Psychology A2 - Media

HUESMAN ET AL - Correlational:

• Longitudinal study over 15 years.
• Found that viewing of TV violence in childhood correlated significantly with adult aggression in both men and women.
• Men who were classed as 'high violence viewers' in childhood had 3 times the crime conviction rate of 'low violence viewers', particularly those who strongly identified with the same sex role models offered by TV.

CHARLTON ET AL - No Link:

• Longitudinal study which looked at aggression before and after the introduction of TV to St Helena.
• Found that there was no overall increase in aggressive behaviour post television.

• However, they did find that those who scored highly on ASB scales chose to watch more violence on TV than those with low ASB scores (strong gender difference).

BANDURA - Causal:

• Found that children who had viewed video footage of an adult 'acting aggressively' towards a bobo doll, imitated the behaviour exactly as it had been shown in the video when they were given access to bobo dolls.

HAGELL & NEWBURN - No Link:

• Found that young offenders watch significantly less TV than their non-offending peers.

DRABMAN & THOMAS - Causal:

• Found that children who had just seen a violent film were less likely to intervene or call for help from an adult when two confederates started to have a fight, than those who had seen a non-violent film.

CONCLUSION: 

The research findings present a very mixed picture, with lots of contradictory findings. Some findings can be used to support a causal link between media and antisocial behaviour, whilst some support a correlation. Equally, however, there are findings to support the view that the media simply cannot be blamed for anti-social behaviour.

1. In highly controlled laboratory conditions the data is cross-sectional, meaning that it examines the immediate effects of the media on behaviour and there is no way of knowing if there will be any lasting long term effects. This criticism applies to Bandura and Drabman & Thomas.

2. Where the long term effects of exposure to media are examined, there is no way of controlling all the other variables which may affect how anti-social someone is. This means that there is no way of knowing if the relationship is causal or simply a correlation. This criticism applies to Huesman et al and Charlton et al. Examples of third variables that may be involved include biological factors and socio-economic background (such as parenting).

3. If we suspect that there is a link between anti-social media and anti-social behaviour, then you could argue that for ethical reasons, we should not expose people to anti-social media if they would not normally choose to view such media.

4. It is difficult to operationalise the DV (ASB) as for ethical reasons researchers cannot provoke participants to harm each other. For this reason, research sometimes measures ASB by asking how aggressive/hostile somebody feels, or seeing how they react to other people being anti-social. Just because somebody has aggressive feelings or is less upset by anti-social behaviour that they see, it does not neccessarily mean that they will behave aggressively themselves. This criticsm applies to Drabman & Thomas.

Through the process of modelling (which involves observation and then imitation), we learn our behaviour from the environment around us. If the media provides us with anti-social role models (people with whom we identify, or aspire to be like) then it is likely that we will be strongly influenced by their behaviour. This is especially true when through vicarious reinforcement, the anti-social behaviour that we observe in the media is somehow rewarded, for example because  it goes unpunished, brings about a successful outcome, or is glorified. We will be highly motivated to imitate behaviour we observe provided or expectation of success is greater than our expectation of punishment, and this will depend on the vicarious reinforcement to which we have been exposed. Exact imitation without thought.

Being repeatedly exposed to violence and other anti-social behaviour desensitises us. This means that we become increasingly less shocked and upset by what we see because we get used to it. Over time, we experience a decrease in arousal in response to anti-social behaviour. This means that we do not have an unpleasant physical response to what we see. In theory, what stops us from behaving anti-socially is the fact that it makes us feel uncomfortable as it elicits a physiological stress response. If we are so used to violence that we no longer experience then there is nothinng to stop us from behaving violently ourselves and/or being generally more accepting of violence in real life.

1. Supporting research suggests SLT may offer a valid explanation:

a). BANDURA found that... when children observed film footage of an adult role model attacking a bobo doll, they imitated the behaviour exactly. When reinforced with sweets, imitation increased. Which supports SLT because... it was the behaviour of the role model in the film that was imitated.

b). STACK found that... in the 4 years after the Austrian media were banned from reporting on subway suicides, actual subway suicides declined by 75%. Which supports SLT because... it suggests an element of imitation, of 'copycat' behaviour when observed in the media.

