of play (eg., “human shoe” on account of it
getting squashed). Affordances can explain how one thing (banana) can be treated as another
(phone) without the need to invoke internal capacities like concept-manipulation, and
explain how objects get their new meanings from use, instead of appealing to intentionality
as the mechanism that fixates meanings. Even in pretend play, where seemingly ‘anything
can stand for anything else’, the affordances of the objects suggest what can be played and
how (Rucinska, forthcoming).
31
In addition, EEC suggests that social structures surrounding the play, or immediate
(dialogical) interactions with the external world, are key factors in the development of
pretence capacities. For example, caregivers initiate novel actions by directly engaging with
children in play and guiding them in it. A caregiver can show to a child how to hold a banana,
which the child then imitates. Imitation of others’ behaviours is a type of direct
responsiveness to the environment; whether such imitative capacities require further
representational capacities is to be explored in more detail, but to the extent that imitation
does not require complex capacities for symbolic representation, neither would pretence.
However, caregivers can guide play indirectly, as spectators, by responding to child’s play.
Reactions like smiles or gestures reassure and encourage the child what is acceptable as play.
As therapists Waltzlawick and Bavelas (1967) claim, “all behaviour, not only the use of
words, is communication” (p. 73).
The game of charades is an exemplar of how pretend meanings can get created during play,
as opposed to them being decided in advance. The actors adjust their behaviors to
accommodate the understanding of the spectators, and explore their means of depicting the
same thing. Such play might be an example of how ‘meaning is created together’, through the
breakdown and rebuilding of communication (similarly to what De Jaegher and Di Paolo
(2009) call ‘participatory sense-making’). There is a potential in this explanation to
understanding how intersubjective cognition can be shaped.
Finally, engagement in narrative practices may function as indirect, external factor in
shaping symbolic play of older children. Although language users, it may be a mistake to
conceive of their creativity as internally produced. A child’s familiarity with various stories
can re-shape the game and create novel scenarios. For example, a child may be recreating the
routines she has seen or participated in in the game of ‘restaurant’, and enhance it with a
story about a restaurant, using pre-established set of characters and settings (‘waiter’,
‘customer’), but also using other stories to improvise with the script (e.g., ‘sword-fighters’ to
turn cutlery into swords). Narratives may provide the necessary structures of going beyond
what is immediately present, enabling elaboration without relying on internal structures or
internalizing what has been learnt as a set of rules for playing ‘restaurant’.
To sum up, what structures pretence and makes it a cognitive act (instead of mere ‘thrashing
about’) can be explained by the EEC tools such as object affordances, social context and
narrative capacities. They help to understand how one can engage in pretence without
invoking full-blown symbolic capacities and higher cognitive mechanisms such as reasoning,
inferring or supposing, notions as ‘imagination’ in their explanatory models (which
presuppose what is to be explained), altogether avoiding the need to invoke mental
mechanisms in the brain (see ichols and Stich’s (2000) ‘pretence box’), which require
further explanation themselves. In this light, the EEC account of pretence is worth further
investigating, as it also promotes a better understanding of the development of social
cognition in general.
32
2.4 Social emotions
A “wide” perspective on emotions urges that emotions are fundamentally social
phenomena.
Dominant theories of emotion in philosophy and cognitive science tend to adopt an
individualistic perspective. Whether modelled as evaluative judgments, appraisal processes,
physiological states of bodily arousal, or something else, emotions are commonly thought of
as private states individuated by their neurobiology, cognitive content, behavioural
expression, or phenomenal character (cf. Damasio 1999; Laird 2007; Nussbaum 2001;
Panskepp 1998; Russell 2009; Schwarz and Clore 1988; Prinz 2004). From this perspective,
the environment is of secondary performance for understanding the inner (i.e., agent-
centric) mechanisms that are the real heart of emotions.
A “wide” perspective on emotions, in contrast, urges that emotions are fundamentally social
phenomena: they are scaffolded and shaped by features of the agent’s social niche, as well as
the various ways the individual—along with other emotional agents—actively modifies and
negotiates this shared niche (cf. Averill 1985; Colombetti and Roberts under review; Krueger
forthcoming; Parkinson et al 2005; Slaby forthcoming; Stephan et al forthcoming).
Accordingly, the role of the environment is seen not merely as providing stimulus inputs and
serving as an arena for behavioural outputs. Rather, the environment—understood broadly
to include not only material features of the agent’s niche, but also sociocultural and
interpersonal aspects as well—plays an active role in shaping emotions on multiple time
scales, including both the moment-to-moment character of emotional episodes, as well as
the long-term development of an individual’s emotional repertoire (cf. Griffiths and
Scarantino 2009). While not looking to entirely jettison a consideration of internal
mechanisms, a wide perspective on emotions thus argues that the larger bodily, social, and
interactive context in which emotions are situated needs to be part of the target
explanandum.
One way to think about the social character of emotions is to emphasize their embodied
character. Here, the idea is that brain alone is not sufficient to generate emotional
experience. Rather, the rest of the (non-neural) body in its biological, physiological,
morphological, and kinematic details makes a non-trivial contribution to the realization of
some emotions. For example, many studies appear to indicate a reciprocal relation between
an emotional experience and its behavioural expression. Subjects induced to adopt an
emotion-specific facial expression or posture report experiencing the corresponding emotion
(Duclos and Laird 2001; Edelman 1984; Flack et al 1999; Marzoli et al forthcoming; see
Laird 2007 for review). Conversely, inhibiting the expression (e.g., suppressing the facial
signature of anger or happiness) diminishes the associated experience (Niedenthal 2007).
Individuals who suffer severe spinal cord injuries and lose the capacity to behaviourally
express emotions report less-intense feelings of high-arousal emotions like fear and anger
(Chwalisz et al 1988; Hoffman 1966; see also Laird 2007, pp.74-76; Mack et al 2005). Still
other studies have found that inhibited facial expressiveness—e.g., following Botox injections
(Davis et al 2010; Havas et al 2010) or due to congenital facial paralysis (Cole and Spalding
2009)—results in diminished emotional phenomenology.
Embodied approaches to emotion emphasize the extent to which emotions depend upon
33
extra-neural factors and feedback. Moreover, if the physical expression of anger, say, is
literally a part of the anger itself—that is, part of its physical realization—some emotions, in
virtue of their embodied character, can be said to have a social face. They are partially
constituted by world-directed features perceptually available to other agents. But this
embodied perspective remains a fairly conservative way of thinking about the social
character of emotions. The environment does not enter into this characterization in any
substantive manner.
