Environmental enrichment is an important requirement for the welfare of farm animals, which are often kept in barren environments. This is also true for pigs. In 2001, the European Commission adopted a Directive (2001/93/EC) laying down minimum standards for the protection of pigs. In addition to requirements on noise levels, light conditions and water supply Directive 2001/93/EC states that: ‘‘Pigs must have permanent access to a sufficient quantity of material to enable proper investigation and manipulation activities, such as straw, hay, wood, sawdust, mushroom compost, peat or a mixture of such, which does not compromise the health of the animals.’’ A problem with the EC Directive is that its leaves too much room for interpretation. It is not clear, for example, whether metal chains, ropes, rubber toys or hard plastic balls are sufficient materials to enable proper investigation and manipulation activities. Although reviews have been published on the welfare of pigs generally (e.g. SVC, 1997) and on enrichment for captive animals generally (e.g. Young, 2003), no recent review on environmental enrichment for pigs is available, despite much scientific activity on the subject. In addition, a method is lacking to translate results of research into a specified directive. The aim of this paper, therefore, was to review the literature on environmental enrichment in pigs using a formalised procedure to start disclosing the scientific basis for the EC Directive as part of a larger project aimed at constructing a science-based model for assessing enrichment value in pigs. In order to help determine what is sufficient material for weaned and growing pigs, a literature review was conducted in a transparent and formalised way, systematically collecting relevant information in a database and translating this information into welfare-relevance. In total, 54 experiments reported in 47 references were selected for analysis. These references contained 200 statistically significant and welfare-relevant findings. A cross-table was constructed showing how classes of enrichment materials significantly affect classes of measured parameters. The classes of enrichment materials were metal objects, rubber, rope, wood, mineral blocks, roughage, substrates, straw and compound materials. The classes of welfare parameters were object-directed behaviour, pen-directed behaviour, tail and ear biting, aggression, (other) harmful social behaviour, activity (including play), fear (of humans), production and ‘health and hygiene’. With a number of important caveats described in the paper the cross-table allows the tentative conclusion that the available scientific evidence indicates that metal objects are not suitable enrichment materials for pigs, that rubber, rope, wood, roughage and substrates may be sufficient and that straw and compound materials are best. The methodology developed here for reviewing the available scientific evidence is recommended for other areas of application. It provided an important first step towards making transparent the scientific basis for legal requirements on enrichment materials for pigs and supporting political decision making in this area.