articulo_rs.pdf

8/10/2019 Articulo_RS.pdf

http://slidepdf.com/reader/full/articulorspdf 1/14

UNIV. PSYCHOL. BOGOTÁ, COLOMBIA V. 11 NO. 3 PP. 729-741 JUL-SEP 2012 ISSN 1657-9267 729

Social Representations: A Review of Theoryand Research from the Structural Approach

Representaciones sociales: una revisión de la teoría ylas investigaciones desde el enfoque estructural

Recibido: mayo 11 de 2010 | Revisado: diciembre 7 de 2010 | Aceptado: julio 8 de 2011

JOAO WACHELKE *

Federal University of Uberlandia, Brazil

A B S T R A C T

The present paper is a review of the theoretical advances and empiricalfindings related to social representations according to the structural ap-proach, a research stream that aims at studying the influence of social fac-

tors in thinking processes through the identification and characterizationof relationship structures. The presentation of the approach begins withthe baseline definitions of social representations according to a structuralapproach, moving on to an overview on the nature of representation ele-ments, the relationships between representations and practices, cognitivescheme dimensions, central core theory, representation transformations andinteraction context effects. In addition to positioning ourselves concerningpolemic topics during the review, in the final section we evaluate brieflythe current state and future perspectives of structural research on socialrepresentations, mostly addressing the problem of defining consensus, thedifficulty of characterizing a collective construct from individual data, andthe secondary importance of content in structural laws. Key words authorsSocial Representations, Structural Approach, Central Core Theory, Basic CognitiveSchemes.

Key words plusSocial Psychology, Review, Interactions.

R E S U M E N

Este artículo es una revisión de los avances teóricos y los hallazgos empíricosrelacionados con las representaciones sociales de acuerdo con el enfoqueestructural, una corriente de investigación que tiene como objetivo estudiarla influencia de factores sociales en los procesos de pensamiento a través de laidentificación y caracterización de relaciones estructurales. La presentacióndel enfoque comienza con las definiciones básicas de las representacionessociales de acuerdo con el enfoque estructural, pasando a la naturaleza delos elementos de representación, las relaciones entre las representaciones

y prácticas, las dimensiones de esquemas cognitivos, la teoría del núcleocentral, las transformaciones de representación y efectos de los contextosde interacción. Además de comentar algunos temas controversiales en

la revisión, en la sección final se evalúa brevemente la situación actual yperspectivas de la investigación estructural de las representaciones socia-les, principalmente relacionadas con el problema de definir el consenso, ladificultad de caracterizar una construcción colectiva a partir de datos indi-viduales y la importancia secundaria del contenido en las leyes estructurales.Palabras clave autoresRepresentaciones sociales, aproximación estructural, teoría de núcleo central,esquemas cognitivos básicos.Palabras clave descriptoresPsicología social, revisión, interacciones.

SICI: 1657-9267(201209)11:3<729:SRRTRS>2.0.TX;2-Q

Para citar este artículo: Wachelke, J. (2012). SocialRepresentations: A Review of Theory and Researchfrom the Structural Approach. Universitas Psycho-

logica, 11(3), 729-741.

* Professor at the Federal University of Uberlandia,Brazil. Email: [email protected]



JOAO WACHELKE

730 UN IV E R SITA S P SY CH OLOGICA V. 11 NO. 3 JU LIO-SE P TIE M BR E 2012

A social representation is a sociopsychological con-struct that performs a symbolic role, representingsomething –an object– to someone –a person orgroup. While doing so, the representation actually

substitutes the object it represents, and thereforebecomes the object itself, for the person or groupthat refers to it (Moscovici, 1961, 1976). As such, asocial representation is a product that results froma process of representing, and always replaces theobject that a social actor links to it. The objectcan only be accessed through a representation; fora given social actor, that representation “is” theobject (Abric, 1994a).

After almost 50 years since the introduction ofsocial representations in social psychology, related

research has formed an impressive body of stud-ies published worldwide, including Latin America(e.g. Parales Quenza, 2006; Magnabosco-Martins,Camargo & Biasus, 2009; Souza Filho & Duran-degui, 2009). There have been various schools thathave proposed different theoretical approachesand methodological strategies to deal with thephenomenon.

The present paper focuses on one of thoseschools, i.e., the structural approach, and aimsat providing a state-of-the-art review of the cur-rent theoretical positions supported by its studies

about social representations. This review also hasa secondary aim of making the structural approachto social representations more available to an in-ternational English-reading audience –so far, thevast majority of studies and essays on it has beenpublished in French only, in spite of the reasonablyglobal diffusion of the approach.

The structural approach to social representationsis a school that originated in France in the 1970s and80s. Its perspective is marked by an experimentaloutlook, which deals with socially shared knowledgeas structures, i.e., systems formed by interconnected

units, the functioning of which is regulated by laws.In the case of social representation, the structureis formed by simple ideas, basic cognitive units ofmeaning that are called cognems (Codol, 1969) orelements. A social representation is a set of cognemsthat refer to a social object and form an integratedknowledge structure shared by a group.

