Despite the fact that attachment working models explain why attachment security in early childhood has predictive power for later adult emotional and relational functioning, little is known about such models. For this reason the current study aimed to explore the content, structure and processes of attachment working models, specifically their affective content, impact on memory processes and semantic associations. The study incorporated four cognitive paradigms: a Chinese characters affective judgments task, a lexical decision task, a free-recall and cued-recognition memory task and a word categorization task based on the Implicit Attitudes Task (Greenwald, McGhee and Schwartz, 1998). Participants also self-reported their attachment status using scales from the Inventory of Parental and Peer Attachment (Armsden and Greenberg, 1987, 1989), the Parental Bonding Instrument (Parker, Tupling and Brown, 1979), the Attachment Style Questionnaire (Feeney, Noller and Hanrahan,1994) and a newly developed instrument, the Parental Attachment Scale (Fouladi, Moller & McCarthy, 2005). The sample consisted of undergraduate students at a southwestern university. For three of the four tasks, the hypotheses for tasks responses and associations between task performance and self-reported attachment security were not met. For the fourth word categorization task, the results did support hypotheses about task response, and evidence was therefore provided for the idea that working models of attachment security incorporate semantic associations with both positive attachment and pleasant words. Implications of the study as well as study limitations and directions for future research are discussed.