Evaluation of animal experiments
Animals will not show demand characteristics, as they do not come to research with a past history or a desire to please, therefore the changes observed as a result of the independent variable are likely to be more experimentally valid.
Animals mature more quickly and therefore the effect of interventions made early in life can be more quickly observed than in humans. This includes the effect of genetic manipulation on future generations.
The high degree of control that is possible with animal subjects makes it easier to draw cause and effect conclusions, we can be more sure that no other variable is responsible for the observed changes in the independent variable because we control all aspects of their lives.
It is possible (within the ethical rules) to do things to animals that it would not be possible to do to humans, such as lesion (deliberately damage) parts of their brains and observe the change in behaviour this causes, as long as the species chosen has similar physiology.
It lays psychology open to charges of speciesism, a form of prejudice that suggests that other species are not as valuable as humans, there are many campaign groups who argue that this is immoral.
Some would argue that humans are too different from other species for us to be able to generalise from animals to us e.g. humans are the only species to have evolved a complex language so our behaviour is always likely to be different from non-humans.
Animals cannot give feedback on how something has affected them, so the researcher has to make assumptions about what is motivating any changes in behaviour, this could be regarded as anthropomorphism where we ascribe human motivation to non-human behaviour.
It can be expensive and difficult to conduct research on non-humans, compliance with the code of ethics requires specialist housing and proper monitoring of their welfare.