The main fields of study related to the struc-tural investigation of social representations will bepresented and discussed in the text. We begin withthe baseline definitions of social representations ac-

cording to a structural approach, and move on togive an overview of the advances on the nature ofrepresentation elements, the relationships betweensocial representations and social practices, the basiccognitive schemes model, central core theory, socialrepresentation transformations, and interaction ef-fects. In addition to positioning ourselves concern-ing polemic topics during the review, in the finalsection we discuss briefly the state of structuralresearch on social representations.

A few things must be made clear: throughout

the text we express our positions concerning po-lemic topics, at times presenting views that are notnecessarily shared by other scholars. This is donemostly for reasons of disambiguation and evaluationof the field directions, and should not be taken asa consensus within the community. Moreover, thecurrent review is restricted to the conceptual andmethodological framework of the “French” struc-tural approach itself; hence, overlaps and relation-ships with other developments from other socialrepresentation approaches are not addressed. Onesuch task has already been conducted by Parales

Quenza (2005), who has identified a large degreeof compatibility between the structural approachand the mainstream social cognition perspective.We try to present the structural approach on socialrepresentation in its own terms. Finally, for spacepurposes, the text has been organized so as topoint out the theoretical contributions of the worksmentioned. Individual studies are not described indetail; the reader is thereby directed to the originalsources in order to obtain methodological details.

Social Representation

Structure: Basic Concepts

When a completely new object appears in societyor when a group faces a stake due to a pre-existingobject, the basic conditions for the genesis of asocial representation are fulfilled (Garnier, 1999).However, not all social configurations allow for



SOCIAL R EPRESENTATIONS : A R EVIEW OF THEORY AND R ESEARCH FROM THE STRUCTURAL APPROACH

UN IV E R SITA S P SY CH OLOGICA V. 11 NO. 3 J U LIO-SE P TIE M BR E 2012 731

the establishment of a social representation struc-ture. Moscovici’s work (1961, 1976) proposes thatthree minimum conditions have to be satisfied ina given context: a social object must be ambigu-

ously defined, people should feel the need to inferabout it, and different aspects of that object shouldbe salient for different groups. Moliner (1993) hasproposed complementary criteria: the object mustbe polymorph, referring to a general class; theremust be an intergroup context, opposing at leasttwo groups regarding the object; and the objectmust be linked to a stake for the group, threateningeither their identity or social cohesion. Flament andRouquette (2003) add that an object must have aconcept function for the group, explaining a set of

occurrences of subordinate phenomena; it must bea communication topic; and it must be associatedwith a level of social practices.

Central core theory is the most establishedtheoretical development on social representationstructure and functioning within the context of thestructural approach. According to it, a representa-tion is formed by two qualitatively different elementsystems: a central core and a peripheral system.In its classic formulations, the theory states thatthe central core includes a few key elements thatgenerate the global meaning of the representation

and organize the whole structure. The elementsfrom the core have strong historical and ideologicalroots and are consensual within a group. It is thecentral core that defines and distinguishes repre-sentations; one can say that two representations aredifferent when at least one element from their coresis not the same. The peripheral system is the flex-ible part of the structure. It is not necessarily sharedwithin the group; it integrates particular informa-tion to the structure, connecting it to environ-mental practices and modulations (Abric, 1994a,1994b). The peripheral elements function as action

scripts adapting the guidelines from the central coreto concrete situations and specific occurrences ofthe social object (Flament, 1994a; Katerelos, 1993).Due to its flexibility, one of its functions is to defendthe central core contents against contradictions; ifthere is a situation that challenges the meaning ofcentral elements, the peripheral system is activated

and tries to justify the contradiction in order toendure it. Those rationalization mechanisms thatfunction as bumpers for the core are called ‘strangeschemes’ (Flament, 1987, 1989).

The functions of social representations includeproviding knowledge about the object to the group,maintaining group identity, guiding action andpractices regarding the object, and justifying thosepractices (Abric, 1994a). According to Flament(1987), a representation with a single central core isto be considered an autonomous social representa-tion, whereas representations without an organizedcore find their meanings in other related repre-sentations, and are classified as non-autonomous.Milland (2001) challenges that view. According

to said author, there is no representation withouta core, but sometimes an object can be interpretedby two different social representations, constitut-ing different reading grids; that would be the casefor representations still being structured, withoutmeaningful associated practices.

Representations and Practices

Social representations are usually found associat-ed with practices employed by a group concerningthe referring social object. The concept of social

practices is prone to multiple interpretations;Flament and Rouquette (2003) distinguishedamong four of them: the performance of an act,as opposed to not doing something; the frequencyor intensity of execution of a given action; theexpertise regarding an action; and the differentways of executing an action. The authors alsoclarify that a practice is not to be understood asphysical behavior only; the discourse concerning asocial object is also included. The broad definitionprovided by Flament (2001) is a good guideline: apractice is defined as a behavioral system that is

socially legitimated. Thus, a few pertinent ques-tions impose themselves: what are the relation-ships between representations and practices like?Do representations determine practices?

