Affective touch is crucial for human development, social bonding, and emotional support, while physical interaction plays a key role in rehabilitation, aiding motor recovery and well-being. Although AI has advanced verbal and non-verbal communication, replicating human-like physical interaction remains a challenge. This paper explores whether robots can convey emotional touch. Using a soft brush as a haptic transfer mediator, we isolated temporal and spatial touch characteristics to assess their impact on human perception. We implemented Cartesian dynamic movement primitives to replicate the human touch and adapt it to individuals by sampling geometric points on the forearm. Questionnaire analysis revealed that robotic touch at 3 cm/s was more effective in replicating the pleasurable sensations of human touch compared to 0.3 cm/s. Our findings indicate that the human-like quality of robotic touch depends more on temporal factors than spatial trajectories. However, robotic touch was still distinguishable from human touch, with 75.71% of participants correctly identifying it as robotic. These results contribute to robotic rehabilitation by improving the understanding of affective touch, with potential applications in motor recovery, sensory stimulation, and emotional well-being.