A stronger (and more genuinely “wide”) way of considering the social nature of emotions is
to look at the different ways the environment actively scaffolds emotional performance and
development on multiple time scales (cf. Griffiths and Scarantion 2009; Parkinson et al
2005). From this perspective, emotions are not pre-formed inner states that only enter the
social arena via behavioural outputs. Rather, the environment is understood to be part of the
very process by which emotions are experienced and expressed.
Consider work on audience effects, which indicate that emotional responses, despite a
constant stimulus, differ with respect to the anticipated recipient of the response. For
example, ten-pin bowlers smile significantly more after producing a positive event (e.g.,
bowling a strike or a spare) when they turn to face their friends than when they are still
facing the pin (Kraut and Johnston 1979). A similar effect has been observed in Spanish
soccer fans who issue authentic (i.e., “Duchenne”) smiles (Ekman 1972) in response to goals
only when facing one another person (Fernandez-Dols and Ruiz-Belda 1997), as well as in
Olympic gold-medal winners who smile almost exclusively when receiving their gold medal
(and not when they are alone prior to the ceremony) (Fernandez-Dols and Ruiz-Belda 1995).
Audience effects have even been observed in young infants (Scanlon Jones et al 1991). Other
work in developmental psychology demonstrates how the repertoire of physical strategies
caregivers use to engage with infants—facial expressions, postural adjustments, exaggerated
gestures and vocalisations, gaze manipulation, etc.—function as real-time scaffolding
supporting the emergence, regulation, and performance of many basic emotions from the
beginning of life (cf. Hobson 2002; Krueger 2013; Murray and Trevarthen 1985; Posner and
Rothbart 1998; Reddy 2008; Rochat 2009; Tronick et al 1978). This research supports the
thesis that certain emotions depend crucially upon the ongoing (i.e., synchronic) feedback
and support of the social-communicative context in which they are situated.
The environment also scaffolds the long-term (i.e., diachronic) development of an
individual’s emotional repertoire. A central part of emotional development involves the
cultivation of various emotional habits—habits which embody various sociocultural factors
(i.e., norms, values, and beliefs governing emotion scripts, displays rules, etc.) relative to
different social contexts and cultural environments (cf. Markus and Kitayama 1991). For
example, consider emotional responses during a wedding ceremony. Myriad material
technologies within this niche (e.g., music, singing, clothing, ritualized aspects of the
ceremony, decorations, food, building structure, etc.) collectively orchestrate the
performance of wedding-appropriate emotions on moment-to-moment basis. But this
orchestration also reflects more enduring cultural values, norms, and practices—what
Parkinson et al (2005) term “ideational factors”—that exert a material impact on the
character of these technologies and the emotions they orchestrate. The affective atmosphere
of a Catholic wedding may differ significantly from a Hindu wedding or a secular ceremony.
Children grow up attending multiple wedding ceremonies and gradually learn what sort of
34
emotions are appropriate within this context and how, exactly, they are to be expressed (or
repressed, as the case may be). This is one example of how culture-specific emotion scripts,
display rules, and material features of the environment converge to shape the child’s
emotional development and enrich their emotional repertoire. Emotional ontogeny reflects
different features of the affective culture in which it develops.
A wide perspective to emotions thus need not deny the importance of the brain or individual
mechanisms. But it does caution against adopting an excessively narrow perspective that
abstracts emotions away from the broader bodily, social, and cultural contexts that play a
critical role in their development and functioning.
3. Challenges
Wide cognition approaches need not, even if sometimes appear to, explain all phenomena
related to social intelligence or mentality broadly conceived. Also, there are a number of
essential open questions that need to be addressed in the future.
Unit of analysis / boundaries of cognition. It is an open question what should be the basic
unit of analysis in the case of distributed cognitive systems: individual agents or groups? Or
maybe the relationships and processes obtaining between the group members? Deciding
such questions in a systematic way is currently not known.
How natural is culture? It remains a challenge to state the exact relationship between what
was traditionally conceived as the biological core of cognition and wide cognition; are
culturally mediated cognitive abilities essentially as natural as biological ones? Traditional
notions of innateness in cognitive science seem to be a source of major confusion (Griffiths
2002, Samuels 2004).
Collective intentionality. A non-controversial account of collective intentionality is needed.
It is clear that the notion is related to the extended mind framework (Tollefsen 2006). There
is also an open question how collective intentionality relates to the idea that minds are social
artifacts (Castelfranchi 2013).
Is distributed cognition a useful concept? The notion of the distributed cognition, though
influential, may be viewed as one of the applications of the concept of the extended mind
(see section 1.6 above); at the same time, all accounts of social cognition presuppose that
cognitive phenomena are distributed among multiple cognitive agents in one way or another.
In this way, there may be no special methodology or theory of distributed cognition: this is
however still unclear.
Time scales. Cognitive phenomena occur at different time scales: traditional cognitive
science presupposed that individual cognitive acts occur at the time of scale of several
hundred milliseconds, though it has been long recognized that social factors – and other
‘bands’ of cognition occur at different time scales ( ewell 1990, Anderson 2002, Papo 2013).
The challenge remains to specify how to investigate cultural and social phenomena so as to
include them all in a unified models of social and cultural cognition. For example, so called
35
network goods have value to someone only if other interactive parties also have them. It
makes no sense to own a fax machine if nobody else owns one. This phenomenon can be
studied on different time scales, and it is an open question how to integrate economical
analyses with the psychological research on joint action, and the ethnographical studies
typical of distributed cognition with computational modeling in evolutionary game theory, as
it is also an example of a co-adaptive behavior.
Non-individual factors: constitutive or causal? There is an unsolved controversy regarding
the status of non-individual factors (Adams & Aizawa 2008, Rupert 2009). Is there a way to
decide which ones are constitutive for cognitive capacities, and which simply have causal
influence? Recent debates in philosophy of mind and cognitive science suggest it is helpful to
keep notions of causal dependence and constitution separate (e.g., Menary, 2010b). For it is
one thing to argue that cognitive processes can be causally dependent on the body and its
interaction with an environment and it is another to argue that the body and its interactions
actually constitute or realize cognition. In this regard, Rowlands (2010: pp. 60-61) argues
that embedded mind is the claim that for some cognitive processes, there is a (perhaps
essential) causal dependence between environmental structures and bodily movement. If,
following Robbins and Aydede (2009: p. 3), we think of situated cognition as consisting in
the general claim that cognition can depend on structures in an agent’s environment (for a
review, see Anderson, 2003), then we can think of embedded mind as the targeted working
out of this claim.