Contrary to what intuitive thinking might sug-gest, the currently accepted theoretical positionis that practices mediate representations and the



JOAO WACHELKE


environment,1 and not the other way around (Fla-ment & Rouquette, 2003). This implies that bothrepresentations and practices are subordinate toenvironmental constraints; it is often difficult to

separate both, as representation and practices findthemselves in a correspondence relationship mostof the time, except when environmental eventsimpose changes in the latter, making them incom-patible with the former (Flament, 1994b, 2001).Nevertheless, social representations predict thecarrying out of social practices in at least two cases:when a social actor faces a situation involving a so-cial object and has significant autonomy to act freefrom strong constraints; and when an affectivelycharged situation activates issues that are shared

within a collectivity. In both cases, it is likely thata pertinent representation will guide practices andbehavior (Abric, 1994c).

Practices and environmental constraints per-form essential roles to bring about transformationsin representations. However, it is necessary to ad-dress two topics before presenting the theoreticalmodels on social representation dynamics: cogni-tive scheme dimensions and structural centrality.

Basic Cognitive Schemes

A key theoretical advance for the understanding ofrelationships between cognems consisted in the ba-sic cognitive schemes model, abbreviated as SCB.2 That model classifies the possible logical relation-ships between two units within a structure; it makesit possible, for example, to frame relationships be-tween a social object and a single representationelement. The two units are coded as A and B andcan be linked by up to 28 connectors, which can begrouped in five basic cognitive schemes accordingto the logical operation domain that they refer to:lexical (3 connectors), neighborhood (3), composi-

tion (3), praxis (12) and attribution (9) (Guimelli& Rouquette, 1992).

1 Environment is understood here as the set of constraints that areexternal to the direct representing connection between a groupand a pertinent social object.

2 From the original French expression, schèmes cognitifs de base.

Rateau (1995a) observed the empirical associa-tions of connector activations with multiple socialrepresentation objects and proposed that the modelbe reduced to three basic meta-schemes, which he

called social representation dimensions: description(the grouping of lexical, neighborhood and compo-sition, summing up to 9 connectors), praxis (12),and evaluation (renaming the attribution scheme).

The contribution of the SCB model consists inthe possibility of understanding the different logicalroles that elements perform within the structure.The model makes it possible to characterize theactivation of a representation or an element in agiven context, going beyond the distinction be-tween central and peripheral elements.

Based on Rateau’s results, Flament (1994b)has pointed out that social representation ele-ments could be conceived of as schemes withnormative, descriptive and functional roles in therepresentation, admitting the possibil ity of mixedroles involving more than one of those dimensionsas well. Likewise, Abric and Tafani (1995) laterdemonstrated that the elements from the centralcore have different functions: some of them pro-vide norms regarding the social object, whereasothers are related to practices, and a third groupperforms both functions.

Structure Centrality

Perhaps the most important assumption of centralcore theory is the existence of a qualitative differ-ence between the central and peripheral systems.Such distinction allows for the identification ofwhat is in fact shared within a representation anddefines its organization. The first experimentalevidence of the validity of the central core wasprovided by Moliner (1989), who verified that inthe absence of certain elements on a representation

specimen, research participants would consensuallydeny that it referred to a given social representa-tion, whereas in the absence of other elementsthey would preserve the representation readinggrid. The latter were peripheral elements, condi-tional and negotiable, while the former were partof the central core, essential elements that defined





the social representation object. That study wasthe first application of a double denial principlenamed “questioning” or “calling-into-question”,usually abbreviated as MEC.3 It asks participants

if, in the absence of a characteristic, a specimenrefers to a representation class. Questioning is cur-rently the most widely accepted and employed tech-nique to identify the central core, being employedwith procedural variations (e.g. Moliner, 2001a).More recently, Lo Monaco, Lheureux and Halimi-Falkowicz (2008) have developed an alternativetechnique to determine the centrality of elementsthrough the measurement of a related principle:independence from context changes.

According to Moliner (1994), central elementshave two distinguishing properties. First, they pos-sess symbolic value regarding the social object ofinterest; central elements are essential to keepingits identity. Techniques grounded on the MECprinciple rely on the diagnosis of that property todetermine the centrality of elements. A secondproperty is related to the associative power of cen-tral elements; these can be associated with moreelements on the structure, being broader, whereasperipheral elements are associated with fewer ones.Two other properties derive from the two alreadymentioned. High symbolic value means that the

element is also salient in discourse, and high as-sociative power implies that central elements arefound connected to a higher number of elements.However, salience and connectivity, while typical ofcentral core cognems, are not exclusive; peripheralelements can be activated by specific contexts andpresent those secondary properties as well (Fla-ment, 1989; Moliner, 1989, 1994).

