Objective   To investigate the effect of different recording methods for orbicularis oculi muscle activity on the acquisition of trace eyeblink conditioning in guinea pigs.     Methods   A 350 ms tone conditioned stimulus (CS) was paired with a 100 ms corneal oxygen-puff unconditioned stimulus (US). The guinea pigs were trained for 10 consecutive sessions of trace eyeblink conditioning. During the experiments

a spring-return potentiometer or a metal microelectrode was used to detect the activity of orbicularis oculi muscle. Homemade software written under the Matlab environment was used to quantitatively analyze the behavioral data.     Results   Both groups of guinea pigs acquired the trace conditioned eyeblink responses (CRs). However

the animals using the spring-return potentiometer acquired the trace CRs more quickly compared with the animals using the microelectrode（P0.05). In addition

the CR peak latency of the animals with potentiometer was significantly later than that of the animals with metal microelectrode on session 10（P0.05).     Conclusion   Relative to the orbicularis oculi electromyogram

the spring-return potentiometer is more suitable for studying the neurobiological mechanisms underlying associative learning in guinea pigs.