Due to the brain9s limited cognitive capacity, simultaneous execution of multiple tasks can lead to performance impairments, especially when the tasks occur closely in time. This limitation is known as dual-task interference. We aimed to investigate the time course of this phenomenon in the brain, utilizing a combination of EEG, multivariate pattern analysis (MVPA), and drift diffusion modeling (DDM). In a simulated driving environment, participants first performed a tone discrimination task, followed by a lane-change task with varying onset time differences (Stimulus Onset Asynchrony, SOA), either short or long. As expected, the dual-task interference resulted in an increase in lane-change reaction time. The DDM analysis indicated that this increase was attributable to changes in both the decision time and the post-decision time. Our MVPA findings revealed a decrease in decoding accuracy of the lane-change task from ~200 to ~800 ms after stimulus onset in short SOA compared to long SOA, suggesting a change in lane-change direction information in both decision and motor processing stages. Moreover, a distinct pattern of generalization emerged in temporal generalization of short SOA condition, coupled with a delayed latency from ~500 ms in conditional generalization. Searchlight analysis illustrated the progression of this information reduction, starting in occipital, parietal, and parieto-occipital leads responsible for visual response and decision making, and then transferring to the frontal leads for mapping decisions onto motor regions. Consistent with the hybrid dual-task theory, our results suggest that processing of the two tasks occurs in partial parallel until the decision bound is reached. After the decision is made, another competition arises between the two tasks in motor areas for execution. Overall, our findings shed light on the intricate mechanisms underlying dual-task interference and provide further insights into the simultaneous processing of multiple tasks in the brain.