Rateau’s (1995b, 1995c) research has led to im-portant advances in the understanding of centralcore structure. His research has shown that there isa hierarchy of elements in the core: some of them,

called prioritary, are truly unconditional and definethe object, whereas a second set of elements, namedadjunct, despite having high symbolic values, areconditional. Their goal is to evaluate or specify the

3 From the original French expression, mise en cause.

object. In MEC tasks, prioritary elements displaypatterns of absolute rejection, while adjunct onesusually generate more diverse responses and condi-tional rejection. Only prioritary elements are essen-

tial to maintaining the identity of the social object.A further productive theoretical perspective

concerning social representation structure hasalso been presented by Moliner (1995), who hasproposed a two-dimensional model: social repre-sentation elements would have a double nature,including two key structural coordinates. The firstdimension involves representation structure itself,in which elements can be either central or periph-eral; their status is determined by the assessment oftheir symbolic values through MEC tasks based on

unconditionality. The second dimension opposesthe roles of description and evaluation performedby elements in the structure; in other words, it is adimension opposed to low and high affective load-ings of elements. Both dimensions are posited asbeing theoretically independent, and their cross-ing allows a classification of four element statuses:definitions (descriptive central elements), norms(evaluative central elements), descriptions (descrip-tive peripheral elements) and prescriptions (norma-tive peripheral elements).

Nevertheless, the two-dimensional model has

been challenged, due to some theoretical limita-tions and new empirical findings. Rizkallah (2003)has indicated a theoretical shortcoming related tothe affective loading dimension: every evaluationpresupposes a description, which means that thedescription modality is present in both poles of thedimension. Another limitation is related to resultswhich show that the structural and descriptive-evaluative dimensions are not independent. In spiteof Flament’s (1994b) early position that central ele-ments were unconditional prescriptions regardinga social object, research had already indicated that

centrality was not always linked to unconditional-ity; at times the symbolic values of central elementsrelied mostly on their normative function withinthe structure, rather than on their unconditionalnature (Moliner, 1992). Results from Gigling andRateau (1999) in research conducted with an ar-tificial object have also shown that the attribution



JOAO WACHELKE


of value to an element might lead it to assume acentral role, pointing out the importance of thenormative function in defining centrality.

Nevertheless, the two-dimensional model was a

major step in pushing structural theory forward. Itsmain contribution, with additional significant im-pact more recently, was perhaps the integration ofaffect into central core theory, thus opening doorsto studies relating social representations to attitudes(e.g. Moliner & Tafani, 1997; Tafani & Souchet,2001). In terms of central core theory, the two-dimensional model was the first theoretical effortthat truly took evaluative variables into accountin characterizing social representation structure.

Advancing on that topic, recent studies from

Lheureux, Rateau and Guimelli (2008) have con-firmed that element centrality and normativityare not independent of each other. Their resultsindicate that social representation elements pos-sess a double component, associated with two roleswithin the structure: semantic and normative. Thesemantic component relates to the goal of definingthe object class, whereas the normative componentjudges object specimens. Results obtained by theseauthors show that such double-dimensional natureis not only found at the core, but throughout thewhole representation. The two dimensions are not

independent, since normativity seems to play a keyrole: peripheral elements that score higher thanothers on normativity indexes are perceived as be-ing less conditional, and furthermore, the symbolicvalues of some central elements are based on nor-mativity rather than unconditionality, as evidencedby conditional rejection rates.

The authors’ model makes it possible to explainhow different sets of the structure are activated ac-cording to contextual demands: each representa-tion consists in a categorization system, and when asocial actor encounters a specimen, the first action

it undertakes is to define what the object is, basedon its prioritary elements. Once the object class isidentified, adjunct elements come into play in orderto evaluate the specific occurrence. Based on theresulting evaluation, conditional peripheral sets areactivated to deal with it according to contextualneeds. Thus, the model from Lheureux, Rateau and

Guimelli (2008) makes it possible to explain themechanisms involving peripheral understructuresidentified by Katerelos (1993).

Finally, an alternative SCB-based perspective of

centrality is grounded on the balanced activationof attributive and practical schemes (Rouquette &Rateau, 1998). According to that model, an elementis central when the valences related to praxis andattribution schemes are both high. When both va-lences are low, it is the case of a ‘regular’ peripheralelement, and when one partial valence is dispro-portionally high over the other, it is the case of aperipheral element activated by a contextual effect.This perspective has the advantage of identifyingover-activated elements, but the inconvenience of

being unable to deal with Rateau’s (1995b) hierar-chical core model.

Social Representation Dynamics

A transformation in a representation involves acentral element becoming conditional and thusattaining peripheral status, or a peripheral elementbeing ‘promoted’ to the central core. Changes inthe salience or activation of peripheral elementsare considered minor changes but not structuraltransformations, since the peripheral system is

flexible by definition. So far, the only way to inducerepresentation changes passes through the carry-ing out of new practices, usually brought about byenvironmental events. Based on research results,Flament (1994b) formulated the general modelfor social-representation dynamics determined bypractices, introducing two key variables. The firstone is the compatibility of new practices with therepresentation. The second is the social actors’ per-ception of the reversibility of new practices.

When new practices are compatible with thecentral core, there is no challenge to the represen-

tational structure, and no transformation takesplace. When, on the other hand, the practices arerelated to a peripheral aspect of the representa-tion, the peripheral schemes involved increase inactivation (Guimelli, 1994). If the implementationof new practices is perceived as reversible, then saidactivation state is temporary, and no transformation





takes place. If, in contrast, the practices are seen aspermanent, then the peripheral schemes becomecentral, and a progressive transformation occurs(Flament, 1994b; Guimelli, 1989).