2. Are we already predisposed to ASB and this is why we are drawn to AS media?

We only imitate role models with whom we identify or aspire to be like. We also only imitate behaviour if we have self-efficacy. This is a specific form of self-confidence (a belief that we can successfully imitate what we see). Maybe people choose role models and types of media which reflect their own character and their own strengths, so maybe already anti-social people are drawn to aggressive role models and anti-social media. This would suggest that some other factors, not the media, made them anti-social in the first place. For example, Friedrich & Stein found that the only children influenced by aggressive cartoons were those who were already aggressive before viewing.

3. It is impossible to isolate anti-social media as the IV:

Maybe there are other third variables which explain why some people are more affected by what they see in the media than others. For example, perhaps media role models are more influential in homes where there is a lack of real life role models, and this, combined with the fact that these children are likely to watch more anti-social TV, makes the effects of anti-social media stronger. This would suggest that the relationship is more complex than SLT suggests.

4. Contradictory evidence:

For every research finding which appears to support SLT, there is contradictory research which challenges the idea that the media can be blamed for anti-social behaviour. For example, Hagell & Newburn found that young offenders watched significantly less TV than their non-offending peers, which would suggest that media is not to blame for the offending behaviour. Furthermore, copycat acts of violence are rare despite the millions of people who watch anti-social media.

1. Research evidence 

FANTI ET AL (2009) found that... repeated exposure to media violence reduces the psychological impact of media violence in the short term, therefore desensitizing viewers to media violence. As a result, viewers tended to feel less sympathetic toward the victims of violence and this increases their enjoyment of the violence portrayed in the media. Supported by Drabman & Thomas.

2. Just because you enjoy it more, and are less upset by it, does this make you anti-social?

Critics argue that just because we are desensitised to anti-social behaviour (through repeated exposure to it), this does not neccessarily mean that we will act in an anti-social way ourselves. It may be true that we experience less of a physical response to witnessing violence, for example, but that does not mean that we will be violent as a result. In fact, some research has suggested that anti-social media can actually be cathartic.

FESBACH proposed the 'catharsis hypothesis' suggesting that anti-social media provides a healthy outlet for emotions (such as frustration and anger) which has a relaxing and calming effect. If you accept this argument, you could argue that being exposed to some anti-social media may actually make us less likely to behave in an anti-social way. Unfortunately, whilst this seems like a logical theory, there is limited empirical evidence to back up Feshbach's theory.

3.Contradictory evidence:

For every research finding which appears to support desensitisation as a theory of why anti-social media is a bad thing, there is contradictory research which challenges the idea that the media can be blamed for anti-social behaviour. For example, Hagell & Newburn found that young offenders watched significantly less TV than their non offending peers, which would suggest that the media is not to blame for the offending behaviour.

FORGE & PHEMISTER - Correlational:

• Found that nursery aged children who were regularly exposed to prosocial media (particularly Sesame Street) behaved in a more altruistic way towards their peers than children who were not exposed to these programmes.

SPRAFKIN ET AL - Causal:

• Children who were exposed to an episode of Lassie in which Lassie helps to rescue puppies, were later more likely to sacrifice winning a game in order to respond to the sound of puppies barking (suggesting they may need help) than those who watched an episode of Lassie which did not feature such helping behaviour.

BARON ET AL - Causal:

• Allocated children aged 7-9 to watch one of three clips from The Waltons.
• The first group watched a clip showing co-operative behaviour. The second watched a clip with non-co-operative behaviour, and the third watched a neutral clip.
• Following this, a confederate walked past and dropped a pile of books. Children who had watched the co-operative clip were quicker and more likely to help than those in the other two groups.

