In many brain regions, tuning to stimuli and behaviour gradually changes over the course of days and weeks in a phenomenon known as representational drift. The rate of drift varies across cells in a population, and it remains unclear what underlies this heterogeneity. In this study, we explored whether drift is influenced by a neuron's propensity to synchronise or desynchronise its activity with other neurons in the population. Using data from parietal cortex (Driscoll et al. 2017) and visual cortex (Marks & Goard 2021), we evaluated pairwise encodings of task information to derive measures for pairwise synergy and redundancy. By defining a baseline measure of zero synergy and redundancy through conditionally shuffled data, we identified a pair's synergistic and redundant interaction. We then assigned redundancy and synergy indices to neurons by averaging their pairwise redundancy with other cells in the population.We found that a neuron’s tuning stability is positively correlated with its redundancy and synergy indices. To demonstrate that this correlation is not trivial, we developed a model simulating data inspired by the recorded neural data and showed that the same drift rate can be achieved with and without synergistic and redundant coupling between signals. Thus, we claim that synergy and redundancy indices capture relevant features of neurons for their drift rate. We hypothesise that pairwise synergy and redundancy signify co-tuning and counter-tuning to task-related features that may vary from trial to trial. Furthermore, their correlation to drift indicates mechanistic interactions between synergistically and redundantly coupled neuron pairs.