Background/aim: The functional Tactile Object Recognition Test (fTORT) is a measure of haptic object recognition capacity recently adapted for use with children with neurological impairment. The current study aimed to investigate preliminary evidence of construct validity and responsiveness of the fTORT and its association with a measure of upper limb activity. Methods: A cross-sectional study of 28 children with spastic hemiplegic cerebral palsy (CP) (mean age 10 years 8 months; SD two years four months; 16 male) and 39 typically developing (TD) children (mean age 11 years; SD two years nine months; 19 male) was utilised to investigate construct validity and association between measures. Sixteen children with CP (mean age 10 years 10 months; SD two years 8 months; 9 male) who were randomly allocated to either a treatment (n = 6) or control group (n = 10) were assessed at four time points to assess test responsiveness. Results: There was a very significant difference (P value <0.0001) indicating greater haptic object recognition ability for the TD group (n = 39; median: 40; range: 33–42) than the group with CP (n = 28; median: 32.5; range: 3–41). fTORT scores demonstrated a significant association with scores on the activity measure (Pearson's r: 0.68; P = 0.0001). There were no significant changes over time in fTORT scores (P = 0.22) and no significant difference between the treatment and control groups (P = 0.47). Conclusion: The fTORT demonstrated preliminary construct validity, and was positively associated with an upper limb activity measure but scores did not change significantly following somatosensory training. This preliminary paper supports further research and future psychometric knowledge about the tool.