Background: Although people with Williams syndrome (WS) are often characterized as friendly and sociable with relatively good general language abilities, there is emerging evidence of pragmatic difficulties and trouble comprehending aspects of non-literal language. Aims: The main aim was to investigate the comprehension of sarcasm, metaphor and simile in WS relative to typically developing controls. A secondary aim was to examine the association between non-literal language comprehension and a range of other cognitive abilities, both in WS and in the typically developing population. Methods & Procedures: Twenty-six participants with WS were compared with 26 typically developing chronological age-matched controls (TDCA) and 26 typically developing mental age-matched controls (TDMA). Participants listened to stories in which characters made non-literal comments. They were then asked what each character meant by their comment. In order to investigate the second aim of the study, cognitive abilities were also assessed using the Woodcock-Johnson (Revised) Tests of Cognitive Ability, including expressive vocabulary, verbal working memory, perceptual integration, inferential reasoning and overall cognitive ability. Outcomes & Results: Comprehension of non-literal language in WS was significantly below TDCA levels, but was not significantly different to TDMA levels. For typically developing controls, each of the cognitive measures was strongly correlated with each of the measures of non-literal language comprehension. The same relationships were not always found for participants with WS. In particular, sarcasm comprehension in WS was not significantly correlated with any of the assessed cognitive abilities, and expressive vocabulary was not significantly correlated with any measure of non-literal comprehension. Conclusions & Implications: Comprehension of simile in WS is below TDCA levels but seems on par with their mental age level. It appears that comprehension of sarcasm and metaphors is above the cognitive capabilities and mental age level of most individuals with WS. Further, the pattern of correlations between non-literal comprehension and cognitive abilities in WS relative to controls suggests that perhaps the linguistic and cognitive systems that underpin non-literal language comprehension in typically developing individuals interact and integrate in different ways in WS.