Conscious attentional demands of encoding and retrieval from long-term memory.

The importance of allocating conscious attention to encoding and retrieval from long-term memory was investigated. Encoding, retrieval, and reactivation hypotheses were tested. College students were shown schematic faces in a secondary task and numerical problems as a primary task. The primary task was imposed both when the faces were encoded on study trials and when they were retrieved on recognition test trials. Difficulty of the primary task (easy vs. hard) was varied on both study and test trials. The easy problems permitted conscious attention to be given to the secondary faces whereas the hard problems prevented such allocation. Instructions about encoding the faces (incidental vs. intentional) were given to different groups. The degree of conscious attention that was allocated to the faces on study trials was assessed by means of interference, intention, and retrospection criteria of attended processing. Recognition of the faces was better when subjects allocated conscious attention to encoding than when they did not, in consistent support of the encoding hypothesis. The data ruled out a reactivation explanation. Finally, at times, paying special attention at retrieval improved recognition performance, providing weak support for the retrieval hypothesis.