But when practices oppose both central andperipheral elements, then new sets of peripheral ele-ments called “strange schemes” are formed in orderto try to accommodate the contradiction (Flament,1989). If the situation is perceived as being revers-ible, then the contradiction is successfully neutral-ized and the representation remains unchanged;but if it is permanent, then strange schemes cannotmaintain the integrity of the structure and eventu-ally there is a transformation in the central core toadapt to the new social context (Flament, 1994b).

The understanding of social representation dy-namics usually comes from results obtained in fieldstudies. According to Flament (2001), it is not likelythat a representation be truly transformed in thelaboratory, because even if a person’s beliefs changedue to an experimental setting, true representa-tion-change implies opposing beliefs and practicesshared by one’s group, and that is socially undesir-able by definition. Nevertheless, social influenceparadigms have proven very useful for understand-ing the interaction and communication processesinvolving changes and resistance of the structural

status of social representation elements for situatedsamples, thus accounting for pertinent instances ofrepresentation transformation. Mugny, Souchet,Quiamzade and Codaccioni (2009) have providedan overview of the variables that have been studiedas factors of representation transformation: majorityand minority influence; the influence of epistemicauthorities and asymmetry in intergroup status;and the cases in which representations regardingthe influence-context situation modulate influenceprocesses.

Presenting an important innovation, Tafani

and Souchet (2001) has made use of counter-representational essays, i.e., tasks which force par-ticipants to provide opinions that contradict theshared representation. Later, Souchet and Tafani(2004) managed to reproduce Flament’s (1994b)complete dynamics model in a laboratory context,even including reversibility perception: changes last

longer when contradictory practices are perceivedas irreversible. In conclusion, even if true socialrepresentation dynamics cannot take place in ar-tificial contexts, there are promising possibilities in

the laboratory of contributing to the understandingof a variety of processes in representational dynam-ics, to say the least.

If social representations do change, one inevi-tably comes to the question of how they evolve. Itis essential to stress that representations do have ahistory, and that they adapt to the environment,even if it does take years or generations for themto change; the characterization of a representationstructure is always the description of a representa-tional state, an heir of preceding states (Rouquette

& Guimelli, 1994). A social representation can befound in three chronological phases: emergence, orits birth as the social object appears in communica-tion practices for the first time; stability, in whichthe representation becomes stable with a clear-cutcore; and transformation, when environmentalconstraints bring about the already-mentionedprocess of change (Moliner, 2001). Stability andtransformation alternate until the representationis no longer pertinent in its social context, a situa-tion which one can arguably refer to as the ‘death’of a representation.

Representations in Action:Interaction Context Effects

The actualization of representations in people’severyday lives is linked to the influence of contextvariables. There are two basic types of context: theglobal social context and the immediate situationalone. The global context comprises the intergroupstakes and historical heritage that activate thecentral core and are responsible for its formation.The situational context is related to the multiple

and particular interaction conditions in which thesame social object comes into play, modulating theaction of the peripheral system (Abric, 1994c; Abric& Guimelli, 1998).

Global context effects have already been ad-dressed through the mechanisms of representation-structuring processes and dynamics, but what



JOAO WACHELKE


about interaction context effects? How do indi-viduals employ social representation knowledge inparticular situations?

There are three topics that have guided research

on context effects so far. The first one is dedicatedto understanding the connection of people to socialobjects. Rouquette (1996) has formulated threetheoretically-independent dimensions that could ac-count for the personal implication regarding an ob-ject: personal identification, or the extent to whichan object is related to a specific individual and notto everyone in general; social valuation, or the stakevalue linked to the object; and perceived possibilityof taking action concerning the object. Flament andRouquette (2003) view those implication dimen-

sions as intermediate factors that could account forinterpersonal and situational differences within agroup in terms of behavior, opinions and attitudesrelated to a social object; they refer to different de-grees of involvement with the object. Implicationis a condition for the transformation of representa-tions and adhesion to related beliefs. Additionally,different implication levels usually mean differencesin the use of social thinking processes: high and lowlevels implied that people employ different modes ofreasoning when facing contradictions to a represen-tational core (Guimelli, 2002), and might be associ-

ated with the activation of different basic cognitiveschemes (Gruev-Vintila & Rouquette, 2007).More recently, Guimelli and Abric (2007) have

suggested that knowledge of the social object couldbe a fourth implication dimension. Nevertheless,such a dimension is questionable, as it overlaps withdefinitions of social practices: since the discourseconcerning an object is considered one kind ofpractice (Flament & Rouquette, 2003), differentlevels of knowledge could then be related to differ-ent levels of practices. As an example, research bySalesses (2005) evaluates the role of knowledge of

an object in a way that is at the very least close tothe understanding of social practices. A further in-novation of Guimelli and Abric (2007) has been adifferentiation between the collective and personalpossibility of action. That distinction has provenuseful in a study about social implication and col-lective risk (Ernst-Vintila, 2009).

