A research team conducts a retrospective case–control study to investigate the association between pesticide exposure and Parkinson disease. Cases are 200 individuals diagnosed with Parkinson disease within the past five years; controls are 200 age- and sex-matched individuals without Parkinson disease. Investigators use a structured telephone survey to assess participants’ lifetime exposure to pesticides. Analysis reveals that cases report a significantly higher frequency of pesticide exposure before age 50 compared to controls (OR = 2.5, 95% CI 1.6–3.8). However, investigators note that a greater proportion of cases than controls consulted detailed diaries or spoke with family members when recalling past job duties. Which of the following biases is most likely affecting the study results?