Autism Spectrum Disorder (ASD) is a mental condition involving genetic and environmental components. The underlying circuit mechanisms are unclear, but behaviorally, aversion towards unfamiliarity, a hallmark of autism, might be involved. Understanding and treating autism would likely benefit from mechanistic studies in model organisms, but whether bona fide ASD-like deficits can be modeled and investigated in mice has remained unclear. I will discuss recent results from our lab providing evidence that in Shank3-/- ASD model mice, exposure to novel environments produces long-lasting failure to engage and repetitive behaviors upon re-exposure, two hallmarks of autism. In Shank3-/- mice, a prelimbic cortex (PreL)-dependent memory upon first exposure led to failure to recruit appropriate PreL->Tail of striatum activity required for engagement at re-exposure. Inclusion of familiar features at first exposure prevented subsequent failure to engage in Shank3-/- mice. Therefore, novel context experience has a key role to trigger ASD-like phenotypes in genetically predisposed mice, and behavioral therapies involving familiarity and enrichment might prevent the emergence of ASD phenotypes.