Another topic that has deserved significant at-tention is the influence of normative pressure onthe expression of social representation contents,especially when research participants complete

questionnaires. The existence of normative pres-sures caused by characteristics of the experimenter,or involvement of the participants and their groupsin task instructions, can lead people to respond ina way that is socially desirable, so as to provide agood image of themselves to the experimenter or incomparison with their group members (Chokier &Moliner, 2006; Chokier & Rateau, 2009; Flament,Guimelli & Abric, 2006; Guimelli, 2009; Guimelli& Deschamps, 2000).

Social representations also perform an impor-

tant role concerning social identity, as they arethe final product of the action of identity pro-cesses involving the interaction of self, intergroupand collective representation with categorization,comparison and attribution processes (Deschamps& Moliner, 2008). Therefore, a third topic oninteraction-context effects derives from a basiccharacteristic of social identity, and this presents aproblem: since individuals belong to multiple socialgroups (Tajfel, 1973), how does the knowledge ofsocial representation come into play in a specificsituation? Are there contextual cues that activate

single representations, or is there an interactionbetween different social representations shared bya single individual (Breakwell, 1993)? Such prob-lems have not been the explicit object of structuralstudies, but results from Wachelke and Camargo(2008) point out that when group membership issalient, the expression of elements related to grouppractices is favored.

Final Remarks: General Challengesand Future Possibilities

Keeping in mind that the organization of this text,including the emphasis on and interpretation ofsome previous efforts, in an attempt to make senseof relationships between different studies inevitablyreveals some of our positions regarding the field,we would like to close this review by addressingsome general topics that pervade the study of social





representations from a structural perspective asa whole. Perhaps the first thing that can be con-cluded from an analysis of the literature related tothe structural approach is that the research phase

responsible for significant baseline discoveries andformulations, such as classic central-core theory,relationships between representations and prac-tices and formulation of a basic cognitive-schemesmodel has come to an end. The basic models andconceptions regarding social representation struc-tures have already been proposed, developed andestablished; as happens with any science, new dataevidence theoretical contradictions and shortcom-ings calling for smaller scale-model adjustments andrefinement; a new phase marked mostly by more

specialized research problems is underway. In anycase, this comment obviously refers to a case thatdoes not rely on radical change brought about bygroundbreaking research; and yet, once in a while,such cornerstone studies do come about and havevery strong implications for the whole field, provok-ing dramatic turns in research direction.

Another key point refers to differentiation ofthe social representation construct from othermore classic socio-psychological notions such asattitudes, stereotypes, prototypes, beliefs, and soon. The specificity of the concept of social repre-

sentation lies supposedly in the fact that it is a sym-bolic structure shared by a collective and activelynegotiated through inter- and within-group com-munication. Moreover, the representation regardsan object that refers to a class of events and has acertain relevance for the group. While this makessense and is theoretically sound, it does entail a fewdifficulties in terms of empirical operationalization,and consequently, of verification. How is it possibleto be sure that a social representation is actuallyshared by a group? A tendency towards consensusamong the participants of a group is usually taken

as a measurement, as for example in MEC tasks.But the other classic symbolic constructs we havejust mentioned are also numerically shared (Jahoda,1988), which makes the distinction somewhat un-clear. To complicate things even more, researchevidence indicates that an awareness of groupconsensus among group members is associated with

different properties of representational elements(Moliner, 2001b).

Outside the structural approach, Wagner (1994)had already criticized the numerical consensus

criterion for identifying social representations andsustained that it is functional consensus that isto be looked for; social representation must playa role for reflexive, self-conscious groups; groupmembers must acknowledge that a common socialrepresentation underlies group practices linkedto a given object. This implies a group signaturein social representation that the author calls ho-lomorphy. Said considerations definitely providean alternative view of consensus in comparison towhat has usually been done in structural research,

but so far they have not been incorporated intostructural-approach studies. They may help to pro-vide greater precision and sophistication to socialrepresentation models and would merit at least aneffort of integration so as to outline better what ismeant by consensus, an essential coordinate of thephenomenon of interest.

Another limitation that is more difficult to over-come refers to the fact that, even if it is theoreticallystated that social representations are constructs of acollective level of analysis, most empirical researchof a structural nature relies on data from individu-

als, and infers social effects through an aggregationof data. If on the one hand, this way of conductingresearch has the benefits of making it possible toachieve precision and compatibility with standardexperimental practice from social psychology; onthe other hand, all that can be assessed are the ef-fects of field and laboratory manipulations regard-ing the cognitive and affective processes of groupmembers. Thus, the social representation itself, asa collective construct, remains inaccessible. This isundoubtedly related to Flament’s (2001) commenton the impossibility of transforming social repre-

sentations experimentally. Nevertheless, althoughhe focused on social desirability, we would consideranother aspect of the problem: if a social representa-tion is by definition a historical construct, then onlygrand-magnitude events affecting the collectivitywhich maintains a social representation can bringabout transformation of the construct. Laboratory



JOAO WACHELKE


settings only make it possible to achieve changethat is restricted to the participants involved, acharacteristic that is unlikely to change unlessradically new paradigms are introduced. In spite of

this peculiarity, research on social representationdynamics is stimulating in suggesting the steps thatsocial representational change might take, startingfrom individual change and then proceeding to col-lective legitimation.