Is there a theoretically significant difference between social and artefactual scaffold? It is
currently unclear whether social scaffolding of cognition should be framed in the same terms
as its technological underpinning. Should distributed cognitive phenomena always
presuppose collective intentionality and extend into multiple agents, or maybe only artefacts
allow for extending the mind?
How to study these phenomena? There is no overarching methodological strategy to decide
what is the best way to deal with social cognition, even if it seems plausible that in many
cases, wide approaches are needed.
4. Dangers
Duplication of effort. Social intelligence can be explained and modeled in various
approaches. There is a danger of duplicating effort; for example, social neuroscience may
inadvertently rediscover sociological explanations.
Isolation. Empirical evidence in various theories may rather constrain other theories rather
than screen them off. Theories completely isolated from what is currently known are likely to
be too unconstrained, and simply false. This is why it is essential to strive for integration; at
the same time, integration need not mean inter-theoretic reduction. Rather, other theories
known to be true should constrain the theory under investigation: it should be possible that
all are true at the same time.
Theoretical manifestos without empirical work. In all emerging fields of science, theoretical
manifestos are needed to make assumptions explicit. However, there is a danger of
36
stagnation if these manifestos are not put to work in empirical research. For example,
multiple Artificial Life (ALife) models show that the basic assumption that complex behavior
may result from simple agents is true. These models are examples of stagnation as far as they
do not show anything else; the result is simply trivial.
Conceptual confusion and misunderstanding. In interdisciplinary research, communication
issues are bound to arise, in particular due to different vocabularies used. For this reason,
there is a special role for philosophy of science to explain that some of the confusions are
merely terminological.
5. Summary
Wide approaches to cognition can be fruitfully applied to the study of social intelligence. The
embodied and grounded approach focuses on sensomotor aspects of social interactions and
accentuates their dynamical and time-related profiles, even in the study of phenomena
traditionally conceived of in a contemplative manner, such as imagining or thinking about
abstract entities. The embedded and situated cognition insists that cognitive and social
processes are situated in the environment that cannot be ignored in the study of the social.
In the perspective of the extended cognition, cognitive agents can expand beyond the
boundaries of their heads and bodies to incorporate technical artifacts, social institutions
such as language, and other agents to form collective intentionality. Enactive approaches
stress that cognitive processes emerge out of interaction of agents and their environments in
a dynamical manner. Finally, the study of distributed cognition focuses on cognitive
processes that require multiple cognitive agents to interact in an orchestrated manner so as
to jointly solve cognitive problems.
As such, wide cognition transcends the traditional methodological individualism of cognitive
science, and offers new perspectives on such social phenomena as joint action, cognitive
artifacts, mindreading, imitation and pretence, as well as social emotions, which we covered
in more detail in Section 2. It does not need to be stressed that there are many more
phenomena in the domain of social intelligence that would benefit from the application of
methodologies specific to wide cognition. For example, complex multi-agent interactions are
studied in economics and sociology, where the benefit or loss to one participant depends on
what other participants do. These situations include Prisoners’ Dilemma-type and Stag
Hunt-type games. Examples include uprisings and insurrections, where the risk/reward
trade-off to an individual changes dramatically if lots of others join in the uprising compared
to when they don’t. However, such interactions have been mostly studied mostly in multi-
agent computer simulations (for a general introduction, see Squazzoni 2012), which
generally abstract away from embodied-related aspects of social interactions. These may play
a role in constraining the behavior, just like standard environments tend to do (Wood & Neal
2007).
A wide perspective to does not deny the importance of the brain or individual mechanisms.
But it does caution against adopting an excessively narrow perspective that abstracts
emotions away from the broader bodily, social, and cultural contexts that play a critical role
in their development and functioning.
37
Nevertheless, there remain important challenges to be met. We listed some of them in
sections 3 and 4; but the overall danger is that the potential benefit—that of unifying various
explanations—may be lost if the research focus were limited to a narrow scope of
phenomena. This would still give rise to microtheories, purely theoretical speculation, and
even further disintegration of the field of the study of social intelligence. For this reason, to
leverage the unificationary potential of wide cognition, bigger research projects are needed
that will bring various approaches under a single umbrella.
38
6. References
Adams, Frederick, and Kenneth Aizawa. 2008. The Bounds of Cognition. Malden, MA: Blackwell Pub. Amorim, M-E., 2003, What is my avatar seeing?: The coordination of "out-of-body" and "embodied"
perspectives for scene recognition across view, in: Visual Cognition, 10/2: 157-199. Anderson, John R. 2002. “Spanning Seven Orders of Magnitude: a Challenge for Cognitive Modeling.”
Cognitive Science 26 (1) (February): 85–112. doi:10.1016/S0364-0213(01)00062-3. Anderson, M. L. 2003. Embodied Cognition: A field guide. Artificial Intelligence, 149, 91-130. Butterfill, Stephen A., and Ian A. Apperly. 2013. “How to Construct a Minimal Theory of Mind.” Mind
& Language 28 (5): 606–637. doi:10.1111/mila.12036. Arbib, Michael A. 2005. “From Monkey-Like Action Recognition to Human Language: An
Evolutionary Framework for eurolinguistics.” Behavioral and Brain Sciences 28 (02): 105–24; discussion 125–67. doi:10.1017/S0140525X05000038.
Ash, J. 2013. Technologies of Captivation: Video Games and the Attunement of Affect, in: Body Society, 19: 127-51.
Averill, James R. 1985. “The Social Construction of Emotion: With Special Reference to Love.” In The Social Construction of the Person, edited by Kenneth J. Gergen and Keith E. Davis, 89–109. New York: Springer Verlag.
Baillargeon, Renée, Rose M. Scott, and Zijing He. 2010. “False-Belief Understanding in Infants.” Trends in Cognitive Sciences 14 (3): 110–118.
