Background: Motor abnormalities have clinical relevance as a component of psychotic illness; they are not only a proxy of altered neurodevelopment, but also intimately related to psychotic risk. We aimed to assess motor development and its association with psychotic experiences in children with familial high risk (FHR) of schizophrenia or bipolar disorder compared with controls. Methods: The Danish High Risk and Resilience Study is a prospective longitudinal cohort study, for which participants were extracted from Danish registers. Children born in Denmark between Sept 1, 2004, and Aug 31, 2009, with no, one, or two parents born in Denmark with schizophrenia or bipolar disorder, could be included in the study. No ethnicity data were collected. Children with no biological parent diagnosed with schizophrenia spectrum disorder or bipolar disorder were matched to children with FHR of schizophrenia (one or two parents with schizophrenia spectrum disorder) on the basis of sex, age, and municipality. Children with FHR of bipolar disorder (one or two parents with bipolar disorder) were included as a non-matched group. We assessed motor function in children with FHR of schizophrenia, children with FHR of bipolar disorder, and children in the control group at approximately age 8 years (baseline; 2013–16) and age 12 years (follow-up; 2017–20) using the Movement Assessment Battery for Children—Second Edition (Movement ABC-2). Psychotic experiences were assessed using the psychosis section of the Kiddie Schedule for Affective Disorders and Schizophrenia for School-Age Children—Present and Lifetime Version. Raters were masked regarding familial risk status. Motor development from baseline to follow-up in the different groups was assessed using a linear mixed model. Logistic regression examined the relationship between definite motor problems (≤5th percentile on Movement ABC-2) and psychotic experiences. Findings: Between March 1, 2017, and June 30, 2020, we studied 437 children (234 [54%] boys, 203 [46%] girls; mean age 11·99 years [SD 0·26, range 11·08–12·86]). Children with FHR of schizophrenia showed stable motor developmental deficits in manual dexterity (difference in intercept –1·62 [95% CI –2·39 to –0·85], p<0·0001; difference in slope 0·17 [–0·48 to 0·81], p=0·61) and balance (difference in intercept –1·58 [–2·34 to –0·82], p<0·0001; difference in slope 0·32 [–0·34 to 0·99], p=0·34), and a developmental lag in aiming and catching (difference in slope –1·07 [–1·72 to –0·41], p=0·0015; difference in intercept –0·59 [–1·35 to 0·17], p=0·13) compared with controls. Children with FHR of bipolar disorder showed no motor developmental differences on a group basis. Compared with controls, children with FHR of schizophrenia were more likely to have definite motor problems (odds ratio [OR] 2·86 [95% CI 1·60 to 5·11], p=0·0004), as were children with FHR of bipolar disorder (OR 2·45 [1·28 to 4·70], p=0·0068). Children with definite motor problems across all groups were more likely (OR 1·90 [1·12 to 3·21, p=0·017] to have had psychotic experiences than children with no definite motor problems. Interpretation: Clinicians should be aware that motor impairment in childhood can reflect neurodevelopmental vulnerability to psychosis. Our findings contribute to the identification of early risk markers for severe mental illness, both for use by clinicians and for establishing a basis for future primary preventive intervention studies in the premorbid phase. Funding: The Independent Research Fund Denmark, the Mental Health Services of the Capital Region of Denmark, the Lundbeck Foundation Initiative for Integrative Psychiatric Research (iPSYCH), Aarhus University, the Beatrice Surovell Haskell Fund, the Tryg Foundation, and the Innovation Fund Denmark. Translation: For the Danish translation of the abstract see Supplementary Materials section.