MARES & WOODARD - Correlational:

• In a meta-analysis involving over 5000 children of the effects of pro-social behaviour found a positive correlation of .27, suggesting some relationship between the media and behaviour.
• However the strength of the effect was different for different children, with children whose parents discussed programme content with them being more influenced by what they saw. This suggests parental mediation has a large role to play.

1. The research which suggests there is a direct, causal relationship is largely laboratory based in which the effects of the media are measured soon after exposure to the pro-social media. This means that it is cross-sectional research and we cannot be certain how long the pro-social media content will continue to influence actual behaviour. This criticism applies to Sprafkin et al and Baron et al.

2. Research which is not cross-sectional can only ever identify a relationship between pro-social media and pro-social behaviour as there cannot be enough control over variables to be certain that the relationship is causal. Correlational research always runs the risk of ignoring possible third variables. In the case of the effects of pro-social media, parental mediation is an important variable. This criticism applies to Forge & Phemister.

Why might parental mediaton be an important variable in how effective pro-social media is?

1. Parents who chat to their children after the prosocial media reinforce the pro-social message.

2. Good parenting means children are only exposed to pro-social media (already pro-social children watch only pro-social media)

Through the process of modelling (which involves observation and then imitation), we learn our behaviour from the environment around us. If the media provides us with pro-social role models (people with whom we identify, or aspire to be like) then it is likely that we will be strongly influenced by their behaviour. This is especially true when through vicarious reinforcement, the pro-social behaviour that we observe in the media is somehow rewarded, for example it brings about a successful outcome, or the role models are rewarded for their actions. We will be motivated to imitate the pro-social behaviour if our expectation of reward for doing so is high. Eg. Watching shows such as Sesame Street encourage children to become more helpful among their peers.

The cognitive approach states that we make sense of the world around us using schemas. Schemas are chunks of stored information gained from our experiences which we use as a framework to assimilate new information; sometimes called scripts. For example, pro-social TV can provide positive experiences of people from different faiths, cultures and races, or people with disabilities, which means that children grow up with a positive view of people who are different to them. In the same way, pro-social TV can help provide children with a script for how to behave in certain situations and for dealing with emotions.

1. Supporting research suggests SLT may offer a valid explanation:

a). SPRAFKIN found that... children who watched Lassie helping others were more likely to help others themselves. Which supports SLT because... the IV determined the DV.

b). BARON found that... the children who watched The Waltons being co-operative helped the confederate who dropped the books. Which supports SLT because... the IV determined the DV

2. Are we already predisposed to PSB and this is why we are drawn to PS media?

We only imitate role models with whom we identify or aspire to be like. We will also only imitate if our expectation is that we too will be rewarded for pro-social behaviour. Maybe people choose role models and types of media which reflect their own character and their existing environment, so maybe already pro-social children are drawn to kind/positive role models and pro-social media. This would suggest that it is their existing environment and personality that draws them to pro-social media.

3. It is impossible to isolate pro-social media as the IV:

Maybe there are other third variables which explain why some people are more influenced by what they see in the media than others. For example, there is a link between parental mediation and the effects of pro-social TV. Perhaps pro-social role models are more influential in some families because they reinforce behaviour that the child is used to seeing at home and the messages are consistent with what their parents are saying. Furthermore, perhaps these children do not have as many anti-social role models as their parents have more control over what they are allowed to watch on TV. This suggests that the relationship is more complex than proposed by SLT. In fact, pro-social media may have very little effect if it does not mirror real life.

4. Weakness in the research on which the xplanation is based:

SLT suggests a direct, causal link between what children see in the media and how they behave. The research that supports is labarotory based, and tests behaviour shortly after viewing. This means that the research is cross-sectional, and we cannot be certain that any effects caused by the media actually have any long-term impact. This criticism applies, for example, to Sprafkin's famous 'Lassie' experiment or Baron's 'Waltons' experiment.

TRUGLIA ET AL - found that preschool aged children were aware of racial differences, and that episodes of Sesame Street aimed at promoting racial harmony were effective in helping children feel positively about having friends of a different race.

However... there is a limit to how much a schema can be changed by the media, as schemas are very hard to change.