Ballard, D. 1991. “Animate Vision.” Artificial Intelligence 48, 57-86. Ballard, D. H. & Hayhoe, M. M. & Pook, P. K. & Rajesh, P. N. Rao 1997. Deictic Codes for the
Embodiment of Cognition. Behavioural and Brain Sciences, 20 (4): 723-742. Barandiaran, X., Di Paolo, E., & Rohde, M. 2009. Defining agency:individuality, normativity,
asymmetry and spatio-temporality in action. Adaotive Behavior, 17(4), 1–13. Baron-Cohen, S. 1995. Mindblindness: An essay on autism and theory of mind. MIT Press. Barsalou, L.W. 2008. Grounded cognition. Annual Review of Psychology, 59, 617-645. Bell, C., & Gemmell, J. 2009. Total recall: how the E-memory revolution will change everything:
Dutton. Bickhard, Mark H., and L. Terveen. 1995. Foundational Issues in Artificial Intelligence and Cognitive
Science: Impasse and Solution. North-Holland. Borghi, A.M & Cimatti, F. 2010. Embodied cognition and beyond: Acting and sensing the body.
Neuropsychologia, 48, 763-773. Borghi, A.M., Pecher, D. 2011. Introduction to the special topic Embodied and Grounded Cognition.
Frontiers in Psychology, 2, 187. Borghi, Anna M., Claudia Scorolli, Daniele Caligiore, Gianluca Baldassarre, and Luca Tummolini.
2013. “The Embodied Mind Extended: Using Words as Social Tools.” Frontiers in Psychology 4. doi:10.3389/fpsyg.2013.00214.
Bratman, M. E. 1992. Shared cooperative activity. The Philosophical Review 101(2), 327–341. Brooks, R. 1991. Intelligence without Representation. Artificial Intelligence, 47, 139-159. Beunza, Daniel and David Stark. 2008. Tools of the trade: the socio-technology of arbitrage in a
Wall Street trading room. In Living in a Material World: Economic Sociology Meets Science and Technology Studies, edited by Trevor Pinch and Richard Swedborg, 253-290. Cambridge, MA, USA: MIT Press.
Butterfill, Stephen. 2013. “ ot Just Wide but Shared: Joint Action Is a Core Form of Social Intelligence.” Speech, Kazimierz Dolny, Poland, August 20, 2013.
Carr, N. 2010. The Shallows: How the internet is changing the way we think, read and remember. London: Atlantic Books.
Castelfranchi, Cristiano. 2013. “Minds as Social Institutions.” Phenomenology and the Cognitive Sciences (May 30). doi:10.1007/s11097-013-9324-0.
Chatterjee, A. 2010. Disembodying cognition. Language and Cognition, 2-1, 79-116. Chemero, A. 2009. Radical Embodied Cognitive Science. Cambridge, MA: MIT Press. Chwalisz, K., E. Diener, and D. Gallagher. 1988. “Autonomic Arousal Feedback and Emotional
Experience: Evidence from the Spinal Cord Injured.” Journal of Personality and Social Psychology 54 (5) (May): 820–828.
Clark, A. 1997. Being There: Putting Brain, Body, and World Together Again. Cambridge, MA: The MIT Press.
Clark, A. 2008. Supersizing the Mind: Embodiment, Action and Cognitive Extension. Oxford: Oxford University Press.
Clark, A., & Chalmers, D. 1998. The Extended Mind. Analysis, 58, 10-23.
39
Clowes, R. W. 2013. The Cognitive Integration of E-Memory. Review of Philosophy and Psychology(4), 107-133.
Cole, Jonathan, and Henrietta Spalding. 2009. The Invisible Smile: Living Without Facial Expression. Oxford: Oxford University Press.
Colombetti, Giovanna, and Tom Roberts. “Extending the Extended Mind: The Case for Extended Affectivity.”
Cowley, Stephen J. 2009. “Distributed Language and Dynamics.” Pragmatics Cognition 17 (3): 495–507. doi:10.1075/p&c.17.3.01cow.
Currie, G. & Ravenscroft, I. 2002. Recreative Minds: Imagination in Philosophy and Psychology. Oxford: Oxford University Press.
Damasio, Antonio. 1999. The Feeling of What Happens: Body and Emotion in the Making of Consciousness. New York: Harcourt Brace.
Davis, J. I. & Markman, A. B. 2012, Embodied Cognition as a Practical Paradigm: Introduction to the Topic, The Future of Embodied Cognition. Topics in Cognitive Science, 4: 685–691.
Davis, Joshua Ian, Ann Senghas, Fredric Brandt, and Kevin . Ochsner. 2010. “The Effects of BOTOX Injections on Emotional Experience.” Emotion 10 (3): 433–440. doi:10.1037/a0018690.
De Jaegher, H., Di Paolo, E. 2007. “Participatory sense-making.” Phenomenology and the Cognitive Sciences, 6(4), 485–507. doi:10.1007/s11097-007-9076-9
De Jeagher, H., Gallagher, S.,Di Paolo, E., 2010, Can social interaction constitute social cognition?, in: Trends in cognitive sciences, 10/31: 441-447
Di Paolo, E. A. 2005. Autopoiesis, Adaptivity, Teleology, Agency. Phenomenology and the Cognitive Sciences, 4(4), 429–452.
Di Paolo, E. A. 2009. “Extended life.” Topoi, 28, 9–21. Di Paolo, E. A., Rohde, M., & DeJaegher, H. 2010. Horizons for the enactive mind: values, social
interaction and play. In J. Stewart, O. Gapenne, & E. Di Paolo (Eds.), Enaction: Towards a new paradigm of cognitive science. Cambridge, Mass: MIT Press.
Di Paolo, E. A., Thompson, E. Forthcoming. In The Enactive Approach, edited by Lawrence Shapiro, The Routledge Handbook of Embodied Cognition, London: Routledge.
Dove, G. 2009. Beyond Conceptual Symbols. A Call for Representational Pluralism. Cognition, 110, 412-431.
Dror, I. E., & Harnad, S. 2008. Offloading cognition onto cognitive technology. In I. E. Dror & S. Harnad (Eds.), Cognition Distributed: How Cognitive Technology Extends Our Minds (pp. 1-23). Amsterdam: John Benjamins Publishing.
Dror, I. E., & Harnad, S., eds. 2008. Cognition Distributed: How Cognitive Technology Extends Our Minds. Amsterdam: John Benjamins.
Duclos, Sandra E., and James D. Laird. 2001. “The Deliberate Control of Emotional Experience through Control of Expressions.” Cognition & Emotion 15 (1): 27–56. doi:10.1080/02699930126057.
Edelman, B. 1984. “A Multiple-factor of Body Weight Control.” Journal of General Psychology 110: 99–114.
Ekman, Paul, Richard J. Davidson, and Wallace V. Friesen. 1990. “The Duchenne Smile: Emotional Expression and Brain Physiology. II.” Journal of Personality and Social Psychology 58 (2) (February): 342–353.
