This study examined the association between trait depression and information-processing biases. Thirtyparticipants were divided into high- and low-trait depressive groups based on the median of their depressive subscale scores according to the Basic Personality Inventory. Information-processing biaseswere measured using a deployment-of-attention task (DOAT) and a recognition memory task (RMT). For the DOAT, participants saw one emotional face paired with a neutral face of the same person, and then were forced to choose on which face the color patch had first occurred. The percentage of participants’ choices favoring the happy, angry, or sad faces represented the selective attentional bias score for each emotion, respectively. For the RMT, participants rated different types of emotional faces and subsequently discriminated old faces from new faces. The memory strength for each type of face was calculated from hit and false-positive rates, based on the signal detection theory. Compared with the low-trait depressive group, the high-trait depressive group showed a negative cognitive style. This was an enhanced recognition memory for sad faces and a weakened inhibition of attending to sad faces, suggesting that those with high depressive trait may be vulnerable to interpersonal withdrawal