Formation

It is debated whether we form a relationship using cognitive factors (Social Exchange Theory; SET,Thibaut and Kelly) or behavioural factors (Reward and Need for Satisfaction model;RN,Byrne and Clore).

Arguing for the role of behavioural factors, according to the RN we form a relationship with someone who is likely to fulfil our needs and provide us with satisfaction either directly, through operant conditioning (Skinner) whereby a person is rewarding in themselves (e.g. A good sense of humour); or indirectly, through classical conditioning (Pavlov) where we associate the person with pleasurable events that we do with them (e.g. Going to the cinema). We are unlikely to form a relationship with someone we find punishing.

Giving support to the behavioural argument, thereis limited evidence to support the role of RN in relationship formationfrom Griffiff and Guay who found that participants rated an experimenter more positively if he had praised them for their performance in the task. This shows the RN does play a role in liking (which may play a role in relationship formation) as we are most likely to feel positive towards those who provide us with most direct positive reinforcement.

Further weak evidence to support the role of behavioural factors in relationship formation comes from Argyle, who found that individuals who showed most reinforcing characteristics were most liked (e.g. Smiling). This again shows the role of RN in relationship formation,as we are more likely to be attracted to a potential partner who shows most direct positive reinforcement.

Adding further weight to the argument, being from the behavioural approach a strength of the RN is that it is parsimonious because it provides a justifiably simplistic explanation of the formation of relationships, which is supported by research by Griffiff and Guay. Therefore it is perhaps likely that we should accept Behavioural factors in the formation of relationships.

However, this is undermined because the research used to support the theory can be criticised for its low internal validity as it did not successfully measure relationship formation but instead more closely investigated liking or attraction, which are relevant to but not the same as relationship formation.  Therefore the RN can be further criticised for being unscientific because of it is impossible to actually test its predictions as relationships cannot be formed in experimental conditions.

Conversely, arguing for the role of cognitive factors in relationship formation, SET suggests that we will form a relationship if we anticipate that its future benefits are likely to outweigh its costs on two levels: on the Comparison Level we compare the anticipated future costs and benefits of the potential relationship within itself; and on the Comparison Level for Alternatives we make a similar cost/benefit analysis but this timeto alternative relationships (e.g. friends or family relationships). We form a relationship in the presence of a positive pay off matrix.

However, entirely undermining the role of cognitive factors in the formation of relationships, there is convincing against SET from Rusbult, who found that costs and benefits are ignored in the ‘honeymoon phase’. This shows that cognitive factors do not play a role in relationship formation as costs and benefits were only used in the later stages of a relationship.

As a result, the theory can be further criticised for being simplisticbecause it unfairly simplifies the formation of a relationship by ignoring important influences.For example Duck suggests that the theory ignores the influence of family pressures in the decision to form a relationship. Furthermore the evidence from Argyle to suggest the importance of non-verbal factors shows that cognitive factors cannot explain our choice to form a relationship.

Adding to this inadequacy, another limitation of SET is that it is unfalsifiable because it cannot be tested using an I.V or D.V. This is because you cannot test private cognitive thoughts about the costs and benefits of relationship formation. Indeed, this has been argued by Argyle, who has criticised the artificial methods used to measure costs and benefits in cognitive research.

Whilst both theories are compelling, due to the overwhelming evidence against cognitive factors, it is likely that we form a relationship through Behavioural factors and use Cognitive factors in the later stages of relationships. Indeed due to the excitement during the formation stage it is unlikely that we think rationally about the costs and benefits of relationships.