Evans, Jonathan St.B.T. 2003. “In Two Minds: Dual-Process Accounts of Reasoning.” Trends in Cognitive Sciences 7 (10): 454–459. doi:10.1016/j.tics.2003.08.012.
Feigenson, L., Dehane, S., Spelke, E. S. 2004. “Core systems of number.” Trends in Cognitive Science, 8, 307–314.
Fernández-Dols, José-Miguel, and María-Angeles Ruiz-Belda. 1995. “Are Smiles a Sign of Happiness? Gold Medal Winners at the Olympic Games.” Journal of Personality and Social Psychology 69 (6) (December): 1113–1119.
Fernández-Dols, José-Miguel, and María-Angeles Ruiz-Belda. 1997. “Spontaneous Facial Behavior During Intense Emotional Episodes: Artistic Truth and Optical Truth.” In The Psychology of Facial Expression, edited by James A. Russell and José-Miguel Fernández-Dols, 255–294. Cambridge: Cambridge University Press.
Fischer, M. & Zwaan, R. 2008. Embodied language: A review of the role of the motor system in language comprehension. The Quarterly Journal of Experimental Psychology, 61, 825-850.
Gallagher, S. Hutto, D. 2008. ‘Primary Interaction and arrative Practice’, in J. Zlatev, T. Racine, C. Sinha & E. Itkonen (ed.), The Shared Mind: Perspectives on Intersubjectivity. Amsterdam: John Benjamins, pp. 17–38.
Gallese, V. Sinigaglia, C. 2011. “What is so special about embodied simulation?” Trends in Cognitive Sciences 15 (11): 512-519.
40
Galloway, A. 2005, Gaming: Essays on Algorithmic Culture, Minnessota (University of Minnessota Pres)
Gentner, D. 2010. Psychology in Cognitive Science: 1978–2038. Topics in Cognitive Science 2, 328-344.
Gergely, György, Harold Bekkering, and Ildikó Király. 2002. “Rational Imitation in Preverbal Infants.” Nature 415 (6873): 755. doi:10.1038/415755a.
Gergely, György, and Gergely Csibra. 2003. “Teleological Reasoning in Infancy: The aive Theory of Rational Action.” Trends in Cognitive Sciences 7 (7): 287–292.
Gibbs, R. W. 2005. Embodiment and cognitive science. Cambridge: Cambridge University Press. Gibson, James J. 1986. The Ecological Approach to Visual Perception. Hove: Psychology Press. Giere, R. N. 2004. The Problem of Agency in Scientific Distributed Cognitive Systems. Journal of
Cognition & Culture 4, no. 3/4, 759-774. Gilbert, M. 1992. On Social Facts. Princeton, NJ: Princeton University Press. Goldman, A. 2005. Imitation, mindreading, and simulation, in S. Hurley and N. Chater (eds.)
(2005), Perspectives on Imitation (Vol. 2, pp. 79-93). Cambridge, MA MIT Press. Goldman, A. 2006. Simulating Minds: The Philosophy, Psychology and Neuroscience of
Mindreading, New York: Oxford University Press. Goldman, A.I. de Vignemont, F. 2009. “Is social cognition embodied?” Trends in Cognitive Sciences
13(4):154-9. Griffiths, Paul, and Andrea Scarantino. 2009. “Emotions in the Wild: The Situated Perspective on
Emotion.” In The Cambridge Handbook of Situated Cognition, edited by Philip Robbins and Murat Aydede, 437–453. New York: Cambridge University Press.
Griffiths, Paul. 2002. “What Is Innateness?” The Monist 85 (1). Harris, P.L. & Kavanaugh, R.D. 1993. Young children's understanding of pretense. Monographs of the
Society for Research in Child Development 58, No. 1. Harris, P.L. 1994. Understanding pretense. In C. Lewis, & P. Mitchell, Children's early understanding
of mind: origins and development, p. 235-259. Hillsdale, NJ: Lawrence Erlbaum Associates. Havas, D. A., A. M. Glenberg, K. A. Gutowski, M. J. Lucarelli, and R. J. Davidson. 2010. “Cosmetic Use
of Botulinum Toxin-A Affects Processing of Emotional Language.” Psychological Science 21 (7) (June): 895–900. doi:10.1177/0956797610374742.
Heersmink, R. 2013, A Taxonomy of Cognitive Artifacts: Function, Information, and Categories, Review of Philosophy and Psychology,
Heras-Escribano, M., Noble, J. & De Pinedo, M. 2013. The only wrong cell is the dead one: On the enactive approach to normativity. ECAL 2013. Cambridge, Mass.: MIT Press, pp. 665-670.
Hobson, Peter. 2002. The Cradle of Thought: Exploring the Origins of Thinking. London: MacMillan. Hurley, S. 2001. Perception and action: Alternative views. Synthese, 129(1), 3-40. Hutchins, E. 1995a. Cognition in the Wild. Cambridge, Mass.: The MIT Press. Hutchins, E. 1995b. “How a Cockpit Remembers Its Speeds.” Cognitive Science 19, no. 3, 265–288. Hutto, Daniel. 2008. Folk Psychological Narratives: The Sociocultural Basis of Understanding
Reasons. Cambridge Mass.: MIT Press. Hutto, D.D. & Myin, E. 2013. Radicalizing Enactivism: Basic Minds without Content. Cambridge:
MIT Press. Jirak, D., Menz, M., Buccino, G., Borghi, A.M., Binkofski, F. 2010. “Grasping language. A short story
on embodiment.” Consciousness and Cognition, 19, 711-720 Jones, Susan Scanlon, Kimberly Collins, and Hye-Won Hong. 1991. “An Audience Effect on Smile
Production in 10-month-old Infants.” Psychological Science 2 (1): 45–49. Kelso, J. A. Scott. 1995. Dynamic Patterns: The Self-Organization of Brain and Behavior. Cambridge,
Mass.: MIT Press. Kilner, J., Y. Paulignan, and S. Blakemore 2003. An interference effect of observed biological
movement on action. Current Biology 13(6), 522–525. Kim, Jinah, Tony Wigram, and Christian Gold. 2008. “The Effects of Improvisational Music Therapy
on Joint Attention Behaviors in Autistic Children: a Randomized Controlled Study.” Journal of Autism and Developmental Disorders 38 (9): 1758–66. doi:10.1007/s10803-008-0566-6.