We tend to use existing schemas to make sense of new information. This means that if our previous experiences contradict the pro-social message we get from the media, then we are likely to discard the pro-social message as 'unusual' and stick to our existing schema.  In other words, it is unlikely that pro-social media can override rigid schemas if they are already in place. Pro-social media will only therefore really have an effect if it is consistent with messages from previous experiences.

INCREASE IN PHYSIOLOGICAL AROUSAL: 

Gaming raises our arousal levels, heightening our emotional response to what is going on around us.

There is plenty of evidence that gaming does indeed lead to an increase in arousal levels. However, it seems more likely  that it is the content of the game, not the fact that it increases arousal levels, that it is likely to be linked to any negative effects of playing the game. Exciting games which are not violent lead to a rise in physiological arousal, but do not appear to have any negative effects on behaviour.

INCREASE IN AGGRESSIVE BEHAVIOUR, THOUGHTS AND FEELINGS:

Exposure to violent computer games may lead to an increase in aggressive behaviour by gamers in real life.

BUSHMAN found... that those who played the realistic video games averaged the highest level of noise. Those who identified themselves as an aggressive character made more noise than those who didn't.

REDUCTION IN PROSOCIAL BEHAVIOUR SUCH AS HELPING OTHERS:

People who play violent video games may be less co-operative with others.

SHEESE gave participants the choice to co-operate or exploit other gamers during an experiment. He found those just exposed to a violent scene were more likely to choose to exploit others compared to those exposed to a non-violent scene.

DESENSITISATION TO VIOLENCE:

We may get used to violence, and as a result we will be less shocked and upset by it. Repeated exposure to violent games means that gamers may become comfortable with violence.

CARNEGY found... heart rate increased after playing the games. During the movie, the heart rates for those who played non-violent  games increased significantly, whereas there was a small drop for the violent gamers.Violent gamers were less aroused than non-violent gamers by the violent images.

FINDINGS:

As expected the boys who had taken part in the realistic violent games averaged the highest level of noise. Within this group the boys who had identified most strongly with the aggressive character they were playing gave significantly higher noise levels than less identified individuals.

• Good control over variables (lab studies)
• Cross-sectional
• May be already aggressive (direction of causality)
• Third variables?

FINDINGS:

As expected, average heart rates increased after playing the game, whether it was a violent game like Carmaggedon or Duke Nukem, or a non-violent game like Glider Pro or Tetra madness. During the violent movie, the heart rates for those playing non-violent games increased significantly, while there was a small drop in those playing violent games.

Galvanic skin response is a measure of arousal. During the movie, there was a significant drop in galvanic skin response for the violent gamers, and a non-significant increase for the non-violent gamers. Violent gamers were significantly less aroused by the violent images than non-violent gamers.

• Good control over variables
• Large sample size

• Demand characteristics
• Cross-sectional as arousal was measured directly afterwards.
• Volunteers = may really enjoy playing violent video games (direction of causality)
• Third variables?

• GEE (2003): Encourage the players to reflect on their choices - encourages perserverance, independent thinking and commitment.
• DURKIN & BARBER: Gaming is an important part of healthy adolescence.
• OLSEN ET AL: Found that for young adolescent boys, gaming was frequent focus for conversation with their peers and was useful tool for structuring time with friends. Gaming increases sociability
• KESTENBAUM & WEINSTEIN: Found that gaming is a healthy outlet for frustration and aggression that helps the gamer discharge aggression. Gaming is cathartic. 

Effects of long term exposure not examined:

NEGATIVE:

• Some of the research is laboratory based. This means it tests behaviour shortly after playing games and therefore it can be criticised for being cross-sectional. There is no follow up to see if the negative effects of the game have a lasting effect on a gamer's personality.

POSITIVE:

• Some of the research which suggests gaming is a positive feature of adolescence (eg. Olsen 2007) does not follow participants into adulthood to see if these effects are long lasting. Most research is conducted on adolescents and so claims about the long term effects are speculative.

