BACKGROUND:Altered TNF levels are associated with cognitive impairment in depression, schizophrenia, bipolar disorder, and Alzheimer's disease (AD). Exercise improves cognition-like behaviours, reduces the expression of tumour necrosis factor alpha (TNF), and increases expression of the soluble TNF receptors soluble TNFR1 (sTNFR1) and sTNFR2. We suggest TNF and its receptors are involved in cognitive function and dysfunction, and investigate whether exercise mediates its effects on cognitive function via TNF and its receptors. METHODS:We utilised C57BL/6, TNF-/-, TNFR1-/-, and TNFR2-/- mice to compare exercise to non-exercise control groups to investigate whether exercise exerts its effects on various types of cognition-like behaviours via TNF and its receptors. RESULTS:Recognition memory improved with exercise in WT mice, was impaired in TNFR1-/- exercise mice, showed non-significant impairment with exercise in TNF-/- mice, and no changes in TNFR2-/- mice. In spatial learning there were exercise related improvements in WT mice, non-significant but meaningful impairments evident in TNFR1-/- exercise mice, modest improvement in TNF-/- exercise mice, and potentially meaningful non-significant improvements in TNFR2-/- exercise mice. Moreover, WT and TNFR2-/- mice displayed noteworthy non-significant improvements in spatial memory, whereas TNFR1-/- exercise mice demonstrated non-significant spatial memory impairment. Exercise did not alter cognitive flexibility in any strain. DISCUSSION:TNF receptor signalling via the TNFR1 and TNFR2 appears to mediate the effects of exercise on cognitive-like behaviours. The potential for exercise to regulate human TNF and TNF signalling and cognitive dysfunction needs investigation under inflammatory conditions including depression and neuropsychiatric disorders.