Kirsh, D. 1996. “Adapting the environment instead of oneself.” Adaptive Behavior, 4(3-4), 415-452. Knoblich, G., S. Butterfill, and N. Sebanz 2010. Psychological research on joint action: Theory and
data. In B. Ross (Ed.), Psychology of Learning and Motivation, Volume 51. Academic Press. Koelsch. S., Offermanns, K., Franzke, P. 2010. Music in the treatment of affective disorders: an
exploratory investigation of a new method for music therapeutic research. Music Perception 27:307-316.
Kourtis, D., N. Sebanz, and G. Knoblich 2010. Favouritism in the motor system: Social interaction modulates action simulation. Biology Letters 6(6), 758–761.
41
Kourtis, D., . Sebanz, and G. Knoblich 2013. Predictive representation of other people’s actions in joint action planning: An EEG study. Social Neuroscience 8(1), 31–42.
Kraut, Robert E., and Robert E. Johnston. 1979. “Social and Emotional Messages of Smiling: An Ethological Approach.” Journal of Personality and Social Psychology 37 (9): 1539–1553.
Krueger, Joel 2013b. “Ontogenesis of the Socially Extended Mind.” Cognitive Systems Research 25-26 (December): 40–46. doi:10.1016/j.cogsys.2013.03.001.
Krueger, Joel. 2013a. “Emotions and the Social iche.” In Collective Emotions, edited by Christian Von Scheve and Mikko Salmela. New York: Oxford University Press.
Kyselo, M. and Di Paolo, E. 2013. “Locked-in Syndrome: a Challenge for Embodied Cognitive Science.” Phenomenology and the Cognitive Sciences. doi: 10.1007/s11097-013-9344-9
Laird, James D. 2007. Feelings: The Perception of Self. Oxford: Oxford University Press. Lakoff, G. 2012. “Explaining Embodied Cognition Results.” Topics in Cognitive Science, 1-13. Leslie, A. 1987. Pretense and representation: The origins of "theory of mind." Psychological Review
94, 412-426. Leslie, A. 2005. “Developmental parallels in understanding minds and bodies.” Trends in Cognitive
Sciences, 9(10), 459–62. Lillard, A. 1994. Making sense of pretense. In Children’s early understanding of mind: Origins and
development, edited by C. Lewis and P. Mitchell, 211-234. Hove, UK: Lawrence Erlbaum. Loehr, J. D. and C. Palmer 2011. Temporal coordination between performing musicians. The
Quarterly Journal of Experimental Psychology 64(11), 2153–2167. Loehr, J. D., D. Kourtis, C. Vesper, N. Sebanz, G. Knoblich, et al. 2013. Monitoring individual and
joint action outcomes in duet music performance. Journal of cognitive neuroscience forthcoming.
Louwerse, M. M., 2011. “Symbol interdependency in symbolic and embodied cognition.” TopiCS in Cognitive Science, 3, 273-302.
Lyons, D. 2009. The rational continuum of human imitation. In Mirror Neuron Systems: The Role of Mirror Processes in Social Cognition, pp. 77-106. New York: Humana Press.
MacKenzie, Donald. 2009. Material Markets: How Economic Agents are Constructed. Oxford, UK: Oxford University Press.
Mahon, B. & Caramazza, A. 2008. A critical look at the embodied cognition hypothesis and a new proposal for grounding conceptual content, Journal of Physiology, 102, 59-70.
Markus, Hazel R., and Shinobu Kitayama. 1994. “The Cultural Construction of Self and Emoion: Implications for Social Behavior.” In Emotion and Culture: Empirical Studies of Mutual Influence, edited by Shinobu Kitayama and Hazel R. Markus, 89–130. Washington, D.C.: American Psychological Association.
Marsh, L., Pearson, A, Ropar, D., Hamilton, A. 2013. “Children with autism don t overimitate.” Current Biology, 23(7), R266-R268.
Martin, A. 2007. The representation of object concepts in the brain. Annual Review of Psychology, 58, 25–45.
Marzoli, Daniele, Mariagrazia Custodero, Alessandra Pagliara, and Luca Tommasi. 2013. “Sun-induced Frowning Fosters Aggressive Feelings.” Cognition & Emotion 0 (0): 1–9. doi:10.1080/02699931.2013.801338.
Mayer-Schönberger, V. 2011. Delete: The virtue of forgetting in the digital age. Princeton University Press.
McGann, M., De Jaegher, H., Di Paolo, E. 2013. “Enaction and psychology.” Review of General Psychology, 17(2), 203–209. doi:10.1037/a0032935
Meltzoff, A. 1995. “Understanding the intentions of others: Re-enactment of intended acts by 18-month-old children.” Developmental Psychology, 31, 838–850.
Meltzoff, A. N. 2005. Imitation and Other Minds: The “Like Me” Hypothesis. In Perspectives on Imitation, vol. 2, 55-77, edited by Hurley, S. & N. Chater. Cambridge, MA: MIT Press.
Meltzoff, A. N. (2011). Social cognition and the origins of imitation, empathy, and theory of mind. In The Wiley-Blackwell handbook of childhood cognitive development (2nd ed.), 49-75, ed. by U. Goswami. Malden, MA: Wiley-Blackwell.
Menary, R. (2010a). Cognitive integration and the extended mind. In R. Menary (Ed.), The Extended Mind (pp. 227-244). London, England: Bradford Book, MIT Press.
Menary, R. (eds) (2010b). The Extended Mind. The MIT Press: Bradford Books. Ménoret, M., L. Varnet, R. Fargier, A. Cheylus, A. Curie, V. des Portes, T. azir, and P. Y. (submitted).
Neural basis of non-verbal social interactions: a dual-eeg study. Michaelian, K., & Sutton, J. 2013. Distributed Cognition and Memory Research: History and Current
Directions. Review of Philosophy and Psychology(4), 1-. More, M., Vita-More N. 2013. The Transhumanist Reader: MIT Press, Cambridge MA.
42
Newell, Allen. 1990. Unified Theories of Cognition. Cambridge, Mass. and London: Harvard University Press.
ichols, S Stich, S. 2000. “A cognitive theory of pretense.” Cognition 74, 115-147. Niedenthal, Paula M. 2007. “Embodying Emotion.” Science 316: 1002–1005. ielsen, M., Tomaselli, K. 2010. “Overimitation in Kalahari Bushman children and the origins of
human cultural cognition.” Psychological Science, 21(5), 729-736. Norman, Donald A. 1990. Cognitive artifacts: Department of Cognitive Science, University of
California, San Diego. ovembre, G., L. F. Ticini, S. Schütz-Bosbach, and P. E. Keller 2013. Motor simulation and the
coordination of self and other in real-time joint action. Social Cognitive and Affective Neuroscience forthcoming. PMID: 23709353.
