Discuss the merits and limitations of the control theory approach to motivation.

   One of the other major limitations of this theory is that it does not take in to account the effect that external factors have on drinking behaviour. If there are fitness consequences to the animal for attaining the goal of the system, in this case drinking, then this will undoubtedly have an effect, which Toates and Oately’s model does not take in to account. Animals conduct a cost-benefit analysis before carrying out a behaviour, if they decide that the costs of the behaviour out weigh the benefits, for example drinking brings a high risk of predation then the animal will not drink. An experiment by Marwine and collier (1979) demonstrates this well. Rats were required to press a bar a specific number of times before they could gain access to water. They found that when water was made to be particularly costly i.e. the rat had to press the lever 300 times to gain access to water, only one large drink was taken every day. The fluctuation in the internal state at this point must be immense, as the rat is only drinking a large amount once a day. The rat chooses to let the homeostatic level or ‘set point’ drift because of the external costs of receiving the water. If following the control theory model the rat should constantly press the lever so its internal state, in this case optimum level of water in the body, will not shift very far from its ‘set point’, this does not occur showing a limitation to this model. Nevertheless the control theory model of drinking does account for certain existing facts and makes testable predictions that have often proved to be accurate, it therefore has all the properties of a good model but may need some fine tuning.

  Historically a clear distinction was made between homeostatic aspects of behaviour such as drinking feeding and temperature regulation and non-homeostatic aspects such as aggression and sexual behaviour. However more recently it has been shown that the processes that control sexual behaviour and aggression are not essentially differently to that of feeding and drinking (McFarland). They all involve the similar control principles, therefore a similar model using the same terminology and concepts can be devised for aggression (McFarland). Archer (1979 cited in Toates 1980) developed a model of aggression using the control theory principles. Archer believes that fear and aggression are caused by the same types of stimuli and that danger is any large discrepancy between observed and expected stimuli. Essentially animals build models or hypothesis’ of the expected states of various features in their environment, and fear is produced when the actual state differs considerably from that of the expected state (Toates 1980). However this idea in itself causes a problem as the control theory maintains that all behaviour is controlled by the departure of an internal variable from a ‘set point’ (Huntingford 1984). It is clear that aggression is not controlled by the departure of an internal variable from a set point but rather an expectation of what state the surrounding environment should be in. This model is also based on general observations of a range of vertebrate and relies on assumptions of the physiology behind motivation of aggression.

  The control theory’s ability to allow models to be quantifiably tested is as mentioned earlier one of its greatest strengths, but this is very difficult to do in the area of aggression. The very nature of attack and escape behaviour makes it very hard to quantify. It has not yet been possible to build computer simulations of control theory models of aggression, and until actual numbers are found to measure aggression and fear it is difficult to state how accurate this model is above the descriptive level (Toates 1980).

   Archer’s model of aggression is without a doubt more specific and objective than others in this area, and has provided a much needed unifying structure from which many facts of aggression can be understood (Toates 1980). However it still does not explain how fear and aggression can be activated at the same time in several species, and fails to explain some of the important details of animal fighting- such as explaining the exact sequence of events that occur during a fight (Huntingford 1984). These are issues that may need looking at to improve and develop this model.

  The control theory assumes that when a discrepancy-activated behaviour is performed it automatically reduces the amount of that the particular behaviour, known as negative feedback; however behaviour can be self-enhancing (Huntingford 1984). There are examples of this in several areas of motivation, for example in certain species rate of food intake increases during the beginning of the meal, and attack rate increases in the early stages of a fight. Control theory models have developed to integrate this phenomenon in to their systems. They have done this by the addition of a positive feedback loop, which has only added to the strength of this theory and made it even more powerful. This is demonstrated by the control theory model of mating in rats by Toates and O’Rourke (1978 cited in Toates 1980). In rats, stimuli received by the males have a positive effect on arousal, meaning there is a positive feedback loop. Performance of intromission increases the level of arousal and consequently the probability of further intromissions (Huntingford 1984). The physiological basis for the variable that determines the level of arousal, therefore the direction of the behaviour sequence is unknown, and it is obvious that it does not correspond to any simple physiological factor whose level needs monitoring. However this model has produced a well-designed explanation of the main features of sexual behaviour in rats by employing the principles of both negative and positive feedback.  

   One major limitation in this area is that most research is conducted with laboratory rats, with admittedly much success, but we have to be careful not to take interspecific comparisons too far. It is wrong to view humans as sophisticated versions of rodents although there general similarities i.e. penile insertion, ejaculation, there is undoubtedly a large difference in their patterns of copulation (Toates 1980). Perhaps this model will have to be reviewed taking in to account the wide range of species differences.

   Finally the control theory has been applied to the social interaction of courtship in newts. Male newts sexual behaviour is presumed to be controlled by three variables in this control model, initial amount of sperm, apparent sexual state of female, and oxygen requirements. This model does not show the simplicity of some of the other models we have discussed but its complexity is only equal to that of the behaviour it aims to explain. Computer simulations of this model have provided a remarkable match with observations of the behaviour in real newts. For example the model predicts that the courtship sequence depends on the males oxygen balance, and in real newts increasing the oxygen debt decreases the length of the courtship sequence and decreasing the oxygen debt has the opposite effect (Huntingford 1984). This model does seem to accurately account very skilfully for the complex sequence of sexual behaviour in newts. It shows that when developed the original control theory approach is able provide insight not only in to the motivational bases of simple homeostatic behaviour but complex non-homeostatic behaviour that includes more than one animal. This is a great strength.

   Overall the control theory approach to motivational systems seems to provide accurate and powerful frameworks in a range of areas although they all have there individual criticisms. One of the general criticisms of this theory is that it views each model as an individual entity, and does not explain how the systems interact with one another. For example if a rat is both hungry and thirsty it will have a small drink, then eat, then drink again and so on in a cycle until both systems have returned to their optimum level. The two motivational systems of hunger and thirst must interact as the rat switches behaviours. This theory does not deal with the ability of an animal to smoothly and brilliantly switch between systems. Nevertheless the control theory has had a great influence in the area of animal motivation as discussed, and has generated much research a lot of which offers it substantial support.

WORD COUNT- 2030

References

• Huntingford, F. (1984) The Study of animal behaviour. London. Chapman & Hall.
• Manning, A. & Dawkins, M.S. (1998) An introduction to animal behaviour. 5th edition. Cambridge.
• Marwine, A., & Collier, G. (1979). The rat at the waterhole. Journal of comparative and Physiological Psychology, 93, 391-402.
• McFarland, D. (1999) Animal Behaviour. Psychobiology, Ethology and Evolution. Pitman
• Toates, F.M. (1980) Animal behaviour –A systems approach. Wiley & sons.