<p>Childhood maltreatment is an established risk factor for poor mental health, yet not all maltreatment-exposed children develop psychiatric symptoms. This highlights the need for individualised metrics to detect those at heightened risk. Previous efforts to develop such metrics have largely focused on maltreatment severity and aggregate demographic risk. Although neurocognitive alterations in domains like social cognition and reward processing have been linked to maltreatment, most research has examined cross-sectional group-level differences in isolated domains. This proof-of-concept study examines whether indexing combined neurocognitive vulnerability can help predict which maltreatment-exposed children are most at risk of future symptom increases. A sample (mean age = 12.8) of maltreatment-exposed children (MT = 85) and propensity-score matched (PSM) non-maltreated peers (NMT = 90) were assessed at baseline using five neurocognitive tasks spanning various domains of social cognition and reward processing. A combined neurocognitive vulnerability metric was created, and symptom changes were measured 1.5 years later for a subset of individuals (MT = 57, NMT = 47). Combined neurocognitive vulnerability predicted future mental health symptom increases in the maltreatment-exposed group, even after controlling for baseline symptoms and clinical status. Maltreatment-exposed children in the high neurocognitive vulnerability category had a 92.1% adjusted predicted probability of experiencing worsening symptoms over time. Furthermore, we found that children in the maltreatment-exposed group were more likely to fall into medium and high neurocognitive vulnerability categories than non-maltreated peers. These preliminary findings underscore the potential of using combined neurocognitive vulnerability metrics to identify maltreatment-exposed children at the increased risk for future mental health problems, supporting its role in prevention strategies.</p>