Nature and Prevention of BullyingTMR NETWORK PROJECT


Theory of Mind in Bullying: A Methodological Reassessment

Report of the Working Group chaired by Andrea Smorti, University of Firenze

In collaboration with, S.Bisaccia & S.Pagnucci, University of Firenze, L. Morley-Williams, University of Surrey, Roehampton,
R. Ortega & J. Ortega, University of Sevilla, P.K. Smith & S Kaipianen, Goldsmiths College, University of London,
and with the contribution of Sonia Guidi, University of Firenze

December 1999


Contents

Abstract
Introduction
First Approach: Level of Social Understanding
Second Approach: Biased Social Cognitions
Summing Up
A Project of Research
References


Abstract

What kind of theory of mind is used by bullies and victims both in school and the workplace? This working party assessed methodologies and measures used in studies of bullying and theory of mind. Existing research was divided into two groups.

The first included studies which assessed: (i) the level of social understanding through perspective-taking tasks; (ii) the level of social understanding through the construct of empathy; (iii) bullies and victims' capacity of reconstructing the antecedents of their social experiences and in particular, of bullying.

The second included studies which made use of (i) self report inventories capable of assessing the role played in bullying of constructs such as self efficacy and moral disengagement (ii) hypothetical situations, described in stories, cartoons, videos. Some recent studies, especially those aimed to assess bullies' and victims' level of social understanding with narratives, have produced interesting results. It seems that bullies do reason on the basis of mental states more than victims. On the basis of this review a cross national project of research on social cognition in bullying is outlined.

^ to Top

Introduction

This working party assessed methodologies and measures used in studies of bullying and theory of mind. Methodologies and measures implemented in studies focusing on bullying were reassessed consulting Medline and Psyclit data banks and journals. Theory of mind OR cognition OR thought OR narrative AND Bully OR aggression OR social rejection were used as key words. All studies dealing with the problem of social cognition or understanding of mental states in social maladjusted children or adolescents (rejected, bullies, victims, aggressive children) were considered for the present analysis. Moreover all the instruments used in these studies gave significant results in comparing social cognition in aggressive and non aggressive children.

The studies taken into consideration were divided into two large groups according to their main approach. In the first group the main approach takes into consideration the "level" of social understanding, in the second group the approach focuses on attitudes or biased social cognitions.

^ to Top

First Approach: Level of Social Understanding

  1. Construct: cognitive perspective taking

    Methods: perspective-taking and "theory of mind" tasks (see for example: Happé, 1995; Happé & Frith 1996; Tager-Flusberg & Sullivan, 1995; Sutton, Smith and Swettenham, in 1999 a and 1999 b)

    Examples:
    Advantages: These methods have the advantage of deeply evaluating a child's level of social reasoning (at a metarapresentational level). Some studies have shown, for example that bullies, more than victims, were capable of understanding mental states and perspectives of others.

    Problems: Many of the theory of Mind tests have been devised for children below 4-5 years of age. In addition, these tasks measure a cognitive perspective-taking capacity (cold cognition). However bullies have been identified as having cold cognitions, therefore a more focused test on empathic capacities seems necessary.

  2. Construct: Empathy

    Methods: Stories or cartoons to interpret, self reports

    Examples:

    Advantages: These methods permit us to assess the effective link existing between cognitions and aggression. This is conceptually relevant because the potentially aggressive person, observing a victim's distress, feels a vicarious distress which in turn reduces aggression (Richardson, Hammock & al.,1994). In this sense empathy tests seem useful in studying bullying and social rejection. Moreover many of these tests have been standardised and have a good consistency and reliability.

    Problems: The relation between empathy and aggression is complex. Empathy is affected by anger and emotional arousal and by the other's provocation. Therefore it is difficult to predict what the affective response of the child will be in situations of provocation or in other contexts in which anger or arousal are heightened.

  3. Construct: Narrative thought

    Methods: Stories, micro-autobiographies to make up (See: Mckeough, Yates & Marini 1994; Baumeister, Stillwell & Wotman 1990; Smorti and Pagnucci, 1999).

    Examples:
    Advantages: Narratives have the advantages of focusing research on a respondent's specific experience of bullying. Specific systems of analysis permit the study of the respondent's mental state comprehension and mental organisation around bullying episodes. For example in some studies (see: Smorti & Pagnucci, 1999), the results showed how bullies had higher scores than victims both in the completeness of the story scheme and in level of perspective taking.

    Problems: The outcome of these tests is affected by linguistic factors, like written language competence. Therefore, a researcher needs to assess this capacity in order to avoid misleading influences. Moreover these are open ended and qualitative tasks, therefore their analysis may require time and particular attention to evaluator's self consistency. However plenty of software for text analysis can be used to overcome, at least partially, this problem.

^ to Top

Second Approach: Biased Social Cognitions

  1. Construct: Self efficacy, Moral disengagement

    Methods: Self report inventory (see: Bandura, 1997; Bandura, Barbaranelli, Caprara & Pastorelli, 1996; Erdley & Asher, 1996) .

    Examples:
    Advantages: These methods can be very useful to assess bullies' and victims' sense of self confidence, or bullies disengagement from the outcome of their own acts. They have psychometric soundness and are easy to administer.