CAUSATION PROBLEM:

NEGATIVE:

• There is an issue with the direction of causality. It is possible that violent games make people more aggressive (Bushman), although it is equally possible that already aggressive people are drawn to violent games. This makes much of the research flawed; especially as it would be unethical to make someone who doesn't want to play violent games play them for the purposes of research.

POSITIVE:

• There is an issue with the direction of causality. It is possible that computer games make people more persistant and independent in their thinking (Gee). However, it is equally possible that people who are already persistent in their thinking are drawn to gaming as their hobby. This makes much of the research flawed; especially as the kind of people who consent to take part in research are probably already keen gamers.

METHODOLOGICAL PROBLEMS:

NEGATIVE:

• Much of the research into the negative effects of gaming is laboratory based, meaning it lacks ecological validity. For example, in Bushman's study, aggression was measured by getting participants to select the volume of a noise blast. Arguably this is not a valid way to measure aggression which is a difficult variable to measure in lab conditions.

POSITIVE:

• Much of the research into the posative effects of gaming is purely correlational, meaning it can lack internal validity because of a lack of control over potential third variables. For example, Durkin & Barber reported that gaming was a positive feature of a healthy adolescence. However, they measured this by testing things like engagement with school and disobedience to parents. Arguably there is no way they could have controlled all the variables involved in whether someone is engaged with school and argumentative at home!

LIMITED RESEARCH INTO GAMING:

• There is a lack of scientific, empirical evidence in this area and many of the research findings are contradictory. There is also a strong bias towards studying the effects of games on adolescent boys which means there is a lack of population validity.

The Hovland Yale model states that in order to change people's attitudes, you must take them through 4 stages:

A - ATTENTION

C - COMPREHENSION

R- REACTANCE

A - ATTITUDE CHANGE

The model argues that how likely you are to succeed in taking people through these stages, will depend on the decisions you make regarding 4 key factors: SOURCE, MESSAGE, MEDIUM, TARGET AUDIENCE.

SOURCE: Who is doing the persuading?

• The more knowledgeable/expert a communicator is perceived to be, the more persuasive they are.
• Popular (celebrities) and attractive eople are more persuasive, probably because of the Halo Effect.
• People who appear to be arguing against their own interests are perceived to be more trustworthy and therefore more persuasive (eg. someone known for reckless drivig arguing that drink driving is dangerous is more persuasive than a health fanatic arguing the same point).

MESSAGE: The content of what is being said

• There is a decision to make about whether to present a one sided (do not even acknowledge that there is another way of looking at something) or a two-sided argument (acknowledge that there is a different point of view, but state why this view is wrong). A more intelligent audience with prior knowledge are better persuaded with a two-sided argument. A less intelligent audience with a lack of knowledge are better persuaded with a one sided argument.

MEDIUM: How it is put across

• Making messages slightly fearful can be very persuasive, although too much fear can disrupt our ability to process the message properly (we might block it out).
• Presenting messages in a well organised/structured way makes them more persuasive.
• Presenting real life examples/personal stories can be more persuasive than just facts and figures. Perhaps because they arouse empathy.

TARGET AUDIENCE: The person who is being persuaded

• The intelligence and prior knowledge of the audience will affect how you present your message.
• The self-esteem of your audience may affect how you present your message. People with low self-esteem are easier to persuade.

• This model explains that there are actually two routes to persuasion, and so it is an example of a dual processing model.
• This model explains that in some cases and for some people, the actual message is not that important.
• There is a route to persuasion which involves distracting people from the actual message by thinking about more peripheral or superficial factors.
• For example, many people have strong opinions about which political party they would vote for, and yet don't really know what their policies are.