In terms of future trends in the structural re-search investigation of social representations, wewould like to stress a few promising possibilities, inboth an internal sense (theoretical developmentabout representational structure itself) and an ex-ternal one (theory related to the processes connect-

ing social representations with other social thinkingconstructs). Internally, perhaps the main debateconcerns the key topic of structural centrality ofrepresentations: there is a competition betweenconcurring theories that are at times incompatible;such as the case concerning Moliner’s (1995) two-dimensional model and Lheureux et al.’s (2008)double-component approach. A refinement in cen-trality models is of capital importance in redefiningthe conceptions of social representation structureas a whole. In addition, there is still much room fordeveloping research involving the basic cognitive

schemes model (Guimelli & Rouquette, 1992),which has all the characteristics of a conceptualframework into which the various advances thathave been achieved by the structural approach canbe integrated. Research has also advanced consid-erably in tackling the various ways through whichsocial representation content is expressed accordingto context modulation (Chokier & Rateau, 2009;Guimelli, 2009) and the relationships between so-cial representations and social implication.

Externally, the studies continue to extend the“frontiers” of knowledge regarding social represen-

tations. If it is understood that social representa-tions constitute a symbolic structure among othersfrom the viewpoint of a social-thinking architec-ture, such as ideologies and opinions (Rouquette,1996), then there is a need for characterizing thesimilarities, differences and links among those for-mations. In the same vein, another fruitful trend

involves studying the relationships between socialrepresentations and processes studied throughmainstream social cognition, such as stereotypes,causal attribution, in-group bias, commitment and

others (Rateau & Moliner, 2009).A final comment to be made relates to the

specificity of the structural approach among theacademic schools dedicated to the study of socialrepresentations. The structural approach aimsat identifying structural processes and proper-ties typical of social representations, regardless ofrepresentational content. If the explanation of aprocess is determined by or related to differencesin content linked to different representations, thenthat process bears little structural interest;4 the goal,rather, is to achieve formulations that enable a gen-eralization to object classes, rather than limited tothe understanding of single objects. According tothis view, content is considered a secondary qual-ity rather than the focus of analysis (Rouquette& Rateau, 1998). This contrasts with the othersocial representation schools, which tend to givespecial attention to the processes and configura-tions linked to specific objects.

A consequence of giving privilege to structuralprocesses and trying to put contents aside is thatin order to achieve the formulation of laws related

to social-representation functioning mechanismsand identification of effects linked to associatedvariables, one must conduct research on a varietyof objects and grasp common processes that can begeneralized to a common representing activity thatcommands the formation and operation of poten-tially all social representations, or of specific andidentifiable varieties of representations. This is theideal procedure that guides and evaluates the basicresearch in social-representation theory accordingto a structural perspective, and that has made itpossible to construct a solid, verifiable and evolving

body of knowledge, of which we have tried to pro-vide a comprehensive summary through this review.

4 Let it be clear, though, that if some regularity in contents orconfiguration in a content taxonomy accounts for identifiableand replicable patterns in representation processes, it is nothingother than a structural manifestation of some sort, and not anisolated case.





References

Abric, J. C. (1994a). Les représentations sociales: aspectsthéoriques. In J. C. Abric (Ed.), Pratiques sociales

et representations. Paris: PUF.Abric, J. C. (1994b). L’organisation interne des repre-

sentations sociales: système central et systèmepériphérique. In C. Guimelli (Ed.), Structures et

transformations des représentations sociales. Lau-sanne: Delachaux et Niestlé.

Abric, J. C. (1994c). Pratiques sociales, représentationssociales. In J. C. Abric (Ed.), Pratiques sociales et

représentations. Paris: PUF.Abric, J. C. & Guimelli, C. (1998). Représentations

sociales et effets de contexte. Connexions, 72(2),

23-38.Abric, J. C. & Tafani, E. (1995). Nature et fonction-nement du noyau central d’une représentationsociale: la représentation de l’entreprise. Cahiers

Internationaux de Psychologie Sociale, 28, 22-31l.Breakwell, G. M. (1993). Social representations and

social identity. Papers on Social Representations,

2, 1-20.Chokier, N. & Moliner, P. (2006). La “zone muette”

des représentations sociales, pression normativee/ou comparaison sociale? Bulletin de Psychologie,

59(3), 281-286.

Chokier, N. & Rateau, P. (2009). Représentations d’ob-jets “sensibles” et processus de comparaison soi/au-trui. In P. Rateau, P. Moliner (Eds.), Représentations

sociales et processus sociocognitifs. Rennes: PUR.Codol, J. P. (1969). Note terminologique sur l’emploi de

quelques expressions concernant les activités etprocessus cognitifs en psychologie sociale. Bulletin

de Psychologie, 23(280), 63-71.Deschamps, J. C. & Moliner, P. (2008). L’identité en

psychologie sociale. Paris: Armand Colin.Flament, C. (1987). Pratiques et représentations sociales.

In J.L. Beauvois ; R.V. Joule & J. M. Monteil (Eds.).Perspectives cognitives et conduites sociales. V. 1.Cousset: Del Val.

