AbstractObjective:Patients with an equivalent clinical background may show unexpected interindividual differences in their outcome. The cognitive reserve (CR) model has been proposed to account for such discrepancies, but its role after acquired severe injuries is still being debated. We hypothesize that inappropriate investigative methods might have been used when dealing with severe patients, which have very likely reduced the possibility of observing meaningful influences in recovery from severe traumas.Methods:To overcome this issue, the potential neuroprotective role of CR was investigated, considering a wider spectrum of clinical symptoms ranging from low-level brain stem functions necessary for life to more complex motor and cognitive skills. In the present study, data from 50 severe patients, 20 suffering from post-anoxic encephalopathy (PAE) and 30 with traumatic brain injury (TBI), were collected and retrospectively analyzed.Results:We found that CR, diagnosis, time of hospitalization, and their interaction had an effect on the clinical indexes. When the predictive power of CR was investigated by means of two machine learning classifier algorithms, CR, together with age, emerged as the strongest factor in discriminating between patients who reached or did not reach successful recovery.Conclusions:Overall, the present study highlights a possible role of CR in shaping the recovery of severe patients suffering from either PAE or TBI. The practical implications underlying the need to routinely considered CR in the clinical practice are discussed.