Animal models should be viewed only as generating hypotheses about human behaviour, because there are differences between animal and human behaviour as well as similarities. There are two ways that gene-based behavioural variants of animals can be used for this. The first way is in identifying and mapping genes with effects on human behaviour. This is considered for genes with effects on alcohol consumption, tolerance and withdrawal in mice. The second way is in developing hypotheses about the biological causes of human behaviour. This is considered for the role of the cyclic amp response element-binding protein (CREB) gene and protein in learning and memory of a mollusc, fruitflies and mice. CREB is a transcription factor. The success and relevance of these and other gene-based animal models of human behaviour depends on the occurrence of homologous genes in animals and humans and on the occurrence of similar biological mechanism for the animal and human behaviour.

Key Concepts:

• Animal and humans are both similar and different biologically and behaviourally.
  
• Gene-based animal models of human behaviour may be due to homologous genes acting on homologous behaviours or to homologous genes acting on analogous behaviours.
  
• Gene-based animal models of human behaviour are best considered as hypothesis generators.
  
• Animal models generate hypotheses as to which genes may be involved in a human behaviour and how to find those genes from either comparative gene maps or molecular similarity.
  
• Animal models generate hypotheses as to the developmental, molecular, cellular or physiological effects of a gene on human behaviour.