CENTRAL ROUTE TO PERSUASION:

• Your audience must be motivated to engage with your message.
• Your audience must have the time to process/think deeply about your message
• Your argument must be strong and convincing
• Most likely to be convinced by the actual argument you are making

PERIPHERAL ROUTE TO PERSUASION:

• Your audience may be less willing to think deeply about your message
• Your audience do not have the time to process/think deeply about the message
• it doesn't matter whether your argument is strong and convincing
• Most likely to be convinced by peripheral/context cues and not the actual argument

Affective conditioning is a form of classical conditioning in which advertisers pair their product or brand with a positive emotional response. For example, it may be 'cute', make us laugh or feel good about ourselves. In theory, if faced with a decision in the future about which product to buy, we will unconsciously gravitate towards the one that makes us feel good, even if we can't really explain why we prefer that brand. This tactic does not involve us having to think about the product, and so works best when there is a low need cognition. This is often the case if a product is a luxury item we don't really need, or a brand which is not particularly good value for money.

• Advertising brands that are linked with factors that make us feel positive, are more successful as we feel safer and more confident if we feel good about something.
• There is evidence to support the effectiveness of affective conditioning from Dempsey & Mitchell who found that even though they knew the other pen was better, participants chose the one that made them feel good, supporting the idea that we choose the brand that is positive.

This is a psychological phenomenon whereby people feel a preference for people or things simply because they are familiar, also known as the exposure effect and the familiarity principle. This may well have an evolutionary basis to it since we may be hardwired to believe that familiar thigs are less likely to be dangerous. However, when it comes to TV advertising, MEE has no basic in logic. Research suggests that being frequently exposed to a product tricks us into thinking that it is more trustworthy and superior to other brands to which we have had little or no exposure.

• This is a psychological phenomenon whereby people feel a preference for people or things simply because they are familiar, also known as the exposure effect and the familiarity principle. This may well have an evolutionary basis to it since we may be hardwired to believe that familiar thigs are less likely to be dangerous. However, when it comes to TV advertising, MEE has no basic in logic. Research suggests that being frequently exposed to a product tricks us into thinking that it is more trustworthy and superior to other brands to which we have had little or no exposure.
• The effectiveness of the MEE is supported by research from Brasel & Gips who found... the MEE works even on fast-forward as all the audience needs to do is register the brand/logo enough for it to feel familiar

1. Not everyone is as easily 'tricked' by TV adverts: Some people have a higher 'need for cognition' than others. Those with a naturally high need for cognition are less easy to persuade and like to think deeply about the choices they make, including what they spend their money on! Research suggests that the need for cognition correlates with higher self-esteem, masculine sex-role orientation, and is inversely related to social anxiety.

2. Methodological problems with research into the effectiveness of TV advertising: It is very difficult to measure the effectiveness of TV advertising because of methodological problems:

• Lab experiments - People are placed in hypothesised money-spending scenarios, but in real life they would be spending real money e.g. Dempsey & Mitchell
• Correlational studies - Possible third variables e.g. in a time where unemployment was low and inflation was low.

• This explanation revolves around the Celebrity Attitude Scale (CAS), which measures the strength of a person's attraction to celebrities on a scale of 1 to 3. At Level 1 (Entertainment/Social), where most people remain, a low level interest in celebrities can be healthy because it can provide a topic of conversation, and because it can help us develop our own sense of personal identity.
• According to McCutcheon, people who move beyond Level 1 into Level 2 (Intense/Personal), do so because they are unhappy in their own lives and parasocial relationships are an attempt to cope with or escape from reality. According to the AA model, absorbing yourself in the lives of celebrities beyond the entertainment/social level is indicative of poor psychological adjustment and/or a weak sense of identity.
• This model also states that absorbing yourself in the life of a celebrity can become addictive, so an individual at Level 2 feels an increasingly stronger sense of involvement with their celebrity. Only those with the poorest mental health will move to Level 3 (Borderline/Pathological) where there is a risk of stalking behaviour. Uncontrollable behaviours and fantasies.

1. Giles found that we use parasocial relationships as a way to gossip in offices and bars, supporting the view that a Level 1 interest in celebrities can be positive.

2. Maltby found that the personality traits associated with being at Level 1 of the CAS included being sociable, lively, active and venturesome, further supporting the view that a low level interest in celebrities can be perfectly.

