H567/02 Section A - Moray (1959): auditory attention
We can follow one voice while missing much of another conversation. Moray tested whether repetition, a person's own name and instructions could make unattended messages reach awareness, using three distinct listening experiments.
Selective listening and shadowing
Selective attention prioritises some information while other information receives little conscious processing. Cherry's earlier work used dichotic listening, presenting a different message to each ear. Participants shadowed one message by repeating it aloud as they heard it, giving an observable sign that they were following that stream.
Rejected message
The message presented to the ear the participant is instructed not to shadow. “Rejected” describes the task instruction, not complete absence of all processing.
Moray asked how strong this attentional barrier was. Repetition might produce learning without deliberate attention; personally meaningful information, such as one's own name, might break through; an instruction to remember neutral digits might change what was selected.
All three were controlled laboratory experiments. A modified stereophonic tape recorder supplied separate headphone outputs, approximately matched for perceived loudness. One male speaker recorded the messages. Participants received four practice shadowing passages. The general sample consisted of undergraduates and research workers of both sexes. The primary paper does not separately state Experiment 1's sample size; it reports 12 participants for Experiment 2 and two groups of 14 for Experiment 3.
The cognitive area concerns the internal selection of information inferred from repetitions, recognition and reports. The attention theme asks what reaches awareness under competing demands; it does not equate failing to report a message with being physically unable to hear sound.
Experiment 1: repeated unattended words
The first experiment asked whether repeated exposure produces recognition without attention. While a participant shadowed prose in one ear, a short simple-word list was repeated 35 times in the other. The list was faded in after shadowing began and faded out at the end, reducing the chance that a sudden start or finish would draw attention.
After reporting anything remembered from the rejected message, participants took a recognition test about 30 seconds later. It included seven words from the shadowed message, seven from the rejected list and seven new control words. Comparing material categories within participants makes this a repeated measures comparison of recognition. The dependent variable was the number of words selected as recognised.
| Word category | Mean selected as recognised, out of 7 |
|---|---|
| Shadowed message | 4.9 |
| Rejected list | 1.9 |
| New control words | 2.6 |
The new words reveal false recognition: selecting a word is not automatically evidence that it was learned. Recognition of the rejected list was not better than recognition of new material, despite 35 repetitions. Shadowed words were recognised more than new words, a statistically significant difference.
The result supports a strong barrier to remembering unattended verbal content. It does not show that every unattended stimulus is blocked, that no unconscious influence is possible, or that a reported mean of 1.9 represents accurate learning of 1.9 rejected words.
Experiment 2: personally meaningful cues
The second experiment tested whether an instruction preceded by a participant's name was more likely to be heard in the rejected message. Twelve students and research workers shadowed ten short fiction passages. Instructions to stop or change ears sometimes occurred with the participant's own name and sometimes without it; two passages warned of a forthcoming change-ear instruction. Conditions were compared within the same participants.
The non-name member of each instruction pair preceded the name version, to reduce the possibility that first hearing the name version would make the later neutral instruction expected. The speaker used a steady delivery, and checks showed no significant increase in intensity when names occurred. Responses were tape-recorded and later examined; participants also gave reports between passages or at the end.
| Rejected-stream instruction | Presentations | Reported heard |
|---|---|---|
| Preceded by own name | 39 | 20 |
| Without own name | 36 | 4 |
These are pooled instruction occasions, not 39 or 36 different participants. The name total exceeds the intended 36 because three participants switched ears, placing additional instructions in their rejected stream. The comparison of mean hearing frequencies was significant. Only four of the twenty heard name-instruction occasions resulted in switching ears: hearing and obeying are different outcomes.
A warning before the passage gave a higher mean hearing frequency than unannounced name instructions, but that difference was not significant. Thus it did not establish that instructions alone reliably overcame the barrier.
Continue this lesson.
You’re previewing NovaLearn without an account. Create a free account to read the rest.
Create free account