Studies regarding attention demonstrate that subjects shadow, or repeat, a message from the attended channel or ear. This phenomenon has been studied with dichotic listening; messages are played in both ears with only one attended to or shadowed. Findings are inconsistent, some show that subjects hear very little from the unattended message, others show that they hear different amounts or types of material. Words that have personal importance or meaning were found to be noticed. It appears that ignoring or blocking is an active mechanism that contributes to this selective listening process. Triesman attempted to demonstrate that there are contextual cues involved in selective listening and that subjects will hear the unattended message without knowing it, in certain (contextual) circumstances.
Triesman hypothesized that probability or expectancy based on transition probabilities between words would override the dichotic localization cues. Words made highly probable instead of important would be attended to without realizing it and would be allowed through the selective attention filter, coming from the rejected ear. Results of the study confirmed that contextual cues were not sufficient to cause subjects to change permanently to the unattended to channel; subjects remained virtually completely unaware of the content of the rejected passage. However, one or two highly probable words were allowed through from the rejected channel, by the selective filter, when the transition probabilities on the attended to channel were suddenly contradicted.
Triesman's experiment did not show that contextual cues would cause subjects to change to a second channel to make sense of a passage; it did not explain why subjects sometimes shadowed the wrong ear without knowing it. Subjects believed that the words they heard (from the unattended to channel) came from the accepted channel. Therefore it was concluded that it seems unlikely that th...