Nussbaum, Martha. 2001. Upheavals of Thought: The Intelligence of Emotions. Cambridge: Cambridge University Press.
O’Regan, J. K. o , A. 2001. “A sensorimotor account of vision and visual consciousness.” Behavioral and Brain Sciences, 24(5), pp. 939–1031.
Panksepp, Jaak. 1998. “The Periconscious Substrates of Consciousness: Affective States and the Evolutionary Origins of the SELF.” Journal of Consciousness Studies 5 (5-6): 566–582.
Papo, David. 2013. “Time Scales in Cognitive euroscience.” Frontiers in Physiology 4 (January): 86. doi:10.3389/fphys.2013.00086.
Parkinson, Brian, Agneta H. Fischer, and Antony S.R. Manstead. 2005. Emotions in Social Relations. New York: Psychology Press.
Pecher, Diane, Inge Boot, and Saskia Van Dantzig. 2011. “Abstract Concepts: Sensory-Motor Grounding, Metaphors, and Beyond.” In The Psychology of Learning and Motivation. Vol. 54, edited by Brian Ross, 54:217–248. Burlington: Academic Press. doi:10.1016/B978-0-12-385527-5.00007-3.
Perry, M. Distributed Cognition. 2003. In: HCI Models, Theories, and Frameworks: Toward an Interdisciplinary Science, edited by John M. Carroll, 193-223. Amsterdam: Morgan Kaufmann.
Pezzulo, G., Barsalou, L.W., Cangelosi, A., Fischer, M.A., McRae, K., Spivey, M. 2011. The mechanics of embodiment: A dialogue on embodiment and computational modeling. Frontiers in Cognition, 2(5), 1-21.
Piaget, J. 1952. The origins of intelligence in children. International Universities Press. Posner, Michael, and Marky K. Rothbart. 1998. “Attention, Self-Regulation, and Consciousness.”
Philosophical Transactions of the Royal Society B: Biological Sciences 353: 1915–1927. Prinz, Jesse J. 2004. Gut Reactions: A Perceptual Theory of Emotions. Oxford: Oxford University
Press. Ramenzoni, V. C., T. J. Davis, M. A. Riley, K. Shockley, and A. A. Baker 2011. Joint action in a
cooperative precision task: nested processes of intrapersonal and interpersonal coordination. Experimental Brain Research 211(3-4), 447–57.
Rączaszek-Leonardi, Joanna, and J. A. Scott Kelso. 2008. “Reconciling Symbolic and Dynamic Aspects of Language: Toward a Dynamic Psycholinguistics.” New Ideas in Psychology 26 (2): 193–207. doi:10.1016/j.newideapsych.2007.07.003.
Reddy, Vasudevi. 2008. How Infants Know Minds. Cambridge: Harvard University Press. Richardson, M. J., K. L. Marsh, and R. M. Baron 2007. Judging and actualizing intrapersonal and
interpersonal affordances. Journal of Experimental Psychology: Human Perception and Performance. Vol. 33(4) 33(4), 845–859.
Rizzolatti, Giacomo, and Laila Craighero. 2004. “The Mirror- euron System.” Annual Review of Neuroscience 27: 169–92. doi:10.1146/annurev.neuro.27.070203.144230.
Robbins, P. & Aydede, M. 2009. The Cambridge Handbook of Situated Cognition. Cambridge University Press.
Rochat, Philippe. 2009. “Social-Affective Origins of Mindreading and Metacognition.” Behavioral and Brain Sciences 32 (02): 160–161. doi:10.1017/S0140525X0900079X.
Ross, D., Spurrett, D., Kincaid, H. & Stephens, G.L., eds. 2007. Distributed Cognition and the Will: Individual Volition in Social Context. Cambridge, Mass: The MIT Press.
Rowlands, M. 2010. The New Science of the Mind: From Extended Mind to Embodied Phenomenology. The MIT Press: Bradford Books.
Rupert, Robert D. 2009. Cognitive Systems and the Extended Mind. Oxford: Oxford University Press. Russell, James A. 2009. “Emotion, Core Affect, and Psychological Construction.” Cognition &
Emotion 23 (7): 1259. doi:10.1080/02699930902809375. Salomon, G., ed. 1997. Distributed Cognitions: Psychological and Educational Considerations.
Cambridge, Mass.: Cambridge University Press. Samuels, Richard. 2004. “Innateness in Cognitive Science.” Trends in Cognitive Sciences 8 (3)
43
(March): 136–41. doi:10.1016/j.tics.2004.01.010. Sartori, L. and A. Cavallo 2011. From simulation to reciprocity: the case of complementary actions.
Social neuroscience 7(2), 146–58. Schwarz, N., and G.L. Clore. 1988. “How Do I Feel About It? Informative Functions of Affective
States.” In Affect, Cognition, and Social Behavior, edited by K. Fiedler and J.P. Forgas, 44–62. Toronto: Hogrefe.
Searle, J. 1980. Minds, Brains and Programs. Behavioral and Brain Sciences 3, no. 3, 417–457. Searle, J. R. 1990. Collective intentions and actions. In P. Cohen, J. Morgan, and M. Pollack (Eds.),
Intentions in Communication, pp. 90–105. Cambridge: Cambridge University Press. Reprinted in Searle, J. R. 2002. Consciousness and Language. Cambridge: Cambridge University Press (pp. 90–105).
Sebanz, N., Bekkering, H. & Knoblich, G. (2006b), Joint action: bodies and minds moving together. Trends in Cognitive Sciences, 10(2), 70-76.
Sebanz, N., G. Knoblich, W. Prinz, and E. Wascher (2006a). Twin peaks: An ERP study of action planning and control in coacting individuals. Journal of Cognitive Neuroscience 18(5), 859–870.
Semin, G.R., & Smith, E.R. (Eds.) 2008. Embodied grounding: Social, cognitive, affective, and neuroscientific approaches. New York: Cambridge University Press.
Slaby, Jan. 2013. “Emotions and the Extended Mind.” In Collective Emotions, edited by Mikko Salmela and Christian Von Scheve. Oxford University Press.
