Abstract: The rapid advancement of Industry 4.0 has prompted manufacturing firms worldwide—especially in emerging economies like China—to embrace high levels of automation in pursuit of greater efficiency,consistency,and scalability. While automation streamlines production and reduces dependency on manual labor,it also imposes new cognitive and coordination demands on frontline employees. In particular,unanticipated events such as peer turnover can destabilize these tightly coupled human-machine systems,raising critical questions about how automation interacts with organizational shocks to affect individual productivity over time. However,existing literature has yet to fully investigate the dynamic and contingent nature of this interaction. To address this gap,this study draws on Information Processing Theory (IPT) to examine how peer turnover influences employee performance under varying levels of job automation,and how workers recover from such disruptions. Using a unique panel dataset from a leading heavy-equipment manufacturing firm in North China,we track monthly productivity data for 333 frontline employees across a 5-year period (2019—2023),yielding 7087 worker-month observations. This rich longitudinal design enables us to isolate and analyze productivity patterns before,during,and after peer turnover events. We adopt a mixed-effects discontinuous growth model to estimate short- and medium-term changes in productivity trajectories,and triangulate our findings through 18 semi-structured interviews with workers and supervisors to uncover the underlying mechanisms of disruption and adaptation. Our empirical analysis reveals a U-shaped trajectory in employee productivity surrounding peer turnover. Specifically,productivity drops significantly in the month of a coworkers departure and begins a gradual recovery thereafter. This pattern is consistent with IPTs prediction that unanticipated organizational change temporarily exceeds employees information processing capacity,triggering performance setbacks until adaptation processes are activated. Importantly,our moderation analysis shows that job automation intensifies the negative effects of turnover and slows the pace of recovery. Workers assigned to highly automated production lines—characterized by rigid process interdependence,real-time system control,and limited task discretion—experience both a steeper decline and a more protracted rebound in performance compared to their counterparts in less automated roles. These findings challenge the prevailing view of automation as a stabilizing buffer against labor-related disruptions,and instead suggest that automation can amplify disruption effects by constraining adaptive flexibility and increasing information load. The qualitative evidence supports these findings by identifying three key breakdowns in information processing during the post-turnover period. Firstly,task overload—remaining workers must absorb additional responsibilities without preparation,stretching their cognitive and physical capacity; Secondly,disrupted informal coordination—peer-to-peer knowledge exchange,tacit guidance,and situational troubleshooting diminish,hampering effective task calibration; Thirdly,low system transparency—automated systems often lack intuitive feedback channels,making it difficult for workers to interpret problems or adjust their behaviors in real-time. In highly automated contexts,these breakdowns persist longer and are harder to correct due to the inherent rigidity of the system and the limited affordance for improvisation. During the recovery phase,the three gradual adaptation mechanisms also take effect. Workers develop operational fluency,engage in trial-and-error learning,and collaboratively restructure task responsibilities within their teams. Yet these mechanisms unfold at a slower pace in automated environments,where employees have less room to experiment and reconfigure workflows. This study makes three theoretical contributions. Firstly，It extends Information Processing Theory by demonstrating how automation not only alters the structure of task execution but also reshapes the flow and burden of information processing during organizational transitions. Secondly,it introduces a dynamic and context-sensitive perspective to understanding the performance effects of peer turnover,highlighting the role of technological context as a key moderator. Thirdly,it contributes to the literature on socio-technical systems by detailing how automation conditions the interplay between human cognition,coordination mechanisms,and adaptive capacity. From a practical perspective,the study offers several implications for firms engaged in production automation transformation. Managers should recognize that automation does not uniformly mitigate disruptions,and may in fact increase organizational fragility in response to labor shocks. Therefore,firms should invest in modular task design,cross-functional training,and supportive automation systems that empower employees to respond flexibly to unexpected changes. Furthermore,organizations should prioritize building redundant communication channels and enhancing workforce-system integration to ensure resilience during critical personnel transitions. In summary,this study provides a comprehensive,data-driven exploration of how peer turnover interacts with automation to shape productivity dynamics on the factory floor. By integrating panel data econometrics with qualitative field insights,it advances theoretical and practical understanding of human adaptability in technology-intensive work settings. The findings hold broad relevance for researchers and practitioners interested in organizational resilience,workforce management,and the human implications of industrial automation in emerging markets.