Semantic verbal fluency (SVF) tasks involve the activation of language, semantic memory, working memory and the executive functions. The second half of the test increases the demands on working memory and search for lexical items, which can make it more difficult than the first half.To conduct a pilot study in order to standardise the two halves of the SVF test ('animals' category) in adults over 50 years of age.Two verbal fluency tasks (semantic and phonological) and the Spanish version of the Mini-Mental State Examination with working memory were applied in 122 controls (range: 50-89 years), who were distributed according to age and schooling.There is a significant difference between the two halves of the test adjusted for age and schooling (p0.0001). Scoring in the second half showed a higher correlation with age, schooling and working memory than the first half. Performance in the two halves was standardised for three age groups (50-59, 60-79 and 80-89 years) and two levels of schooling (high and low).The number of names in the second half of the SVF test was lower than in the first half in all the groups, which supports the hypothesis of a greater cognitive demand, probably on working memory and the search for lexical items in semantic memory. Having normative values available for the two halves of the SVF test allows simultaneous and independent interpretation of the performance at two levels of intra-task difficulty. This model of analysis complements the traditional assessment and can easily be applied in day-to-day clinical practice.