Autosomal-recessive spastic ataxia of Charlevoix-Saguenay (ARSACS) is caused by loss-of-function mutation in the SACS gene, which encodes sacsin, a putative HSP70-HSP90 co-chaperone. Previous studies with Sacs knock-out (KO) mice and patient-derived fibroblasts suggested that SACSIN mutations inhibit the function of the mitochondrial fission enzyme dynamin-related protein 1 (Drp1). This in turn resulted in mitochondrial hyperfusion and dysfunction. We experimentally tested this hypothesis by genetically manipulating the mitochondrial fission/fusion equilibrium, creating double KO (DKO) mice that also lack positive (PP2A/Bβ2) and negative (PKA/AKAP1) regulators of Drp1. Neither promoting mitochondrial fusion (Bβ2KO) nor fission (Akap1 KO) influenced progression of motor symptoms in Sacs KO mice. However, our studies identified profound learning and memory deficits in aged Sacs KO mice. Moreover, this cognitive impairment was rescued in a gene dose-dependent manner by deletion of the Drp1 inhibitor PKA/Akap1. Our results are inconsistent with mitochondrial dysfunction as a primary pathogenic mechanism in ARSACS. Instead, they imply that promoting mitochondrial fission may be beneficial at later stages of the disease when pathology extends to brain regions subserving learning and memory.