Flament, C. (1989). Structure et dynamique des repré-sentations sociales. In D. Jodelet (Ed.), Les repré-

sentations sociales. Paris: PUF.Flament, C. (1994a). Aspects périphériques des repre-

sentations sociales. In C. Guimelli (Ed.),Structures

et transformations des représentations sociales. Lau-sanne: Delachaux et Niestlé.

Flament, C. (1994b). Structure, dynamique et trans-formation des representations sociales. In J. C.

Abric (Ed.), Pratiques sociales et representations. Paris: PUF.

Flament, C. (2001). Pratiques sociales et dynamique desreprésentations. In P. Moliner (Ed.), La dynamique

des représentations sociales. Grenoble: PUG.Flament, C., Guimelli, C. & Abric, J. C. (2006). Effets

de masquage dans l’expression d’une représenta-tion sociale. Cahiers Internationaux de Psychologie

Sociale, 69, 15-31.Flament, C. & Rouquette, M. L. (2003). Anatomie des

idées ordinaires. Paris: Armand Colin.Ernst-Vintila, A. (2009). Le rôle de l’implication person-

nelle dans l’expression de la pensée sociale sur lesrisques. In M. L. Rouquette (Ed.), La pensée sociale:

perspectives fondamentales et recherches appliqués.Toulouse: Érès.

Garnier, C. (1999). La genèse des représentations so-ciales dans une perspective développementale. InM. L. Rouquette & C. Garnier (Eds.). La genèse

des représentations sociales. Montreal: Nouvelles.Gigling, M. & Rateau, P. (1999). Les six épreuves des

Gopas. Une étude expérimentale du processusd’ancrage. In M. L. Rouquette & C. Garnier (Eds.).

La genèse des représentations sociales. Montreal:Nouvelles.

Gruev-Vintila, A. & Rouquette, M. L. (2007). Socialthinking about collective risk: how do risk-relatedpractice and personal involvement impact its socialrepresentations? Journal of Risk Research, 10(4),555-581.

Guimelli, C. (1989). Pratiques nouvelles et transfor-mation sans rupture d’une représentation sociale.La représentation de la chasse et de la nature. In. J. L. Beauvois, R.V. Joule & J. M. Monteil (Eds.).Perspectives cognitives et conduites sociales. V. 2.

Cousset.: Del Val.Guimelli, C. (1994). La fonction d’ infirmière: pratiques

et représentations sociales. In J. C. Abric (Ed.),Pratiques sociales et représentations. Paris: PUF.

Guimelli, C. (2002). Étude expérimentale du rôle del’implication de soi dans les modalités de raisonne-ment intervenant dans le cadre des représentations





Rateau, P. (1995c). Hiérarchie du système central desreprésentations sociales et processus de rationali-sation de la mise en cause de ses éléments. Bulletin

de Psychologie, 49(422), 73-87.

Rateau, P. & Moliner, P. (Eds.) (2009). Représentations

sociales et processus sociocognitifs. Rennes: PUR.Rizkallah, E. (2003). Le modèle bi-dimmensionnel des

RS: une critique théorique. Journal International

sur les Représentations Sociales, 1(1), 1-10.Rouquette, M. L. (1996). La communication sociale.

Paris: Dunod.Rouquette, M. L. & Guimelli, C. (1994). Sul la cognition

sociale, l’histoire et le temps. In C. Guimelli (Ed.),Structures et transformations des représentations so-

ciales. Lausanne: Delachaux et Niestlé.Rouquette, M. L. & Rateau, P. (1998). Introduction à

l’étude des représentations sociales. Grenoble: PUG.Salesses, L. (2005). Rôle du niveau de connaissance dans

le processus de structuration d’une représentationsociale. Cahiers Internationaux de Psychologie So-

ciale, 66, 25-42.Souchet, L. & Tafani, E. (2004). Pratiques, engagement

et représentations sociales : contribution expéri-mentale à un modèle de la dynamique représen-

tationelle. Cahiers Internationaux de Psychologie

Sociale, 63, 81-92.Souza Filho, E.A. & Durandegui, A. B. (2009). The

contextual analysis in social representations of the

body among ethnic groups in Rio de Janeiro, Brazil.Universitas Psychologica, 8(3), 771-783.

Tafani, E. & Souchet, L. (2001). Changement d’attitudeet dynamique représentationelle. In P. Moliner(Ed.), La dynamique des représentations sociales.Grenoble: PUG, pp. 59-88.

Tajfel, H. (1973). La catégorisation sociale. In S. Mos-covici (Ed.), Introduction à la psychologie sociale, v.

1. Paris: Larousse.Wachelke, J. & Camargo, B. (2008). What you ask is

what you get: effects of question types and groupsalience in instructions on the activation of so-cial representation elements. In 10th Transfer of

Knowledge Conference, Program and Abstracts. Volterra: ESCON.

Wagner, W. (1994). Fields of research and socio-genesisof social representations: a discussion of criteriaand diagnostics. Social Science Information, 33(2),199-228.

articulo_rs.pdf

Documents