    Problems: The main problem is that these tests tend to entail a generic attitude of a respondent and lack attention to situational differences. Moreover they may lead a respondent to give a "good" self presentation.

  2. Construct: causal or intent attribution

    Methods: finding out appropriate antecedents to a problem story (See Dodge & Frame, 1982; Dodge & Coi, 1987; Crick & Dodge, 1996; Ray-Cohen , 1997; Cassidy et al. (1996) Cutrona and Feshbach (1979); Smorti and Ciucci (in press).

    Examples:
    Advantages: Hypothetical situations (stories on social interaction of bullying or neutral type) involving a double source of information that are discrepant permit the assessment of the role of schemata and social cues. This encompasses whether respondents are driven by their schemata to interpret social interaction or are capable (and to what extent) of detecting new signals from the environment that can contradict those schemata and eventually transform them. Moreover these tasks permit an analysis of a respondent's social understanding in terms of the type of antecedents used to interpret others' behaviour. For exemple bullies, more than victims, were dispositional, that is they used a narrative strategy that took into account mental states of the protagonist of the story, while victims, more frequently than bullies, were situational, that is they used a strategy that made use of other people's actions towards the protagonist and situational information.

    Problems: Highly situation-specific responses are elicited; a respondent is encouraged to think reflexively. Therefore these tasks furnish information on more abstract cognition rather than "on-line" children's interpretations of these specific situations. Since the aim is to assess the effective way of interpreting bullying events or social interaction episodes, we don't know whether or how a respondent uses this abstract cognition in everyday interactions.

^ to Top

Summing Up

This methodological reassessment of studies on theory of mind in Bullying showed two main outcomes:

  1. the necessity of considering theory of mind capacities according to approaches: Level of social understanding and strategies of social understanding. The first approach entails a construct of maturity of thought, the second a construct of personal style or attitude in interpreting social events;

  2. many of the tasks used are stories that a respondent has to construct or to interpret. Narratives were revealed to be effective tools to assess, in aggressive children, in bullies and in victims, both their level of social understanding and the strategies they use to reconstruct antecedents.

^ to Top

A Project of Research

Comprehension of others' mental states in bullies and in victims: a cross national comparison

Starting from these considerations a cross national research has been planned.

Aim

To assess the differences existing among bullies victims and control children as to:

  1. the maturational level of social understanding.

  2. the strategies of social understanding.

    To this aim a set of narrative tasks was chosen. Three main aspects of these capacities were taken into consideration.

  3. Capacity of using a story scheme that gives a coherent account of the events (for this problem see also: Fonagy et al., 1996; Stein & Liwag, 1997; Shapiro & Hudson 1997)

  4. Capacity of cognitive and affective perspective taking that permits an understanding of characters' mental states and perspectives (for this problem see also: Locroft & Teglasi, 1997; Fox, 1991; McKeough, 1996).

  5. Capacity of finding out appropriate antecedents to a problem, whether these are stable or accidental, situational or dispositional, entailing an action scenario (causal-mechanical material) or psychological one (for these problems see also: Ray-Cohen, 1997; Dodge & Frame, 1982; Feldman, 1991; Feldman, Bruner, Renderer & Spitzer, 1991).

Sample

Bullies, victims and control children will be drawn from a wider sample aged 12-14 yrs. in three different countries: Italy, Spain and possibly England. Each cell (gender x role x country) will consist of at least 20-30 children.

Mediator variables

SES
In order to check social class and family composition, samples will be chosen from schools that are located in different areas of the city (Low-Medium-High).

Linguistic ability
In respect to the impact of linguistic ability entailed in narrative tasks, a teacher's assessment will be used. Teachers will rate children's language competence.

Measure of bullying

A peer nomination instrument (Ortega & Ortega, 1999)

Narrative tasks

  1. Autobiographical memory stories

    Three type of autobiographical accounts will be proposed to pupils: a victim story, a bully story and a friendship story. In the victim story, pupils are encouraged to write about an incident in which someone has angered them. In the bully story, pupils are encouraged to write about an incident in which he/she has angered someone else. In the friendship story they are encouraged to describe an event about a time when he/she and some of his/her classmates were involved in a friendship episode.

    These instruments will be analysed in order to assess:
    1. The level of coherence present in the story.
    2. The level of social thinking.
    3. The level and the type of the story scheme used.

  2. Discrepant Stories Task

    The task used consists of 11 stories dealing with themes of social interaction between two peers, the protagonist (person A) and the partner (person B). There are two elements: the first part consists of the description of a typical piece of behaviour (script) on the part of the protagonist (A) towards his/her partner (B) at school. In the second part (violation) an episode was described in which the protagonist carried out a violation of the script.

    This instrument will be analysed in order to assess the nature of the antecedents and the consequences, in particular whether these are:
    1. stable or accidental;
    2. dispositional or situational;
    3. action or psychological verbs used;
    4. present past antecedent.

    The partner's reaction is also evaluated. On this basis eight different strategies of interpretation are calculated.

^ to Top

References


^ to Top


TMR NETWORK PROJECT