3. In the same study, Maltby found that the personality traits associated with being at Levels 2 and 3 of the CAS included being tense, moody, impulsive and anti-social, supporting the view that Levels 2 and 3 on the CAS are indicative of poor psychological adjustment.

4. McCutcheon found that there was a negative correlation between scores on the CAS and scores on cognitive ability tests, suggesting that celebrity worship is correlated with poor functioning. This supports the view that people may be attracted to celebrities when they lack the skills to deal with their own lives, and so absorb themselves into the life of a celebrity which may seem much easier. This might explain why a personal crisis is often the triger for someone to move further up the CAS.

However, these research findings can be criticised because...

• They are correlational, meaning... we don't know if poor mental health is the reason for the worship or because of it.
• They may raise ethical issues... people at levels 2 and 3 are more vulnerable and may become distressed at the psychological research into them.
• They rely heavily on self-report techniques which means... people will give socially-desirable responses. This means that the research lacks internal validity, as it is difficult to gain empirical evidence in this field.

A second social psychological theory which can explain the attraction of celebrities is provided by McCann, who sees the need for parasocial relationships as a problem which can be traced back to early childhood and the relationship between an infant and the main caregiver.

According to this theory, those with an insecure attachment type (which is set in infancy and stays with us throughout our adult life) are more likely to become strongly attached to celebrities than those with a secure attachment type.

For an insecure resistant adult, parasocial relationships are attractive because... real relationships might keep failing (clingy, controlling, jealous, possessive) whereas celebrities will not let them down.

For an insecure avoidant adult, parasocial relationships are attractive because... there is no demand for intimacy or closeness, which avoidant types find difficult.

According to this theory, stalking of celebrities is directly related to insecure attachment patterns. Those who are insecure resistant may especially try to contact their idol to seek approval. They may feel let down by their celebrity, meaning they want to harm them.

1. McCutcheon found that adults classified as insecure were more likely to sympathise with/condone celebrity stalking, suggesting that those with insecure attachments are more likely to think that stalking a celebrity is acceptable.

2. Roberts found a significant correlation between insecure attachment type and attempts to contact celebrities, again suggesting that at the more serious end, stalking behaviour may be linked to attachment type.

3. Tonin found that 21 stalkers (detained under the mental health act) were significantly likely to have insecure attachment types than control groups (who were detained for other reasons). This suggests that stalking behaviour may be a manifestation of insecure attachments.

4. McCutcheon found no support for the hypothesis that insecurely attached adults would be more likely to be attracted to celebrity culture. So whilst attachment type may not explain why someone becomes interested in a celebrity in the first, it seems that once involved in the life of a celebrity, it is those with an insecure attachment type who pose the greatest risk of extreme fan behaviour such as stalking.

Problems with this research include:

• It is only correlational, which means... we cannot determine the direction of causality or third variables such as losing your job/home/family etc.
• It relies heavily on self-report techniques which is a problem because... people will give socially-desirable answers, making it lack internal validity.
• It is socially sensitive because... parents are being blamed for their relationships with their child.

• "Exchange of information about people not present"
• Knowledge of who was doing what, who was a potential threat, who was reproducing with who, enhanced your own chances of survival.
• The 'need' to gossip is hard-wired, and celebrities offer a safe outlet for this because they aren't going to know what you said.

• Paying close attention to the most successful individuals within your group increases your own chances of survival and 'reproductive fitness'.
• It made more sense to keep a close check on what high profile people are doing/how they are looking, because imitation ('general copying') may bring about the same rewards for you.
• The desire to keep up to date with high status individuals (celebrities) is therefore hard-wired into us.

• Presents a more optimistic view of celebrity culture than social psychology because... it does not see an interest in celebrities as negative.

• Very difficult to falsify because... we can't test it and we can't go back in time.
• Reductionist for failing to take into account the evidence (from social psychology) which indicates that... actually some get obssessed and get psychological illness.
• If following celebrities is a means of learning (eg. general copying) so that we increase our own reproductive fitness, then it is difficult to explain why some popular celebrities are gay.