Squazzoni, Flaminio. 2012. Agent-Based Computational Sociology. Chichester: Wiley-Blackwell. Stanovich, Keith E. 2011. Rationality and the Reflective Mind. Oxford: Oxford University Press. Steels, L. Brooks, R. 1995. The “Artificial Life” route to “Artificial Intelligence”: Building Situated
Embodied Agents. Hillsdale, NJ: Lawrence Eribaum Associates. Stephan, Achim, Sven Walter, and Wendy Wilutzky. 2013. “Emotions Beyond Brain and Body.”
Philosophical Psychology (September 6): 1–17. doi:10.1080/09515089.2013.828376. Sterelny, K. 2010. Minds: extended or scaffolded? Phenomenology and the Cognitive Sciences, 9(4),
465-481. Sutton, J. 2010. Exograms and interdisciplinarity: history, the extended mind, and the civilizing
process. In R. Menary (Ed.), The Extended Mind (pp. 189-225). London, England: Bradford Book, MIT Press.
Thelen, Esther, Gregor Schöner, Christian Scheier, and Linda B. Smith. 2001. “The Dynamics of Embodiment: A Field Theory of Infant Perseverative Reaching.” Behavioral and Brain Sciences 24 (1) (February): 1–34. doi:10.1017/S0140525X01003910.
Thompson, E. 2007. Mind in Life Biology, Phenomenology and the Sciences of Mind (1st ed.). Harvard University Press.
Thompson, E., & Stapleton, M. 2009. Making Sense of Sense-Making: Reflections on Enactive and Extended Mind Theories. Topoi, 28 (1), 23 -30.
Tollefsen, Deborah Perron. 2006. “From Extended Mind to Collective Mind.” Cognitive Systems Research 7 (2-3) (June): 140–150. doi:10.1016/j.cogsys.2006.01.001.
Tomasello, M. & Carpenter, M. 2005. Intention reading and imitative learning. In Perspectives on imitation: From neuroscience to social science: Vol. 2. Imitation, human development, and culture, 133-148), edited by S. Hurley & N. Chater. Cambridge, MA: MIT Press.
Tomasello, M., M. Carpenter, J. Call, T. Behne, and H. Moll 2005. Understanding and sharing intentions: The origins of cultural cognition. Behavioral and Brain Sciences 28, 675–735.
Toni, I., de Lange, F.P., Noordzij, M.L., Hagoort, P. 2008. Language beyond action. Journal of Physiology, Paris, 102, 71-79.
Tribble, E. Sutton, J. 2011. “Cognitive Ecology as a Framework for Shakespearean Studies”, Shakespeare Studies 39, 94-103.
Tribble, E. B. 2005. Distributing Cognition in the Globe. Shakespeare Quarterly, 56(2), 135-155. Tronick, Edward, Heidelise Als, Lauren Adamson, Susan Wise, and T. Berry Brazelton. 1978. “The
Infant’s Response to Entrapment Between Contradictory Messages in Face-to-Face Interaction.” Journal of the American Academy of Child Psychiatry 17 (1): 1–13. doi:10.1016/S0002-7138(09)62273-1.
Tsai, J. C.-C., N. Sebanz, and G. Knoblich (2011, January). The GROOP effect: Groups mimic group actions. Cognition 118(1), 135–140.
Turkle, S. Constructions and Reconstructions of Self in Virtual Reality - Playing MUDs, retrieved from: http://web.mit.edu/sturkle/www/constructions.html on 30.09.2013.
van Elk, M., Slors, M. & Bekkering, H. 2010. Embodied language understanding requires an enactivist paradigm of cognition. Frontiers in Psychology 1, 234.
Varela, F. 1997. Patterns of life: Intertwining identity and cognition. Brain and Cognition 34(1), 72–87.
44
Varela, F. J. 1979. Principles of Biological Autonomy. Norwalk, CT: Appleton & Lange. Varela, F. J., Thompson, E., & Rosch, E. 1991. The Embodied Mind. Cambridge, Mass: MIT Press. Vesper, C., G. Knoblich, and N. Sebanz 2013. Our actions in my mind: Motor imagery of joint action.
Neuropsychologia forthcoming. Vesper, C., R. P. R. D. van der Wel, G. Knoblich, and N. Sebanz 2013. Are you ready to jump?
predictive mechanisms in interpersonal coordination. Journal of Experimental Psychology: Human Perception and Performance 39(1), 48–61.
Vesper, C., S. Butterfill, G. Knoblich, and N. Sebanz 2010. A minimal architecture for joint action. Neural Networks 23(8-9), 998–1003.
Walter, Sven. 2013. “Situated Cognition: A Field Guide to Some Open Conceptual and Ontological Issues.” Review of Philosophy and Psychology. doi:10.1007/s13164-013-0167-y.
Watzlawick, P., Bavelas, J. B., & Jackson, D. D. 1967. Pragmatics of human communication: A study of inter-actional patterns, pathologies and paradoxes. New York: Norton.
Weber, A., & Varela, F. J. 2002. Life after Kant: Natural purposes and the autopoietic foundations of biological individuality. Phenomenology and the Cognitive Sciences, 1(2), 97–125.
Wiemer-Hastings, Katja, and Lawrence W. Barsalou. 2005. “Situating Abstract Concepts.” In Grounding Cognition: The Role of Perception and Action in Memory, Language, and Thought, edited by Diane Pecher and R. Zwaan, 129–163. New York: Cambridge University Press.
Willems, R.M., & Hagoort, P. 2007. Neural evidence for the interplay between language, gesture and action: a review. Brain and Language, 101, 278-289.
Wood, Wendy, and David T eal. 2007. “A ew Look at Habits and the Habit-Goal Interface.” Psychological Review 114 (4): 843–63. doi:10.1037/0033-295X.114.4.843.
Woods, D. 2003. Discovering How Distributed Cognitive Systems Work. In Handbook of Cognitive Task Design, edited by Erik Hollnagel, 37–53. Mahwah, New Jersey: Lawrence Erlbaum Associates.
Zawidzki, Tadeusz. 2011. ‘How to interpret infant socio-cognitive competence’ Review of Philosophy and Psychology, 2(3), 483-497. Zawidzki, Tadeusz. 2013. Mindshaping: a New Framework for Understanding Human Social
Cognition. Cambridge MA: MIT Press. Zhang, J., & Norman, D.A. 1994. Representations in Distributed Cognitive Tasks. Cognitive Science
18, no. 1, 87–122.